Comparison of Sample Size by Bootstrap and by Formulas Based on Normal Distribution Assumption.

PubMed

Wang, Zuozhen

2018-01-01

Bootstrapping technique is distribution-independent, which provides an indirect way to estimate the sample size for a clinical trial based on a relatively smaller sample. In this paper, sample size estimation to compare two parallel-design arms for continuous data by bootstrap procedure are presented for various test types (inequality, non-inferiority, superiority, and equivalence), respectively. Meanwhile, sample size calculation by mathematical formulas (normal distribution assumption) for the identical data are also carried out. Consequently, power difference between the two calculation methods is acceptably small for all the test types. It shows that the bootstrap procedure is a credible technique for sample size estimation. After that, we compared the powers determined using the two methods based on data that violate the normal distribution assumption. To accommodate the feature of the data, the nonparametric statistical method of Wilcoxon test was applied to compare the two groups in the data during the process of bootstrap power estimation. As a result, the power estimated by normal distribution-based formula is far larger than that by bootstrap for each specific sample size per group. Hence, for this type of data, it is preferable that the bootstrap method be applied for sample size calculation at the beginning, and that the same statistical method as used in the subsequent statistical analysis is employed for each bootstrap sample during the course of bootstrap sample size estimation, provided there is historical true data available that can be well representative of the population to which the proposed trial is planning to extrapolate.

Four hundred or more participants needed for stable contingency table estimates of clinical prediction rule performance.

PubMed

Kent, Peter; Boyle, Eleanor; Keating, Jennifer L; Albert, Hanne B; Hartvigsen, Jan

2017-02-01

To quantify variability in the results of statistical analyses based on contingency tables and discuss the implications for the choice of sample size for studies that derive clinical prediction rules. An analysis of three pre-existing sets of large cohort data (n = 4,062-8,674) was performed. In each data set, repeated random sampling of various sample sizes, from n = 100 up to n = 2,000, was performed 100 times at each sample size and the variability in estimates of sensitivity, specificity, positive and negative likelihood ratios, posttest probabilities, odds ratios, and risk/prevalence ratios for each sample size was calculated. There were very wide, and statistically significant, differences in estimates derived from contingency tables from the same data set when calculated in sample sizes below 400 people, and typically, this variability stabilized in samples of 400-600 people. Although estimates of prevalence also varied significantly in samples below 600 people, that relationship only explains a small component of the variability in these statistical parameters. To reduce sample-specific variability, contingency tables should consist of 400 participants or more when used to derive clinical prediction rules or test their performance. Copyright © 2016 Elsevier Inc. All rights reserved.

Robust functional statistics applied to Probability Density Function shape screening of sEMG data.

PubMed

Boudaoud, S; Rix, H; Al Harrach, M; Marin, F

2014-01-01

Recent studies pointed out possible shape modifications of the Probability Density Function (PDF) of surface electromyographical (sEMG) data according to several contexts like fatigue and muscle force increase. Following this idea, criteria have been proposed to monitor these shape modifications mainly using High Order Statistics (HOS) parameters like skewness and kurtosis. In experimental conditions, these parameters are confronted with small sample size in the estimation process. This small sample size induces errors in the estimated HOS parameters restraining real-time and precise sEMG PDF shape monitoring. Recently, a functional formalism, the Core Shape Model (CSM), has been used to analyse shape modifications of PDF curves. In this work, taking inspiration from CSM method, robust functional statistics are proposed to emulate both skewness and kurtosis behaviors. These functional statistics combine both kernel density estimation and PDF shape distances to evaluate shape modifications even in presence of small sample size. Then, the proposed statistics are tested, using Monte Carlo simulations, on both normal and Log-normal PDFs that mimic observed sEMG PDF shape behavior during muscle contraction. According to the obtained results, the functional statistics seem to be more robust than HOS parameters to small sample size effect and more accurate in sEMG PDF shape screening applications.

Reexamining Sample Size Requirements for Multivariate, Abundance-Based Community Research: When Resources are Limited, the Research Does Not Have to Be.

PubMed

Forcino, Frank L; Leighton, Lindsey R; Twerdy, Pamela; Cahill, James F

2015-01-01

Community ecologists commonly perform multivariate techniques (e.g., ordination, cluster analysis) to assess patterns and gradients of taxonomic variation. A critical requirement for a meaningful statistical analysis is accurate information on the taxa found within an ecological sample. However, oversampling (too many individuals counted per sample) also comes at a cost, particularly for ecological systems in which identification and quantification is substantially more resource consuming than the field expedition itself. In such systems, an increasingly larger sample size will eventually result in diminishing returns in improving any pattern or gradient revealed by the data, but will also lead to continually increasing costs. Here, we examine 396 datasets: 44 previously published and 352 created datasets. Using meta-analytic and simulation-based approaches, the research within the present paper seeks (1) to determine minimal sample sizes required to produce robust multivariate statistical results when conducting abundance-based, community ecology research. Furthermore, we seek (2) to determine the dataset parameters (i.e., evenness, number of taxa, number of samples) that require larger sample sizes, regardless of resource availability. We found that in the 44 previously published and the 220 created datasets with randomly chosen abundances, a conservative estimate of a sample size of 58 produced the same multivariate results as all larger sample sizes. However, this minimal number varies as a function of evenness, where increased evenness resulted in increased minimal sample sizes. Sample sizes as small as 58 individuals are sufficient for a broad range of multivariate abundance-based research. In cases when resource availability is the limiting factor for conducting a project (e.g., small university, time to conduct the research project), statistically viable results can still be obtained with less of an investment.

Methods for flexible sample-size design in clinical trials: Likelihood, weighted, dual test, and promising zone approaches.

PubMed

Shih, Weichung Joe; Li, Gang; Wang, Yining

2016-03-01

Sample size plays a crucial role in clinical trials. Flexible sample-size designs, as part of the more general category of adaptive designs that utilize interim data, have been a popular topic in recent years. In this paper, we give a comparative review of four related methods for such a design. The likelihood method uses the likelihood ratio test with an adjusted critical value. The weighted method adjusts the test statistic with given weights rather than the critical value. The dual test method requires both the likelihood ratio statistic and the weighted statistic to be greater than the unadjusted critical value. The promising zone approach uses the likelihood ratio statistic with the unadjusted value and other constraints. All four methods preserve the type-I error rate. In this paper we explore their properties and compare their relationships and merits. We show that the sample size rules for the dual test are in conflict with the rules of the promising zone approach. We delineate what is necessary to specify in the study protocol to ensure the validity of the statistical procedure and what can be kept implicit in the protocol so that more flexibility can be attained for confirmatory phase III trials in meeting regulatory requirements. We also prove that under mild conditions, the likelihood ratio test still preserves the type-I error rate when the actual sample size is larger than the re-calculated one. Copyright © 2015 Elsevier Inc. All rights reserved.

Speeding Up Non-Parametric Bootstrap Computations for Statistics Based on Sample Moments in Small/Moderate Sample Size Applications

PubMed Central

Chaibub Neto, Elias

2015-01-01

In this paper we propose a vectorized implementation of the non-parametric bootstrap for statistics based on sample moments. Basically, we adopt the multinomial sampling formulation of the non-parametric bootstrap, and compute bootstrap replications of sample moment statistics by simply weighting the observed data according to multinomial counts instead of evaluating the statistic on a resampled version of the observed data. Using this formulation we can generate a matrix of bootstrap weights and compute the entire vector of bootstrap replications with a few matrix multiplications. Vectorization is particularly important for matrix-oriented programming languages such as R, where matrix/vector calculations tend to be faster than scalar operations implemented in a loop. We illustrate the application of the vectorized implementation in real and simulated data sets, when bootstrapping Pearson’s sample correlation coefficient, and compared its performance against two state-of-the-art R implementations of the non-parametric bootstrap, as well as a straightforward one based on a for loop. Our investigations spanned varying sample sizes and number of bootstrap replications. The vectorized bootstrap compared favorably against the state-of-the-art implementations in all cases tested, and was remarkably/considerably faster for small/moderate sample sizes. The same results were observed in the comparison with the straightforward implementation, except for large sample sizes, where the vectorized bootstrap was slightly slower than the straightforward implementation due to increased time expenditures in the generation of weight matrices via multinomial sampling. PMID:26125965

The Statistical Power of Planned Comparisons.

ERIC Educational Resources Information Center

Benton, Roberta L.

Basic principles underlying statistical power are examined; and issues pertaining to effect size, sample size, error variance, and significance level are highlighted via the use of specific hypothetical examples. Analysis of variance (ANOVA) and related methods remain popular, although other procedures sometimes have more statistical power against…

Introduction to Sample Size Choice for Confidence Intervals Based on "t" Statistics

ERIC Educational Resources Information Center

Liu, Xiaofeng Steven; Loudermilk, Brandon; Simpson, Thomas

2014-01-01

Sample size can be chosen to achieve a specified width in a confidence interval. The probability of obtaining a narrow width given that the confidence interval includes the population parameter is defined as the power of the confidence interval, a concept unfamiliar to many practitioners. This article shows how to utilize the Statistical Analysis…

Robust Covariate-Adjusted Log-Rank Statistics and Corresponding Sample Size Formula for Recurrent Events Data

PubMed Central

Song, Rui; Kosorok, Michael R.; Cai, Jianwen

2009-01-01

Summary Recurrent events data are frequently encountered in clinical trials. This article develops robust covariate-adjusted log-rank statistics applied to recurrent events data with arbitrary numbers of events under independent censoring and the corresponding sample size formula. The proposed log-rank tests are robust with respect to different data-generating processes and are adjusted for predictive covariates. It reduces to the Kong and Slud (1997, Biometrika 84, 847–862) setting in the case of a single event. The sample size formula is derived based on the asymptotic normality of the covariate-adjusted log-rank statistics under certain local alternatives and a working model for baseline covariates in the recurrent event data context. When the effect size is small and the baseline covariates do not contain significant information about event times, it reduces to the same form as that of Schoenfeld (1983, Biometrics 39, 499–503) for cases of a single event or independent event times within a subject. We carry out simulations to study the control of type I error and the comparison of powers between several methods in finite samples. The proposed sample size formula is illustrated using data from an rhDNase study. PMID:18162107

A normative inference approach for optimal sample sizes in decisions from experience

PubMed Central

Ostwald, Dirk; Starke, Ludger; Hertwig, Ralph

2015-01-01

“Decisions from experience” (DFE) refers to a body of work that emerged in research on behavioral decision making over the last decade. One of the major experimental paradigms employed to study experience-based choice is the “sampling paradigm,” which serves as a model of decision making under limited knowledge about the statistical structure of the world. In this paradigm respondents are presented with two payoff distributions, which, in contrast to standard approaches in behavioral economics, are specified not in terms of explicit outcome-probability information, but by the opportunity to sample outcomes from each distribution without economic consequences. Participants are encouraged to explore the distributions until they feel confident enough to decide from which they would prefer to draw from in a final trial involving real monetary payoffs. One commonly employed measure to characterize the behavior of participants in the sampling paradigm is the sample size, that is, the number of outcome draws which participants choose to obtain from each distribution prior to terminating sampling. A natural question that arises in this context concerns the “optimal” sample size, which could be used as a normative benchmark to evaluate human sampling behavior in DFE. In this theoretical study, we relate the DFE sampling paradigm to the classical statistical decision theoretic literature and, under a probabilistic inference assumption, evaluate optimal sample sizes for DFE. In our treatment we go beyond analytically established results by showing how the classical statistical decision theoretic framework can be used to derive optimal sample sizes under arbitrary, but numerically evaluable, constraints. Finally, we critically evaluate the value of deriving optimal sample sizes under this framework as testable predictions for the experimental study of sampling behavior in DFE. PMID:26441720

Sampling stratospheric aerosols with impactors

NASA Technical Reports Server (NTRS)

Oberbeck, Verne R.

1989-01-01

Derivation of statistically significant size distributions from impactor samples of rarefield stratospheric aerosols imposes difficult sampling constraints on collector design. It is shown that it is necessary to design impactors of different size for each range of aerosol size collected so as to obtain acceptable levels of uncertainty with a reasonable amount of data reduction.

Regression modeling of particle size distributions in urban storm water: advancements through improved sample collection methods

USGS Publications Warehouse

Fienen, Michael N.; Selbig, William R.

2012-01-01

A new sample collection system was developed to improve the representation of sediment entrained in urban storm water by integrating water quality samples from the entire water column. The depth-integrated sampler arm (DISA) was able to mitigate sediment stratification bias in storm water, thereby improving the characterization of suspended-sediment concentration and particle size distribution at three independent study locations. Use of the DISA decreased variability, which improved statistical regression to predict particle size distribution using surrogate environmental parameters, such as precipitation depth and intensity. The performance of this statistical modeling technique was compared to results using traditional fixed-point sampling methods and was found to perform better. When environmental parameters can be used to predict particle size distributions, environmental managers have more options when characterizing concentrations, loads, and particle size distributions in urban runoff.

Sample Size in Clinical Cardioprotection Trials Using Myocardial Salvage Index, Infarct Size, or Biochemical Markers as Endpoint.

PubMed

Engblom, Henrik; Heiberg, Einar; Erlinge, David; Jensen, Svend Eggert; Nordrehaug, Jan Erik; Dubois-Randé, Jean-Luc; Halvorsen, Sigrun; Hoffmann, Pavel; Koul, Sasha; Carlsson, Marcus; Atar, Dan; Arheden, Håkan

2016-03-09

Cardiac magnetic resonance (CMR) can quantify myocardial infarct (MI) size and myocardium at risk (MaR), enabling assessment of myocardial salvage index (MSI). We assessed how MSI impacts the number of patients needed to reach statistical power in relation to MI size alone and levels of biochemical markers in clinical cardioprotection trials and how scan day affect sample size. Controls (n=90) from the recent CHILL-MI and MITOCARE trials were included. MI size, MaR, and MSI were assessed from CMR. High-sensitivity troponin T (hsTnT) and creatine kinase isoenzyme MB (CKMB) levels were assessed in CHILL-MI patients (n=50). Utilizing distribution of these variables, 100 000 clinical trials were simulated for calculation of sample size required to reach sufficient power. For a treatment effect of 25% decrease in outcome variables, 50 patients were required in each arm using MSI compared to 93, 98, 120, 141, and 143 for MI size alone, hsTnT (area under the curve [AUC] and peak), and CKMB (AUC and peak) in order to reach a power of 90%. If average CMR scan day between treatment and control arms differed by 1 day, sample size needs to be increased by 54% (77 vs 50) to avoid scan day bias masking a treatment effect of 25%. Sample size in cardioprotection trials can be reduced 46% to 65% without compromising statistical power when using MSI by CMR as an outcome variable instead of MI size alone or biochemical markers. It is essential to ensure lack of bias in scan day between treatment and control arms to avoid compromising statistical power. © 2016 The Authors. Published on behalf of the American Heart Association, Inc., by Wiley Blackwell.

Post-stratified estimation: with-in strata and total sample size recommendations

Treesearch

James A. Westfall; Paul L. Patterson; John W. Coulston

2011-01-01

Post-stratification is used to reduce the variance of estimates of the mean. Because the stratification is not fixed in advance, within-strata sample sizes can be quite small. The survey statistics literature provides some guidance on minimum within-strata sample sizes; however, the recommendations and justifications are inconsistent and apply broadly for many...

Statistical aspects of genetic association testing in small samples, based on selective DNA pooling data in the arctic fox.

PubMed

Szyda, Joanna; Liu, Zengting; Zatoń-Dobrowolska, Magdalena; Wierzbicki, Heliodor; Rzasa, Anna

2008-01-01

We analysed data from a selective DNA pooling experiment with 130 individuals of the arctic fox (Alopex lagopus), which originated from 2 different types regarding body size. The association between alleles of 6 selected unlinked molecular markers and body size was tested by using univariate and multinomial logistic regression models, applying odds ratio and test statistics from the power divergence family. Due to the small sample size and the resulting sparseness of the data table, in hypothesis testing we could not rely on the asymptotic distributions of the tests. Instead, we tried to account for data sparseness by (i) modifying confidence intervals of odds ratio; (ii) using a normal approximation of the asymptotic distribution of the power divergence tests with different approaches for calculating moments of the statistics; and (iii) assessing P values empirically, based on bootstrap samples. As a result, a significant association was observed for 3 markers. Furthermore, we used simulations to assess the validity of the normal approximation of the asymptotic distribution of the test statistics under the conditions of small and sparse samples.

The endothelial sample size analysis in corneal specular microscopy clinical examinations.

PubMed

Abib, Fernando C; Holzchuh, Ricardo; Schaefer, Artur; Schaefer, Tania; Godois, Ronialci

2012-05-01

To evaluate endothelial cell sample size and statistical error in corneal specular microscopy (CSM) examinations. One hundred twenty examinations were conducted with 4 types of corneal specular microscopes: 30 with each BioOptics, CSO, Konan, and Topcon corneal specular microscopes. All endothelial image data were analyzed by respective instrument software and also by the Cells Analyzer software with a method developed in our lab. A reliability degree (RD) of 95% and a relative error (RE) of 0.05 were used as cut-off values to analyze images of the counted endothelial cells called samples. The sample size mean was the number of cells evaluated on the images obtained with each device. Only examinations with RE < 0.05 were considered statistically correct and suitable for comparisons with future examinations. The Cells Analyzer software was used to calculate the RE and customized sample size for all examinations. Bio-Optics: sample size, 97 ± 22 cells; RE, 6.52 ± 0.86; only 10% of the examinations had sufficient endothelial cell quantity (RE < 0.05); customized sample size, 162 ± 34 cells. CSO: sample size, 110 ± 20 cells; RE, 5.98 ± 0.98; only 16.6% of the examinations had sufficient endothelial cell quantity (RE < 0.05); customized sample size, 157 ± 45 cells. Konan: sample size, 80 ± 27 cells; RE, 10.6 ± 3.67; none of the examinations had sufficient endothelial cell quantity (RE > 0.05); customized sample size, 336 ± 131 cells. Topcon: sample size, 87 ± 17 cells; RE, 10.1 ± 2.52; none of the examinations had sufficient endothelial cell quantity (RE > 0.05); customized sample size, 382 ± 159 cells. A very high number of CSM examinations had sample errors based on Cells Analyzer software. The endothelial sample size (examinations) needs to include more cells to be reliable and reproducible. The Cells Analyzer tutorial routine will be useful for CSM examination reliability and reproducibility.

Statistical considerations for agroforestry studies

Treesearch

James A. Baldwin

1993-01-01

Statistical topics that related to agroforestry studies are discussed. These included study objectives, populations of interest, sampling schemes, sample sizes, estimation vs. hypothesis testing, and P-values. In addition, a relatively new and very much improved histogram display is described.

The quantitative LOD score: test statistic and sample size for exclusion and linkage of quantitative traits in human sibships.

PubMed

Page, G P; Amos, C I; Boerwinkle, E

1998-04-01

We present a test statistic, the quantitative LOD (QLOD) score, for the testing of both linkage and exclusion of quantitative-trait loci in randomly selected human sibships. As with the traditional LOD score, the boundary values of 3, for linkage, and -2, for exclusion, can be used for the QLOD score. We investigated the sample sizes required for inferring exclusion and linkage, for various combinations of linked genetic variance, total heritability, recombination distance, and sibship size, using fixed-size sampling. The sample sizes required for both linkage and exclusion were not qualitatively different and depended on the percentage of variance being linked or excluded and on the total genetic variance. Information regarding linkage and exclusion in sibships larger than size 2 increased as approximately all possible pairs n(n-1)/2 up to sibships of size 6. Increasing the recombination (theta) distance between the marker and the trait loci reduced empirically the power for both linkage and exclusion, as a function of approximately (1-2theta)4.

Sample Size Calculations for Precise Interval Estimation of the Eta-Squared Effect Size

ERIC Educational Resources Information Center

Shieh, Gwowen

2015-01-01

Analysis of variance is one of the most frequently used statistical analyses in the behavioral, educational, and social sciences, and special attention has been paid to the selection and use of an appropriate effect size measure of association in analysis of variance. This article presents the sample size procedures for precise interval estimation…

Tests of Independence in Contingency Tables with Small Samples: A Comparison of Statistical Power.

ERIC Educational Resources Information Center

Parshall, Cynthia G.; Kromrey, Jeffrey D.

1996-01-01

Power and Type I error rates were estimated for contingency tables with small sample sizes for the following four types of tests: (1) Pearson's chi-square; (2) chi-square with Yates's continuity correction; (3) the likelihood ratio test; and (4) Fisher's Exact Test. Various marginal distributions, sample sizes, and effect sizes were examined. (SLD)

[An investigation of the statistical power of the effect size in randomized controlled trials for the treatment of patients with type 2 diabetes mellitus using Chinese medicine].

PubMed

Ma, Li-Xin; Liu, Jian-Ping

2012-01-01

To investigate whether the power of the effect size was based on adequate sample size in randomized controlled trials (RCTs) for the treatment of patients with type 2 diabetes mellitus (T2DM) using Chinese medicine. China Knowledge Resource Integrated Database (CNKI), VIP Database for Chinese Technical Periodicals (VIP), Chinese Biomedical Database (CBM), and Wangfang Data were systematically recruited using terms like "Xiaoke" or diabetes, Chinese herbal medicine, patent medicine, traditional Chinese medicine, randomized, controlled, blinded, and placebo-controlled. Limitation was set on the intervention course > or = 3 months in order to identify the information of outcome assessement and the sample size. Data collection forms were made according to the checking lists found in the CONSORT statement. Independent double data extractions were performed on all included trials. The statistical power of the effects size for each RCT study was assessed using sample size calculation equations. (1) A total of 207 RCTs were included, including 111 superiority trials and 96 non-inferiority trials. (2) Among the 111 superiority trials, fasting plasma glucose (FPG) and glycosylated hemoglobin HbA1c (HbA1c) outcome measure were reported in 9% and 12% of the RCTs respectively with the sample size > 150 in each trial. For the outcome of HbA1c, only 10% of the RCTs had more than 80% power. For FPG, 23% of the RCTs had more than 80% power. (3) In the 96 non-inferiority trials, the outcomes FPG and HbA1c were reported as 31% and 36% respectively. These RCTs had a samples size > 150. For HbA1c only 36% of the RCTs had more than 80% power. For FPG, only 27% of the studies had more than 80% power. The sample size for statistical analysis was distressingly low and most RCTs did not achieve 80% power. In order to obtain a sufficient statistic power, it is recommended that clinical trials should establish clear research objective and hypothesis first, and choose scientific and evidence-based study design and outcome measurements. At the same time, calculate required sample size to ensure a precise research conclusion.

A U-statistics based approach to sample size planning of two-arm trials with discrete outcome criterion aiming to establish either superiority or noninferiority.

PubMed

Wellek, Stefan

2017-02-28

In current practice, the most frequently applied approach to the handling of ties in the Mann-Whitney-Wilcoxon (MWW) test is based on the conditional distribution of the sum of mid-ranks, given the observed pattern of ties. Starting from this conditional version of the testing procedure, a sample size formula was derived and investigated by Zhao et al. (Stat Med 2008). In contrast, the approach we pursue here is a nonconditional one exploiting explicit representations for the variances of and the covariance between the two U-statistics estimators involved in the Mann-Whitney form of the test statistic. The accuracy of both ways of approximating the sample sizes required for attaining a prespecified level of power in the MWW test for superiority with arbitrarily tied data is comparatively evaluated by means of simulation. The key qualitative conclusions to be drawn from these numerical comparisons are as follows: With the sample sizes calculated by means of the respective formula, both versions of the test maintain the level and the prespecified power with about the same degree of accuracy. Despite the equivalence in terms of accuracy, the sample size estimates obtained by means of the new formula are in many cases markedly lower than that calculated for the conditional test. Perhaps, a still more important advantage of the nonconditional approach based on U-statistics is that it can be also adopted for noninferiority trials. Copyright © 2016 John Wiley & Sons, Ltd. Copyright © 2016 John Wiley & Sons, Ltd.

How conservative is Fisher's exact test? A quantitative evaluation of the two-sample comparative binomial trial.

PubMed

Crans, Gerald G; Shuster, Jonathan J

2008-08-15

The debate as to which statistical methodology is most appropriate for the analysis of the two-sample comparative binomial trial has persisted for decades. Practitioners who favor the conditional methods of Fisher, Fisher's exact test (FET), claim that only experimental outcomes containing the same amount of information should be considered when performing analyses. Hence, the total number of successes should be fixed at its observed level in hypothetical repetitions of the experiment. Using conditional methods in clinical settings can pose interpretation difficulties, since results are derived using conditional sample spaces rather than the set of all possible outcomes. Perhaps more importantly from a clinical trial design perspective, this test can be too conservative, resulting in greater resource requirements and more subjects exposed to an experimental treatment. The actual significance level attained by FET (the size of the test) has not been reported in the statistical literature. Berger (J. R. Statist. Soc. D (The Statistician) 2001; 50:79-85) proposed assessing the conservativeness of conditional methods using p-value confidence intervals. In this paper we develop a numerical algorithm that calculates the size of FET for sample sizes, n, up to 125 per group at the two-sided significance level, alpha = 0.05. Additionally, this numerical method is used to define new significance levels alpha(*) = alpha+epsilon, where epsilon is a small positive number, for each n, such that the size of the test is as close as possible to the pre-specified alpha (0.05 for the current work) without exceeding it. Lastly, a sample size and power calculation example are presented, which demonstrates the statistical advantages of implementing the adjustment to FET (using alpha(*) instead of alpha) in the two-sample comparative binomial trial. 2008 John Wiley & Sons, Ltd

Multiple category-lot quality assurance sampling: a new classification system with application to schistosomiasis control.

PubMed

Olives, Casey; Valadez, Joseph J; Brooker, Simon J; Pagano, Marcello

2012-01-01

Originally a binary classifier, Lot Quality Assurance Sampling (LQAS) has proven to be a useful tool for classification of the prevalence of Schistosoma mansoni into multiple categories (≤10%, >10 and <50%, ≥50%), and semi-curtailed sampling has been shown to effectively reduce the number of observations needed to reach a decision. To date the statistical underpinnings for Multiple Category-LQAS (MC-LQAS) have not received full treatment. We explore the analytical properties of MC-LQAS, and validate its use for the classification of S. mansoni prevalence in multiple settings in East Africa. We outline MC-LQAS design principles and formulae for operating characteristic curves. In addition, we derive the average sample number for MC-LQAS when utilizing semi-curtailed sampling and introduce curtailed sampling in this setting. We also assess the performance of MC-LQAS designs with maximum sample sizes of n=15 and n=25 via a weighted kappa-statistic using S. mansoni data collected in 388 schools from four studies in East Africa. Overall performance of MC-LQAS classification was high (kappa-statistic of 0.87). In three of the studies, the kappa-statistic for a design with n=15 was greater than 0.75. In the fourth study, where these designs performed poorly (kappa-statistic less than 0.50), the majority of observations fell in regions where potential error is known to be high. Employment of semi-curtailed and curtailed sampling further reduced the sample size by as many as 0.5 and 3.5 observations per school, respectively, without increasing classification error. This work provides the needed analytics to understand the properties of MC-LQAS for assessing the prevalance of S. mansoni and shows that in most settings a sample size of 15 children provides a reliable classification of schools.

Item Analysis Appropriate for Domain-Referenced Classroom Testing. (Project Technical Report Number 1).

ERIC Educational Resources Information Center

Nitko, Anthony J.; Hsu, Tse-chi

Item analysis procedures appropriate for domain-referenced classroom testing are described. A conceptual framework within which item statistics can be considered and promising statistics in light of this framework are presented. The sampling fluctuations of the more promising item statistics for sample sizes comparable to the typical classroom…

Sample size in psychological research over the past 30 years.

PubMed

Marszalek, Jacob M; Barber, Carolyn; Kohlhart, Julie; Holmes, Cooper B

2011-04-01

The American Psychological Association (APA) Task Force on Statistical Inference was formed in 1996 in response to a growing body of research demonstrating methodological issues that threatened the credibility of psychological research, and made recommendations to address them. One issue was the small, even dramatically inadequate, size of samples used in studies published by leading journals. The present study assessed the progress made since the Task Force's final report in 1999. Sample sizes reported in four leading APA journals in 1955, 1977, 1995, and 2006 were compared using nonparametric statistics, while data from the last two waves were fit to a hierarchical generalized linear growth model for more in-depth analysis. Overall, results indicate that the recommendations for increasing sample sizes have not been integrated in core psychological research, although results slightly vary by field. This and other implications are discussed in the context of current methodological critique and practice.

Designing Intervention Studies: Selected Populations, Range Restrictions, and Statistical Power

PubMed Central

Miciak, Jeremy; Taylor, W. Pat; Stuebing, Karla K.; Fletcher, Jack M.; Vaughn, Sharon

2016-01-01

An appropriate estimate of statistical power is critical for the design of intervention studies. Although the inclusion of a pretest covariate in the test of the primary outcome can increase statistical power, samples selected on the basis of pretest performance may demonstrate range restriction on the selection measure and other correlated measures. This can result in attenuated pretest-posttest correlations, reducing the variance explained by the pretest covariate. We investigated the implications of two potential range restriction scenarios: direct truncation on a selection measure and indirect range restriction on correlated measures. Empirical and simulated data indicated direct range restriction on the pretest covariate greatly reduced statistical power and necessitated sample size increases of 82%–155% (dependent on selection criteria) to achieve equivalent statistical power to parameters with unrestricted samples. However, measures demonstrating indirect range restriction required much smaller sample size increases (32%–71%) under equivalent scenarios. Additional analyses manipulated the correlations between measures and pretest-posttest correlations to guide planning experiments. Results highlight the need to differentiate between selection measures and potential covariates and to investigate range restriction as a factor impacting statistical power. PMID:28479943

Designing Intervention Studies: Selected Populations, Range Restrictions, and Statistical Power.

PubMed

Miciak, Jeremy; Taylor, W Pat; Stuebing, Karla K; Fletcher, Jack M; Vaughn, Sharon

2016-01-01

An appropriate estimate of statistical power is critical for the design of intervention studies. Although the inclusion of a pretest covariate in the test of the primary outcome can increase statistical power, samples selected on the basis of pretest performance may demonstrate range restriction on the selection measure and other correlated measures. This can result in attenuated pretest-posttest correlations, reducing the variance explained by the pretest covariate. We investigated the implications of two potential range restriction scenarios: direct truncation on a selection measure and indirect range restriction on correlated measures. Empirical and simulated data indicated direct range restriction on the pretest covariate greatly reduced statistical power and necessitated sample size increases of 82%-155% (dependent on selection criteria) to achieve equivalent statistical power to parameters with unrestricted samples. However, measures demonstrating indirect range restriction required much smaller sample size increases (32%-71%) under equivalent scenarios. Additional analyses manipulated the correlations between measures and pretest-posttest correlations to guide planning experiments. Results highlight the need to differentiate between selection measures and potential covariates and to investigate range restriction as a factor impacting statistical power.

Automated sampling assessment for molecular simulations using the effective sample size

PubMed Central

Zhang, Xin; Bhatt, Divesh; Zuckerman, Daniel M.

2010-01-01

To quantify the progress in the development of algorithms and forcefields used in molecular simulations, a general method for the assessment of the sampling quality is needed. Statistical mechanics principles suggest the populations of physical states characterize equilibrium sampling in a fundamental way. We therefore develop an approach for analyzing the variances in state populations, which quantifies the degree of sampling in terms of the effective sample size (ESS). The ESS estimates the number of statistically independent configurations contained in a simulated ensemble. The method is applicable to both traditional dynamics simulations as well as more modern (e.g., multi–canonical) approaches. Our procedure is tested in a variety of systems from toy models to atomistic protein simulations. We also introduce a simple automated procedure to obtain approximate physical states from dynamic trajectories: this allows sample–size estimation in systems for which physical states are not known in advance. PMID:21221418

Sample size and power considerations in network meta-analysis

PubMed Central

2012-01-01

Background Network meta-analysis is becoming increasingly popular for establishing comparative effectiveness among multiple interventions for the same disease. Network meta-analysis inherits all methodological challenges of standard pairwise meta-analysis, but with increased complexity due to the multitude of intervention comparisons. One issue that is now widely recognized in pairwise meta-analysis is the issue of sample size and statistical power. This issue, however, has so far only received little attention in network meta-analysis. To date, no approaches have been proposed for evaluating the adequacy of the sample size, and thus power, in a treatment network. Findings In this article, we develop easy-to-use flexible methods for estimating the ‘effective sample size’ in indirect comparison meta-analysis and network meta-analysis. The effective sample size for a particular treatment comparison can be interpreted as the number of patients in a pairwise meta-analysis that would provide the same degree and strength of evidence as that which is provided in the indirect comparison or network meta-analysis. We further develop methods for retrospectively estimating the statistical power for each comparison in a network meta-analysis. We illustrate the performance of the proposed methods for estimating effective sample size and statistical power using data from a network meta-analysis on interventions for smoking cessation including over 100 trials. Conclusion The proposed methods are easy to use and will be of high value to regulatory agencies and decision makers who must assess the strength of the evidence supporting comparative effectiveness estimates. PMID:22992327

Confidence Interval Coverage for Cohen's Effect Size Statistic

ERIC Educational Resources Information Center

Algina, James; Keselman, H. J.; Penfield, Randall D.

2006-01-01

Kelley compared three methods for setting a confidence interval (CI) around Cohen's standardized mean difference statistic: the noncentral-"t"-based, percentile (PERC) bootstrap, and biased-corrected and accelerated (BCA) bootstrap methods under three conditions of nonnormality, eight cases of sample size, and six cases of population…

75 FR 48815 - Medicaid Program and Children's Health Insurance Program (CHIP); Revisions to the Medicaid...

Federal Register 2010, 2011, 2012, 2013, 2014

2010-08-11

... size may be reduced by the finite population correction factor. The finite population correction is a statistical formula utilized to determine sample size where the population is considered finite rather than... program may notify us and the annual sample size will be reduced by the finite population correction...

Sample sizes and model comparison metrics for species distribution models

Treesearch

B.B. Hanberry; H.S. He; D.C. Dey

2012-01-01

Species distribution models use small samples to produce continuous distribution maps. The question of how small a sample can be to produce an accurate model generally has been answered based on comparisons to maximum sample sizes of 200 observations or fewer. In addition, model comparisons often are made with the kappa statistic, which has become controversial....

Effect size and statistical power in the rodent fear conditioning literature - A systematic review.

PubMed

Carneiro, Clarissa F D; Moulin, Thiago C; Macleod, Malcolm R; Amaral, Olavo B

2018-01-01

Proposals to increase research reproducibility frequently call for focusing on effect sizes instead of p values, as well as for increasing the statistical power of experiments. However, it is unclear to what extent these two concepts are indeed taken into account in basic biomedical science. To study this in a real-case scenario, we performed a systematic review of effect sizes and statistical power in studies on learning of rodent fear conditioning, a widely used behavioral task to evaluate memory. Our search criteria yielded 410 experiments comparing control and treated groups in 122 articles. Interventions had a mean effect size of 29.5%, and amnesia caused by memory-impairing interventions was nearly always partial. Mean statistical power to detect the average effect size observed in well-powered experiments with significant differences (37.2%) was 65%, and was lower among studies with non-significant results. Only one article reported a sample size calculation, and our estimated sample size to achieve 80% power considering typical effect sizes and variances (15 animals per group) was reached in only 12.2% of experiments. Actual effect sizes correlated with effect size inferences made by readers on the basis of textual descriptions of results only when findings were non-significant, and neither effect size nor power correlated with study quality indicators, number of citations or impact factor of the publishing journal. In summary, effect sizes and statistical power have a wide distribution in the rodent fear conditioning literature, but do not seem to have a large influence on how results are described or cited. Failure to take these concepts into consideration might limit attempts to improve reproducibility in this field of science.

Effect size and statistical power in the rodent fear conditioning literature – A systematic review

PubMed Central

Macleod, Malcolm R.

2018-01-01

Proposals to increase research reproducibility frequently call for focusing on effect sizes instead of p values, as well as for increasing the statistical power of experiments. However, it is unclear to what extent these two concepts are indeed taken into account in basic biomedical science. To study this in a real-case scenario, we performed a systematic review of effect sizes and statistical power in studies on learning of rodent fear conditioning, a widely used behavioral task to evaluate memory. Our search criteria yielded 410 experiments comparing control and treated groups in 122 articles. Interventions had a mean effect size of 29.5%, and amnesia caused by memory-impairing interventions was nearly always partial. Mean statistical power to detect the average effect size observed in well-powered experiments with significant differences (37.2%) was 65%, and was lower among studies with non-significant results. Only one article reported a sample size calculation, and our estimated sample size to achieve 80% power considering typical effect sizes and variances (15 animals per group) was reached in only 12.2% of experiments. Actual effect sizes correlated with effect size inferences made by readers on the basis of textual descriptions of results only when findings were non-significant, and neither effect size nor power correlated with study quality indicators, number of citations or impact factor of the publishing journal. In summary, effect sizes and statistical power have a wide distribution in the rodent fear conditioning literature, but do not seem to have a large influence on how results are described or cited. Failure to take these concepts into consideration might limit attempts to improve reproducibility in this field of science. PMID:29698451

Estimation of sample size and testing power (Part 4).

PubMed

Hu, Liang-ping; Bao, Xiao-lei; Guan, Xue; Zhou, Shi-guo

2012-01-01

Sample size estimation is necessary for any experimental or survey research. An appropriate estimation of sample size based on known information and statistical knowledge is of great significance. This article introduces methods of sample size estimation of difference test for data with the design of one factor with two levels, including sample size estimation formulas and realization based on the formulas and the POWER procedure of SAS software for quantitative data and qualitative data with the design of one factor with two levels. In addition, this article presents examples for analysis, which will play a leading role for researchers to implement the repetition principle during the research design phase.

[Formal sample size calculation and its limited validity in animal studies of medical basic research].

PubMed

Mayer, B; Muche, R

2013-01-01

Animal studies are highly relevant for basic medical research, although their usage is discussed controversially in public. Thus, an optimal sample size for these projects should be aimed at from a biometrical point of view. Statistical sample size calculation is usually the appropriate methodology in planning medical research projects. However, required information is often not valid or only available during the course of an animal experiment. This article critically discusses the validity of formal sample size calculation for animal studies. Within the discussion, some requirements are formulated to fundamentally regulate the process of sample size determination for animal experiments.

"Magnitude-based inference": a statistical review.

PubMed

Welsh, Alan H; Knight, Emma J

2015-04-01

We consider "magnitude-based inference" and its interpretation by examining in detail its use in the problem of comparing two means. We extract from the spreadsheets, which are provided to users of the analysis (http://www.sportsci.org/), a precise description of how "magnitude-based inference" is implemented. We compare the implemented version of the method with general descriptions of it and interpret the method in familiar statistical terms. We show that "magnitude-based inference" is not a progressive improvement on modern statistics. The additional probabilities introduced are not directly related to the confidence interval but, rather, are interpretable either as P values for two different nonstandard tests (for different null hypotheses) or as approximate Bayesian calculations, which also lead to a type of test. We also discuss sample size calculations associated with "magnitude-based inference" and show that the substantial reduction in sample sizes claimed for the method (30% of the sample size obtained from standard frequentist calculations) is not justifiable so the sample size calculations should not be used. Rather than using "magnitude-based inference," a better solution is to be realistic about the limitations of the data and use either confidence intervals or a fully Bayesian analysis.

Methodological quality of behavioural weight loss studies: a systematic review

PubMed Central

Lemon, S. C.; Wang, M. L.; Haughton, C. F.; Estabrook, D. P.; Frisard, C. F.; Pagoto, S. L.

2018-01-01

Summary This systematic review assessed the methodological quality of behavioural weight loss intervention studies conducted among adults and associations between quality and statistically significant weight loss outcome, strength of intervention effectiveness and sample size. Searches for trials published between January, 2009 and December, 2014 were conducted using PUBMED, MEDLINE and PSYCINFO and identified ninety studies. Methodological quality indicators included study design, anthropometric measurement approach, sample size calculations, intent-to-treat (ITT) analysis, loss to follow-up rate, missing data strategy, sampling strategy, report of treatment receipt and report of intervention fidelity (mean = 6.3). Indicators most commonly utilized included randomized design (100%), objectively measured anthropometrics (96.7%), ITT analysis (86.7%) and reporting treatment adherence (76.7%). Most studies (62.2%) had a follow-up rate >75% and reported a loss to follow-up analytic strategy or minimal missing data (69.9%). Describing intervention fidelity (34.4%) and sampling from a known population (41.1%) were least common. Methodological quality was not associated with reporting a statistically significant result, effect size or sample size. This review found the published literature of behavioural weight loss trials to be of high quality for specific indicators, including study design and measurement. Identified for improvement include utilization of more rigorous statistical approaches to loss to follow up and better fidelity reporting. PMID:27071775

Précis of statistical significance: rationale, validity, and utility.

PubMed

Chow, S L

1998-04-01

The null-hypothesis significance-test procedure (NHSTP) is defended in the context of the theory-corroboration experiment, as well as the following contrasts: (a) substantive hypotheses versus statistical hypotheses, (b) theory corroboration versus statistical hypothesis testing, (c) theoretical inference versus statistical decision, (d) experiments versus nonexperimental studies, and (e) theory corroboration versus treatment assessment. The null hypothesis can be true because it is the hypothesis that errors are randomly distributed in data. Moreover, the null hypothesis is never used as a categorical proposition. Statistical significance means only that chance influences can be excluded as an explanation of data; it does not identify the nonchance factor responsible. The experimental conclusion is drawn with the inductive principle underlying the experimental design. A chain of deductive arguments gives rise to the theoretical conclusion via the experimental conclusion. The anomalous relationship between statistical significance and the effect size often used to criticize NHSTP is more apparent than real. The absolute size of the effect is not an index of evidential support for the substantive hypothesis. Nor is the effect size, by itself, informative as to the practical importance of the research result. Being a conditional probability, statistical power cannot be the a priori probability of statistical significance. The validity of statistical power is debatable because statistical significance is determined with a single sampling distribution of the test statistic based on H0, whereas it takes two distributions to represent statistical power or effect size. Sample size should not be determined in the mechanical manner envisaged in power analysis. It is inappropriate to criticize NHSTP for nonstatistical reasons. At the same time, neither effect size, nor confidence interval estimate, nor posterior probability can be used to exclude chance as an explanation of data. Neither can any of them fulfill the nonstatistical functions expected of them by critics.

Interpretation of correlations in clinical research.

PubMed

Hung, Man; Bounsanga, Jerry; Voss, Maren Wright

2017-11-01

Critically analyzing research is a key skill in evidence-based practice and requires knowledge of research methods, results interpretation, and applications, all of which rely on a foundation based in statistics. Evidence-based practice makes high demands on trained medical professionals to interpret an ever-expanding array of research evidence. As clinical training emphasizes medical care rather than statistics, it is useful to review the basics of statistical methods and what they mean for interpreting clinical studies. We reviewed the basic concepts of correlational associations, violations of normality, unobserved variable bias, sample size, and alpha inflation. The foundations of causal inference were discussed and sound statistical analyses were examined. We discuss four ways in which correlational analysis is misused, including causal inference overreach, over-reliance on significance, alpha inflation, and sample size bias. Recent published studies in the medical field provide evidence of causal assertion overreach drawn from correlational findings. The findings present a primer on the assumptions and nature of correlational methods of analysis and urge clinicians to exercise appropriate caution as they critically analyze the evidence before them and evaluate evidence that supports practice. Critically analyzing new evidence requires statistical knowledge in addition to clinical knowledge. Studies can overstate relationships, expressing causal assertions when only correlational evidence is available. Failure to account for the effect of sample size in the analyses tends to overstate the importance of predictive variables. It is important not to overemphasize the statistical significance without consideration of effect size and whether differences could be considered clinically meaningful.

Generalized SAMPLE SIZE Determination Formulas for Investigating Contextual Effects by a Three-Level Random Intercept Model.

PubMed

Usami, Satoshi

2017-03-01

Behavioral and psychological researchers have shown strong interests in investigating contextual effects (i.e., the influences of combinations of individual- and group-level predictors on individual-level outcomes). The present research provides generalized formulas for determining the sample size needed in investigating contextual effects according to the desired level of statistical power as well as width of confidence interval. These formulas are derived within a three-level random intercept model that includes one predictor/contextual variable at each level to simultaneously cover various kinds of contextual effects that researchers can show interest. The relative influences of indices included in the formulas on the standard errors of contextual effects estimates are investigated with the aim of further simplifying sample size determination procedures. In addition, simulation studies are performed to investigate finite sample behavior of calculated statistical power, showing that estimated sample sizes based on derived formulas can be both positively and negatively biased due to complex effects of unreliability of contextual variables, multicollinearity, and violation of assumption regarding the known variances. Thus, it is advisable to compare estimated sample sizes under various specifications of indices and to evaluate its potential bias, as illustrated in the example.

Sample size determination for estimating antibody seroconversion rate under stable malaria transmission intensity.

PubMed

Sepúlveda, Nuno; Drakeley, Chris

2015-04-03

In the last decade, several epidemiological studies have demonstrated the potential of using seroprevalence (SP) and seroconversion rate (SCR) as informative indicators of malaria burden in low transmission settings or in populations on the cusp of elimination. However, most of studies are designed to control ensuing statistical inference over parasite rates and not on these alternative malaria burden measures. SP is in essence a proportion and, thus, many methods exist for the respective sample size determination. In contrast, designing a study where SCR is the primary endpoint, is not an easy task because precision and statistical power are affected by the age distribution of a given population. Two sample size calculators for SCR estimation are proposed. The first one consists of transforming the confidence interval for SP into the corresponding one for SCR given a known seroreversion rate (SRR). The second calculator extends the previous one to the most common situation where SRR is unknown. In this situation, data simulation was used together with linear regression in order to study the expected relationship between sample size and precision. The performance of the first sample size calculator was studied in terms of the coverage of the confidence intervals for SCR. The results pointed out to eventual problems of under or over coverage for sample sizes ≤250 in very low and high malaria transmission settings (SCR ≤ 0.0036 and SCR ≥ 0.29, respectively). The correct coverage was obtained for the remaining transmission intensities with sample sizes ≥ 50. Sample size determination was then carried out for cross-sectional surveys using realistic SCRs from past sero-epidemiological studies and typical age distributions from African and non-African populations. For SCR < 0.058, African studies require a larger sample size than their non-African counterparts in order to obtain the same precision. The opposite happens for the remaining transmission intensities. With respect to the second sample size calculator, simulation unravelled the likelihood of not having enough information to estimate SRR in low transmission settings (SCR ≤ 0.0108). In that case, the respective estimates tend to underestimate the true SCR. This problem is minimized by sample sizes of no less than 500 individuals. The sample sizes determined by this second method highlighted the prior expectation that, when SRR is not known, sample sizes are increased in relation to the situation of a known SRR. In contrast to the first sample size calculation, African studies would now require lesser individuals than their counterparts conducted elsewhere, irrespective of the transmission intensity. Although the proposed sample size calculators can be instrumental to design future cross-sectional surveys, the choice of a particular sample size must be seen as a much broader exercise that involves weighting statistical precision with ethical issues, available human and economic resources, and possible time constraints. Moreover, if the sample size determination is carried out on varying transmission intensities, as done here, the respective sample sizes can also be used in studies comparing sites with different malaria transmission intensities. In conclusion, the proposed sample size calculators are a step towards the design of better sero-epidemiological studies. Their basic ideas show promise to be applied to the planning of alternative sampling schemes that may target or oversample specific age groups.

Fragment size distribution statistics in dynamic fragmentation of laser shock-loaded tin

NASA Astrophysics Data System (ADS)

He, Weihua; Xin, Jianting; Zhao, Yongqiang; Chu, Genbai; Xi, Tao; Shui, Min; Lu, Feng; Gu, Yuqiu

2017-06-01

This work investigates the geometric statistics method to characterize the size distribution of tin fragments produced in the laser shock-loaded dynamic fragmentation process. In the shock experiments, the ejection of the tin sample with etched V-shape groove in the free surface are collected by the soft recovery technique. Subsequently, the produced fragments are automatically detected with the fine post-shot analysis techniques including the X-ray micro-tomography and the improved watershed method. To characterize the size distributions of the fragments, a theoretical random geometric statistics model based on Poisson mixtures is derived for dynamic heterogeneous fragmentation problem, which reveals linear combinational exponential distribution. The experimental data related to fragment size distributions of the laser shock-loaded tin sample are examined with the proposed theoretical model, and its fitting performance is compared with that of other state-of-the-art fragment size distribution models. The comparison results prove that our proposed model can provide far more reasonable fitting result for the laser shock-loaded tin.

Weighting by Inverse Variance or by Sample Size in Random-Effects Meta-Analysis

ERIC Educational Resources Information Center

Marin-Martinez, Fulgencio; Sanchez-Meca, Julio

2010-01-01

Most of the statistical procedures in meta-analysis are based on the estimation of average effect sizes from a set of primary studies. The optimal weight for averaging a set of independent effect sizes is the inverse variance of each effect size, but in practice these weights have to be estimated, being affected by sampling error. When assuming a…

[A comparison of convenience sampling and purposive sampling].

PubMed

Suen, Lee-Jen Wu; Huang, Hui-Man; Lee, Hao-Hsien

2014-06-01

Convenience sampling and purposive sampling are two different sampling methods. This article first explains sampling terms such as target population, accessible population, simple random sampling, intended sample, actual sample, and statistical power analysis. These terms are then used to explain the difference between "convenience sampling" and purposive sampling." Convenience sampling is a non-probabilistic sampling technique applicable to qualitative or quantitative studies, although it is most frequently used in quantitative studies. In convenience samples, subjects more readily accessible to the researcher are more likely to be included. Thus, in quantitative studies, opportunity to participate is not equal for all qualified individuals in the target population and study results are not necessarily generalizable to this population. As in all quantitative studies, increasing the sample size increases the statistical power of the convenience sample. In contrast, purposive sampling is typically used in qualitative studies. Researchers who use this technique carefully select subjects based on study purpose with the expectation that each participant will provide unique and rich information of value to the study. As a result, members of the accessible population are not interchangeable and sample size is determined by data saturation not by statistical power analysis.

Technology Tips: Sample Too Small? Probably Not!

ERIC Educational Resources Information Center

Strayer, Jeremy F.

2013-01-01

Statistical studies are referenced in the news every day, so frequently that people are sometimes skeptical of reported results. Often, no matter how large a sample size researchers use in their studies, people believe that the sample size is too small to make broad generalizations. The tasks presented in this article use simulations of repeated…

Performance of digital RGB reflectance color extraction for plaque lesion

NASA Astrophysics Data System (ADS)

Hashim, Hadzli; Taib, Mohd Nasir; Jailani, Rozita; Sulaiman, Saadiah; Baba, Roshidah

2005-01-01

Several clinical psoriasis lesion groups are been studied for digital RGB color features extraction. Previous works have used samples size that included all the outliers lying beyond the standard deviation factors from the peak histograms. This paper described the statistical performances of the RGB model with and without removing these outliers. Plaque lesion is experimented with other types of psoriasis. The statistical tests are compared with respect to three samples size; the original 90 samples, the first size reduction by removing outliers from 2 standard deviation distances (2SD) and the second size reduction by removing outliers from 1 standard deviation distance (1SD). Quantification of data images through the normal/direct and differential of the conventional reflectance method is considered. Results performances are concluded by observing the error plots with 95% confidence interval and findings of the inference T-tests applied. The statistical tests outcomes have shown that B component for conventional differential method can be used to distinctively classify plaque from the other psoriasis groups in consistent with the error plots finding with an improvement in p-value greater than 0.5.

The relationship between national-level carbon dioxide emissions and population size: an assessment of regional and temporal variation, 1960-2005.

PubMed

Jorgenson, Andrew K; Clark, Brett

2013-01-01

This study examines the regional and temporal differences in the statistical relationship between national-level carbon dioxide emissions and national-level population size. The authors analyze panel data from 1960 to 2005 for a diverse sample of nations, and employ descriptive statistics and rigorous panel regression modeling techniques. Initial descriptive analyses indicate that all regions experienced overall increases in carbon emissions and population size during the 45-year period of investigation, but with notable differences. For carbon emissions, the sample of countries in Asia experienced the largest percent increase, followed by countries in Latin America, Africa, and lastly the sample of relatively affluent countries in Europe, North America, and Oceania combined. For population size, the sample of countries in Africa experienced the largest percent increase, followed countries in Latin America, Asia, and the combined sample of countries in Europe, North America, and Oceania. Findings for two-way fixed effects panel regression elasticity models of national-level carbon emissions indicate that the estimated elasticity coefficient for population size is much smaller for nations in Africa than for nations in other regions of the world. Regarding potential temporal changes, from 1960 to 2005 the estimated elasticity coefficient for population size decreased by 25% for the sample of Africa countries, 14% for the sample of Asia countries, 6.5% for the sample of Latin America countries, but remained the same in size for the sample of countries in Europe, North America, and Oceania. Overall, while population size continues to be the primary driver of total national-level anthropogenic carbon dioxide emissions, the findings for this study highlight the need for future research and policies to recognize that the actual impacts of population size on national-level carbon emissions differ across both time and region.

Does size matter? Statistical limits of paleomagnetic field reconstruction from small rock specimens

NASA Astrophysics Data System (ADS)

Berndt, Thomas; Muxworthy, Adrian R.; Fabian, Karl

2016-01-01

As samples of ever decreasing sizes are being studied paleomagnetically, care has to be taken that the underlying assumptions of statistical thermodynamics (Maxwell-Boltzmann statistics) are being met. Here we determine how many grains and how large a magnetic moment a sample needs to have to be able to accurately record an ambient field. It is found that for samples with a thermoremanent magnetic moment larger than 10-11Am2 the assumption of a sufficiently large number of grains is usually given. Standard 25 mm diameter paleomagnetic samples usually contain enough magnetic grains such that statistical errors are negligible, but "single silicate crystal" works on, for example, zircon, plagioclase, and olivine crystals are approaching the limits of what is physically possible, leading to statistic errors in both the angular deviation and paleointensity that are comparable to other sources of error. The reliability of nanopaleomagnetic imaging techniques capable of resolving individual grains (used, for example, to study the cloudy zone in meteorites), however, is questionable due to the limited area of the material covered.

Chi-Squared Test of Fit and Sample Size-A Comparison between a Random Sample Approach and a Chi-Square Value Adjustment Method.

PubMed

Bergh, Daniel

2015-01-01

Chi-square statistics are commonly used for tests of fit of measurement models. Chi-square is also sensitive to sample size, which is why several approaches to handle large samples in test of fit analysis have been developed. One strategy to handle the sample size problem may be to adjust the sample size in the analysis of fit. An alternative is to adopt a random sample approach. The purpose of this study was to analyze and to compare these two strategies using simulated data. Given an original sample size of 21,000, for reductions of sample sizes down to the order of 5,000 the adjusted sample size function works as good as the random sample approach. In contrast, when applying adjustments to sample sizes of lower order the adjustment function is less effective at approximating the chi-square value for an actual random sample of the relevant size. Hence, the fit is exaggerated and misfit under-estimated using the adjusted sample size function. Although there are big differences in chi-square values between the two approaches at lower sample sizes, the inferences based on the p-values may be the same.

Relative efficiency and sample size for cluster randomized trials with variable cluster sizes.

PubMed

You, Zhiying; Williams, O Dale; Aban, Inmaculada; Kabagambe, Edmond Kato; Tiwari, Hemant K; Cutter, Gary

2011-02-01

The statistical power of cluster randomized trials depends on two sample size components, the number of clusters per group and the numbers of individuals within clusters (cluster size). Variable cluster sizes are common and this variation alone may have significant impact on study power. Previous approaches have taken this into account by either adjusting total sample size using a designated design effect or adjusting the number of clusters according to an assessment of the relative efficiency of unequal versus equal cluster sizes. This article defines a relative efficiency of unequal versus equal cluster sizes using noncentrality parameters, investigates properties of this measure, and proposes an approach for adjusting the required sample size accordingly. We focus on comparing two groups with normally distributed outcomes using t-test, and use the noncentrality parameter to define the relative efficiency of unequal versus equal cluster sizes and show that statistical power depends only on this parameter for a given number of clusters. We calculate the sample size required for an unequal cluster sizes trial to have the same power as one with equal cluster sizes. Relative efficiency based on the noncentrality parameter is straightforward to calculate and easy to interpret. It connects the required mean cluster size directly to the required sample size with equal cluster sizes. Consequently, our approach first determines the sample size requirements with equal cluster sizes for a pre-specified study power and then calculates the required mean cluster size while keeping the number of clusters unchanged. Our approach allows adjustment in mean cluster size alone or simultaneous adjustment in mean cluster size and number of clusters, and is a flexible alternative to and a useful complement to existing methods. Comparison indicated that we have defined a relative efficiency that is greater than the relative efficiency in the literature under some conditions. Our measure of relative efficiency might be less than the measure in the literature under some conditions, underestimating the relative efficiency. The relative efficiency of unequal versus equal cluster sizes defined using the noncentrality parameter suggests a sample size approach that is a flexible alternative and a useful complement to existing methods.

GLIMMPSE Lite: Calculating Power and Sample Size on Smartphone Devices

PubMed Central

Munjal, Aarti; Sakhadeo, Uttara R.; Muller, Keith E.; Glueck, Deborah H.; Kreidler, Sarah M.

2014-01-01

Researchers seeking to develop complex statistical applications for mobile devices face a common set of difficult implementation issues. In this work, we discuss general solutions to the design challenges. We demonstrate the utility of the solutions for a free mobile application designed to provide power and sample size calculations for univariate, one-way analysis of variance (ANOVA), GLIMMPSE Lite. Our design decisions provide a guide for other scientists seeking to produce statistical software for mobile platforms. PMID:25541688

HYPERSAMP - HYPERGEOMETRIC ATTRIBUTE SAMPLING SYSTEM BASED ON RISK AND FRACTION DEFECTIVE

NASA Technical Reports Server (NTRS)

De, Salvo L. J.

1994-01-01

HYPERSAMP is a demonstration of an attribute sampling system developed to determine the minimum sample size required for any preselected value for consumer's risk and fraction of nonconforming. This statistical method can be used in place of MIL-STD-105E sampling plans when a minimum sample size is desirable, such as when tests are destructive or expensive. HYPERSAMP utilizes the Hypergeometric Distribution and can be used for any fraction nonconforming. The program employs an iterative technique that circumvents the obstacle presented by the factorial of a non-whole number. HYPERSAMP provides the required Hypergeometric sample size for any equivalent real number of nonconformances in the lot or batch under evaluation. Many currently used sampling systems, such as the MIL-STD-105E, utilize the Binomial or the Poisson equations as an estimate of the Hypergeometric when performing inspection by attributes. However, this is primarily because of the difficulty in calculation of the factorials required by the Hypergeometric. Sampling plans based on the Binomial or Poisson equations will result in the maximum sample size possible with the Hypergeometric. The difference in the sample sizes between the Poisson or Binomial and the Hypergeometric can be significant. For example, a lot size of 400 devices with an error rate of 1.0% and a confidence of 99% would require a sample size of 400 (all units would need to be inspected) for the Binomial sampling plan and only 273 for a Hypergeometric sampling plan. The Hypergeometric results in a savings of 127 units, a significant reduction in the required sample size. HYPERSAMP is a demonstration program and is limited to sampling plans with zero defectives in the sample (acceptance number of zero). Since it is only a demonstration program, the sample size determination is limited to sample sizes of 1500 or less. The Hypergeometric Attribute Sampling System demonstration code is a spreadsheet program written for IBM PC compatible computers running DOS and Lotus 1-2-3 or Quattro Pro. This program is distributed on a 5.25 inch 360K MS-DOS format diskette, and the program price includes documentation. This statistical method was developed in 1992.

Normal Approximations to the Distributions of the Wilcoxon Statistics: Accurate to What "N"? Graphical Insights

ERIC Educational Resources Information Center

Bellera, Carine A.; Julien, Marilyse; Hanley, James A.

2010-01-01

The Wilcoxon statistics are usually taught as nonparametric alternatives for the 1- and 2-sample Student-"t" statistics in situations where the data appear to arise from non-normal distributions, or where sample sizes are so small that we cannot check whether they do. In the past, critical values, based on exact tail areas, were…

"What If" Analyses: Ways to Interpret Statistical Significance Test Results Using EXCEL or "R"

ERIC Educational Resources Information Center

Ozturk, Elif

2012-01-01

The present paper aims to review two motivations to conduct "what if" analyses using Excel and "R" to understand the statistical significance tests through the sample size context. "What if" analyses can be used to teach students what statistical significance tests really do and in applied research either prospectively to estimate what sample size…

Orphan therapies: making best use of postmarket data.

PubMed

Maro, Judith C; Brown, Jeffrey S; Dal Pan, Gerald J; Li, Lingling

2014-08-01

Postmarket surveillance of the comparative safety and efficacy of orphan therapeutics is challenging, particularly when multiple therapeutics are licensed for the same orphan indication. To make best use of product-specific registry data collected to fulfill regulatory requirements, we propose the creation of a distributed electronic health data network among registries. Such a network could support sequential statistical analyses designed to detect early warnings of excess risks. We use a simulated example to explore the circumstances under which a distributed network may prove advantageous. We perform sample size calculations for sequential and non-sequential statistical studies aimed at comparing the incidence of hepatotoxicity following initiation of two newly licensed therapies for homozygous familial hypercholesterolemia. We calculate the sample size savings ratio, or the proportion of sample size saved if one conducted a sequential study as compared to a non-sequential study. Then, using models to describe the adoption and utilization of these therapies, we simulate when these sample sizes are attainable in calendar years. We then calculate the analytic calendar time savings ratio, analogous to the sample size savings ratio. We repeat these analyses for numerous scenarios. Sequential analyses detect effect sizes earlier or at the same time as non-sequential analyses. The most substantial potential savings occur when the market share is more imbalanced (i.e., 90% for therapy A) and the effect size is closest to the null hypothesis. However, due to low exposure prevalence, these savings are difficult to realize within the 30-year time frame of this simulation for scenarios in which the outcome of interest occurs at or more frequently than one event/100 person-years. We illustrate a process to assess whether sequential statistical analyses of registry data performed via distributed networks may prove a worthwhile infrastructure investment for pharmacovigilance.

Application of binomial and multinomial probability statistics to the sampling design process of a global grain tracing and recall system

USDA-ARS?s Scientific Manuscript database

Small, coded, pill-sized tracers embedded in grain are proposed as a method for grain traceability. A sampling process for a grain traceability system was designed and investigated by applying probability statistics using a science-based sampling approach to collect an adequate number of tracers fo...

Statistical Methods in Assembly Quality Management of Multi-Element Products on Automatic Rotor Lines

NASA Astrophysics Data System (ADS)

Pries, V. V.; Proskuriakov, N. E.

2018-04-01

To control the assembly quality of multi-element mass-produced products on automatic rotor lines, control methods with operational feedback are required. However, due to possible failures in the operation of the devices and systems of automatic rotor line, there is always a real probability of getting defective (incomplete) products into the output process stream. Therefore, a continuous sampling control of the products completeness, based on the use of statistical methods, remains an important element in managing the quality of assembly of multi-element mass products on automatic rotor lines. The feature of continuous sampling control of the multi-element products completeness in the assembly process is its breaking sort, which excludes the possibility of returning component parts after sampling control to the process stream and leads to a decrease in the actual productivity of the assembly equipment. Therefore, the use of statistical procedures for continuous sampling control of the multi-element products completeness when assembled on automatic rotor lines requires the use of such sampling plans that ensure a minimum size of control samples. Comparison of the values of the limit of the average output defect level for the continuous sampling plan (CSP) and for the automated continuous sampling plan (ACSP) shows the possibility of providing lower limit values for the average output defects level using the ACSP-1. Also, the average sample size when using the ACSP-1 plan is less than when using the CSP-1 plan. Thus, the application of statistical methods in the assembly quality management of multi-element products on automatic rotor lines, involving the use of proposed plans and methods for continuous selective control, will allow to automating sampling control procedures and the required level of quality of assembled products while minimizing sample size.

A random-sum Wilcoxon statistic and its application to analysis of ROC and LROC data.

PubMed

Tang, Liansheng Larry; Balakrishnan, N

2011-01-01

The Wilcoxon-Mann-Whitney statistic is commonly used for a distribution-free comparison of two groups. One requirement for its use is that the sample sizes of the two groups are fixed. This is violated in some of the applications such as medical imaging studies and diagnostic marker studies; in the former, the violation occurs since the number of correctly localized abnormal images is random, while in the latter the violation is due to some subjects not having observable measurements. For this reason, we propose here a random-sum Wilcoxon statistic for comparing two groups in the presence of ties, and derive its variance as well as its asymptotic distribution for large sample sizes. The proposed statistic includes the regular Wilcoxon rank-sum statistic. Finally, we apply the proposed statistic for summarizing location response operating characteristic data from a liver computed tomography study, and also for summarizing diagnostic accuracy of biomarker data.

Selecting the optimum plot size for a California design-based stream and wetland mapping program.

PubMed

Lackey, Leila G; Stein, Eric D

2014-04-01

Accurate estimates of the extent and distribution of wetlands and streams are the foundation of wetland monitoring, management, restoration, and regulatory programs. Traditionally, these estimates have relied on comprehensive mapping. However, this approach is prohibitively resource-intensive over large areas, making it both impractical and statistically unreliable. Probabilistic (design-based) approaches to evaluating status and trends provide a more cost-effective alternative because, compared with comprehensive mapping, overall extent is inferred from mapping a statistically representative, randomly selected subset of the target area. In this type of design, the size of sample plots has a significant impact on program costs and on statistical precision and accuracy; however, no consensus exists on the appropriate plot size for remote monitoring of stream and wetland extent. This study utilized simulated sampling to assess the performance of four plot sizes (1, 4, 9, and 16 km(2)) for three geographic regions of California. Simulation results showed smaller plot sizes (1 and 4 km(2)) were most efficient for achieving desired levels of statistical accuracy and precision. However, larger plot sizes were more likely to contain rare and spatially limited wetland subtypes. Balancing these considerations led to selection of 4 km(2) for the California status and trends program.

OCT Amplitude and Speckle Statistics of Discrete Random Media.

PubMed

Almasian, Mitra; van Leeuwen, Ton G; Faber, Dirk J

2017-11-01

Speckle, amplitude fluctuations in optical coherence tomography (OCT) images, contains information on sub-resolution structural properties of the imaged sample. Speckle statistics could therefore be utilized in the characterization of biological tissues. However, a rigorous theoretical framework relating OCT speckle statistics to structural tissue properties has yet to be developed. As a first step, we present a theoretical description of OCT speckle, relating the OCT amplitude variance to size and organization for samples of discrete random media (DRM). Starting the calculations from the size and organization of the scattering particles, we analytically find expressions for the OCT amplitude mean, amplitude variance, the backscattering coefficient and the scattering coefficient. We assume fully developed speckle and verify the validity of this assumption by experiments on controlled samples of silica microspheres suspended in water. We show that the OCT amplitude variance is sensitive to sub-resolution changes in size and organization of the scattering particles. Experimentally determined and theoretically calculated optical properties are compared and in good agreement.

Analysis of Longitudinal Outcome Data with Missing Values in Total Knee Arthroplasty.

PubMed

Kang, Yeon Gwi; Lee, Jang Taek; Kang, Jong Yeal; Kim, Ga Hye; Kim, Tae Kyun

2016-01-01

We sought to determine the influence of missing data on the statistical results, and to determine which statistical method is most appropriate for the analysis of longitudinal outcome data of TKA with missing values among repeated measures ANOVA, generalized estimating equation (GEE) and mixed effects model repeated measures (MMRM). Data sets with missing values were generated with different proportion of missing data, sample size and missing-data generation mechanism. Each data set was analyzed with three statistical methods. The influence of missing data was greater with higher proportion of missing data and smaller sample size. MMRM tended to show least changes in the statistics. When missing values were generated by 'missing not at random' mechanism, no statistical methods could fully avoid deviations in the results. Copyright © 2016 Elsevier Inc. All rights reserved.

Seven ways to increase power without increasing N.

PubMed

Hansen, W B; Collins, L M

1994-01-01

Many readers of this monograph may wonder why a chapter on statistical power was included. After all, by now the issue of statistical power is in many respects mundane. Everyone knows that statistical power is a central research consideration, and certainly most National Institute on Drug Abuse grantees or prospective grantees understand the importance of including a power analysis in research proposals. However, there is ample evidence that, in practice, prevention researchers are not paying sufficient attention to statistical power. If they were, the findings observed by Hansen (1992) in a recent review of the prevention literature would not have emerged. Hansen (1992) examined statistical power based on 46 cohorts followed longitudinally, using nonparametric assumptions given the subjects' age at posttest and the numbers of subjects. Results of this analysis indicated that, in order for a study to attain 80-percent power for detecting differences between treatment and control groups, the difference between groups at posttest would need to be at least 8 percent (in the best studies) and as much as 16 percent (in the weakest studies). In order for a study to attain 80-percent power for detecting group differences in pre-post change, 22 of the 46 cohorts would have needed relative pre-post reductions of greater than 100 percent. Thirty-three of the 46 cohorts had less than 50-percent power to detect a 50-percent relative reduction in substance use. These results are consistent with other review findings (e.g., Lipsey 1990) that have shown a similar lack of power in a broad range of research topics. Thus, it seems that, although researchers are aware of the importance of statistical power (particularly of the necessity for calculating it when proposing research), they somehow are failing to end up with adequate power in their completed studies. This chapter argues that the failure of many prevention studies to maintain adequate statistical power is due to an overemphasis on sample size (N) as the only, or even the best, way to increase statistical power. It is easy to see how this overemphasis has come about. Sample size is easy to manipulate, has the advantage of being related to power in a straight-forward way, and usually is under the direct control of the researcher, except for limitations imposed by finances or subject availability. Another option for increasing power is to increase the alpha used for hypothesis-testing but, as very few researchers seriously consider significance levels much larger than the traditional .05, this strategy seldom is used. Of course, sample size is important, and the authors of this chapter are not recommending that researchers cease choosing sample sizes carefully. Rather, they argue that researchers should not confine themselves to increasing N to enhance power. It is important to take additional measures to maintain and improve power over and above making sure the initial sample size is sufficient. The authors recommend two general strategies. One strategy involves attempting to maintain the effective initial sample size so that power is not lost needlessly. The other strategy is to take measures to maximize the third factor that determines statistical power: effect size.

Multiple Category-Lot Quality Assurance Sampling: A New Classification System with Application to Schistosomiasis Control

PubMed Central

Olives, Casey; Valadez, Joseph J.; Brooker, Simon J.; Pagano, Marcello

2012-01-01

Background Originally a binary classifier, Lot Quality Assurance Sampling (LQAS) has proven to be a useful tool for classification of the prevalence of Schistosoma mansoni into multiple categories (≤10%, >10 and <50%, ≥50%), and semi-curtailed sampling has been shown to effectively reduce the number of observations needed to reach a decision. To date the statistical underpinnings for Multiple Category-LQAS (MC-LQAS) have not received full treatment. We explore the analytical properties of MC-LQAS, and validate its use for the classification of S. mansoni prevalence in multiple settings in East Africa. Methodology We outline MC-LQAS design principles and formulae for operating characteristic curves. In addition, we derive the average sample number for MC-LQAS when utilizing semi-curtailed sampling and introduce curtailed sampling in this setting. We also assess the performance of MC-LQAS designs with maximum sample sizes of n = 15 and n = 25 via a weighted kappa-statistic using S. mansoni data collected in 388 schools from four studies in East Africa. Principle Findings Overall performance of MC-LQAS classification was high (kappa-statistic of 0.87). In three of the studies, the kappa-statistic for a design with n = 15 was greater than 0.75. In the fourth study, where these designs performed poorly (kappa-statistic less than 0.50), the majority of observations fell in regions where potential error is known to be high. Employment of semi-curtailed and curtailed sampling further reduced the sample size by as many as 0.5 and 3.5 observations per school, respectively, without increasing classification error. Conclusion/Significance This work provides the needed analytics to understand the properties of MC-LQAS for assessing the prevalance of S. mansoni and shows that in most settings a sample size of 15 children provides a reliable classification of schools. PMID:22970333

A reliability evaluation methodology for memory chips for space applications when sample size is small

NASA Technical Reports Server (NTRS)

Chen, Y.; Nguyen, D.; Guertin, S.; Berstein, J.; White, M.; Menke, R.; Kayali, S.

2003-01-01

This paper presents a reliability evaluation methodology to obtain the statistical reliability information of memory chips for space applications when the test sample size needs to be kept small because of the high cost of the radiation hardness memories.

Experimental design, power and sample size for animal reproduction experiments.

PubMed

Chapman, Phillip L; Seidel, George E

2008-01-01

The present paper concerns statistical issues in the design of animal reproduction experiments, with emphasis on the problems of sample size determination and power calculations. We include examples and non-technical discussions aimed at helping researchers avoid serious errors that may invalidate or seriously impair the validity of conclusions from experiments. Screen shots from interactive power calculation programs and basic SAS power calculation programs are presented to aid in understanding statistical power and computing power in some common experimental situations. Practical issues that are common to most statistical design problems are briefly discussed. These include one-sided hypothesis tests, power level criteria, equality of within-group variances, transformations of response variables to achieve variance equality, optimal specification of treatment group sizes, 'post hoc' power analysis and arguments for the increased use of confidence intervals in place of hypothesis tests.

“Magnitude-based Inference”: A Statistical Review

PubMed Central

Welsh, Alan H.; Knight, Emma J.

2015-01-01

ABSTRACT Purpose We consider “magnitude-based inference” and its interpretation by examining in detail its use in the problem of comparing two means. Methods We extract from the spreadsheets, which are provided to users of the analysis (http://www.sportsci.org/), a precise description of how “magnitude-based inference” is implemented. We compare the implemented version of the method with general descriptions of it and interpret the method in familiar statistical terms. Results and Conclusions We show that “magnitude-based inference” is not a progressive improvement on modern statistics. The additional probabilities introduced are not directly related to the confidence interval but, rather, are interpretable either as P values for two different nonstandard tests (for different null hypotheses) or as approximate Bayesian calculations, which also lead to a type of test. We also discuss sample size calculations associated with “magnitude-based inference” and show that the substantial reduction in sample sizes claimed for the method (30% of the sample size obtained from standard frequentist calculations) is not justifiable so the sample size calculations should not be used. Rather than using “magnitude-based inference,” a better solution is to be realistic about the limitations of the data and use either confidence intervals or a fully Bayesian analysis. PMID:25051387

Study samples are too small to produce sufficiently precise reliability coefficients.

PubMed

Charter, Richard A

2003-04-01

In a survey of journal articles, test manuals, and test critique books, the author found that a mean sample size (N) of 260 participants had been used for reliability studies on 742 tests. The distribution was skewed because the median sample size for the total sample was only 90. The median sample sizes for the internal consistency, retest, and interjudge reliabilities were 182, 64, and 36, respectively. The author presented sample size statistics for the various internal consistency methods and types of tests. In general, the author found that the sample sizes that were used in the internal consistency studies were too small to produce sufficiently precise reliability coefficients, which in turn could cause imprecise estimates of examinee true-score confidence intervals. The results also suggest that larger sample sizes have been used in the last decade compared with those that were used in earlier decades.

Maximum type I error rate inflation from sample size reassessment when investigators are blind to treatment labels.

PubMed

Żebrowska, Magdalena; Posch, Martin; Magirr, Dominic

2016-05-30

Consider a parallel group trial for the comparison of an experimental treatment to a control, where the second-stage sample size may depend on the blinded primary endpoint data as well as on additional blinded data from a secondary endpoint. For the setting of normally distributed endpoints, we demonstrate that this may lead to an inflation of the type I error rate if the null hypothesis holds for the primary but not the secondary endpoint. We derive upper bounds for the inflation of the type I error rate, both for trials that employ random allocation and for those that use block randomization. We illustrate the worst-case sample size reassessment rule in a case study. For both randomization strategies, the maximum type I error rate increases with the effect size in the secondary endpoint and the correlation between endpoints. The maximum inflation increases with smaller block sizes if information on the block size is used in the reassessment rule. Based on our findings, we do not question the well-established use of blinded sample size reassessment methods with nuisance parameter estimates computed from the blinded interim data of the primary endpoint. However, we demonstrate that the type I error rate control of these methods relies on the application of specific, binding, pre-planned and fully algorithmic sample size reassessment rules and does not extend to general or unplanned sample size adjustments based on blinded data. © 2015 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd. © 2015 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

Evidence for a Global Sampling Process in Extraction of Summary Statistics of Item Sizes in a Set.

PubMed

Tokita, Midori; Ueda, Sachiyo; Ishiguchi, Akira

2016-01-01

Several studies have shown that our visual system may construct a "summary statistical representation" over groups of visual objects. Although there is a general understanding that human observers can accurately represent sets of a variety of features, many questions on how summary statistics, such as an average, are computed remain unanswered. This study investigated sampling properties of visual information used by human observers to extract two types of summary statistics of item sets, average and variance. We presented three models of ideal observers to extract the summary statistics: a global sampling model without sampling noise, global sampling model with sampling noise, and limited sampling model. We compared the performance of an ideal observer of each model with that of human observers using statistical efficiency analysis. Results suggest that summary statistics of items in a set may be computed without representing individual items, which makes it possible to discard the limited sampling account. Moreover, the extraction of summary statistics may not necessarily require the representation of individual objects with focused attention when the sets of items are larger than 4.

Modified Distribution-Free Goodness-of-Fit Test Statistic.

PubMed

Chun, So Yeon; Browne, Michael W; Shapiro, Alexander

2018-03-01

Covariance structure analysis and its structural equation modeling extensions have become one of the most widely used methodologies in social sciences such as psychology, education, and economics. An important issue in such analysis is to assess the goodness of fit of a model under analysis. One of the most popular test statistics used in covariance structure analysis is the asymptotically distribution-free (ADF) test statistic introduced by Browne (Br J Math Stat Psychol 37:62-83, 1984). The ADF statistic can be used to test models without any specific distribution assumption (e.g., multivariate normal distribution) of the observed data. Despite its advantage, it has been shown in various empirical studies that unless sample sizes are extremely large, this ADF statistic could perform very poorly in practice. In this paper, we provide a theoretical explanation for this phenomenon and further propose a modified test statistic that improves the performance in samples of realistic size. The proposed statistic deals with the possible ill-conditioning of the involved large-scale covariance matrices.

Designing image segmentation studies: Statistical power, sample size and reference standard quality.

PubMed

Gibson, Eli; Hu, Yipeng; Huisman, Henkjan J; Barratt, Dean C

2017-12-01

Segmentation algorithms are typically evaluated by comparison to an accepted reference standard. The cost of generating accurate reference standards for medical image segmentation can be substantial. Since the study cost and the likelihood of detecting a clinically meaningful difference in accuracy both depend on the size and on the quality of the study reference standard, balancing these trade-offs supports the efficient use of research resources. In this work, we derive a statistical power calculation that enables researchers to estimate the appropriate sample size to detect clinically meaningful differences in segmentation accuracy (i.e. the proportion of voxels matching the reference standard) between two algorithms. Furthermore, we derive a formula to relate reference standard errors to their effect on the sample sizes of studies using lower-quality (but potentially more affordable and practically available) reference standards. The accuracy of the derived sample size formula was estimated through Monte Carlo simulation, demonstrating, with 95% confidence, a predicted statistical power within 4% of simulated values across a range of model parameters. This corresponds to sample size errors of less than 4 subjects and errors in the detectable accuracy difference less than 0.6%. The applicability of the formula to real-world data was assessed using bootstrap resampling simulations for pairs of algorithms from the PROMISE12 prostate MR segmentation challenge data set. The model predicted the simulated power for the majority of algorithm pairs within 4% for simulated experiments using a high-quality reference standard and within 6% for simulated experiments using a low-quality reference standard. A case study, also based on the PROMISE12 data, illustrates using the formulae to evaluate whether to use a lower-quality reference standard in a prostate segmentation study. Copyright © 2017 The Authors. Published by Elsevier B.V. All rights reserved.

Sampling surface and subsurface particle-size distributions in wadable gravel-and cobble-bed streams for analyses in sediment transport, hydraulics, and streambed monitoring

Treesearch

Kristin Bunte; Steven R. Abt

2001-01-01

This document provides guidance for sampling surface and subsurface sediment from wadable gravel-and cobble-bed streams. After a short introduction to streams types and classifications in gravel-bed rivers, the document explains the field and laboratory measurement of particle sizes and the statistical analysis of particle-size distributions. Analysis of particle...

Sample size considerations when groups are the appropriate unit of analyses

PubMed Central

Sadler, Georgia Robins; Ko, Celine Marie; Alisangco, Jennifer; Rosbrook, Bradley P.; Miller, Eric; Fullerton, Judith

2007-01-01

This paper discusses issues to be considered by nurse researchers when groups should be used as a unit of randomization. Advantages and disadvantages are presented, with statistical calculations needed to determine effective sample size. Examples of these concepts are presented using data from the Black Cosmetologists Promoting Health Program. Different hypothetical scenarios and their impact on sample size are presented. Given the complexity of calculating sample size when using groups as a unit of randomization, it’s advantageous for researchers to work closely with statisticians when designing and implementing studies that anticipate the use of groups as the unit of randomization. PMID:17693219

Optimum sample size allocation to minimize cost or maximize power for the two-sample trimmed mean test.

PubMed

Guo, Jiin-Huarng; Luh, Wei-Ming

2009-05-01

When planning a study, sample size determination is one of the most important tasks facing the researcher. The size will depend on the purpose of the study, the cost limitations, and the nature of the data. By specifying the standard deviation ratio and/or the sample size ratio, the present study considers the problem of heterogeneous variances and non-normality for Yuen's two-group test and develops sample size formulas to minimize the total cost or maximize the power of the test. For a given power, the sample size allocation ratio can be manipulated so that the proposed formulas can minimize the total cost, the total sample size, or the sum of total sample size and total cost. On the other hand, for a given total cost, the optimum sample size allocation ratio can maximize the statistical power of the test. After the sample size is determined, the present simulation applies Yuen's test to the sample generated, and then the procedure is validated in terms of Type I errors and power. Simulation results show that the proposed formulas can control Type I errors and achieve the desired power under the various conditions specified. Finally, the implications for determining sample sizes in experimental studies and future research are discussed.

Automated grain mapping using wide angle convergent beam electron diffraction in transmission electron microscope for nanomaterials.

PubMed

Kumar, Vineet

2011-12-01

The grain size statistics, commonly derived from the grain map of a material sample, are important microstructure characteristics that greatly influence its properties. The grain map for nanomaterials is usually obtained manually by visual inspection of the transmission electron microscope (TEM) micrographs because automated methods do not perform satisfactorily. While the visual inspection method provides reliable results, it is a labor intensive process and is often prone to human errors. In this article, an automated grain mapping method is developed using TEM diffraction patterns. The presented method uses wide angle convergent beam diffraction in the TEM. The automated technique was applied on a platinum thin film sample to obtain the grain map and subsequently derive grain size statistics from it. The grain size statistics obtained with the automated method were found in good agreement with the visual inspection method.

Assessment of sampling stability in ecological applications of discriminant analysis

USGS Publications Warehouse

Williams, B.K.; Titus, K.

1988-01-01

A simulation study was undertaken to assess the sampling stability of the variable loadings in linear discriminant function analysis. A factorial design was used for the factors of multivariate dimensionality, dispersion structure, configuration of group means, and sample size. A total of 32,400 discriminant analyses were conducted, based on data from simulated populations with appropriate underlying statistical distributions. A review of 60 published studies and 142 individual analyses indicated that sample sizes in ecological studies often have met that requirement. However, individual group sample sizes frequently were very unequal, and checks of assumptions usually were not reported. The authors recommend that ecologists obtain group sample sizes that are at least three times as large as the number of variables measured.

The Relationship between National-Level Carbon Dioxide Emissions and Population Size: An Assessment of Regional and Temporal Variation, 1960–2005

PubMed Central

Jorgenson, Andrew K.; Clark, Brett

2013-01-01

This study examines the regional and temporal differences in the statistical relationship between national-level carbon dioxide emissions and national-level population size. The authors analyze panel data from 1960 to 2005 for a diverse sample of nations, and employ descriptive statistics and rigorous panel regression modeling techniques. Initial descriptive analyses indicate that all regions experienced overall increases in carbon emissions and population size during the 45-year period of investigation, but with notable differences. For carbon emissions, the sample of countries in Asia experienced the largest percent increase, followed by countries in Latin America, Africa, and lastly the sample of relatively affluent countries in Europe, North America, and Oceania combined. For population size, the sample of countries in Africa experienced the largest percent increase, followed countries in Latin America, Asia, and the combined sample of countries in Europe, North America, and Oceania. Findings for two-way fixed effects panel regression elasticity models of national-level carbon emissions indicate that the estimated elasticity coefficient for population size is much smaller for nations in Africa than for nations in other regions of the world. Regarding potential temporal changes, from 1960 to 2005 the estimated elasticity coefficient for population size decreased by 25% for the sample of Africa countries, 14% for the sample of Asia countries, 6.5% for the sample of Latin America countries, but remained the same in size for the sample of countries in Europe, North America, and Oceania. Overall, while population size continues to be the primary driver of total national-level anthropogenic carbon dioxide emissions, the findings for this study highlight the need for future research and policies to recognize that the actual impacts of population size on national-level carbon emissions differ across both time and region. PMID:23437323

The Statistics and Mathematics of High Dimension Low Sample Size Asymptotics.

PubMed

Shen, Dan; Shen, Haipeng; Zhu, Hongtu; Marron, J S

2016-10-01

The aim of this paper is to establish several deep theoretical properties of principal component analysis for multiple-component spike covariance models. Our new results reveal an asymptotic conical structure in critical sample eigendirections under the spike models with distinguishable (or indistinguishable) eigenvalues, when the sample size and/or the number of variables (or dimension) tend to infinity. The consistency of the sample eigenvectors relative to their population counterparts is determined by the ratio between the dimension and the product of the sample size with the spike size. When this ratio converges to a nonzero constant, the sample eigenvector converges to a cone, with a certain angle to its corresponding population eigenvector. In the High Dimension, Low Sample Size case, the angle between the sample eigenvector and its population counterpart converges to a limiting distribution. Several generalizations of the multi-spike covariance models are also explored, and additional theoretical results are presented.

Interpreting and Reporting Effect Sizes in Research Investigations.

ERIC Educational Resources Information Center

Tapia, Martha; Marsh, George E., II

Since 1994, the American Psychological Association (APA) has advocated the inclusion of effect size indices in reporting research to elucidate the statistical significance of studies based on sample size. In 2001, the fifth edition of the APA "Publication Manual" stressed the importance of including an index of effect size to clarify…

A Bayesian nonparametric method for prediction in EST analysis

PubMed Central

Lijoi, Antonio; Mena, Ramsés H; Prünster, Igor

2007-01-01

Background Expressed sequence tags (ESTs) analyses are a fundamental tool for gene identification in organisms. Given a preliminary EST sample from a certain library, several statistical prediction problems arise. In particular, it is of interest to estimate how many new genes can be detected in a future EST sample of given size and also to determine the gene discovery rate: these estimates represent the basis for deciding whether to proceed sequencing the library and, in case of a positive decision, a guideline for selecting the size of the new sample. Such information is also useful for establishing sequencing efficiency in experimental design and for measuring the degree of redundancy of an EST library. Results In this work we propose a Bayesian nonparametric approach for tackling statistical problems related to EST surveys. In particular, we provide estimates for: a) the coverage, defined as the proportion of unique genes in the library represented in the given sample of reads; b) the number of new unique genes to be observed in a future sample; c) the discovery rate of new genes as a function of the future sample size. The Bayesian nonparametric model we adopt conveys, in a statistically rigorous way, the available information into prediction. Our proposal has appealing properties over frequentist nonparametric methods, which become unstable when prediction is required for large future samples. EST libraries, previously studied with frequentist methods, are analyzed in detail. Conclusion The Bayesian nonparametric approach we undertake yields valuable tools for gene capture and prediction in EST libraries. The estimators we obtain do not feature the kind of drawbacks associated with frequentist estimators and are reliable for any size of the additional sample. PMID:17868445

Got power? A systematic review of sample size adequacy in health professions education research.

PubMed

Cook, David A; Hatala, Rose

2015-03-01

Many education research studies employ small samples, which in turn lowers statistical power. We re-analyzed the results of a meta-analysis of simulation-based education to determine study power across a range of effect sizes, and the smallest effect that could be plausibly excluded. We systematically searched multiple databases through May 2011, and included all studies evaluating simulation-based education for health professionals in comparison with no intervention or another simulation intervention. Reviewers working in duplicate abstracted information to calculate standardized mean differences (SMD's). We included 897 original research studies. Among the 627 no-intervention-comparison studies the median sample size was 25. Only two studies (0.3%) had ≥80% power to detect a small difference (SMD > 0.2 standard deviations) and 136 (22%) had power to detect a large difference (SMD > 0.8). 110 no-intervention-comparison studies failed to find a statistically significant difference, but none excluded a small difference and only 47 (43%) excluded a large difference. Among 297 studies comparing alternate simulation approaches the median sample size was 30. Only one study (0.3%) had ≥80% power to detect a small difference and 79 (27%) had power to detect a large difference. Of the 128 studies that did not detect a statistically significant effect, 4 (3%) excluded a small difference and 91 (71%) excluded a large difference. In conclusion, most education research studies are powered only to detect effects of large magnitude. For most studies that do not reach statistical significance, the possibility of large and important differences still exists.

An Investigation of Sample Size Splitting on ATFIND and DIMTEST

ERIC Educational Resources Information Center

Socha, Alan; DeMars, Christine E.

2013-01-01

Modeling multidimensional test data with a unidimensional model can result in serious statistical errors, such as bias in item parameter estimates. Many methods exist for assessing the dimensionality of a test. The current study focused on DIMTEST. Using simulated data, the effects of sample size splitting for use with the ATFIND procedure for…

State Estimates of Disability in America. Disability Statistics Report 3.

ERIC Educational Resources Information Center

LaPlante, Mitchell P.

This study presents and discusses existing data on disability by state, from the 1980 and 1990 censuses, the Current Population Survey (CPS), and the National Health Interview Survey (NHIS). The study used direct methods for states with large sample sizes and synthetic estimates for states with low sample sizes. The study's highlighted findings…

Equivalent statistics and data interpretation.

PubMed

Francis, Gregory

2017-08-01

Recent reform efforts in psychological science have led to a plethora of choices for scientists to analyze their data. A scientist making an inference about their data must now decide whether to report a p value, summarize the data with a standardized effect size and its confidence interval, report a Bayes Factor, or use other model comparison methods. To make good choices among these options, it is necessary for researchers to understand the characteristics of the various statistics used by the different analysis frameworks. Toward that end, this paper makes two contributions. First, it shows that for the case of a two-sample t test with known sample sizes, many different summary statistics are mathematically equivalent in the sense that they are based on the very same information in the data set. When the sample sizes are known, the p value provides as much information about a data set as the confidence interval of Cohen's d or a JZS Bayes factor. Second, this equivalence means that different analysis methods differ only in their interpretation of the empirical data. At first glance, it might seem that mathematical equivalence of the statistics suggests that it does not matter much which statistic is reported, but the opposite is true because the appropriateness of a reported statistic is relative to the inference it promotes. Accordingly, scientists should choose an analysis method appropriate for their scientific investigation. A direct comparison of the different inferential frameworks provides some guidance for scientists to make good choices and improve scientific practice.

Sample Size Requirements for Studies of Treatment Effects on Beta-Cell Function in Newly Diagnosed Type 1 Diabetes

PubMed Central

Lachin, John M.; McGee, Paula L.; Greenbaum, Carla J.; Palmer, Jerry; Gottlieb, Peter; Skyler, Jay

2011-01-01

Preservation of -cell function as measured by stimulated C-peptide has recently been accepted as a therapeutic target for subjects with newly diagnosed type 1 diabetes. In recently completed studies conducted by the Type 1 Diabetes Trial Network (TrialNet), repeated 2-hour Mixed Meal Tolerance Tests (MMTT) were obtained for up to 24 months from 156 subjects with up to 3 months duration of type 1 diabetes at the time of study enrollment. These data provide the information needed to more accurately determine the sample size needed for future studies of the effects of new agents on the 2-hour area under the curve (AUC) of the C-peptide values. The natural log(), log(+1) and square-root transformations of the AUC were assessed. In general, a transformation of the data is needed to better satisfy the normality assumptions for commonly used statistical tests. Statistical analysis of the raw and transformed data are provided to estimate the mean levels over time and the residual variation in untreated subjects that allow sample size calculations for future studies at either 12 or 24 months of follow-up and among children 8–12 years of age, adolescents (13–17 years) and adults (18+ years). The sample size needed to detect a given relative (percentage) difference with treatment versus control is greater at 24 months than at 12 months of follow-up, and differs among age categories. Owing to greater residual variation among those 13–17 years of age, a larger sample size is required for this age group. Methods are also described for assessment of sample size for mixtures of subjects among the age categories. Statistical expressions are presented for the presentation of analyses of log(+1) and transformed values in terms of the original units of measurement (pmol/ml). Analyses using different transformations are described for the TrialNet study of masked anti-CD20 (rituximab) versus masked placebo. These results provide the information needed to accurately evaluate the sample size for studies of new agents to preserve C-peptide levels in newly diagnosed type 1 diabetes. PMID:22102862

Sample size requirements for studies of treatment effects on beta-cell function in newly diagnosed type 1 diabetes.

PubMed

Lachin, John M; McGee, Paula L; Greenbaum, Carla J; Palmer, Jerry; Pescovitz, Mark D; Gottlieb, Peter; Skyler, Jay

2011-01-01

Preservation of β-cell function as measured by stimulated C-peptide has recently been accepted as a therapeutic target for subjects with newly diagnosed type 1 diabetes. In recently completed studies conducted by the Type 1 Diabetes Trial Network (TrialNet), repeated 2-hour Mixed Meal Tolerance Tests (MMTT) were obtained for up to 24 months from 156 subjects with up to 3 months duration of type 1 diabetes at the time of study enrollment. These data provide the information needed to more accurately determine the sample size needed for future studies of the effects of new agents on the 2-hour area under the curve (AUC) of the C-peptide values. The natural log(x), log(x+1) and square-root (√x) transformations of the AUC were assessed. In general, a transformation of the data is needed to better satisfy the normality assumptions for commonly used statistical tests. Statistical analysis of the raw and transformed data are provided to estimate the mean levels over time and the residual variation in untreated subjects that allow sample size calculations for future studies at either 12 or 24 months of follow-up and among children 8-12 years of age, adolescents (13-17 years) and adults (18+ years). The sample size needed to detect a given relative (percentage) difference with treatment versus control is greater at 24 months than at 12 months of follow-up, and differs among age categories. Owing to greater residual variation among those 13-17 years of age, a larger sample size is required for this age group. Methods are also described for assessment of sample size for mixtures of subjects among the age categories. Statistical expressions are presented for the presentation of analyses of log(x+1) and √x transformed values in terms of the original units of measurement (pmol/ml). Analyses using different transformations are described for the TrialNet study of masked anti-CD20 (rituximab) versus masked placebo. These results provide the information needed to accurately evaluate the sample size for studies of new agents to preserve C-peptide levels in newly diagnosed type 1 diabetes.

Power of tests for comparing trend curves with application to national immunization survey (NIS).

PubMed

Zhao, Zhen

2011-02-28

To develop statistical tests for comparing trend curves of study outcomes between two socio-demographic strata across consecutive time points, and compare statistical power of the proposed tests under different trend curves data, three statistical tests were proposed. For large sample size with independent normal assumption among strata and across consecutive time points, the Z and Chi-square test statistics were developed, which are functions of outcome estimates and the standard errors at each of the study time points for the two strata. For small sample size with independent normal assumption, the F-test statistic was generated, which is a function of sample size of the two strata and estimated parameters across study period. If two trend curves are approximately parallel, the power of Z-test is consistently higher than that of both Chi-square and F-test. If two trend curves cross at low interaction, the power of Z-test is higher than or equal to the power of both Chi-square and F-test; however, at high interaction, the powers of Chi-square and F-test are higher than that of Z-test. The measurement of interaction of two trend curves was defined. These tests were applied to the comparison of trend curves of vaccination coverage estimates of standard vaccine series with National Immunization Survey (NIS) 2000-2007 data. Copyright © 2011 John Wiley & Sons, Ltd.

Confidence crisis of results in biomechanics research.

PubMed

Knudson, Duane

2017-11-01

Many biomechanics studies have small sample sizes and incorrect statistical analyses, so reporting of inaccurate inferences and inflated magnitude of effects are common in the field. This review examines these issues in biomechanics research and summarises potential solutions from research in other fields to increase the confidence in the experimental effects reported in biomechanics. Authors, reviewers and editors of biomechanics research reports are encouraged to improve sample sizes and the resulting statistical power, improve reporting transparency, improve the rigour of statistical analyses used, and increase the acceptance of replication studies to improve the validity of inferences from data in biomechanics research. The application of sports biomechanics research results would also improve if a larger percentage of unbiased effects and their uncertainty were reported in the literature.

Inferring Population Size History from Large Samples of Genome-Wide Molecular Data - An Approximate Bayesian Computation Approach

PubMed Central

Boitard, Simon; Rodríguez, Willy; Jay, Flora; Mona, Stefano; Austerlitz, Frédéric

2016-01-01

Inferring the ancestral dynamics of effective population size is a long-standing question in population genetics, which can now be tackled much more accurately thanks to the massive genomic data available in many species. Several promising methods that take advantage of whole-genome sequences have been recently developed in this context. However, they can only be applied to rather small samples, which limits their ability to estimate recent population size history. Besides, they can be very sensitive to sequencing or phasing errors. Here we introduce a new approximate Bayesian computation approach named PopSizeABC that allows estimating the evolution of the effective population size through time, using a large sample of complete genomes. This sample is summarized using the folded allele frequency spectrum and the average zygotic linkage disequilibrium at different bins of physical distance, two classes of statistics that are widely used in population genetics and can be easily computed from unphased and unpolarized SNP data. Our approach provides accurate estimations of past population sizes, from the very first generations before present back to the expected time to the most recent common ancestor of the sample, as shown by simulations under a wide range of demographic scenarios. When applied to samples of 15 or 25 complete genomes in four cattle breeds (Angus, Fleckvieh, Holstein and Jersey), PopSizeABC revealed a series of population declines, related to historical events such as domestication or modern breed creation. We further highlight that our approach is robust to sequencing errors, provided summary statistics are computed from SNPs with common alleles. PMID:26943927

Performance of Bootstrapping Approaches To Model Test Statistics and Parameter Standard Error Estimation in Structural Equation Modeling.

ERIC Educational Resources Information Center

Nevitt, Jonathan; Hancock, Gregory R.

2001-01-01

Evaluated the bootstrap method under varying conditions of nonnormality, sample size, model specification, and number of bootstrap samples drawn from the resampling space. Results for the bootstrap suggest the resampling-based method may be conservative in its control over model rejections, thus having an impact on the statistical power associated…

Only pick the right grains: Modelling the bias due to subjective grain-size interval selection for chronometric and fingerprinting approaches.

NASA Astrophysics Data System (ADS)

Dietze, Michael; Fuchs, Margret; Kreutzer, Sebastian

2016-04-01

Many modern approaches of radiometric dating or geochemical fingerprinting rely on sampling sedimentary deposits. A key assumption of most concepts is that the extracted grain-size fraction of the sampled sediment adequately represents the actual process to be dated or the source area to be fingerprinted. However, these assumptions are not always well constrained. Rather, they have to align with arbitrary, method-determined size intervals, such as "coarse grain" or "fine grain" with partly even different definitions. Such arbitrary intervals violate principal process-based concepts of sediment transport and can thus introduce significant bias to the analysis outcome (i.e., a deviation of the measured from the true value). We present a flexible numerical framework (numOlum) for the statistical programming language R that allows quantifying the bias due to any given analysis size interval for different types of sediment deposits. This framework is applied to synthetic samples from the realms of luminescence dating and geochemical fingerprinting, i.e. a virtual reworked loess section. We show independent validation data from artificially dosed and subsequently mixed grain-size proportions and we present a statistical approach (end-member modelling analysis, EMMA) that allows accounting for the effect of measuring the compound dosimetric history or geochemical composition of a sample. EMMA separates polymodal grain-size distributions into the underlying transport process-related distributions and their contribution to each sample. These underlying distributions can then be used to adjust grain-size preparation intervals to minimise the incorporation of "undesired" grain-size fractions.

Biostatistics Series Module 5: Determining Sample Size

PubMed Central

Hazra, Avijit; Gogtay, Nithya

2016-01-01

Determining the appropriate sample size for a study, whatever be its type, is a fundamental aspect of biomedical research. An adequate sample ensures that the study will yield reliable information, regardless of whether the data ultimately suggests a clinically important difference between the interventions or elements being studied. The probability of Type 1 and Type 2 errors, the expected variance in the sample and the effect size are the essential determinants of sample size in interventional studies. Any method for deriving a conclusion from experimental data carries with it some risk of drawing a false conclusion. Two types of false conclusion may occur, called Type 1 and Type 2 errors, whose probabilities are denoted by the symbols σ and β. A Type 1 error occurs when one concludes that a difference exists between the groups being compared when, in reality, it does not. This is akin to a false positive result. A Type 2 error occurs when one concludes that difference does not exist when, in reality, a difference does exist, and it is equal to or larger than the effect size defined by the alternative to the null hypothesis. This may be viewed as a false negative result. When considering the risk of Type 2 error, it is more intuitive to think in terms of power of the study or (1 − β). Power denotes the probability of detecting a difference when a difference does exist between the groups being compared. Smaller α or larger power will increase sample size. Conventional acceptable values for power and α are 80% or above and 5% or below, respectively, when calculating sample size. Increasing variance in the sample tends to increase the sample size required to achieve a given power level. The effect size is the smallest clinically important difference that is sought to be detected and, rather than statistical convention, is a matter of past experience and clinical judgment. Larger samples are required if smaller differences are to be detected. Although the principles are long known, historically, sample size determination has been difficult, because of relatively complex mathematical considerations and numerous different formulas. However, of late, there has been remarkable improvement in the availability, capability, and user-friendliness of power and sample size determination software. Many can execute routines for determination of sample size and power for a wide variety of research designs and statistical tests. With the drudgery of mathematical calculation gone, researchers must now concentrate on determining appropriate sample size and achieving these targets, so that study conclusions can be accepted as meaningful. PMID:27688437

Testing homogeneity of proportion ratios for stratified correlated bilateral data in two-arm randomized clinical trials.

PubMed

Pei, Yanbo; Tian, Guo-Liang; Tang, Man-Lai

2014-11-10

Stratified data analysis is an important research topic in many biomedical studies and clinical trials. In this article, we develop five test statistics for testing the homogeneity of proportion ratios for stratified correlated bilateral binary data based on an equal correlation model assumption. Bootstrap procedures based on these test statistics are also considered. To evaluate the performance of these statistics and procedures, we conduct Monte Carlo simulations to study their empirical sizes and powers under various scenarios. Our results suggest that the procedure based on score statistic performs well generally and is highly recommended. When the sample size is large, procedures based on the commonly used weighted least square estimate and logarithmic transformation with Mantel-Haenszel estimate are recommended as they do not involve any computation of maximum likelihood estimates requiring iterative algorithms. We also derive approximate sample size formulas based on the recommended test procedures. Finally, we apply the proposed methods to analyze a multi-center randomized clinical trial for scleroderma patients. Copyright © 2014 John Wiley & Sons, Ltd.

Determining sample size for tree utilization surveys

Treesearch

Stanley J. Zarnoch; James W. Bentley; Tony G. Johnson

2004-01-01

The U.S. Department of Agriculture Forest Service has conducted many studies to determine what proportion of the timber harvested in the South is actually utilized. This paper describes the statistical methods used to determine required sample sizes for estimating utilization ratios for a required level of precision. The data used are those for 515 hardwood and 1,557...

A New Sample Size Formula for Regression.

ERIC Educational Resources Information Center

Brooks, Gordon P.; Barcikowski, Robert S.

The focus of this research was to determine the efficacy of a new method of selecting sample sizes for multiple linear regression. A Monte Carlo simulation was used to study both empirical predictive power rates and empirical statistical power rates of the new method and seven other methods: those of C. N. Park and A. L. Dudycha (1974); J. Cohen…

Using Sieving and Unknown Sand Samples for a Sedimentation-Stratigraphy Class Project with Linkage to Introductory Courses

ERIC Educational Resources Information Center

Videtich, Patricia E.; Neal, William J.

2012-01-01

Using sieving and sample "unknowns" for instructional grain-size analysis and interpretation of sands in undergraduate sedimentology courses has advantages over other techniques. Students (1) learn to calculate and use statistics; (2) visually observe differences in the grain-size fractions, thereby developing a sense of specific size…

A note on sample size calculation for mean comparisons based on noncentral t-statistics.

PubMed

Chow, Shein-Chung; Shao, Jun; Wang, Hansheng

2002-11-01

One-sample and two-sample t-tests are commonly used in analyzing data from clinical trials in comparing mean responses from two drug products. During the planning stage of a clinical study, a crucial step is the sample size calculation, i.e., the determination of the number of subjects (patients) needed to achieve a desired power (e.g., 80%) for detecting a clinically meaningful difference in the mean drug responses. Based on noncentral t-distributions, we derive some sample size calculation formulas for testing equality, testing therapeutic noninferiority/superiority, and testing therapeutic equivalence, under the popular one-sample design, two-sample parallel design, and two-sample crossover design. Useful tables are constructed and some examples are given for illustration.

Statistical characterization of a large geochemical database and effect of sample size

USGS Publications Warehouse

Zhang, C.; Manheim, F.T.; Hinde, J.; Grossman, J.N.

2005-01-01

The authors investigated statistical distributions for concentrations of chemical elements from the National Geochemical Survey (NGS) database of the U.S. Geological Survey. At the time of this study, the NGS data set encompasses 48,544 stream sediment and soil samples from the conterminous United States analyzed by ICP-AES following a 4-acid near-total digestion. This report includes 27 elements: Al, Ca, Fe, K, Mg, Na, P, Ti, Ba, Ce, Co, Cr, Cu, Ga, La, Li, Mn, Nb, Nd, Ni, Pb, Sc, Sr, Th, V, Y and Zn. The goal and challenge for the statistical overview was to delineate chemical distributions in a complex, heterogeneous data set spanning a large geographic range (the conterminous United States), and many different geological provinces and rock types. After declustering to create a uniform spatial sample distribution with 16,511 samples, histograms and quantile-quantile (Q-Q) plots were employed to delineate subpopulations that have coherent chemical and mineral affinities. Probability groupings are discerned by changes in slope (kinks) on the plots. Major rock-forming elements, e.g., Al, Ca, K and Na, tend to display linear segments on normal Q-Q plots. These segments can commonly be linked to petrologic or mineralogical associations. For example, linear segments on K and Na plots reflect dilution of clay minerals by quartz sand (low in K and Na). Minor and trace element relationships are best displayed on lognormal Q-Q plots. These sensitively reflect discrete relationships in subpopulations within the wide range of the data. For example, small but distinctly log-linear subpopulations for Pb, Cu, Zn and Ag are interpreted to represent ore-grade enrichment of naturally occurring minerals such as sulfides. None of the 27 chemical elements could pass the test for either normal or lognormal distribution on the declustered data set. Part of the reasons relate to the presence of mixtures of subpopulations and outliers. Random samples of the data set with successively smaller numbers of data points showed that few elements passed standard statistical tests for normality or log-normality until sample size decreased to a few hundred data points. Large sample size enhances the power of statistical tests, and leads to rejection of most statistical hypotheses for real data sets. For large sample sizes (e.g., n > 1000), graphical methods such as histogram, stem-and-leaf, and probability plots are recommended for rough judgement of probability distribution if needed. ?? 2005 Elsevier Ltd. All rights reserved.

A knowledge-based T2-statistic to perform pathway analysis for quantitative proteomic data

PubMed Central

Chen, Yi-Hau

2017-01-01

Approaches to identify significant pathways from high-throughput quantitative data have been developed in recent years. Still, the analysis of proteomic data stays difficult because of limited sample size. This limitation also leads to the practice of using a competitive null as common approach; which fundamentally implies genes or proteins as independent units. The independent assumption ignores the associations among biomolecules with similar functions or cellular localization, as well as the interactions among them manifested as changes in expression ratios. Consequently, these methods often underestimate the associations among biomolecules and cause false positives in practice. Some studies incorporate the sample covariance matrix into the calculation to address this issue. However, sample covariance may not be a precise estimation if the sample size is very limited, which is usually the case for the data produced by mass spectrometry. In this study, we introduce a multivariate test under a self-contained null to perform pathway analysis for quantitative proteomic data. The covariance matrix used in the test statistic is constructed by the confidence scores retrieved from the STRING database or the HitPredict database. We also design an integrating procedure to retain pathways of sufficient evidence as a pathway group. The performance of the proposed T2-statistic is demonstrated using five published experimental datasets: the T-cell activation, the cAMP/PKA signaling, the myoblast differentiation, and the effect of dasatinib on the BCR-ABL pathway are proteomic datasets produced by mass spectrometry; and the protective effect of myocilin via the MAPK signaling pathway is a gene expression dataset of limited sample size. Compared with other popular statistics, the proposed T2-statistic yields more accurate descriptions in agreement with the discussion of the original publication. We implemented the T2-statistic into an R package T2GA, which is available at https://github.com/roqe/T2GA. PMID:28622336

A knowledge-based T2-statistic to perform pathway analysis for quantitative proteomic data.

PubMed

Lai, En-Yu; Chen, Yi-Hau; Wu, Kun-Pin

2017-06-01

Approaches to identify significant pathways from high-throughput quantitative data have been developed in recent years. Still, the analysis of proteomic data stays difficult because of limited sample size. This limitation also leads to the practice of using a competitive null as common approach; which fundamentally implies genes or proteins as independent units. The independent assumption ignores the associations among biomolecules with similar functions or cellular localization, as well as the interactions among them manifested as changes in expression ratios. Consequently, these methods often underestimate the associations among biomolecules and cause false positives in practice. Some studies incorporate the sample covariance matrix into the calculation to address this issue. However, sample covariance may not be a precise estimation if the sample size is very limited, which is usually the case for the data produced by mass spectrometry. In this study, we introduce a multivariate test under a self-contained null to perform pathway analysis for quantitative proteomic data. The covariance matrix used in the test statistic is constructed by the confidence scores retrieved from the STRING database or the HitPredict database. We also design an integrating procedure to retain pathways of sufficient evidence as a pathway group. The performance of the proposed T2-statistic is demonstrated using five published experimental datasets: the T-cell activation, the cAMP/PKA signaling, the myoblast differentiation, and the effect of dasatinib on the BCR-ABL pathway are proteomic datasets produced by mass spectrometry; and the protective effect of myocilin via the MAPK signaling pathway is a gene expression dataset of limited sample size. Compared with other popular statistics, the proposed T2-statistic yields more accurate descriptions in agreement with the discussion of the original publication. We implemented the T2-statistic into an R package T2GA, which is available at https://github.com/roqe/T2GA.

What is a species? A new universal method to measure differentiation and assess the taxonomic rank of allopatric populations, using continuous variables

PubMed Central

Donegan, Thomas M.

2018-01-01

Abstract Existing models for assigning species, subspecies, or no taxonomic rank to populations which are geographically separated from one another were analyzed. This was done by subjecting over 3,000 pairwise comparisons of vocal or biometric data based on birds to a variety of statistical tests that have been proposed as measures of differentiation. One current model which aims to test diagnosability (Isler et al. 1998) is highly conservative, applying a hard cut-off, which excludes from consideration differentiation below diagnosis. It also includes non-overlap as a requirement, a measure which penalizes increases to sample size. The “species scoring” model of Tobias et al. (2010) involves less drastic cut-offs, but unlike Isler et al. (1998), does not control adequately for sample size and attributes scores in many cases to differentiation which is not statistically significant. Four different models of assessing effect sizes were analyzed: using both pooled and unpooled standard deviations and controlling for sample size using t-distributions or omitting to do so. Pooled standard deviations produced more conservative effect sizes when uncontrolled for sample size but less conservative effect sizes when so controlled. Pooled models require assumptions to be made that are typically elusive or unsupported for taxonomic studies. Modifications to improving these frameworks are proposed, including: (i) introducing statistical significance as a gateway to attributing any weighting to findings of differentiation; (ii) abandoning non-overlap as a test; (iii) recalibrating Tobias et al. (2010) scores based on effect sizes controlled for sample size using t-distributions. A new universal method is proposed for measuring differentiation in taxonomy using continuous variables and a formula is proposed for ranking allopatric populations. This is based first on calculating effect sizes using unpooled standard deviations, controlled for sample size using t-distributions, for a series of different variables. All non-significant results are excluded by scoring them as zero. Distance between any two populations is calculated using Euclidian summation of non-zeroed effect size scores. If the score of an allopatric pair exceeds that of a related sympatric pair, then the allopatric population can be ranked as species and, if not, then at most subspecies rank should be assigned. A spreadsheet has been programmed and is being made available which allows this and other tests of differentiation and rank studied in this paper to be rapidly analyzed. PMID:29780266

Sample size requirements for estimating effective dose from computed tomography using solid-state metal-oxide-semiconductor field-effect transistor dosimetry

PubMed Central

Trattner, Sigal; Cheng, Bin; Pieniazek, Radoslaw L.; Hoffmann, Udo; Douglas, Pamela S.; Einstein, Andrew J.

2014-01-01

Purpose: Effective dose (ED) is a widely used metric for comparing ionizing radiation burden between different imaging modalities, scanners, and scan protocols. In computed tomography (CT), ED can be estimated by performing scans on an anthropomorphic phantom in which metal-oxide-semiconductor field-effect transistor (MOSFET) solid-state dosimeters have been placed to enable organ dose measurements. Here a statistical framework is established to determine the sample size (number of scans) needed for estimating ED to a desired precision and confidence, for a particular scanner and scan protocol, subject to practical limitations. Methods: The statistical scheme involves solving equations which minimize the sample size required for estimating ED to desired precision and confidence. It is subject to a constrained variation of the estimated ED and solved using the Lagrange multiplier method. The scheme incorporates measurement variation introduced both by MOSFET calibration, and by variation in MOSFET readings between repeated CT scans. Sample size requirements are illustrated on cardiac, chest, and abdomen–pelvis CT scans performed on a 320-row scanner and chest CT performed on a 16-row scanner. Results: Sample sizes for estimating ED vary considerably between scanners and protocols. Sample size increases as the required precision or confidence is higher and also as the anticipated ED is lower. For example, for a helical chest protocol, for 95% confidence and 5% precision for the ED, 30 measurements are required on the 320-row scanner and 11 on the 16-row scanner when the anticipated ED is 4 mSv; these sample sizes are 5 and 2, respectively, when the anticipated ED is 10 mSv. Conclusions: Applying the suggested scheme, it was found that even at modest sample sizes, it is feasible to estimate ED with high precision and a high degree of confidence. As CT technology develops enabling ED to be lowered, more MOSFET measurements are needed to estimate ED with the same precision and confidence. PMID:24694150

Methodological issues with adaptation of clinical trial design.

PubMed

Hung, H M James; Wang, Sue-Jane; O'Neill, Robert T

2006-01-01

Adaptation of clinical trial design generates many issues that have not been resolved for practical applications, though statistical methodology has advanced greatly. This paper focuses on some methodological issues. In one type of adaptation such as sample size re-estimation, only the postulated value of a parameter for planning the trial size may be altered. In another type, the originally intended hypothesis for testing may be modified using the internal data accumulated at an interim time of the trial, such as changing the primary endpoint and dropping a treatment arm. For sample size re-estimation, we make a contrast between an adaptive test weighting the two-stage test statistics with the statistical information given by the original design and the original sample mean test with a properly corrected critical value. We point out the difficulty in planning a confirmatory trial based on the crude information generated by exploratory trials. In regards to selecting a primary endpoint, we argue that the selection process that allows switching from one endpoint to the other with the internal data of the trial is not very likely to gain a power advantage over the simple process of selecting one from the two endpoints by testing them with an equal split of alpha (Bonferroni adjustment). For dropping a treatment arm, distributing the remaining sample size of the discontinued arm to other treatment arms can substantially improve the statistical power of identifying a superior treatment arm in the design. A common difficult methodological issue is that of how to select an adaptation rule in the trial planning stage. Pre-specification of the adaptation rule is important for the practicality consideration. Changing the originally intended hypothesis for testing with the internal data generates great concerns to clinical trial researchers.

Relation Between Intelligence and Family Size, Position, and Income in Adolescent Girls in Saudi Arabia.

PubMed

Osman, Habab; Alahmadi, Maryam; Bakhiet, Salaheldin; Lynn, Richard

2016-12-01

Data are reported showing statistically significant negative correlations between intelligence and family size, position, and income in a sample of 604 adolescent girls in Saudi Arabia. There were no statistically significant correlations or associations between whether the mother or father were deceased or both parents were alive, and whether the parents were living together or were divorced. © The Author(s) 2016.

The relation between statistical power and inference in fMRI

PubMed Central

Wager, Tor D.; Yarkoni, Tal

2017-01-01

Statistically underpowered studies can result in experimental failure even when all other experimental considerations have been addressed impeccably. In fMRI the combination of a large number of dependent variables, a relatively small number of observations (subjects), and a need to correct for multiple comparisons can decrease statistical power dramatically. This problem has been clearly addressed yet remains controversial—especially in regards to the expected effect sizes in fMRI, and especially for between-subjects effects such as group comparisons and brain-behavior correlations. We aimed to clarify the power problem by considering and contrasting two simulated scenarios of such possible brain-behavior correlations: weak diffuse effects and strong localized effects. Sampling from these scenarios shows that, particularly in the weak diffuse scenario, common sample sizes (n = 20–30) display extremely low statistical power, poorly represent the actual effects in the full sample, and show large variation on subsequent replications. Empirical data from the Human Connectome Project resembles the weak diffuse scenario much more than the localized strong scenario, which underscores the extent of the power problem for many studies. Possible solutions to the power problem include increasing the sample size, using less stringent thresholds, or focusing on a region-of-interest. However, these approaches are not always feasible and some have major drawbacks. The most prominent solutions that may help address the power problem include model-based (multivariate) prediction methods and meta-analyses with related synthesis-oriented approaches. PMID:29155843

Statistical power analysis in wildlife research

USGS Publications Warehouse

Steidl, R.J.; Hayes, J.P.

1997-01-01

Statistical power analysis can be used to increase the efficiency of research efforts and to clarify research results. Power analysis is most valuable in the design or planning phases of research efforts. Such prospective (a priori) power analyses can be used to guide research design and to estimate the number of samples necessary to achieve a high probability of detecting biologically significant effects. Retrospective (a posteriori) power analysis has been advocated as a method to increase information about hypothesis tests that were not rejected. However, estimating power for tests of null hypotheses that were not rejected with the effect size observed in the study is incorrect; these power estimates will always be a??0.50 when bias adjusted and have no relation to true power. Therefore, retrospective power estimates based on the observed effect size for hypothesis tests that were not rejected are misleading; retrospective power estimates are only meaningful when based on effect sizes other than the observed effect size, such as those effect sizes hypothesized to be biologically significant. Retrospective power analysis can be used effectively to estimate the number of samples or effect size that would have been necessary for a completed study to have rejected a specific null hypothesis. Simply presenting confidence intervals can provide additional information about null hypotheses that were not rejected, including information about the size of the true effect and whether or not there is adequate evidence to 'accept' a null hypothesis as true. We suggest that (1) statistical power analyses be routinely incorporated into research planning efforts to increase their efficiency, (2) confidence intervals be used in lieu of retrospective power analyses for null hypotheses that were not rejected to assess the likely size of the true effect, (3) minimum biologically significant effect sizes be used for all power analyses, and (4) if retrospective power estimates are to be reported, then the I?-level, effect sizes, and sample sizes used in calculations must also be reported.

Knowledge level of effect size statistics, confidence intervals and meta-analysis in Spanish academic psychologists.

PubMed

Badenes-Ribera, Laura; Frias-Navarro, Dolores; Pascual-Soler, Marcos; Monterde-I-Bort, Héctor

2016-11-01

The statistical reform movement and the American Psychological Association (APA) defend the use of estimators of the effect size and its confidence intervals, as well as the interpretation of the clinical significance of the findings. A survey was conducted in which academic psychologists were asked about their behavior in designing and carrying out their studies. The sample was composed of 472 participants (45.8% men). The mean number of years as a university professor was 13.56 years (SD= 9.27). The use of effect-size estimators is becoming generalized, as well as the consideration of meta-analytic studies. However, several inadequate practices still persist. A traditional model of methodological behavior based on statistical significance tests is maintained, based on the predominance of Cohen’s d and the unadjusted R2/η2, which are not immune to outliers or departure from normality and the violations of statistical assumptions, and the under-reporting of confidence intervals of effect-size statistics. The paper concludes with recommendations for improving statistical practice.

SEDPAK—A comprehensive operational system and data-processing package in APPLESOFT BASIC for a settling tube, sediment analyzer

NASA Astrophysics Data System (ADS)

Goldbery, R.; Tehori, O.

SEDPAK provides a comprehensive software package for operation of a settling tube and sand analyzer (2-0.063 mm) and includes data-processing programs for statistical and graphic output of results. The programs are menu-driven and written in APPLESOFT BASIC, conforming with APPLE 3.3 DOS. Data storage and retrieval from disc is an important feature of SEDPAK. Additional features of SEDPAK include condensation of raw settling data via standard size-calibration curves to yield statistical grain-size parameters, plots of grain-size frequency distributions and cumulative log/probability curves. The program also has a module for processing of grain-size frequency data from sieved samples. An addition feature of SEDPAK is the option for automatic data processing and graphic output of a sequential or nonsequential array of samples on one side of a disc.

Samples in applied psychology: over a decade of research in review.

PubMed

Shen, Winny; Kiger, Thomas B; Davies, Stacy E; Rasch, Rena L; Simon, Kara M; Ones, Deniz S

2011-09-01

This study examines sample characteristics of articles published in Journal of Applied Psychology (JAP) from 1995 to 2008. At the individual level, the overall median sample size over the period examined was approximately 173, which is generally adequate for detecting the average magnitude of effects of primary interest to researchers who publish in JAP. Samples using higher units of analyses (e.g., teams, departments/work units, and organizations) had lower median sample sizes (Mdn ≈ 65), yet were arguably robust given typical multilevel design choices of JAP authors despite the practical constraints of collecting data at higher units of analysis. A substantial proportion of studies used student samples (~40%); surprisingly, median sample sizes for student samples were smaller than working adult samples. Samples were more commonly occupationally homogeneous (~70%) than occupationally heterogeneous. U.S. and English-speaking participants made up the vast majority of samples, whereas Middle Eastern, African, and Latin American samples were largely unrepresented. On the basis of study results, recommendations are provided for authors, editors, and readers, which converge on 3 themes: (a) appropriateness and match between sample characteristics and research questions, (b) careful consideration of statistical power, and (c) the increased popularity of quantitative synthesis. Implications are discussed in terms of theory building, generalizability of research findings, and statistical power to detect effects. PsycINFO Database Record (c) 2011 APA, all rights reserved

Chance-corrected classification for use in discriminant analysis: Ecological applications

USGS Publications Warehouse

Titus, K.; Mosher, J.A.; Williams, B.K.

1984-01-01

A method for evaluating the classification table from a discriminant analysis is described. The statistic, kappa, is useful to ecologists in that it removes the effects of chance. It is useful even with equal group sample sizes although the need for a chance-corrected measure of prediction becomes greater with more dissimilar group sample sizes. Examples are presented.

Got Power? A Systematic Review of Sample Size Adequacy in Health Professions Education Research

ERIC Educational Resources Information Center

Cook, David A.; Hatala, Rose

2015-01-01

Many education research studies employ small samples, which in turn lowers statistical power. We re-analyzed the results of a meta-analysis of simulation-based education to determine study power across a range of effect sizes, and the smallest effect that could be plausibly excluded. We systematically searched multiple databases through May 2011,…

A visual basic program to generate sediment grain-size statistics and to extrapolate particle distributions

USGS Publications Warehouse

Poppe, L.J.; Eliason, A.H.; Hastings, M.E.

2004-01-01

Measures that describe and summarize sediment grain-size distributions are important to geologists because of the large amount of information contained in textural data sets. Statistical methods are usually employed to simplify the necessary comparisons among samples and quantify the observed differences. The two statistical methods most commonly used by sedimentologists to describe particle distributions are mathematical moments (Krumbein and Pettijohn, 1938) and inclusive graphics (Folk, 1974). The choice of which of these statistical measures to use is typically governed by the amount of data available (Royse, 1970). If the entire distribution is known, the method of moments may be used; if the next to last accumulated percent is greater than 95, inclusive graphics statistics can be generated. Unfortunately, earlier programs designed to describe sediment grain-size distributions statistically do not run in a Windows environment, do not allow extrapolation of the distribution's tails, or do not generate both moment and graphic statistics (Kane and Hubert, 1963; Collias et al., 1963; Schlee and Webster, 1967; Poppe et al., 2000)1.Owing to analytical limitations, electro-resistance multichannel particle-size analyzers, such as Coulter Counters, commonly truncate the tails of the fine-fraction part of grain-size distributions. These devices do not detect fine clay in the 0.6–0.1 μm range (part of the 11-phi and all of the 12-phi and 13-phi fractions). Although size analyses performed down to 0.6 μm microns are adequate for most freshwater and near shore marine sediments, samples from many deeper water marine environments (e.g. rise and abyssal plain) may contain significant material in the fine clay fraction, and these analyses benefit from extrapolation.The program (GSSTAT) described herein generates statistics to characterize sediment grain-size distributions and can extrapolate the fine-grained end of the particle distribution. It is written in Microsoft Visual Basic 6.0 and provides a window to facilitate program execution. The input for the sediment fractions is weight percentages in whole-phi notation (Krumbein, 1934; Inman, 1952), and the program permits the user to select output in either method of moments or inclusive graphics statistics (Fig. 1). Users select options primarily with mouse-click events, or through interactive dialogue boxes.

Fast and accurate imputation of summary statistics enhances evidence of functional enrichment

PubMed Central

Pasaniuc, Bogdan; Zaitlen, Noah; Shi, Huwenbo; Bhatia, Gaurav; Gusev, Alexander; Pickrell, Joseph; Hirschhorn, Joel; Strachan, David P.; Patterson, Nick; Price, Alkes L.

2014-01-01

Motivation: Imputation using external reference panels (e.g. 1000 Genomes) is a widely used approach for increasing power in genome-wide association studies and meta-analysis. Existing hidden Markov models (HMM)-based imputation approaches require individual-level genotypes. Here, we develop a new method for Gaussian imputation from summary association statistics, a type of data that is becoming widely available. Results: In simulations using 1000 Genomes (1000G) data, this method recovers 84% (54%) of the effective sample size for common (>5%) and low-frequency (1–5%) variants [increasing to 87% (60%) when summary linkage disequilibrium information is available from target samples] versus the gold standard of 89% (67%) for HMM-based imputation, which cannot be applied to summary statistics. Our approach accounts for the limited sample size of the reference panel, a crucial step to eliminate false-positive associations, and it is computationally very fast. As an empirical demonstration, we apply our method to seven case–control phenotypes from the Wellcome Trust Case Control Consortium (WTCCC) data and a study of height in the British 1958 birth cohort (1958BC). Gaussian imputation from summary statistics recovers 95% (105%) of the effective sample size (as quantified by the ratio of χ2 association statistics) compared with HMM-based imputation from individual-level genotypes at the 227 (176) published single nucleotide polymorphisms (SNPs) in the WTCCC (1958BC height) data. In addition, for publicly available summary statistics from large meta-analyses of four lipid traits, we publicly release imputed summary statistics at 1000G SNPs, which could not have been obtained using previously published methods, and demonstrate their accuracy by masking subsets of the data. We show that 1000G imputation using our approach increases the magnitude and statistical evidence of enrichment at genic versus non-genic loci for these traits, as compared with an analysis without 1000G imputation. Thus, imputation of summary statistics will be a valuable tool in future functional enrichment analyses. Availability and implementation: Publicly available software package available at http://bogdan.bioinformatics.ucla.edu/software/. Contact: bpasaniuc@mednet.ucla.edu or aprice@hsph.harvard.edu Supplementary information: Supplementary materials are available at Bioinformatics online. PMID:24990607

Statistical inference for tumor growth inhibition T/C ratio.

PubMed

Wu, Jianrong

2010-09-01

The tumor growth inhibition T/C ratio is commonly used to quantify treatment effects in drug screening tumor xenograft experiments. The T/C ratio is converted to an antitumor activity rating using an arbitrary cutoff point and often without any formal statistical inference. Here, we applied a nonparametric bootstrap method and a small sample likelihood ratio statistic to make a statistical inference of the T/C ratio, including both hypothesis testing and a confidence interval estimate. Furthermore, sample size and power are also discussed for statistical design of tumor xenograft experiments. Tumor xenograft data from an actual experiment were analyzed to illustrate the application.

Particle size analysis of sediments, soils and related particulate materials for forensic purposes using laser granulometry.

PubMed

Pye, Kenneth; Blott, Simon J

2004-08-11

Particle size is a fundamental property of any sediment, soil or dust deposit which can provide important clues to nature and provenance. For forensic work, the particle size distribution of sometimes very small samples requires precise determination using a rapid and reliable method with a high resolution. The Coulter trade mark LS230 laser granulometer offers rapid and accurate sizing of particles in the range 0.04-2000 microm for a variety of sample types, including soils, unconsolidated sediments, dusts, powders and other particulate materials. Reliable results are possible for sample weights of just 50 mg. Discrimination between samples is performed on the basis of the shape of the particle size curves and statistical measures of the size distributions. In routine forensic work laser granulometry data can rarely be used in isolation and should be considered in combination with results from other techniques to reach an overall conclusion.

Dispersion and sampling of adult Dermacentor andersoni in rangeland in Western North America.

PubMed

Rochon, K; Scoles, G A; Lysyk, T J

2012-03-01

A fixed precision sampling plan was developed for off-host populations of adult Rocky Mountain wood tick, Dermacentor andersoni (Stiles) based on data collected by dragging at 13 locations in Alberta, Canada; Washington; and Oregon. In total, 222 site-date combinations were sampled. Each site-date combination was considered a sample, and each sample ranged in size from 86 to 250 10 m2 quadrats. Analysis of simulated quadrats ranging in size from 10 to 50 m2 indicated that the most precise sample unit was the 10 m2 quadrat. Samples taken when abundance < 0.04 ticks per 10 m2 were more likely to not depart significantly from statistical randomness than samples taken when abundance was greater. Data were grouped into ten abundance classes and assessed for fit to the Poisson and negative binomial distributions. The Poisson distribution fit only data in abundance classes < 0.02 ticks per 10 m2, while the negative binomial distribution fit data from all abundance classes. A negative binomial distribution with common k = 0.3742 fit data in eight of the 10 abundance classes. Both the Taylor and Iwao mean-variance relationships were fit and used to predict sample sizes for a fixed level of precision. Sample sizes predicted using the Taylor model tended to underestimate actual sample sizes, while sample sizes estimated using the Iwao model tended to overestimate actual sample sizes. Using a negative binomial with common k provided estimates of required sample sizes closest to empirically calculated sample sizes.

Using Candy Samples to Learn about Sampling Techniques and Statistical Data Evaluation

ERIC Educational Resources Information Center

Canaes, Larissa S.; Brancalion, Marcel L.; Rossi, Adriana V.; Rath, Susanne

2008-01-01

A classroom exercise for undergraduate and beginning graduate students that takes about one class period is proposed and discussed. It is an easy, interesting exercise that demonstrates important aspects of sampling techniques (sample amount, particle size, and the representativeness of the sample in relation to the bulk material). The exercise…

RnaSeqSampleSize: real data based sample size estimation for RNA sequencing.

PubMed

Zhao, Shilin; Li, Chung-I; Guo, Yan; Sheng, Quanhu; Shyr, Yu

2018-05-30

One of the most important and often neglected components of a successful RNA sequencing (RNA-Seq) experiment is sample size estimation. A few negative binomial model-based methods have been developed to estimate sample size based on the parameters of a single gene. However, thousands of genes are quantified and tested for differential expression simultaneously in RNA-Seq experiments. Thus, additional issues should be carefully addressed, including the false discovery rate for multiple statistic tests, widely distributed read counts and dispersions for different genes. To solve these issues, we developed a sample size and power estimation method named RnaSeqSampleSize, based on the distributions of gene average read counts and dispersions estimated from real RNA-seq data. Datasets from previous, similar experiments such as the Cancer Genome Atlas (TCGA) can be used as a point of reference. Read counts and their dispersions were estimated from the reference's distribution; using that information, we estimated and summarized the power and sample size. RnaSeqSampleSize is implemented in R language and can be installed from Bioconductor website. A user friendly web graphic interface is provided at http://cqs.mc.vanderbilt.edu/shiny/RnaSeqSampleSize/ . RnaSeqSampleSize provides a convenient and powerful way for power and sample size estimation for an RNAseq experiment. It is also equipped with several unique features, including estimation for interested genes or pathway, power curve visualization, and parameter optimization.

DOE Office of Scientific and Technical Information (OSTI.GOV)

Zhang, Fangyan; Zhang, Song; Chung Wong, Pak

Effectively visualizing large graphs and capturing the statistical properties are two challenging tasks. To aid in these two tasks, many sampling approaches for graph simplification have been proposed, falling into three categories: node sampling, edge sampling, and traversal-based sampling. It is still unknown which approach is the best. We evaluate commonly used graph sampling methods through a combined visual and statistical comparison of graphs sampled at various rates. We conduct our evaluation on three graph models: random graphs, small-world graphs, and scale-free graphs. Initial results indicate that the effectiveness of a sampling method is dependent on the graph model, themore » size of the graph, and the desired statistical property. This benchmark study can be used as a guideline in choosing the appropriate method for a particular graph sampling task, and the results presented can be incorporated into graph visualization and analysis tools.« less

Soil carbon inventories under a bioenergy crop (switchgrass): Measurement limitations

DOE Office of Scientific and Technical Information (OSTI.GOV)

Garten, C.T. Jr.; Wullschleger, S.D.

Approximately 5 yr after planting, coarse root carbon (C) and soil organic C (SOC) inventories were compared under different types of plant cover at four switchgrass (Panicum virgatum L.) production field trials in the southeastern USA. There was significantly more coarse root C under switchgrass (Alamo variety) and forest cover than tall fescue (Festuca arundinacea Schreb.), corn (Zea mays L.), or native pastures of mixed grasses. Inventories of SOC under switchgrass were not significantly greater than SOC inventories under other plant covers. At some locations the statistical power associated with ANOVA of SOC inventories was low, which raised questions aboutmore » whether differences in SOC could be detected statistically. A minimum detectable difference (MDD) for SOC inventories was calculated. The MDD is the smallest detectable difference between treatment means once the variation, significance level, statistical power, and sample size are specified. The analysis indicated that a difference of {approx}50 mg SOC/cm{sup 2} or 5 Mg SOC/ha, which is {approx}10 to 15% of existing SOC, could be detected with reasonable sample sizes and good statistical power. The smallest difference in SOC inventories that can be detected, and only with exceedingly large sample sizes, is {approx}2 to 3%. These measurement limitations have implications for monitoring and verification of proposals to ameliorate increasing global atmospheric CO{sub 2} concentrations by sequestering C in soils.« less

Sample size requirements for estimating effective dose from computed tomography using solid-state metal-oxide-semiconductor field-effect transistor dosimetry

DOE Office of Scientific and Technical Information (OSTI.GOV)

Trattner, Sigal; Cheng, Bin; Pieniazek, Radoslaw L.

2014-04-15

Purpose: Effective dose (ED) is a widely used metric for comparing ionizing radiation burden between different imaging modalities, scanners, and scan protocols. In computed tomography (CT), ED can be estimated by performing scans on an anthropomorphic phantom in which metal-oxide-semiconductor field-effect transistor (MOSFET) solid-state dosimeters have been placed to enable organ dose measurements. Here a statistical framework is established to determine the sample size (number of scans) needed for estimating ED to a desired precision and confidence, for a particular scanner and scan protocol, subject to practical limitations. Methods: The statistical scheme involves solving equations which minimize the sample sizemore » required for estimating ED to desired precision and confidence. It is subject to a constrained variation of the estimated ED and solved using the Lagrange multiplier method. The scheme incorporates measurement variation introduced both by MOSFET calibration, and by variation in MOSFET readings between repeated CT scans. Sample size requirements are illustrated on cardiac, chest, and abdomen–pelvis CT scans performed on a 320-row scanner and chest CT performed on a 16-row scanner. Results: Sample sizes for estimating ED vary considerably between scanners and protocols. Sample size increases as the required precision or confidence is higher and also as the anticipated ED is lower. For example, for a helical chest protocol, for 95% confidence and 5% precision for the ED, 30 measurements are required on the 320-row scanner and 11 on the 16-row scanner when the anticipated ED is 4 mSv; these sample sizes are 5 and 2, respectively, when the anticipated ED is 10 mSv. Conclusions: Applying the suggested scheme, it was found that even at modest sample sizes, it is feasible to estimate ED with high precision and a high degree of confidence. As CT technology develops enabling ED to be lowered, more MOSFET measurements are needed to estimate ED with the same precision and confidence.« less

The size of a pilot study for a clinical trial should be calculated in relation to considerations of precision and efficiency.

PubMed

Sim, Julius; Lewis, Martyn

2012-03-01

To investigate methods to determine the size of a pilot study to inform a power calculation for a randomized controlled trial (RCT) using an interval/ratio outcome measure. Calculations based on confidence intervals (CIs) for the sample standard deviation (SD). Based on CIs for the sample SD, methods are demonstrated whereby (1) the observed SD can be adjusted to secure the desired level of statistical power in the main study with a specified level of confidence; (2) the sample for the main study, if calculated using the observed SD, can be adjusted, again to obtain the desired level of statistical power in the main study; (3) the power of the main study can be calculated for the situation in which the SD in the pilot study proves to be an underestimate of the true SD; and (4) an "efficient" pilot size can be determined to minimize the combined size of the pilot and main RCT. Trialists should calculate the appropriate size of a pilot study, just as they should the size of the main RCT, taking into account the twin needs to demonstrate efficiency in terms of recruitment and to produce precise estimates of treatment effect. Copyright © 2012 Elsevier Inc. All rights reserved.

A Sorting Statistic with Application in Neurological Magnetic Resonance Imaging of Autism.

PubMed

Levman, Jacob; Takahashi, Emi; Forgeron, Cynthia; MacDonald, Patrick; Stewart, Natalie; Lim, Ashley; Martel, Anne

2018-01-01

Effect size refers to the assessment of the extent of differences between two groups of samples on a single measurement. Assessing effect size in medical research is typically accomplished with Cohen's d statistic. Cohen's d statistic assumes that average values are good estimators of the position of a distribution of numbers and also assumes Gaussian (or bell-shaped) underlying data distributions. In this paper, we present an alternative evaluative statistic that can quantify differences between two data distributions in a manner that is similar to traditional effect size calculations; however, the proposed approach avoids making assumptions regarding the shape of the underlying data distribution. The proposed sorting statistic is compared with Cohen's d statistic and is demonstrated to be capable of identifying feature measurements of potential interest for which Cohen's d statistic implies the measurement would be of little use. This proposed sorting statistic has been evaluated on a large clinical autism dataset from Boston Children's Hospital , Harvard Medical School , demonstrating that it can potentially play a constructive role in future healthcare technologies.

A Sorting Statistic with Application in Neurological Magnetic Resonance Imaging of Autism

PubMed Central

Takahashi, Emi; Lim, Ashley; Martel, Anne

2018-01-01

Effect size refers to the assessment of the extent of differences between two groups of samples on a single measurement. Assessing effect size in medical research is typically accomplished with Cohen's d statistic. Cohen's d statistic assumes that average values are good estimators of the position of a distribution of numbers and also assumes Gaussian (or bell-shaped) underlying data distributions. In this paper, we present an alternative evaluative statistic that can quantify differences between two data distributions in a manner that is similar to traditional effect size calculations; however, the proposed approach avoids making assumptions regarding the shape of the underlying data distribution. The proposed sorting statistic is compared with Cohen's d statistic and is demonstrated to be capable of identifying feature measurements of potential interest for which Cohen's d statistic implies the measurement would be of little use. This proposed sorting statistic has been evaluated on a large clinical autism dataset from Boston Children's Hospital, Harvard Medical School, demonstrating that it can potentially play a constructive role in future healthcare technologies. PMID:29796236

Optimal spatial sampling techniques for ground truth data in microwave remote sensing of soil moisture

NASA Technical Reports Server (NTRS)

Rao, R. G. S.; Ulaby, F. T.

1977-01-01

The paper examines optimal sampling techniques for obtaining accurate spatial averages of soil moisture, at various depths and for cell sizes in the range 2.5-40 acres, with a minimum number of samples. Both simple random sampling and stratified sampling procedures are used to reach a set of recommended sample sizes for each depth and for each cell size. Major conclusions from statistical sampling test results are that (1) the number of samples required decreases with increasing depth; (2) when the total number of samples cannot be prespecified or the moisture in only one single layer is of interest, then a simple random sample procedure should be used which is based on the observed mean and SD for data from a single field; (3) when the total number of samples can be prespecified and the objective is to measure the soil moisture profile with depth, then stratified random sampling based on optimal allocation should be used; and (4) decreasing the sensor resolution cell size leads to fairly large decreases in samples sizes with stratified sampling procedures, whereas only a moderate decrease is obtained in simple random sampling procedures.

Scale Comparability between Nonaccommodated and Accommodated Forms of a Statewide High School Assessment: Assessment Using "l[subscript z]" Person-Fit

ERIC Educational Resources Information Center

Seo, Dong Gi; Hao, Shiqi

2016-01-01

Differential item/test functioning (DIF/DTF) are routine procedures to detect item/test unfairness as an explanation for group performance difference. However, unequal sample sizes and small sample sizes have an impact on the statistical power of the DIF/DTF detection procedures. Furthermore, DIF/DTF cannot be used for two test forms without…

Across-cohort QC analyses of GWAS summary statistics from complex traits.

PubMed

Chen, Guo-Bo; Lee, Sang Hong; Robinson, Matthew R; Trzaskowski, Maciej; Zhu, Zhi-Xiang; Winkler, Thomas W; Day, Felix R; Croteau-Chonka, Damien C; Wood, Andrew R; Locke, Adam E; Kutalik, Zoltán; Loos, Ruth J F; Frayling, Timothy M; Hirschhorn, Joel N; Yang, Jian; Wray, Naomi R; Visscher, Peter M

2016-01-01

Genome-wide association studies (GWASs) have been successful in discovering SNP trait associations for many quantitative traits and common diseases. Typically, the effect sizes of SNP alleles are very small and this requires large genome-wide association meta-analyses (GWAMAs) to maximize statistical power. A trend towards ever-larger GWAMA is likely to continue, yet dealing with summary statistics from hundreds of cohorts increases logistical and quality control problems, including unknown sample overlap, and these can lead to both false positive and false negative findings. In this study, we propose four metrics and visualization tools for GWAMA, using summary statistics from cohort-level GWASs. We propose methods to examine the concordance between demographic information, and summary statistics and methods to investigate sample overlap. (I) We use the population genetics F st statistic to verify the genetic origin of each cohort and their geographic location, and demonstrate using GWAMA data from the GIANT Consortium that geographic locations of cohorts can be recovered and outlier cohorts can be detected. (II) We conduct principal component analysis based on reported allele frequencies, and are able to recover the ancestral information for each cohort. (III) We propose a new statistic that uses the reported allelic effect sizes and their standard errors to identify significant sample overlap or heterogeneity between pairs of cohorts. (IV) To quantify unknown sample overlap across all pairs of cohorts, we propose a method that uses randomly generated genetic predictors that does not require the sharing of individual-level genotype data and does not breach individual privacy.

Across-cohort QC analyses of GWAS summary statistics from complex traits

PubMed Central

Chen, Guo-Bo; Lee, Sang Hong; Robinson, Matthew R; Trzaskowski, Maciej; Zhu, Zhi-Xiang; Winkler, Thomas W; Day, Felix R; Croteau-Chonka, Damien C; Wood, Andrew R; Locke, Adam E; Kutalik, Zoltán; Loos, Ruth J F; Frayling, Timothy M; Hirschhorn, Joel N; Yang, Jian; Wray, Naomi R; Visscher, Peter M

2017-01-01

Genome-wide association studies (GWASs) have been successful in discovering SNP trait associations for many quantitative traits and common diseases. Typically, the effect sizes of SNP alleles are very small and this requires large genome-wide association meta-analyses (GWAMAs) to maximize statistical power. A trend towards ever-larger GWAMA is likely to continue, yet dealing with summary statistics from hundreds of cohorts increases logistical and quality control problems, including unknown sample overlap, and these can lead to both false positive and false negative findings. In this study, we propose four metrics and visualization tools for GWAMA, using summary statistics from cohort-level GWASs. We propose methods to examine the concordance between demographic information, and summary statistics and methods to investigate sample overlap. (I) We use the population genetics Fst statistic to verify the genetic origin of each cohort and their geographic location, and demonstrate using GWAMA data from the GIANT Consortium that geographic locations of cohorts can be recovered and outlier cohorts can be detected. (II) We conduct principal component analysis based on reported allele frequencies, and are able to recover the ancestral information for each cohort. (III) We propose a new statistic that uses the reported allelic effect sizes and their standard errors to identify significant sample overlap or heterogeneity between pairs of cohorts. (IV) To quantify unknown sample overlap across all pairs of cohorts, we propose a method that uses randomly generated genetic predictors that does not require the sharing of individual-level genotype data and does not breach individual privacy. PMID:27552965

Statistics 101 for Radiologists.

PubMed

Anvari, Arash; Halpern, Elkan F; Samir, Anthony E

2015-10-01

Diagnostic tests have wide clinical applications, including screening, diagnosis, measuring treatment effect, and determining prognosis. Interpreting diagnostic test results requires an understanding of key statistical concepts used to evaluate test efficacy. This review explains descriptive statistics and discusses probability, including mutually exclusive and independent events and conditional probability. In the inferential statistics section, a statistical perspective on study design is provided, together with an explanation of how to select appropriate statistical tests. Key concepts in recruiting study samples are discussed, including representativeness and random sampling. Variable types are defined, including predictor, outcome, and covariate variables, and the relationship of these variables to one another. In the hypothesis testing section, we explain how to determine if observed differences between groups are likely to be due to chance. We explain type I and II errors, statistical significance, and study power, followed by an explanation of effect sizes and how confidence intervals can be used to generalize observed effect sizes to the larger population. Statistical tests are explained in four categories: t tests and analysis of variance, proportion analysis tests, nonparametric tests, and regression techniques. We discuss sensitivity, specificity, accuracy, receiver operating characteristic analysis, and likelihood ratios. Measures of reliability and agreement, including κ statistics, intraclass correlation coefficients, and Bland-Altman graphs and analysis, are introduced. © RSNA, 2015.

Estimation of within-stratum variance for sample allocation: Foreign commodity production forecasting

NASA Technical Reports Server (NTRS)

Chhikara, R. S.; Perry, C. R., Jr. (Principal Investigator)

1980-01-01

The problem of determining the stratum variances required for an optimum sample allocation for remotely sensed crop surveys is investigated with emphasis on an approach based on the concept of stratum variance as a function of the sampling unit size. A methodology using the existing and easily available information of historical statistics is developed for obtaining initial estimates of stratum variances. The procedure is applied to variance for wheat in the U.S. Great Plains and is evaluated based on the numerical results obtained. It is shown that the proposed technique is viable and performs satisfactorily with the use of a conservative value (smaller than the expected value) for the field size and with the use of crop statistics from the small political division level.

Combining censored and uncensored data in a U-statistic: design and sample size implications for cell therapy research.

PubMed

Moyé, Lemuel A; Lai, Dejian; Jing, Kaiyan; Baraniuk, Mary Sarah; Kwak, Minjung; Penn, Marc S; Wu, Colon O

2011-01-01

The assumptions that anchor large clinical trials are rooted in smaller, Phase II studies. In addition to specifying the target population, intervention delivery, and patient follow-up duration, physician-scientists who design these Phase II studies must select the appropriate response variables (endpoints). However, endpoint measures can be problematic. If the endpoint assesses the change in a continuous measure over time, then the occurrence of an intervening significant clinical event (SCE), such as death, can preclude the follow-up measurement. Finally, the ideal continuous endpoint measurement may be contraindicated in a fraction of the study patients, a change that requires a less precise substitution in this subset of participants.A score function that is based on the U-statistic can address these issues of 1) intercurrent SCE's and 2) response variable ascertainments that use different measurements of different precision. The scoring statistic is easy to apply, clinically relevant, and provides flexibility for the investigators' prospective design decisions. Sample size and power formulations for this statistic are provided as functions of clinical event rates and effect size estimates that are easy for investigators to identify and discuss. Examples are provided from current cardiovascular cell therapy research.

Statistical analyses to support guidelines for marine avian sampling. Final report

USGS Publications Warehouse

Kinlan, Brian P.; Zipkin, Elise; O'Connell, Allan F.; Caldow, Chris

2012-01-01

Interest in development of offshore renewable energy facilities has led to a need for high-quality, statistically robust information on marine wildlife distributions. A practical approach is described to estimate the amount of sampling effort required to have sufficient statistical power to identify species-specific “hotspots” and “coldspots” of marine bird abundance and occurrence in an offshore environment divided into discrete spatial units (e.g., lease blocks), where “hotspots” and “coldspots” are defined relative to a reference (e.g., regional) mean abundance and/or occurrence probability for each species of interest. For example, a location with average abundance or occurrence that is three times larger the mean (3x effect size) could be defined as a “hotspot,” and a location that is three times smaller than the mean (1/3x effect size) as a “coldspot.” The choice of the effect size used to define hot and coldspots will generally depend on a combination of ecological and regulatory considerations. A method is also developed for testing the statistical significance of possible hotspots and coldspots. Both methods are illustrated with historical seabird survey data from the USGS Avian Compendium Database. Our approach consists of five main components: 1. A review of the primary scientific literature on statistical modeling of animal group size and avian count data to develop a candidate set of statistical distributions that have been used or may be useful to model seabird counts. 2. Statistical power curves for one-sample, one-tailed Monte Carlo significance tests of differences of observed small-sample means from a specified reference distribution. These curves show the power to detect "hotspots" or "coldspots" of occurrence and abundance at a range of effect sizes, given assumptions which we discuss. 3. A model selection procedure, based on maximum likelihood fits of models in the candidate set, to determine an appropriate statistical distribution to describe counts of a given species in a particular region and season. 4. Using a large database of historical at-sea seabird survey data, we applied this technique to identify appropriate statistical distributions for modeling a variety of species, allowing the distribution to vary by season. For each species and season, we used the selected distribution to calculate and map retrospective statistical power to detect hotspots and coldspots, and map pvalues from Monte Carlo significance tests of hotspots and coldspots, in discrete lease blocks designated by the U.S. Department of Interior, Bureau of Ocean Energy Management (BOEM). 5. Because our definition of hotspots and coldspots does not explicitly include variability over time, we examine the relationship between the temporal scale of sampling and the proportion of variance captured in time series of key environmental correlates of marine bird abundance, as well as available marine bird abundance time series, and use these analyses to develop recommendations for the temporal distribution of sampling to adequately represent both shortterm and long-term variability. We conclude by presenting a schematic “decision tree” showing how this power analysis approach would fit in a general framework for avian survey design, and discuss implications of model assumptions and results. We discuss avenues for future development of this work, and recommendations for practical implementation in the context of siting and wildlife assessment for offshore renewable energy development projects.

ADEQUACY OF VISUALLY CLASSIFIED PARTICLE COUNT STATISTICS FROM REGIONAL STREAM HABITAT SURVEYS

EPA Science Inventory

Streamlined sampling procedures must be used to achieve a sufficient sample size with limited resources in studies undertaken to evaluate habitat status and potential management-related habitat degradation at a regional scale. At the same time, these sampling procedures must achi...

Explanation of Two Anomalous Results in Statistical Mediation Analysis.

PubMed

Fritz, Matthew S; Taylor, Aaron B; Mackinnon, David P

2012-01-01

Previous studies of different methods of testing mediation models have consistently found two anomalous results. The first result is elevated Type I error rates for the bias-corrected and accelerated bias-corrected bootstrap tests not found in nonresampling tests or in resampling tests that did not include a bias correction. This is of special concern as the bias-corrected bootstrap is often recommended and used due to its higher statistical power compared with other tests. The second result is statistical power reaching an asymptote far below 1.0 and in some conditions even declining slightly as the size of the relationship between X and M , a , increased. Two computer simulations were conducted to examine these findings in greater detail. Results from the first simulation found that the increased Type I error rates for the bias-corrected and accelerated bias-corrected bootstrap are a function of an interaction between the size of the individual paths making up the mediated effect and the sample size, such that elevated Type I error rates occur when the sample size is small and the effect size of the nonzero path is medium or larger. Results from the second simulation found that stagnation and decreases in statistical power as a function of the effect size of the a path occurred primarily when the path between M and Y , b , was small. Two empirical mediation examples are provided using data from a steroid prevention and health promotion program aimed at high school football players (Athletes Training and Learning to Avoid Steroids; Goldberg et al., 1996), one to illustrate a possible Type I error for the bias-corrected bootstrap test and a second to illustrate a loss in power related to the size of a . Implications of these findings are discussed.

Inflammation response and cytotoxic effects in human THP-1 cells of size-fractionated PM10 extracts in a polluted urban site.

PubMed

Schilirò, T; Alessandria, L; Bonetta, S; Carraro, E; Gilli, G

2016-02-01

To contribute to a greater characterization of the airborne particulate matter's toxicity, size-fractionated PM10 was sampled during different seasons in a polluted urban site in Torino, a northern Italian city. Three main size fractions (PM10 - 3 μm; PM3 - 0.95 μm; PM < 0.95 μm) extracts (organic and aqueous) were assayed with THP-1 cells to evaluate their effects on cell proliferation, LDH activity, TNFα, IL-8 and CYP1A1 expression. The mean PM10 concentrations were statistically different in summer and in winter and the finest fraction PM<0.95 was always higher than the others. Size-fractionated PM10 extracts, sampled in an urban traffic meteorological-chemical station produced size-related toxicological effects in relation to season and particles extraction. The PM summer extracts induced a significant release of LDH compared to winter and produced a size-related effect, with higher values measured with PM10-3. Exposure to size-fractionated PM10 extracts did not induce significant expression of TNFα. IL-8 expression was influenced by exposure to size-fractionated PM10 extracts and statistically significant differences were found between kind of extracts for both seasons. The mean fold increases in CYP1A1 expression were statistically different in summer and in winter; winter fraction extracts produced a size-related effect, in particular for organic samples with higher values measured with PM<0.95 extracts. Our results confirm that the only measure of PM can be misleading for the assessment of air quality moreover we support efforts toward identifying potential effect-based tools (e.g. in vitro test) that could be used in the context of the different monitoring programs. Copyright © 2015 Elsevier Ltd. All rights reserved.

Outlier Removal and the Relation with Reporting Errors and Quality of Psychological Research

PubMed Central

Bakker, Marjan; Wicherts, Jelte M.

2014-01-01

Background The removal of outliers to acquire a significant result is a questionable research practice that appears to be commonly used in psychology. In this study, we investigated whether the removal of outliers in psychology papers is related to weaker evidence (against the null hypothesis of no effect), a higher prevalence of reporting errors, and smaller sample sizes in these papers compared to papers in the same journals that did not report the exclusion of outliers from the analyses. Methods and Findings We retrieved a total of 2667 statistical results of null hypothesis significance tests from 153 articles in main psychology journals, and compared results from articles in which outliers were removed (N = 92) with results from articles that reported no exclusion of outliers (N = 61). We preregistered our hypotheses and methods and analyzed the data at the level of articles. Results show no significant difference between the two types of articles in median p value, sample sizes, or prevalence of all reporting errors, large reporting errors, and reporting errors that concerned the statistical significance. However, we did find a discrepancy between the reported degrees of freedom of t tests and the reported sample size in 41% of articles that did not report removal of any data values. This suggests common failure to report data exclusions (or missingness) in psychological articles. Conclusions We failed to find that the removal of outliers from the analysis in psychological articles was related to weaker evidence (against the null hypothesis of no effect), sample size, or the prevalence of errors. However, our control sample might be contaminated due to nondisclosure of excluded values in articles that did not report exclusion of outliers. Results therefore highlight the importance of more transparent reporting of statistical analyses. PMID:25072606

Influence of sampling window size and orientation on parafoveal cone packing density

PubMed Central

Lombardo, Marco; Serrao, Sebastiano; Ducoli, Pietro; Lombardo, Giuseppe

2013-01-01

We assessed the agreement between sampling windows of different size and orientation on packing density estimates in images of the parafoveal cone mosaic acquired using a flood-illumination adaptive optics retinal camera. Horizontal and vertical oriented sampling windows of different size (320x160 µm, 160x80 µm and 80x40 µm) were selected in two retinal locations along the horizontal meridian in one eye of ten subjects. At each location, cone density tended to decline with decreasing sampling area. Although the differences in cone density estimates were not statistically significant, Bland-Altman plots showed that the agreement between cone density estimated within the different sampling window conditions was moderate. The percentage of the preferred packing arrangements of cones by Voronoi tiles was slightly affected by window size and orientation. The results illustrated the high importance of specifying the size and orientation of the sampling window used to derive cone metric estimates to facilitate comparison of different studies. PMID:24009995

Characterizing the Joint Effect of Diverse Test-Statistic Correlation Structures and Effect Size on False Discovery Rates in a Multiple-Comparison Study of Many Outcome Measures

NASA Technical Reports Server (NTRS)

Feiveson, Alan H.; Ploutz-Snyder, Robert; Fiedler, James

2011-01-01

In their 2009 Annals of Statistics paper, Gavrilov, Benjamini, and Sarkar report the results of a simulation assessing the robustness of their adaptive step-down procedure (GBS) for controlling the false discovery rate (FDR) when normally distributed test statistics are serially correlated. In this study we extend the investigation to the case of multiple comparisons involving correlated non-central t-statistics, in particular when several treatments or time periods are being compared to a control in a repeated-measures design with many dependent outcome measures. In addition, we consider several dependence structures other than serial correlation and illustrate how the FDR depends on the interaction between effect size and the type of correlation structure as indexed by Foerstner s distance metric from an identity. The relationship between the correlation matrix R of the original dependent variables and R, the correlation matrix of associated t-statistics is also studied. In general R depends not only on R, but also on sample size and the signed effect sizes for the multiple comparisons.

Approximate sample sizes required to estimate length distributions

USGS Publications Warehouse

Miranda, L.E.

2007-01-01

The sample sizes required to estimate fish length were determined by bootstrapping from reference length distributions. Depending on population characteristics and species-specific maximum lengths, 1-cm length-frequency histograms required 375-1,200 fish to estimate within 10% with 80% confidence, 2.5-cm histograms required 150-425 fish, proportional stock density required 75-140 fish, and mean length required 75-160 fish. In general, smaller species, smaller populations, populations with higher mortality, and simpler length statistics required fewer samples. Indices that require low sample sizes may be suitable for monitoring population status, and when large changes in length are evident, additional sampling effort may be allocated to more precisely define length status with more informative estimators. ?? Copyright by the American Fisheries Society 2007.

Grain size statistics and depositional pattern of the Ecca Group sandstones, Karoo Supergroup in the Eastern Cape Province, South Africa

NASA Astrophysics Data System (ADS)

Baiyegunhi, Christopher; Liu, Kuiwu; Gwavava, Oswald

2017-11-01

Grain size analysis is a vital sedimentological tool used to unravel the hydrodynamic conditions, mode of transportation and deposition of detrital sediments. In this study, detailed grain-size analysis was carried out on thirty-five sandstone samples from the Ecca Group in the Eastern Cape Province of South Africa. Grain-size statistical parameters, bivariate analysis, linear discriminate functions, Passega diagrams and log-probability curves were used to reveal the depositional processes, sedimentation mechanisms, hydrodynamic energy conditions and to discriminate different depositional environments. The grain-size parameters show that most of the sandstones are very fine to fine grained, moderately well sorted, mostly near-symmetrical and mesokurtic in nature. The abundance of very fine to fine grained sandstones indicate the dominance of low energy environment. The bivariate plots show that the samples are mostly grouped, except for the Prince Albert samples that show scattered trend, which is due to the either mixture of two modes in equal proportion in bimodal sediments or good sorting in unimodal sediments. The linear discriminant function analysis is dominantly indicative of turbidity current deposits under shallow marine environments for samples from the Prince Albert, Collingham and Ripon Formations, while those samples from the Fort Brown Formation are lacustrine or deltaic deposits. The C-M plots indicated that the sediments were deposited mainly by suspension and saltation, and graded suspension. Visher diagrams show that saltation is the major process of transportation, followed by suspension.

Individualized statistical learning from medical image databases: application to identification of brain lesions.

PubMed

Erus, Guray; Zacharaki, Evangelia I; Davatzikos, Christos

2014-04-01

This paper presents a method for capturing statistical variation of normal imaging phenotypes, with emphasis on brain structure. The method aims to estimate the statistical variation of a normative set of images from healthy individuals, and identify abnormalities as deviations from normality. A direct estimation of the statistical variation of the entire volumetric image is challenged by the high-dimensionality of images relative to smaller sample sizes. To overcome this limitation, we iteratively sample a large number of lower dimensional subspaces that capture image characteristics ranging from fine and localized to coarser and more global. Within each subspace, a "target-specific" feature selection strategy is applied to further reduce the dimensionality, by considering only imaging characteristics present in a test subject's images. Marginal probability density functions of selected features are estimated through PCA models, in conjunction with an "estimability" criterion that limits the dimensionality of estimated probability densities according to available sample size and underlying anatomy variation. A test sample is iteratively projected to the subspaces of these marginals as determined by PCA models, and its trajectory delineates potential abnormalities. The method is applied to segmentation of various brain lesion types, and to simulated data on which superiority of the iterative method over straight PCA is demonstrated. Copyright © 2014 Elsevier B.V. All rights reserved.

Individualized Statistical Learning from Medical Image Databases: Application to Identification of Brain Lesions

PubMed Central

Erus, Guray; Zacharaki, Evangelia I.; Davatzikos, Christos

2014-01-01

This paper presents a method for capturing statistical variation of normal imaging phenotypes, with emphasis on brain structure. The method aims to estimate the statistical variation of a normative set of images from healthy individuals, and identify abnormalities as deviations from normality. A direct estimation of the statistical variation of the entire volumetric image is challenged by the high-dimensionality of images relative to smaller sample sizes. To overcome this limitation, we iteratively sample a large number of lower dimensional subspaces that capture image characteristics ranging from fine and localized to coarser and more global. Within each subspace, a “target-specific” feature selection strategy is applied to further reduce the dimensionality, by considering only imaging characteristics present in a test subject’s images. Marginal probability density functions of selected features are estimated through PCA models, in conjunction with an “estimability” criterion that limits the dimensionality of estimated probability densities according to available sample size and underlying anatomy variation. A test sample is iteratively projected to the subspaces of these marginals as determined by PCA models, and its trajectory delineates potential abnormalities. The method is applied to segmentation of various brain lesion types, and to simulated data on which superiority of the iterative method over straight PCA is demonstrated. PMID:24607564

On the analysis of very small samples of Gaussian repeated measurements: an alternative approach.

PubMed

Westgate, Philip M; Burchett, Woodrow W

2017-03-15

The analysis of very small samples of Gaussian repeated measurements can be challenging. First, due to a very small number of independent subjects contributing outcomes over time, statistical power can be quite small. Second, nuisance covariance parameters must be appropriately accounted for in the analysis in order to maintain the nominal test size. However, available statistical strategies that ensure valid statistical inference may lack power, whereas more powerful methods may have the potential for inflated test sizes. Therefore, we explore an alternative approach to the analysis of very small samples of Gaussian repeated measurements, with the goal of maintaining valid inference while also improving statistical power relative to other valid methods. This approach uses generalized estimating equations with a bias-corrected empirical covariance matrix that accounts for all small-sample aspects of nuisance correlation parameter estimation in order to maintain valid inference. Furthermore, the approach utilizes correlation selection strategies with the goal of choosing the working structure that will result in the greatest power. In our study, we show that when accurate modeling of the nuisance correlation structure impacts the efficiency of regression parameter estimation, this method can improve power relative to existing methods that yield valid inference. Copyright © 2017 John Wiley & Sons, Ltd. Copyright © 2017 John Wiley & Sons, Ltd.

Influence of size-fractioning techniques on concentrations of selected trace metals in bottom materials from two streams in northeastern Ohio

USGS Publications Warehouse

Koltun, G.F.; Helsel, Dennis R.

1986-01-01

Identical stream-bottom material samples, when fractioned to the same size by different techniques, may contain significantly different trace-metal concentrations. Precision of techniques also may differ, which could affect the ability to discriminate between size-fractioned bottom-material samples having different metal concentrations. Bottom-material samples fractioned to less than 0.020 millimeters by means of three common techniques (air elutriation, sieving, and settling) were analyzed for six trace metals to determine whether the technique used to obtain the desired particle-size fraction affects the ability to discriminate between bottom materials having different trace-metal concentrations. In addition, this study attempts to assess whether median trace-metal concentrations in size-fractioned bottom materials of identical origin differ depending on the size-fractioning technique used. Finally, this study evaluates the efficiency of the three size-fractioning techniques in terms of time, expense, and effort involved. Bottom-material samples were collected at two sites in northeastern Ohio: One is located in an undeveloped forested basin, and the other is located in a basin having a mixture of industrial and surface-mining land uses. The sites were selected for their close physical proximity, similar contributing drainage areas, and the likelihood that trace-metal concentrations in the bottom materials would be significantly different. Statistically significant differences in the concentrations of trace metals were detected between bottom-material samples collected at the two sites when the samples had been size-fractioned by means of air elutriation or sieving. Statistical analyses of samples that had been size fractioned by settling in native water were not measurably different in any of the six trace metals analyzed. Results of multiple comparison tests suggest that differences related to size-fractioning technique were evident in median copper, lead, and iron concentrations. Technique-related differences in copper concentrations most likely resulted from contamination of air-elutriated samples by a feed tip on the elutriator apparatus. No technique-related differences were observed in chromium, manganese, or zinc concentrations. Although air elutriation was the most expensive sizefractioning technique investigated, samples fractioned by this technique appeared to provide a superior level of discrimination between metal concentrations present in the bottom materials of the two sites. Sieving was an adequate lower-cost but more laborintensive alternative.

How Sample Size Affects a Sampling Distribution

ERIC Educational Resources Information Center

Mulekar, Madhuri S.; Siegel, Murray H.

2009-01-01

If students are to understand inferential statistics successfully, they must have a profound understanding of the nature of the sampling distribution. Specifically, they must comprehend the determination of the expected value and standard error of a sampling distribution as well as the meaning of the central limit theorem. Many students in a high…

Power calculation for overall hypothesis testing with high-dimensional commensurate outcomes.

PubMed

Chi, Yueh-Yun; Gribbin, Matthew J; Johnson, Jacqueline L; Muller, Keith E

2014-02-28

The complexity of system biology means that any metabolic, genetic, or proteomic pathway typically includes so many components (e.g., molecules) that statistical methods specialized for overall testing of high-dimensional and commensurate outcomes are required. While many overall tests have been proposed, very few have power and sample size methods. We develop accurate power and sample size methods and software to facilitate study planning for high-dimensional pathway analysis. With an account of any complex correlation structure between high-dimensional outcomes, the new methods allow power calculation even when the sample size is less than the number of variables. We derive the exact (finite-sample) and approximate non-null distributions of the 'univariate' approach to repeated measures test statistic, as well as power-equivalent scenarios useful to generalize our numerical evaluations. Extensive simulations of group comparisons support the accuracy of the approximations even when the ratio of number of variables to sample size is large. We derive a minimum set of constants and parameters sufficient and practical for power calculation. Using the new methods and specifying the minimum set to determine power for a study of metabolic consequences of vitamin B6 deficiency helps illustrate the practical value of the new results. Free software implementing the power and sample size methods applies to a wide range of designs, including one group pre-intervention and post-intervention comparisons, multiple parallel group comparisons with one-way or factorial designs, and the adjustment and evaluation of covariate effects. Copyright © 2013 John Wiley & Sons, Ltd.

Studies in Support of the Application of Statistical Theory to Design and Evaluation of Operational Tests. Annex D. An Application of Bayesian Statistical Methods in the Determination of Sample Size for Operational Testing in the U.S. Army

DTIC Science & Technology

1977-07-01

SIZE C XNI. C UE2 - UTILITY OF EXPERIMENT OF SIZE C XN2. C ICHECK - VARIABLE USLD TO CHECK FOR C TERMINATION, C~C DIMENSION SUBLIM{20),UPLIM(20),UEI(20...1J=UPLIM(K4-I)-XNI (K+1)+SU8LIt1(K+i*. C CHECK FOR TERMINATION. 944 ICHECK =SUBLIM(K)+2 IFIICHECK.GEUPLiHMK.,OR.K.G1.20’ GO TO 930 GO TO 920 930

Sample size determination for mediation analysis of longitudinal data.

PubMed

Pan, Haitao; Liu, Suyu; Miao, Danmin; Yuan, Ying

2018-03-27

Sample size planning for longitudinal data is crucial when designing mediation studies because sufficient statistical power is not only required in grant applications and peer-reviewed publications, but is essential to reliable research results. However, sample size determination is not straightforward for mediation analysis of longitudinal design. To facilitate planning the sample size for longitudinal mediation studies with a multilevel mediation model, this article provides the sample size required to achieve 80% power by simulations under various sizes of the mediation effect, within-subject correlations and numbers of repeated measures. The sample size calculation is based on three commonly used mediation tests: Sobel's method, distribution of product method and the bootstrap method. Among the three methods of testing the mediation effects, Sobel's method required the largest sample size to achieve 80% power. Bootstrapping and the distribution of the product method performed similarly and were more powerful than Sobel's method, as reflected by the relatively smaller sample sizes. For all three methods, the sample size required to achieve 80% power depended on the value of the ICC (i.e., within-subject correlation). A larger value of ICC typically required a larger sample size to achieve 80% power. Simulation results also illustrated the advantage of the longitudinal study design. The sample size tables for most encountered scenarios in practice have also been published for convenient use. Extensive simulations study showed that the distribution of the product method and bootstrapping method have superior performance to the Sobel's method, but the product method was recommended to use in practice in terms of less computation time load compared to the bootstrapping method. A R package has been developed for the product method of sample size determination in mediation longitudinal study design.

Statistical computation of tolerance limits

NASA Technical Reports Server (NTRS)

Wheeler, J. T.

1993-01-01

Based on a new theory, two computer codes were developed specifically to calculate the exact statistical tolerance limits for normal distributions within unknown means and variances for the one-sided and two-sided cases for the tolerance factor, k. The quantity k is defined equivalently in terms of the noncentral t-distribution by the probability equation. Two of the four mathematical methods employ the theory developed for the numerical simulation. Several algorithms for numerically integrating and iteratively root-solving the working equations are written to augment the program simulation. The program codes generate some tables of k's associated with the varying values of the proportion and sample size for each given probability to show accuracy obtained for small sample sizes.

Expected values and variances of Bragg peak intensities measured in a nanocrystalline powder diffraction experiment

DOE Office of Scientific and Technical Information (OSTI.GOV)

Öztürk, Hande; Noyan, I. Cevdet

A rigorous study of sampling and intensity statistics applicable for a powder diffraction experiment as a function of crystallite size is presented. Our analysis yields approximate equations for the expected value, variance and standard deviations for both the number of diffracting grains and the corresponding diffracted intensity for a given Bragg peak. The classical formalism published in 1948 by Alexander, Klug & Kummer [J. Appl. Phys.(1948),19, 742–753] appears as a special case, limited to large crystallite sizes, here. It is observed that both the Lorentz probability expression and the statistics equations used in the classical formalism are inapplicable for nanocrystallinemore » powder samples.« less

Expected values and variances of Bragg peak intensities measured in a nanocrystalline powder diffraction experiment

DOE PAGES

Öztürk, Hande; Noyan, I. Cevdet

2017-08-24

A rigorous study of sampling and intensity statistics applicable for a powder diffraction experiment as a function of crystallite size is presented. Our analysis yields approximate equations for the expected value, variance and standard deviations for both the number of diffracting grains and the corresponding diffracted intensity for a given Bragg peak. The classical formalism published in 1948 by Alexander, Klug & Kummer [J. Appl. Phys.(1948),19, 742–753] appears as a special case, limited to large crystallite sizes, here. It is observed that both the Lorentz probability expression and the statistics equations used in the classical formalism are inapplicable for nanocrystallinemore » powder samples.« less

Sample sizes needed for specified margins of relative error in the estimates of the repeatability and reproducibility standard deviations.

PubMed

McClure, Foster D; Lee, Jung K

2005-01-01

Sample size formulas are developed to estimate the repeatability and reproducibility standard deviations (Sr and S(R)) such that the actual error in (Sr and S(R)) relative to their respective true values, sigmar and sigmaR, are at predefined levels. The statistical consequences associated with AOAC INTERNATIONAL required sample size to validate an analytical method are discussed. In addition, formulas to estimate the uncertainties of (Sr and S(R)) were derived and are provided as supporting documentation. Formula for the Number of Replicates Required for a Specified Margin of Relative Error in the Estimate of the Repeatability Standard Deviation.

Sampling designs for contaminant temporal trend analyses using sedentary species exemplified by the snails Bellamya aeruginosa and Viviparus viviparus.

PubMed

Yin, Ge; Danielsson, Sara; Dahlberg, Anna-Karin; Zhou, Yihui; Qiu, Yanling; Nyberg, Elisabeth; Bignert, Anders

2017-10-01

Environmental monitoring typically assumes samples and sampling activities to be representative of the population being studied. Given a limited budget, an appropriate sampling strategy is essential to support detecting temporal trends of contaminants. In the present study, based on real chemical analysis data on polybrominated diphenyl ethers in snails collected from five subsites in Tianmu Lake, computer simulation is performed to evaluate three sampling strategies by the estimation of required sample size, to reach a detection of an annual change of 5% with a statistical power of 80% and 90% with a significant level of 5%. The results showed that sampling from an arbitrarily selected sampling spot is the worst strategy, requiring much more individual analyses to achieve the above mentioned criteria compared with the other two approaches. A fixed sampling site requires the lowest sample size but may not be representative for the intended study object e.g. a lake and is also sensitive to changes of that particular sampling site. In contrast, sampling at multiple sites along the shore each year, and using pooled samples when the cost to collect and prepare individual specimens are much lower than the cost for chemical analysis, would be the most robust and cost efficient strategy in the long run. Using statistical power as criterion, the results demonstrated quantitatively the consequences of various sampling strategies, and could guide users with respect of required sample sizes depending on sampling design for long term monitoring programs. Copyright © 2017 Elsevier Ltd. All rights reserved.

Double asymptotics for the chi-square statistic.

PubMed

Rempała, Grzegorz A; Wesołowski, Jacek

2016-12-01

Consider distributional limit of the Pearson chi-square statistic when the number of classes m n increases with the sample size n and [Formula: see text]. Under mild moment conditions, the limit is Gaussian for λ = ∞, Poisson for finite λ > 0, and degenerate for λ = 0.

Role of microstructure on twin nucleation and growth in HCP titanium: A statistical study

DOE Office of Scientific and Technical Information (OSTI.GOV)

Arul Kumar, M.; Wroński, M.; McCabe, Rodney James

In this study, a detailed statistical analysis is performed using Electron Back Scatter Diffraction (EBSD) to establish the effect of microstructure on twin nucleation and growth in deformed commercial purity hexagonal close packed (HCP) titanium. Rolled titanium samples are compressed along rolling, transverse and normal directions to establish statistical correlations for {10–12}, {11–21}, and {11–22} twins. A recently developed automated EBSD-twinning analysis software is employed for the statistical analysis. Finally, the analysis provides the following key findings: (I) grain size and strain dependence is different for twin nucleation and growth; (II) twinning statistics can be generalized for the HCP metalsmore » magnesium, zirconium and titanium; and (III) complex microstructure, where grain shape and size distribution is heterogeneous, requires multi-point statistical correlations.« less

Role of microstructure on twin nucleation and growth in HCP titanium: A statistical study

DOE PAGES

Arul Kumar, M.; Wroński, M.; McCabe, Rodney James; ...

2018-02-01

In this study, a detailed statistical analysis is performed using Electron Back Scatter Diffraction (EBSD) to establish the effect of microstructure on twin nucleation and growth in deformed commercial purity hexagonal close packed (HCP) titanium. Rolled titanium samples are compressed along rolling, transverse and normal directions to establish statistical correlations for {10–12}, {11–21}, and {11–22} twins. A recently developed automated EBSD-twinning analysis software is employed for the statistical analysis. Finally, the analysis provides the following key findings: (I) grain size and strain dependence is different for twin nucleation and growth; (II) twinning statistics can be generalized for the HCP metalsmore » magnesium, zirconium and titanium; and (III) complex microstructure, where grain shape and size distribution is heterogeneous, requires multi-point statistical correlations.« less

[Evaluation of the quality of Anales Españoles de Pediatría versus Medicina Clínica].

PubMed

Bonillo Perales, A

2002-08-01

To compare the scientific methodology and quality of articles published in Anales Españoles de Pediatría and Medicina Clínica. A stratified and randomized selection of 40 original articles published in 2001 in Anales Españoles de Pediatría and Medicina Clínica was made. Methodological errors in the critical analysis of original articles (21 items), epidemiological design, sample size, statistical complexity and levels of scientific evidence in both journals were compared using the chi-squared and/or Student's t-test. No differences were found between Anales Españoles de Pediatría and Medicina Clínica in the critical evaluation of original articles (p > 0.2). In original articles published in Anales Españoles de Pediatría, the designs were of lower scientific evidence (a lower proportion of clinical trials, cohort and case-control studies) (17.5 vs 42.5 %, p 0.05), sample sizes were smaller (p 0.003) and there was less statistical complexity in the results section (p 0.03). To improve the scientific quality of Anales Españoles de Pediatría, improved study designs, larger sample sizes and greater statistical complexity are required in its articles.

The N-Pact Factor: Evaluating the Quality of Empirical Journals with Respect to Sample Size and Statistical Power

PubMed Central

Fraley, R. Chris; Vazire, Simine

2014-01-01

The authors evaluate the quality of research reported in major journals in social-personality psychology by ranking those journals with respect to their N-pact Factors (NF)—the statistical power of the empirical studies they publish to detect typical effect sizes. Power is a particularly important attribute for evaluating research quality because, relative to studies that have low power, studies that have high power are more likely to (a) to provide accurate estimates of effects, (b) to produce literatures with low false positive rates, and (c) to lead to replicable findings. The authors show that the average sample size in social-personality research is 104 and that the power to detect the typical effect size in the field is approximately 50%. Moreover, they show that there is considerable variation among journals in sample sizes and power of the studies they publish, with some journals consistently publishing higher power studies than others. The authors hope that these rankings will be of use to authors who are choosing where to submit their best work, provide hiring and promotion committees with a superior way of quantifying journal quality, and encourage competition among journals to improve their NF rankings. PMID:25296159

Stratum variance estimation for sample allocation in crop surveys. [Great Plains Corridor

NASA Technical Reports Server (NTRS)

Perry, C. R., Jr.; Chhikara, R. S. (Principal Investigator)

1980-01-01

The problem of determining stratum variances needed in achieving an optimum sample allocation for crop surveys by remote sensing is investigated by considering an approach based on the concept of stratum variance as a function of the sampling unit size. A methodology using the existing and easily available information of historical crop statistics is developed for obtaining initial estimates of tratum variances. The procedure is applied to estimate stratum variances for wheat in the U.S. Great Plains and is evaluated based on the numerical results thus obtained. It is shown that the proposed technique is viable and performs satisfactorily, with the use of a conservative value for the field size and the crop statistics from the small political subdivision level, when the estimated stratum variances were compared to those obtained using the LANDSAT data.

Foundational Principles for Large-Scale Inference: Illustrations Through Correlation Mining.

PubMed

Hero, Alfred O; Rajaratnam, Bala

2016-01-01

When can reliable inference be drawn in fue "Big Data" context? This paper presents a framework for answering this fundamental question in the context of correlation mining, wifu implications for general large scale inference. In large scale data applications like genomics, connectomics, and eco-informatics fue dataset is often variable-rich but sample-starved: a regime where the number n of acquired samples (statistical replicates) is far fewer than fue number p of observed variables (genes, neurons, voxels, or chemical constituents). Much of recent work has focused on understanding the computational complexity of proposed methods for "Big Data". Sample complexity however has received relatively less attention, especially in the setting when the sample size n is fixed, and the dimension p grows without bound. To address fuis gap, we develop a unified statistical framework that explicitly quantifies the sample complexity of various inferential tasks. Sampling regimes can be divided into several categories: 1) the classical asymptotic regime where fue variable dimension is fixed and fue sample size goes to infinity; 2) the mixed asymptotic regime where both variable dimension and sample size go to infinity at comparable rates; 3) the purely high dimensional asymptotic regime where the variable dimension goes to infinity and the sample size is fixed. Each regime has its niche but only the latter regime applies to exa cale data dimension. We illustrate this high dimensional framework for the problem of correlation mining, where it is the matrix of pairwise and partial correlations among the variables fua t are of interest. Correlation mining arises in numerous applications and subsumes the regression context as a special case. we demonstrate various regimes of correlation mining based on the unifying perspective of high dimensional learning rates and sample complexity for different structured covariance models and different inference tasks.

Volatility measurement with directional change in Chinese stock market: Statistical property and investment strategy

NASA Astrophysics Data System (ADS)

Ma, Junjun; Xiong, Xiong; He, Feng; Zhang, Wei

2017-04-01

The stock price fluctuation is studied in this paper with intrinsic time perspective. The event, directional change (DC) or overshoot, are considered as time scale of price time series. With this directional change law, its corresponding statistical properties and parameter estimation is tested in Chinese stock market. Furthermore, a directional change trading strategy is proposed for invest in the market portfolio in Chinese stock market, and both in-sample and out-of-sample performance are compared among the different method of model parameter estimation. We conclude that DC method can capture important fluctuations in Chinese stock market and gain profit due to the statistical property that average upturn overshoot size is bigger than average downturn directional change size. The optimal parameter of DC method is not fixed and we obtained 1.8% annual excess return with this DC-based trading strategy.

A general approach for sample size calculation for the three-arm 'gold standard' non-inferiority design.

PubMed

Stucke, Kathrin; Kieser, Meinhard

2012-12-10

In the three-arm 'gold standard' non-inferiority design, an experimental treatment, an active reference, and a placebo are compared. This design is becoming increasingly popular, and it is, whenever feasible, recommended for use by regulatory guidelines. We provide a general method to calculate the required sample size for clinical trials performed in this design. As special cases, the situations of continuous, binary, and Poisson distributed outcomes are explored. Taking into account the correlation structure of the involved test statistics, the proposed approach leads to considerable savings in sample size as compared with application of ad hoc methods for all three scale levels. Furthermore, optimal sample size allocation ratios are determined that result in markedly smaller total sample sizes as compared with equal assignment. As optimal allocation makes the active treatment groups larger than the placebo group, implementation of the proposed approach is also desirable from an ethical viewpoint. Copyright © 2012 John Wiley & Sons, Ltd.

Sampling methods for amphibians in streams in the Pacific Northwest.

Treesearch

R. Bruce Bury; Paul Stephen Corn

1991-01-01

Methods describing how to sample aquatic and semiaquatic amphibians in small streams and headwater habitats in the Pacific Northwest are presented. We developed a technique that samples 10-meter stretches of selected streams, which was adequate to detect presence or absence of amphibian species and provided sample sizes statistically sufficient to compare abundance of...

Pedagogical Simulation of Sampling Distributions and the Central Limit Theorem

ERIC Educational Resources Information Center

Hagtvedt, Reidar; Jones, Gregory Todd; Jones, Kari

2007-01-01

Students often find the fact that a sample statistic is a random variable very hard to grasp. Even more mysterious is why a sample mean should become ever more Normal as the sample size increases. This simulation tool is meant to illustrate the process, thereby giving students some intuitive grasp of the relationship between a parent population…

Fast and accurate imputation of summary statistics enhances evidence of functional enrichment.

PubMed

Pasaniuc, Bogdan; Zaitlen, Noah; Shi, Huwenbo; Bhatia, Gaurav; Gusev, Alexander; Pickrell, Joseph; Hirschhorn, Joel; Strachan, David P; Patterson, Nick; Price, Alkes L

2014-10-15

Imputation using external reference panels (e.g. 1000 Genomes) is a widely used approach for increasing power in genome-wide association studies and meta-analysis. Existing hidden Markov models (HMM)-based imputation approaches require individual-level genotypes. Here, we develop a new method for Gaussian imputation from summary association statistics, a type of data that is becoming widely available. In simulations using 1000 Genomes (1000G) data, this method recovers 84% (54%) of the effective sample size for common (>5%) and low-frequency (1-5%) variants [increasing to 87% (60%) when summary linkage disequilibrium information is available from target samples] versus the gold standard of 89% (67%) for HMM-based imputation, which cannot be applied to summary statistics. Our approach accounts for the limited sample size of the reference panel, a crucial step to eliminate false-positive associations, and it is computationally very fast. As an empirical demonstration, we apply our method to seven case-control phenotypes from the Wellcome Trust Case Control Consortium (WTCCC) data and a study of height in the British 1958 birth cohort (1958BC). Gaussian imputation from summary statistics recovers 95% (105%) of the effective sample size (as quantified by the ratio of [Formula: see text] association statistics) compared with HMM-based imputation from individual-level genotypes at the 227 (176) published single nucleotide polymorphisms (SNPs) in the WTCCC (1958BC height) data. In addition, for publicly available summary statistics from large meta-analyses of four lipid traits, we publicly release imputed summary statistics at 1000G SNPs, which could not have been obtained using previously published methods, and demonstrate their accuracy by masking subsets of the data. We show that 1000G imputation using our approach increases the magnitude and statistical evidence of enrichment at genic versus non-genic loci for these traits, as compared with an analysis without 1000G imputation. Thus, imputation of summary statistics will be a valuable tool in future functional enrichment analyses. Publicly available software package available at http://bogdan.bioinformatics.ucla.edu/software/. bpasaniuc@mednet.ucla.edu or aprice@hsph.harvard.edu Supplementary materials are available at Bioinformatics online. © The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.

Min and Max Exponential Extreme Interval Values and Statistics

ERIC Educational Resources Information Center

Jance, Marsha; Thomopoulos, Nick

2009-01-01

The extreme interval values and statistics (expected value, median, mode, standard deviation, and coefficient of variation) for the smallest (min) and largest (max) values of exponentially distributed variables with parameter ? = 1 are examined for different observation (sample) sizes. An extreme interval value g[subscript a] is defined as a…

Lunar soils grain size catalog

NASA Technical Reports Server (NTRS)

Graf, John C.

1993-01-01

This catalog compiles every available grain size distribution for Apollo surface soils, trench samples, cores, and Luna 24 soils. Original laboratory data are tabled, and cumulative weight distribution curves and histograms are plotted. Standard statistical parameters are calculated using the method of moments. Photos and location comments describe the sample environment and geological setting. This catalog can help researchers describe the geotechnical conditions and site variability of the lunar surface essential to the design of a lunar base.

Brain size growth in wild and captive chimpanzees (Pan troglodytes).

PubMed

Cofran, Zachary

2018-05-24

Despite many studies of chimpanzee brain size growth, intraspecific variation is under-explored. Brain size data from chimpanzees of the Taï Forest and the Yerkes Primate Research Center enable a unique glimpse into brain growth variation as age at death is known for individuals, allowing cross-sectional growth curves to be estimated. Because Taï chimpanzees are from the wild but Yerkes apes are captive, potential environmental effects on neural development can also be explored. Previous research has revealed differences in growth and health between wild and captive primates, but such habitat effects have yet to be investigated for brain growth. Here, I use an iterative curve fitting procedure to estimate brain growth and regression parameters for each population, statistically comparing growth models using bootstrapped confidence intervals. Yerkes and Taï brain sizes overlap at all ages, although the sole Taï newborn is at the low end of captive neonatal variation. Growth rate and duration are statistically indistinguishable between the two populations. Resampling the Yerkes sample to match the Taï sample size and age group composition shows that ontogenetic variation in the two groups are remarkably similar despite the latter's limited size. Best fit growth curves for each sample indicate cessation of brain size growth at around 2 years, earlier than has previously been reported. The overall similarity between wild and captive chimpanzees points to the canalization of brain growth in this species. © 2018 Wiley Periodicals, Inc.

A Third Moment Adjusted Test Statistic for Small Sample Factor Analysis.

PubMed

Lin, Johnny; Bentler, Peter M

2012-01-01

Goodness of fit testing in factor analysis is based on the assumption that the test statistic is asymptotically chi-square; but this property may not hold in small samples even when the factors and errors are normally distributed in the population. Robust methods such as Browne's asymptotically distribution-free method and Satorra Bentler's mean scaling statistic were developed under the presumption of non-normality in the factors and errors. This paper finds new application to the case where factors and errors are normally distributed in the population but the skewness of the obtained test statistic is still high due to sampling error in the observed indicators. An extension of Satorra Bentler's statistic is proposed that not only scales the mean but also adjusts the degrees of freedom based on the skewness of the obtained test statistic in order to improve its robustness under small samples. A simple simulation study shows that this third moment adjusted statistic asymptotically performs on par with previously proposed methods, and at a very small sample size offers superior Type I error rates under a properly specified model. Data from Mardia, Kent and Bibby's study of students tested for their ability in five content areas that were either open or closed book were used to illustrate the real-world performance of this statistic.

An Analytic Solution to the Computation of Power and Sample Size for Genetic Association Studies under a Pleiotropic Mode of Inheritance.

PubMed

Gordon, Derek; Londono, Douglas; Patel, Payal; Kim, Wonkuk; Finch, Stephen J; Heiman, Gary A

2016-01-01

Our motivation here is to calculate the power of 3 statistical tests used when there are genetic traits that operate under a pleiotropic mode of inheritance and when qualitative phenotypes are defined by use of thresholds for the multiple quantitative phenotypes. Specifically, we formulate a multivariate function that provides the probability that an individual has a vector of specific quantitative trait values conditional on having a risk locus genotype, and we apply thresholds to define qualitative phenotypes (affected, unaffected) and compute penetrances and conditional genotype frequencies based on the multivariate function. We extend the analytic power and minimum-sample-size-necessary (MSSN) formulas for 2 categorical data-based tests (genotype, linear trend test [LTT]) of genetic association to the pleiotropic model. We further compare the MSSN of the genotype test and the LTT with that of a multivariate ANOVA (Pillai). We approximate the MSSN for statistics by linear models using a factorial design and ANOVA. With ANOVA decomposition, we determine which factors most significantly change the power/MSSN for all statistics. Finally, we determine which test statistics have the smallest MSSN. In this work, MSSN calculations are for 2 traits (bivariate distributions) only (for illustrative purposes). We note that the calculations may be extended to address any number of traits. Our key findings are that the genotype test usually has lower MSSN requirements than the LTT. More inclusive thresholds (top/bottom 25% vs. top/bottom 10%) have higher sample size requirements. The Pillai test has a much larger MSSN than both the genotype test and the LTT, as a result of sample selection. With these formulas, researchers can specify how many subjects they must collect to localize genes for pleiotropic phenotypes. © 2017 S. Karger AG, Basel.

A Systematic Review of Surgical Randomized Controlled Trials: Part 2. Funding Source, Conflict of Interest, and Sample Size in Plastic Surgery.

PubMed

Voineskos, Sophocles H; Coroneos, Christopher J; Ziolkowski, Natalia I; Kaur, Manraj N; Banfield, Laura; Meade, Maureen O; Chung, Kevin C; Thoma, Achilleas; Bhandari, Mohit

2016-02-01

The authors examined industry support, conflict of interest, and sample size in plastic surgery randomized controlled trials that compared surgical interventions. They hypothesized that industry-funded trials demonstrate statistically significant outcomes more often, and randomized controlled trials with small sample sizes report statistically significant results more frequently. An electronic search identified randomized controlled trials published between 2000 and 2013. Independent reviewers assessed manuscripts and performed data extraction. Funding source, conflict of interest, primary outcome direction, and sample size were examined. Chi-squared and independent-samples t tests were used in the analysis. The search identified 173 randomized controlled trials, of which 100 (58 percent) did not acknowledge funding status. A relationship between funding source and trial outcome direction was not observed. Both funding status and conflict of interest reporting improved over time. Only 24 percent (six of 25) of industry-funded randomized controlled trials reported authors to have independent control of data and manuscript contents. The mean number of patients randomized was 73 per trial (median, 43, minimum, 3, maximum, 936). Small trials were not found to be positive more often than large trials (p = 0.87). Randomized controlled trials with small sample size were common; however, this provides great opportunity for the field to engage in further collaboration and produce larger, more definitive trials. Reporting of trial funding and conflict of interest is historically poor, but it greatly improved over the study period. Underreporting at author and journal levels remains a limitation when assessing the relationship between funding source and trial outcomes. Improved reporting and manuscript control should be goals that both authors and journals can actively achieve.

DESCARTES' RULE OF SIGNS AND THE IDENTIFIABILITY OF POPULATION DEMOGRAPHIC MODELS FROM GENOMIC VARIATION DATA.

PubMed

Bhaskar, Anand; Song, Yun S

2014-01-01

The sample frequency spectrum (SFS) is a widely-used summary statistic of genomic variation in a sample of homologous DNA sequences. It provides a highly efficient dimensional reduction of large-scale population genomic data and its mathematical dependence on the underlying population demography is well understood, thus enabling the development of efficient inference algorithms. However, it has been recently shown that very different population demographies can actually generate the same SFS for arbitrarily large sample sizes. Although in principle this nonidentifiability issue poses a thorny challenge to statistical inference, the population size functions involved in the counterexamples are arguably not so biologically realistic. Here, we revisit this problem and examine the identifiability of demographic models under the restriction that the population sizes are piecewise-defined where each piece belongs to some family of biologically-motivated functions. Under this assumption, we prove that the expected SFS of a sample uniquely determines the underlying demographic model, provided that the sample is sufficiently large. We obtain a general bound on the sample size sufficient for identifiability; the bound depends on the number of pieces in the demographic model and also on the type of population size function in each piece. In the cases of piecewise-constant, piecewise-exponential and piecewise-generalized-exponential models, which are often assumed in population genomic inferences, we provide explicit formulas for the bounds as simple functions of the number of pieces. Lastly, we obtain analogous results for the "folded" SFS, which is often used when there is ambiguity as to which allelic type is ancestral. Our results are proved using a generalization of Descartes' rule of signs for polynomials to the Laplace transform of piecewise continuous functions.

DESCARTES’ RULE OF SIGNS AND THE IDENTIFIABILITY OF POPULATION DEMOGRAPHIC MODELS FROM GENOMIC VARIATION DATA1

PubMed Central

Bhaskar, Anand; Song, Yun S.

2016-01-01

The sample frequency spectrum (SFS) is a widely-used summary statistic of genomic variation in a sample of homologous DNA sequences. It provides a highly efficient dimensional reduction of large-scale population genomic data and its mathematical dependence on the underlying population demography is well understood, thus enabling the development of efficient inference algorithms. However, it has been recently shown that very different population demographies can actually generate the same SFS for arbitrarily large sample sizes. Although in principle this nonidentifiability issue poses a thorny challenge to statistical inference, the population size functions involved in the counterexamples are arguably not so biologically realistic. Here, we revisit this problem and examine the identifiability of demographic models under the restriction that the population sizes are piecewise-defined where each piece belongs to some family of biologically-motivated functions. Under this assumption, we prove that the expected SFS of a sample uniquely determines the underlying demographic model, provided that the sample is sufficiently large. We obtain a general bound on the sample size sufficient for identifiability; the bound depends on the number of pieces in the demographic model and also on the type of population size function in each piece. In the cases of piecewise-constant, piecewise-exponential and piecewise-generalized-exponential models, which are often assumed in population genomic inferences, we provide explicit formulas for the bounds as simple functions of the number of pieces. Lastly, we obtain analogous results for the “folded” SFS, which is often used when there is ambiguity as to which allelic type is ancestral. Our results are proved using a generalization of Descartes’ rule of signs for polynomials to the Laplace transform of piecewise continuous functions. PMID:28018011

A Statistical Analysis Plan to Support the Joint Forward Area Air Defense Test.

DTIC Science & Technology

1984-08-02

hy estahlishing a specific significance level prior to performing the statistical test (traditionally a levels are set at .01 or .05). What is often...undesirable increase in 8. For constant a levels , the power (I - 8) of a statistical test can he increased by Increasing the sample size of the test. fRef...ANOVA Iparison Test on MOP I=--ferences Exist AmongF "Upon MOP "A" Factor I "A" Factor I 1MOP " A " Levels ? I . I I I _ _ ________ IPerform k-Sample Com- I

A rational approach to legacy data validation when transitioning between electronic health record systems.

PubMed

Pageler, Natalie M; Grazier G'Sell, Max Jacob; Chandler, Warren; Mailes, Emily; Yang, Christine; Longhurst, Christopher A

2016-09-01

The objective of this project was to use statistical techniques to determine the completeness and accuracy of data migrated during electronic health record conversion. Data validation during migration consists of mapped record testing and validation of a sample of the data for completeness and accuracy. We statistically determined a randomized sample size for each data type based on the desired confidence level and error limits. The only error identified in the post go-live period was a failure to migrate some clinical notes, which was unrelated to the validation process. No errors in the migrated data were found during the 12- month post-implementation period. Compared to the typical industry approach, we have demonstrated that a statistical approach to sampling size for data validation can ensure consistent confidence levels while maximizing efficiency of the validation process during a major electronic health record conversion. © The Author 2016. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For Permissions, please email: journals.permissions@oup.com.

Sample size considerations for clinical research studies in nuclear cardiology.

PubMed

Chiuzan, Cody; West, Erin A; Duong, Jimmy; Cheung, Ken Y K; Einstein, Andrew J

2015-12-01

Sample size calculation is an important element of research design that investigators need to consider in the planning stage of the study. Funding agencies and research review panels request a power analysis, for example, to determine the minimum number of subjects needed for an experiment to be informative. Calculating the right sample size is crucial to gaining accurate information and ensures that research resources are used efficiently and ethically. The simple question "How many subjects do I need?" does not always have a simple answer. Before calculating the sample size requirements, a researcher must address several aspects, such as purpose of the research (descriptive or comparative), type of samples (one or more groups), and data being collected (continuous or categorical). In this article, we describe some of the most frequent methods for calculating the sample size with examples from nuclear cardiology research, including for t tests, analysis of variance (ANOVA), non-parametric tests, correlation, Chi-squared tests, and survival analysis. For the ease of implementation, several examples are also illustrated via user-friendly free statistical software.

Intercomparison of textural parameters of intertidal sediments generated by different statistical procedures, and implications for a unifying descriptive nomenclature

NASA Astrophysics Data System (ADS)

Fan, Daidu; Tu, Junbiao; Cai, Guofu; Shang, Shuai

2015-06-01

Grain-size analysis is a basic routine in sedimentology and related fields, but diverse methods of sample collection, processing and statistical analysis often make direct comparisons and interpretations difficult or even impossible. In this paper, 586 published grain-size datasets from the Qiantang Estuary (East China Sea) sampled and analyzed by the same procedures were merged and their textural parameters calculated by a percentile and two moment methods. The aim was to explore which of the statistical procedures performed best in the discrimination of three distinct sedimentary units on the tidal flats of the middle Qiantang Estuary. A Gaussian curve-fitting method served to simulate mixtures of two normal populations having different modal sizes, sorting values and size distributions, enabling a better understanding of the impact of finer tail components on textural parameters, as well as the proposal of a unifying descriptive nomenclature. The results show that percentile and moment procedures yield almost identical results for mean grain size, and that sorting values are also highly correlated. However, more complex relationships exist between percentile and moment skewness (kurtosis), changing from positive to negative correlations when the proportions of the finer populations decrease below 35% (10%). This change results from the overweighting of tail components in moment statistics, which stands in sharp contrast to the underweighting or complete amputation of small tail components by the percentile procedure. Intercomparisons of bivariate plots suggest an advantage of the Friedman & Johnson moment procedure over the McManus moment method in terms of the description of grain-size distributions, and over the percentile method by virtue of a greater sensitivity to small variations in tail components. The textural parameter scalings of Folk & Ward were translated into their Friedman & Johnson moment counterparts by application of mathematical functions derived by regression analysis of measured and modeled grain-size data, or by determining the abscissa values of intersections between auxiliary lines running parallel to the x-axis and vertical lines corresponding to the descriptive percentile limits along the ordinate of representative bivariate plots. Twofold limits were extrapolated for the moment statistics in relation to single descriptive terms in the cases of skewness and kurtosis by considering both positive and negative correlations between percentile and moment statistics. The extrapolated descriptive scalings were further validated by examining entire size-frequency distributions simulated by mixing two normal populations of designated modal size and sorting values, but varying in mixing ratios. These were found to match well in most of the proposed scalings, although platykurtic and very platykurtic categories were questionable when the proportion of the finer population was below 5%. Irrespective of the statistical procedure, descriptive nomenclatures should therefore be cautiously used when tail components contribute less than 5% to grain-size distributions.

Using the Sampling Margin of Error to Assess the Interpretative Validity of Student Evaluations of Teaching

ERIC Educational Resources Information Center

James, David E.; Schraw, Gregory; Kuch, Fred

2015-01-01

We present an equation, derived from standard statistical theory, that can be used to estimate sampling margin of error for student evaluations of teaching (SETs). We use the equation to examine the effect of sample size, response rates and sample variability on the estimated sampling margin of error, and present results in four tables that allow…

Sample Size for Tablet Compression and Capsule Filling Events During Process Validation.

PubMed

Charoo, Naseem Ahmad; Durivage, Mark; Rahman, Ziyaur; Ayad, Mohamad Haitham

2017-12-01

During solid dosage form manufacturing, the uniformity of dosage units (UDU) is ensured by testing samples at 2 stages, that is, blend stage and tablet compression or capsule/powder filling stage. The aim of this work is to propose a sample size selection approach based on quality risk management principles for process performance qualification (PPQ) and continued process verification (CPV) stages by linking UDU to potential formulation and process risk factors. Bayes success run theorem appeared to be the most appropriate approach among various methods considered in this work for computing sample size for PPQ. The sample sizes for high-risk (reliability level of 99%), medium-risk (reliability level of 95%), and low-risk factors (reliability level of 90%) were estimated to be 299, 59, and 29, respectively. Risk-based assignment of reliability levels was supported by the fact that at low defect rate, the confidence to detect out-of-specification units would decrease which must be supplemented with an increase in sample size to enhance the confidence in estimation. Based on level of knowledge acquired during PPQ and the level of knowledge further required to comprehend process, sample size for CPV was calculated using Bayesian statistics to accomplish reduced sampling design for CPV. Copyright © 2017 American Pharmacists Association®. Published by Elsevier Inc. All rights reserved.

Small sample sizes in the study of ontogenetic allometry; implications for palaeobiology

PubMed Central

Vavrek, Matthew J.

2015-01-01

Quantitative morphometric analyses, particularly ontogenetic allometry, are common methods used in quantifying shape, and changes therein, in both extinct and extant organisms. Due to incompleteness and the potential for restricted sample sizes in the fossil record, palaeobiological analyses of allometry may encounter higher rates of error. Differences in sample size between fossil and extant studies and any resulting effects on allometric analyses have not been thoroughly investigated, and a logical lower threshold to sample size is not clear. Here we show that studies based on fossil datasets have smaller sample sizes than those based on extant taxa. A similar pattern between vertebrates and invertebrates indicates this is not a problem unique to either group, but common to both. We investigate the relationship between sample size, ontogenetic allometric relationship and statistical power using an empirical dataset of skull measurements of modern Alligator mississippiensis. Across a variety of subsampling techniques, used to simulate different taphonomic and/or sampling effects, smaller sample sizes gave less reliable and more variable results, often with the result that allometric relationships will go undetected due to Type II error (failure to reject the null hypothesis). This may result in a false impression of fewer instances of positive/negative allometric growth in fossils compared to living organisms. These limitations are not restricted to fossil data and are equally applicable to allometric analyses of rare extant taxa. No mathematically derived minimum sample size for ontogenetic allometric studies is found; rather results of isometry (but not necessarily allometry) should not be viewed with confidence at small sample sizes. PMID:25780770

Optimal sample sizes for the design of reliability studies: power consideration.

PubMed

Shieh, Gwowen

2014-09-01

Intraclass correlation coefficients are used extensively to measure the reliability or degree of resemblance among group members in multilevel research. This study concerns the problem of the necessary sample size to ensure adequate statistical power for hypothesis tests concerning the intraclass correlation coefficient in the one-way random-effects model. In view of the incomplete and problematic numerical results in the literature, the approximate sample size formula constructed from Fisher's transformation is reevaluated and compared with an exact approach across a wide range of model configurations. These comprehensive examinations showed that the Fisher transformation method is appropriate only under limited circumstances, and therefore it is not recommended as a general method in practice. For advance design planning of reliability studies, the exact sample size procedures are fully described and illustrated for various allocation and cost schemes. Corresponding computer programs are also developed to implement the suggested algorithms.

Random Distribution Pattern and Non-adaptivity of Genome Size in a Highly Variable Population of Festuca pallens

PubMed Central

Šmarda, Petr; Bureš, Petr; Horová, Lucie

2007-01-01

Background and Aims The spatial and statistical distribution of genome sizes and the adaptivity of genome size to some types of habitat, vegetation or microclimatic conditions were investigated in a tetraploid population of Festuca pallens. The population was previously documented to vary highly in genome size and is assumed as a model for the study of the initial stages of genome size differentiation. Methods Using DAPI flow cytometry, samples were measured repeatedly with diploid Festuca pallens as the internal standard. Altogether 172 plants from 57 plots (2·25 m2), distributed in contrasting habitats over the whole locality in South Moravia, Czech Republic, were sampled. The differences in DNA content were confirmed by the double peaks of simultaneously measured samples. Key Results At maximum, a 1·115-fold difference in genome size was observed. The statistical distribution of genome sizes was found to be continuous and best fits the extreme (Gumbel) distribution with rare occurrences of extremely large genomes (positive-skewed), as it is similar for the log-normal distribution of the whole Angiosperms. Even plants from the same plot frequently varied considerably in genome size and the spatial distribution of genome sizes was generally random and unautocorrelated (P > 0·05). The observed spatial pattern and the overall lack of correlations of genome size with recognized vegetation types or microclimatic conditions indicate the absence of ecological adaptivity of genome size in the studied population. Conclusions These experimental data on intraspecific genome size variability in Festuca pallens argue for the absence of natural selection and the selective non-significance of genome size in the initial stages of genome size differentiation, and corroborate the current hypothetical model of genome size evolution in Angiosperms (Bennetzen et al., 2005, Annals of Botany 95: 127–132). PMID:17565968

Evaluation of statistical designs in phase I expansion cohorts: the Dana-Farber/Harvard Cancer Center experience.

PubMed

Dahlberg, Suzanne E; Shapiro, Geoffrey I; Clark, Jeffrey W; Johnson, Bruce E

2014-07-01

Phase I trials have traditionally been designed to assess toxicity and establish phase II doses with dose-finding studies and expansion cohorts but are frequently exceeding the traditional sample size to further assess endpoints in specific patient subsets. The scientific objectives of phase I expansion cohorts and their evolving role in the current era of targeted therapies have yet to be systematically examined. Adult therapeutic phase I trials opened within Dana-Farber/Harvard Cancer Center (DF/HCC) from 1988 to 2012 were identified for sample size details. Statistical designs and study objectives of those submitted in 2011 were reviewed for expansion cohort details. Five hundred twenty-two adult therapeutic phase I trials were identified during the 25 years. The average sample size of a phase I study has increased from 33.8 patients to 73.1 patients over that time. The proportion of trials with planned enrollment of 50 or fewer patients dropped from 93.0% during the time period 1988 to 1992 to 46.0% between 2008 and 2012; at the same time, the proportion of trials enrolling 51 to 100 patients and more than 100 patients increased from 5.3% and 1.8%, respectively, to 40.5% and 13.5% (χ(2) test, two-sided P < .001). Sixteen of the 60 trials (26.7%) in 2011 enrolled patients to three or more sub-cohorts in the expansion phase. Sixty percent of studies provided no statistical justification of the sample size, although 91.7% of trials stated response as an objective. Our data suggest that phase I studies have dramatically changed in size and scientific scope within the last decade. Additional studies addressing the implications of this trend on research processes, ethical concerns, and resource burden are needed. © The Author 2014. Published by Oxford University Press. All rights reserved.

Surrogate and clinical endpoints for studies in peripheral artery occlusive disease: Are statistics the brakes?

PubMed

Waliszewski, Matthias W; Redlich, Ulf; Breul, Victor; Tautenhahn, Jörg

2017-04-30

The aim of this review is to present the available clinical and surrogate endpoints that may be used in future studies performed in patients with peripheral artery occlusive disease (PAOD). Importantly, we describe statistical limitations of the most commonly used endpoints and offer some guidance with respect to study design for a given sample size. The proposed endpoints may be used in studies using surgical or interventional revascularization and/or drug treatments. Considering recently published study endpoints and designs, the usefulness of these endpoints for reimbursement is evaluated. Based on these potential study endpoints and patient sample size estimates with different non-inferiority or tests for difference hypotheses, a rating relative to their corresponding reimbursement values is attempted. As regards the benefit for the patients and for the payers, walking distance and the ankle brachial index (ABI) are the most feasible endpoints in a relatively small study samples given that other non-vascular impact factors can be controlled. Angiographic endpoints such as minimal lumen diameter (MLD) do not seem useful from a reimbursement standpoint despite their intuitiveness. Other surrogate endpoints, such as transcutaneous oxygen tension measurements, have yet to be established as useful endpoints in reasonably sized studies with patients with critical limb ischemia (CLI). From a reimbursement standpoint, WD and ABI are effective endpoints for a moderate study sample size given that non-vascular confounding factors can be controlled.

Monitoring the impact of Bt maize on butterflies in the field: estimation of required sample sizes.

PubMed

Lang, Andreas

2004-01-01

The monitoring of genetically modified organisms (GMOs) after deliberate release is important in order to assess and evaluate possible environmental effects. Concerns have been raised that the transgenic crop, Bt maize, may affect butterflies occurring in field margins. Therefore, a monitoring of butterflies was suggested accompanying the commercial cultivation of Bt maize. In this study, baseline data on the butterfly species and their abundance in maize field margins is presented together with implications for butterfly monitoring. The study was conducted in Bavaria, South Germany, between 2000-2002. A total of 33 butterfly species was recorded in field margins. A small number of species dominated the community, and butterflies observed were mostly common species. Observation duration was the most important factor influencing the monitoring results. Field margin size affected the butterfly abundance, and habitat diversity had a tendency to influence species richness. Sample size and statistical power analyses indicated that a sample size in the range of 75 to 150 field margins for treatment (transgenic maize) and control (conventional maize) would detect (power of 80%) effects larger than 15% in species richness and the butterfly abundance pooled across species. However, a much higher number of field margins must be sampled in order to achieve a higher statistical power, to detect smaller effects, and to monitor single butterfly species.

Using the Bootstrap Method to Evaluate the Critical Range of Misfit for Polytomous Rasch Fit Statistics.

PubMed

Seol, Hyunsoo

2016-06-01

The purpose of this study was to apply the bootstrap procedure to evaluate how the bootstrapped confidence intervals (CIs) for polytomous Rasch fit statistics might differ according to sample sizes and test lengths in comparison with the rule-of-thumb critical value of misfit. A total of 25 simulated data sets were generated to fit the Rasch measurement and then a total of 1,000 replications were conducted to compute the bootstrapped CIs under each of 25 testing conditions. The results showed that rule-of-thumb critical values for assessing the magnitude of misfit were not applicable because the infit and outfit mean square error statistics showed different magnitudes of variability over testing conditions and the standardized fit statistics did not exactly follow the standard normal distribution. Further, they also do not share the same critical range for the item and person misfit. Based on the results of the study, the bootstrapped CIs can be used to identify misfitting items or persons as they offer a reasonable alternative solution, especially when the distributions of the infit and outfit statistics are not well known and depend on sample size. © The Author(s) 2016.

tscvh R Package: Computational of the two samples test on microarray-sequencing data

NASA Astrophysics Data System (ADS)

Fajriyah, Rohmatul; Rosadi, Dedi

2017-12-01

We present a new R package, a tscvh (two samples cross-variance homogeneity), as we called it. This package is a software of the cross-variance statistical test which has been proposed and introduced by Fajriyah ([3] and [4]), based on the cross-variance concept. The test can be used as an alternative test for the significance difference between two means when sample size is small, the situation which is usually appeared in the bioinformatics research. Based on its statistical distribution, the p-value can be also provided. The package is built under a homogeneity of variance between samples.

Sample size considerations for paired experimental design with incomplete observations of continuous outcomes.

PubMed

Zhu, Hong; Xu, Xiaohan; Ahn, Chul

2017-01-01

Paired experimental design is widely used in clinical and health behavioral studies, where each study unit contributes a pair of observations. Investigators often encounter incomplete observations of paired outcomes in the data collected. Some study units contribute complete pairs of observations, while the others contribute either pre- or post-intervention observations. Statistical inference for paired experimental design with incomplete observations of continuous outcomes has been extensively studied in literature. However, sample size method for such study design is sparsely available. We derive a closed-form sample size formula based on the generalized estimating equation approach by treating the incomplete observations as missing data in a linear model. The proposed method properly accounts for the impact of mixed structure of observed data: a combination of paired and unpaired outcomes. The sample size formula is flexible to accommodate different missing patterns, magnitude of missingness, and correlation parameter values. We demonstrate that under complete observations, the proposed generalized estimating equation sample size estimate is the same as that based on the paired t-test. In the presence of missing data, the proposed method would lead to a more accurate sample size estimate comparing with the crude adjustment. Simulation studies are conducted to evaluate the finite-sample performance of the generalized estimating equation sample size formula. A real application example is presented for illustration.

The cost of large numbers of hypothesis tests on power, effect size and sample size.

PubMed

Lazzeroni, L C; Ray, A

2012-01-01

Advances in high-throughput biology and computer science are driving an exponential increase in the number of hypothesis tests in genomics and other scientific disciplines. Studies using current genotyping platforms frequently include a million or more tests. In addition to the monetary cost, this increase imposes a statistical cost owing to the multiple testing corrections needed to avoid large numbers of false-positive results. To safeguard against the resulting loss of power, some have suggested sample sizes on the order of tens of thousands that can be impractical for many diseases or may lower the quality of phenotypic measurements. This study examines the relationship between the number of tests on the one hand and power, detectable effect size or required sample size on the other. We show that once the number of tests is large, power can be maintained at a constant level, with comparatively small increases in the effect size or sample size. For example at the 0.05 significance level, a 13% increase in sample size is needed to maintain 80% power for ten million tests compared with one million tests, whereas a 70% increase in sample size is needed for 10 tests compared with a single test. Relative costs are less when measured by increases in the detectable effect size. We provide an interactive Excel calculator to compute power, effect size or sample size when comparing study designs or genome platforms involving different numbers of hypothesis tests. The results are reassuring in an era of extreme multiple testing.

A Civilian/Military Trauma Institute: National Trauma Coordinating Center

DTIC Science & Technology

2015-12-01

zip codes was used in “proximity to violence” analysis. Data were analyzed using SPSS (version 20.0, SPSS Inc., Chicago, IL). Multivariable linear...number of adverse events and serious events was not statistically higher in one group, the incidence of deep venous thrombosis (DVT) was statistically ...subjects the lack of statistical difference on multivariate analysis may be related to an underpowered sample size. It was recommended that the

Powerful Statistical Inference for Nested Data Using Sufficient Summary Statistics

PubMed Central

Dowding, Irene; Haufe, Stefan

2018-01-01

Hierarchically-organized data arise naturally in many psychology and neuroscience studies. As the standard assumption of independent and identically distributed samples does not hold for such data, two important problems are to accurately estimate group-level effect sizes, and to obtain powerful statistical tests against group-level null hypotheses. A common approach is to summarize subject-level data by a single quantity per subject, which is often the mean or the difference between class means, and treat these as samples in a group-level t-test. This “naive” approach is, however, suboptimal in terms of statistical power, as it ignores information about the intra-subject variance. To address this issue, we review several approaches to deal with nested data, with a focus on methods that are easy to implement. With what we call the sufficient-summary-statistic approach, we highlight a computationally efficient technique that can improve statistical power by taking into account within-subject variances, and we provide step-by-step instructions on how to apply this approach to a number of frequently-used measures of effect size. The properties of the reviewed approaches and the potential benefits over a group-level t-test are quantitatively assessed on simulated data and demonstrated on EEG data from a simulated-driving experiment. PMID:29615885

Computing physical properties with quantum Monte Carlo methods with statistical fluctuations independent of system size.

PubMed

Assaraf, Roland

2014-12-01

We show that the recently proposed correlated sampling without reweighting procedure extends the locality (asymptotic independence of the system size) of a physical property to the statistical fluctuations of its estimator. This makes the approach potentially vastly more efficient for computing space-localized properties in large systems compared with standard correlated methods. A proof is given for a large collection of noninteracting fragments. Calculations on hydrogen chains suggest that this behavior holds not only for systems displaying short-range correlations, but also for systems with long-range correlations.

Incorporating Biological Knowledge into Evaluation of Casual Regulatory Hypothesis

NASA Technical Reports Server (NTRS)

Chrisman, Lonnie; Langley, Pat; Bay, Stephen; Pohorille, Andrew; DeVincenzi, D. (Technical Monitor)

2002-01-01

Biological data can be scarce and costly to obtain. The small number of samples available typically limits statistical power and makes reliable inference of causal relations extremely difficult. However, we argue that statistical power can be increased substantially by incorporating prior knowledge and data from diverse sources. We present a Bayesian framework that combines information from different sources and we show empirically that this lets one make correct causal inferences with small sample sizes that otherwise would be impossible.

Laser Velocimeter Measurements and Analysis in Turbulent Flows with Combustion. Part 2.

DTIC Science & Technology

1983-07-01

sampling error for 63 this sample size. Mean velocities and turbulence intensi- ties were found to be statistically accurate to ± 1 % and 13%, respectively...Although the statist - ical error was found to be rather small (± 1 % for mean velo- cities and 13% for turbulence intensities), there can be additional...34Computational and Experimental Study of a Captive Annular Eddy," Journal of Fluid Mechanics, Vol. 28, pt. 1 , pp. 43-63, 12 April, 1967. 152 REFERENCES (con’d

Sample size re-assessment leading to a raised sample size does not inflate type I error rate under mild conditions.

PubMed

Broberg, Per

2013-07-19

One major concern with adaptive designs, such as the sample size adjustable designs, has been the fear of inflating the type I error rate. In (Stat Med 23:1023-1038, 2004) it is however proven that when observations follow a normal distribution and the interim result show promise, meaning that the conditional power exceeds 50%, type I error rate is protected. This bound and the distributional assumptions may seem to impose undesirable restrictions on the use of these designs. In (Stat Med 30:3267-3284, 2011) the possibility of going below 50% is explored and a region that permits an increased sample size without inflation is defined in terms of the conditional power at the interim. A criterion which is implicit in (Stat Med 30:3267-3284, 2011) is derived by elementary methods and expressed in terms of the test statistic at the interim to simplify practical use. Mathematical and computational details concerning this criterion are exhibited. Under very general conditions the type I error rate is preserved under sample size adjustable schemes that permit a raise. The main result states that for normally distributed observations raising the sample size when the result looks promising, where the definition of promising depends on the amount of knowledge gathered so far, guarantees the protection of the type I error rate. Also, in the many situations where the test statistic approximately follows a normal law, the deviation from the main result remains negligible. This article provides details regarding the Weibull and binomial distributions and indicates how one may approach these distributions within the current setting. There is thus reason to consider such designs more often, since they offer a means of adjusting an important design feature at little or no cost in terms of error rate.

Intuitive statistics by 8-month-old infants

PubMed Central

Xu, Fei; Garcia, Vashti

2008-01-01

Human learners make inductive inferences based on small amounts of data: we generalize from samples to populations and vice versa. The academic discipline of statistics formalizes these intuitive statistical inferences. What is the origin of this ability? We report six experiments investigating whether 8-month-old infants are “intuitive statisticians.” Our results showed that, given a sample, the infants were able to make inferences about the population from which the sample had been drawn. Conversely, given information about the entire population of relatively small size, the infants were able to make predictions about the sample. Our findings provide evidence that infants possess a powerful mechanism for inductive learning, either using heuristics or basic principles of probability. This ability to make inferences based on samples or information about the population develops early and in the absence of schooling or explicit teaching. Human infants may be rational learners from very early in development. PMID:18378901

Statistical inference involving binomial and negative binomial parameters.

PubMed

García-Pérez, Miguel A; Núñez-Antón, Vicente

2009-05-01

Statistical inference about two binomial parameters implies that they are both estimated by binomial sampling. There are occasions in which one aims at testing the equality of two binomial parameters before and after the occurrence of the first success along a sequence of Bernoulli trials. In these cases, the binomial parameter before the first success is estimated by negative binomial sampling whereas that after the first success is estimated by binomial sampling, and both estimates are related. This paper derives statistical tools to test two hypotheses, namely, that both binomial parameters equal some specified value and that both parameters are equal though unknown. Simulation studies are used to show that in small samples both tests are accurate in keeping the nominal Type-I error rates, and also to determine sample size requirements to detect large, medium, and small effects with adequate power. Additional simulations also show that the tests are sufficiently robust to certain violations of their assumptions.

Sampling and counting genome rearrangement scenarios

PubMed Central

2015-01-01

Background Even for moderate size inputs, there are a tremendous number of optimal rearrangement scenarios, regardless what the model is and which specific question is to be answered. Therefore giving one optimal solution might be misleading and cannot be used for statistical inferring. Statistically well funded methods are necessary to sample uniformly from the solution space and then a small number of samples are sufficient for statistical inferring. Contribution In this paper, we give a mini-review about the state-of-the-art of sampling and counting rearrangement scenarios, focusing on the reversal, DCJ and SCJ models. Above that, we also give a Gibbs sampler for sampling most parsimonious labeling of evolutionary trees under the SCJ model. The method has been implemented and tested on real life data. The software package together with example data can be downloaded from http://www.renyi.hu/~miklosi/SCJ-Gibbs/ PMID:26452124

Decisions from Experience: Why Small Samples?

ERIC Educational Resources Information Center

Hertwig, Ralph; Pleskac, Timothy J.

2010-01-01

In many decisions we cannot consult explicit statistics telling us about the risks involved in our actions. In lieu of such data, we can arrive at an understanding of our dicey options by sampling from them. The size of the samples that we take determines, ceteris paribus, how good our choices will be. Studies of decisions from experience have…

Statistical Inference for Data Adaptive Target Parameters.

PubMed

Hubbard, Alan E; Kherad-Pajouh, Sara; van der Laan, Mark J

2016-05-01

Consider one observes n i.i.d. copies of a random variable with a probability distribution that is known to be an element of a particular statistical model. In order to define our statistical target we partition the sample in V equal size sub-samples, and use this partitioning to define V splits in an estimation sample (one of the V subsamples) and corresponding complementary parameter-generating sample. For each of the V parameter-generating samples, we apply an algorithm that maps the sample to a statistical target parameter. We define our sample-split data adaptive statistical target parameter as the average of these V-sample specific target parameters. We present an estimator (and corresponding central limit theorem) of this type of data adaptive target parameter. This general methodology for generating data adaptive target parameters is demonstrated with a number of practical examples that highlight new opportunities for statistical learning from data. This new framework provides a rigorous statistical methodology for both exploratory and confirmatory analysis within the same data. Given that more research is becoming "data-driven", the theory developed within this paper provides a new impetus for a greater involvement of statistical inference into problems that are being increasingly addressed by clever, yet ad hoc pattern finding methods. To suggest such potential, and to verify the predictions of the theory, extensive simulation studies, along with a data analysis based on adaptively determined intervention rules are shown and give insight into how to structure such an approach. The results show that the data adaptive target parameter approach provides a general framework and resulting methodology for data-driven science.

A statistical analysis of seat belt effectiveness in 1973-1975 model cars involved in towaway crashes. Volume 1

DOT National Transportation Integrated Search

1976-09-01

Standardized injury rates and seat belt effectiveness measures are derived from a probability sample of towaway accidents involving 1973-1975 model cars. The data were collected in five different geographic regions. Weighted sample size available for...

An Analysis of Methods Used to Examine Gender Differences in Computer-Related Behavior.

ERIC Educational Resources Information Center

Kay, Robin

1992-01-01

Review of research investigating gender differences in computer-related behavior examines statistical and methodological flaws. Issues addressed include sample selection, sample size, scale development, scale quality, the use of univariate and multivariate analyses, regressional analysis, construct definition, construct testing, and the…

Supervised classification in the presence of misclassified training data: a Monte Carlo simulation study in the three group case.

PubMed

Bolin, Jocelyn Holden; Finch, W Holmes

2014-01-01

Statistical classification of phenomena into observed groups is very common in the social and behavioral sciences. Statistical classification methods, however, are affected by the characteristics of the data under study. Statistical classification can be further complicated by initial misclassification of the observed groups. The purpose of this study is to investigate the impact of initial training data misclassification on several statistical classification and data mining techniques. Misclassification conditions in the three group case will be simulated and results will be presented in terms of overall as well as subgroup classification accuracy. Results show decreased classification accuracy as sample size, group separation and group size ratio decrease and as misclassification percentage increases with random forests demonstrating the highest accuracy across conditions.

Estimating Standardized Linear Contrasts of Means with Desired Precision

ERIC Educational Resources Information Center

Bonett, Douglas G.

2009-01-01

L. Wilkinson and the Task Force on Statistical Inference (1999) recommended reporting confidence intervals for measures of effect sizes. If the sample size is too small, the confidence interval may be too wide to provide meaningful information. Recently, K. Kelley and J. R. Rausch (2006) used an iterative approach to computer-generate tables of…

Statistical issues in reporting quality data: small samples and casemix variation.

PubMed

Zaslavsky, A M

2001-12-01

To present two key statistical issues that arise in analysis and reporting of quality data. Casemix variation is relevant to quality reporting when the units being measured have differing distributions of patient characteristics that also affect the quality outcome. When this is the case, adjustment using stratification or regression may be appropriate. Such adjustments may be controversial when the patient characteristic does not have an obvious relationship to the outcome. Stratified reporting poses problems for sample size and reporting format, but may be useful when casemix effects vary across units. Although there are no absolute standards of reliability, high reliabilities (interunit F > or = 10 or reliability > or = 0.9) are desirable for distinguishing above- and below-average units. When small or unequal sample sizes complicate reporting, precision may be improved using indirect estimation techniques that incorporate auxiliary information, and 'shrinkage' estimation can help to summarize the strength of evidence about units with small samples. With broader understanding of casemix adjustment and methods for analyzing small samples, quality data can be analysed and reported more accurately.

Informal Statistics Help Desk

NASA Technical Reports Server (NTRS)

Young, M.; Koslovsky, M.; Schaefer, Caroline M.; Feiveson, A. H.

2017-01-01

Back by popular demand, the JSC Biostatistics Laboratory and LSAH statisticians are offering an opportunity to discuss your statistical challenges and needs. Take the opportunity to meet the individuals offering expert statistical support to the JSC community. Join us for an informal conversation about any questions you may have encountered with issues of experimental design, analysis, or data visualization. Get answers to common questions about sample size, repeated measures, statistical assumptions, missing data, multiple testing, time-to-event data, and when to trust the results of your analyses.

A pilot randomized trial of two cognitive rehabilitation interventions for mild cognitive impairment: caregiver outcomes.

PubMed

Cuc, Andrea V; Locke, Dona E C; Duncan, Noah; Fields, Julie A; Snyder, Charlene Hoffman; Hanna, Sherrie; Lunde, Angela; Smith, Glenn E; Chandler, Melanie

2017-12-01

This study aims to provide effect size estimates of the impact of two cognitive rehabilitation interventions provided to patients with mild cognitive impairment: computerized brain fitness exercise and memory support system on support partners' outcomes of depression, anxiety, quality of life, and partner burden. A randomized controlled pilot trial was performed. At 6 months, the partners from both treatment groups showed stable to improved depression scores, while partners in an untreated control group showed worsening depression over 6 months. There were no statistically significant differences on anxiety, quality of life, or burden outcomes in this small pilot trial; however, effect sizes were moderate, suggesting that the sample sizes in this pilot study were not adequate to detect statistical significance. Either form of cognitive rehabilitation may help partners' mood, compared with providing no treatment. However, effect size estimates related to other partner outcomes (i.e., burden, quality of life, and anxiety) suggest that follow-up efficacy trials will need sample sizes of at least 30-100 people per group to accurately determine significance. Copyright © 2017 John Wiley & Sons, Ltd. Copyright © 2017 John Wiley & Sons, Ltd.

Sizing for the apparel industry using statistical analysis - a Brazilian case study

NASA Astrophysics Data System (ADS)

Capelassi, C. H.; Carvalho, M. A.; El Kattel, C.; Xu, B.

2017-10-01

The study of the body measurements of Brazilian women used the Kinect Body Imaging system for 3D body scanning. The result of the study aims to meet the needs of the apparel industry for accurate measurements. Data was statistically treated using the IBM SPSS 23 system, with 95% confidence (P<0,05) for the inferential analysis, with the purpose of grouping the measurements in sizes, so that a smaller number of sizes can cover a greater number of people. The sample consisted of 101 volunteers aged between 19 and 62 years. A cluster analysis was performed to identify the main body shapes of the sample. The results were divided between the top and bottom body portions; For the top portion, were used the measurements of the abdomen, waist and bust circumferences, as well as the height; For the bottom portion, were used the measurements of the hip circumference and the height. Three sizing systems were developed for the researched sample from the Abdomen-to-Height Ratio - AHR (top portion): Small (AHR < 0,52), Medium (AHR: 0,52-0,58), Large (AHR > 0,58) and from the Hip-to-Height Ratio - HHR (bottom portion): Small (HHR < 0,62), Medium (HHR: 0,62-0,68), Large (HHR > 0,68).

A Third Moment Adjusted Test Statistic for Small Sample Factor Analysis

PubMed Central

Lin, Johnny; Bentler, Peter M.

2012-01-01

Goodness of fit testing in factor analysis is based on the assumption that the test statistic is asymptotically chi-square; but this property may not hold in small samples even when the factors and errors are normally distributed in the population. Robust methods such as Browne’s asymptotically distribution-free method and Satorra Bentler’s mean scaling statistic were developed under the presumption of non-normality in the factors and errors. This paper finds new application to the case where factors and errors are normally distributed in the population but the skewness of the obtained test statistic is still high due to sampling error in the observed indicators. An extension of Satorra Bentler’s statistic is proposed that not only scales the mean but also adjusts the degrees of freedom based on the skewness of the obtained test statistic in order to improve its robustness under small samples. A simple simulation study shows that this third moment adjusted statistic asymptotically performs on par with previously proposed methods, and at a very small sample size offers superior Type I error rates under a properly specified model. Data from Mardia, Kent and Bibby’s study of students tested for their ability in five content areas that were either open or closed book were used to illustrate the real-world performance of this statistic. PMID:23144511

Random-effects linear modeling and sample size tables for two special crossover designs of average bioequivalence studies: the four-period, two-sequence, two-formulation and six-period, three-sequence, three-formulation designs.

PubMed

Diaz, Francisco J; Berg, Michel J; Krebill, Ron; Welty, Timothy; Gidal, Barry E; Alloway, Rita; Privitera, Michael

2013-12-01

Due to concern and debate in the epilepsy medical community and to the current interest of the US Food and Drug Administration (FDA) in revising approaches to the approval of generic drugs, the FDA is currently supporting ongoing bioequivalence studies of antiepileptic drugs, the EQUIGEN studies. During the design of these crossover studies, the researchers could not find commercial or non-commercial statistical software that quickly allowed computation of sample sizes for their designs, particularly software implementing the FDA requirement of using random-effects linear models for the analyses of bioequivalence studies. This article presents tables for sample-size evaluations of average bioequivalence studies based on the two crossover designs used in the EQUIGEN studies: the four-period, two-sequence, two-formulation design, and the six-period, three-sequence, three-formulation design. Sample-size computations assume that random-effects linear models are used in bioequivalence analyses with crossover designs. Random-effects linear models have been traditionally viewed by many pharmacologists and clinical researchers as just mathematical devices to analyze repeated-measures data. In contrast, a modern view of these models attributes an important mathematical role in theoretical formulations in personalized medicine to them, because these models not only have parameters that represent average patients, but also have parameters that represent individual patients. Moreover, the notation and language of random-effects linear models have evolved over the years. Thus, another goal of this article is to provide a presentation of the statistical modeling of data from bioequivalence studies that highlights the modern view of these models, with special emphasis on power analyses and sample-size computations.

A complete sample of double-lobed radio quasars for VLBI tests of source models - Definition and statistics

NASA Technical Reports Server (NTRS)

Hough, D. H.; Readhead, A. C. S.

1989-01-01

A complete, flux-density-limited sample of double-lobed radio quasars is defined, with nuclei bright enough to be mapped with the Mark III VLBI system. It is shown that the statistics of linear size, nuclear strength, and curvature are consistent with the assumption of random source orientations and simple relativistic beaming in the nuclei. However, these statistics are also consistent with the effects of interaction between the beams and the surrounding medium. The distribution of jet velocities in the nuclei, as measured with VLBI, will provide a powerful test of physical theories of extragalactic radio sources.

Statistical Analysis of a Large Sample Size Pyroshock Test Data Set Including Post Flight Data Assessment. Revision 1

NASA Technical Reports Server (NTRS)

Hughes, William O.; McNelis, Anne M.

2010-01-01

The Earth Observing System (EOS) Terra spacecraft was launched on an Atlas IIAS launch vehicle on its mission to observe planet Earth in late 1999. Prior to launch, the new design of the spacecraft's pyroshock separation system was characterized by a series of 13 separation ground tests. The analysis methods used to evaluate this unusually large amount of shock data will be discussed in this paper, with particular emphasis on population distributions and finding statistically significant families of data, leading to an overall shock separation interface level. The wealth of ground test data also allowed a derivation of a Mission Assurance level for the flight. All of the flight shock measurements were below the EOS Terra Mission Assurance level thus contributing to the overall success of the EOS Terra mission. The effectiveness of the statistical methodology for characterizing the shock interface level and for developing a flight Mission Assurance level from a large sample size of shock data is demonstrated in this paper.

"Adultspan" Publication Patterns: Author and Article Characteristics from 1999 to 2009

ERIC Educational Resources Information Center

Erford, Bradley T.; Clark, Kelly H.; Erford, Breann M.

2011-01-01

Publication patterns of articles in "Adultspan" from 1999 to 2009 were reviewed. Author characteristics and article content were analyzed to determine trends over time. Research articles were analyzed specifically for type of research design, classification, sampling method, types of participants, sample size, types of statistics used, and…

Testing the non-unity of rate ratio under inverse sampling.

PubMed

Tang, Man-Lai; Liao, Yi Jie; Ng, Hong Keung Tony; Chan, Ping Shing

2007-08-01

Inverse sampling is considered to be a more appropriate sampling scheme than the usual binomial sampling scheme when subjects arrive sequentially, when the underlying response of interest is acute, and when maximum likelihood estimators of some epidemiologic indices are undefined. In this article, we study various statistics for testing non-unity rate ratios in case-control studies under inverse sampling. These include the Wald, unconditional score, likelihood ratio and conditional score statistics. Three methods (the asymptotic, conditional exact, and Mid-P methods) are adopted for P-value calculation. We evaluate the performance of different combinations of test statistics and P-value calculation methods in terms of their empirical sizes and powers via Monte Carlo simulation. In general, asymptotic score and conditional score tests are preferable for their actual type I error rates are well controlled around the pre-chosen nominal level, and their powers are comparatively the largest. The exact version of Wald test is recommended if one wants to control the actual type I error rate at or below the pre-chosen nominal level. If larger power is expected and fluctuation of sizes around the pre-chosen nominal level are allowed, then the Mid-P version of Wald test is a desirable alternative. We illustrate the methodologies with a real example from a heart disease study. (c) 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

The Probability of Obtaining Two Statistically Different Test Scores as a Test Index

ERIC Educational Resources Information Center

Muller, Jorg M.

2006-01-01

A new test index is defined as the probability of obtaining two randomly selected test scores (PDTS) as statistically different. After giving a concept definition of the test index, two simulation studies are presented. The first analyzes the influence of the distribution of test scores, test reliability, and sample size on PDTS within classical…

The Adequacy of Different Robust Statistical Tests in Comparing Two Independent Groups

ERIC Educational Resources Information Center

Pero-Cebollero, Maribel; Guardia-Olmos, Joan

2013-01-01

In the current study, we evaluated various robust statistical methods for comparing two independent groups. Two scenarios for simulation were generated: one of equality and another of population mean differences. In each of the scenarios, 33 experimental conditions were used as a function of sample size, standard deviation and asymmetry. For each…

The impact of sample size on the reproducibility of voxel-based lesion-deficit mappings.

PubMed

Lorca-Puls, Diego L; Gajardo-Vidal, Andrea; White, Jitrachote; Seghier, Mohamed L; Leff, Alexander P; Green, David W; Crinion, Jenny T; Ludersdorfer, Philipp; Hope, Thomas M H; Bowman, Howard; Price, Cathy J

2018-07-01

This study investigated how sample size affects the reproducibility of findings from univariate voxel-based lesion-deficit analyses (e.g., voxel-based lesion-symptom mapping and voxel-based morphometry). Our effect of interest was the strength of the mapping between brain damage and speech articulation difficulties, as measured in terms of the proportion of variance explained. First, we identified a region of interest by searching on a voxel-by-voxel basis for brain areas where greater lesion load was associated with poorer speech articulation using a large sample of 360 right-handed English-speaking stroke survivors. We then randomly drew thousands of bootstrap samples from this data set that included either 30, 60, 90, 120, 180, or 360 patients. For each resample, we recorded effect size estimates and p values after conducting exactly the same lesion-deficit analysis within the previously identified region of interest and holding all procedures constant. The results show (1) how often small effect sizes in a heterogeneous population fail to be detected; (2) how effect size and its statistical significance varies with sample size; (3) how low-powered studies (due to small sample sizes) can greatly over-estimate as well as under-estimate effect sizes; and (4) how large sample sizes (N ≥ 90) can yield highly significant p values even when effect sizes are so small that they become trivial in practical terms. The implications of these findings for interpreting the results from univariate voxel-based lesion-deficit analyses are discussed. Copyright © 2018 The Author(s). Published by Elsevier Ltd.. All rights reserved.

A basic introduction to statistics for the orthopaedic surgeon.

PubMed

Bertrand, Catherine; Van Riet, Roger; Verstreken, Frederik; Michielsen, Jef

2012-02-01

Orthopaedic surgeons should review the orthopaedic literature in order to keep pace with the latest insights and practices. A good understanding of basic statistical principles is of crucial importance to the ability to read articles critically, to interpret results and to arrive at correct conclusions. This paper explains some of the key concepts in statistics, including hypothesis testing, Type I and Type II errors, testing of normality, sample size and p values.

High throughput nonparametric probability density estimation.

PubMed

Farmer, Jenny; Jacobs, Donald

2018-01-01

In high throughput applications, such as those found in bioinformatics and finance, it is important to determine accurate probability distribution functions despite only minimal information about data characteristics, and without using human subjectivity. Such an automated process for univariate data is implemented to achieve this goal by merging the maximum entropy method with single order statistics and maximum likelihood. The only required properties of the random variables are that they are continuous and that they are, or can be approximated as, independent and identically distributed. A quasi-log-likelihood function based on single order statistics for sampled uniform random data is used to empirically construct a sample size invariant universal scoring function. Then a probability density estimate is determined by iteratively improving trial cumulative distribution functions, where better estimates are quantified by the scoring function that identifies atypical fluctuations. This criterion resists under and over fitting data as an alternative to employing the Bayesian or Akaike information criterion. Multiple estimates for the probability density reflect uncertainties due to statistical fluctuations in random samples. Scaled quantile residual plots are also introduced as an effective diagnostic to visualize the quality of the estimated probability densities. Benchmark tests show that estimates for the probability density function (PDF) converge to the true PDF as sample size increases on particularly difficult test probability densities that include cases with discontinuities, multi-resolution scales, heavy tails, and singularities. These results indicate the method has general applicability for high throughput statistical inference.

High throughput nonparametric probability density estimation

PubMed Central

Farmer, Jenny

2018-01-01

In high throughput applications, such as those found in bioinformatics and finance, it is important to determine accurate probability distribution functions despite only minimal information about data characteristics, and without using human subjectivity. Such an automated process for univariate data is implemented to achieve this goal by merging the maximum entropy method with single order statistics and maximum likelihood. The only required properties of the random variables are that they are continuous and that they are, or can be approximated as, independent and identically distributed. A quasi-log-likelihood function based on single order statistics for sampled uniform random data is used to empirically construct a sample size invariant universal scoring function. Then a probability density estimate is determined by iteratively improving trial cumulative distribution functions, where better estimates are quantified by the scoring function that identifies atypical fluctuations. This criterion resists under and over fitting data as an alternative to employing the Bayesian or Akaike information criterion. Multiple estimates for the probability density reflect uncertainties due to statistical fluctuations in random samples. Scaled quantile residual plots are also introduced as an effective diagnostic to visualize the quality of the estimated probability densities. Benchmark tests show that estimates for the probability density function (PDF) converge to the true PDF as sample size increases on particularly difficult test probability densities that include cases with discontinuities, multi-resolution scales, heavy tails, and singularities. These results indicate the method has general applicability for high throughput statistical inference. PMID:29750803

Grain size analysis and depositional environment of shallow marine to basin floor, Kelantan River Delta

NASA Astrophysics Data System (ADS)

Afifah, M. R. Nurul; Aziz, A. Che; Roslan, M. Kamal

2015-09-01

Sediment samples were collected from the shallow marine from Kuala Besar, Kelantan outwards to the basin floor of South China Sea which consisted of quaternary bottom sediments. Sixty five samples were analysed for their grain size distribution and statistical relationships. Basic statistical analysis like mean, standard deviation, skewness and kurtosis were calculated and used to differentiate the depositional environment of the sediments and to derive the uniformity of depositional environment either from the beach or river environment. The sediments of all areas were varied in their sorting ranging from very well sorted to poorly sorted, strongly negative skewed to strongly positive skewed, and extremely leptokurtic to very platykurtic in nature. Bivariate plots between the grain-size parameters were then interpreted and the Coarsest-Median (CM) pattern showed the trend suggesting relationships between sediments influenced by three ongoing hydrodynamic factors namely turbidity current, littoral drift and waves dynamic, which functioned to control the sediments distribution pattern in various ways.

Adequacy of laser diffraction for soil particle size analysis

PubMed Central

Fisher, Peter; Aumann, Colin; Chia, Kohleth; O'Halloran, Nick; Chandra, Subhash

2017-01-01

Sedimentation has been a standard methodology for particle size analysis since the early 1900s. In recent years laser diffraction is beginning to replace sedimentation as the prefered technique in some industries, such as marine sediment analysis. However, for the particle size analysis of soils, which have a diverse range of both particle size and shape, laser diffraction still requires evaluation of its reliability. In this study, the sedimentation based sieve plummet balance method and the laser diffraction method were used to measure the particle size distribution of 22 soil samples representing four contrasting Australian Soil Orders. Initially, a precise wet riffling methodology was developed capable of obtaining representative samples within the recommended obscuration range for laser diffraction. It was found that repeatable results were obtained even if measurements were made at the extreme ends of the manufacturer’s recommended obscuration range. Results from statistical analysis suggested that the use of sample pretreatment to remove soil organic carbon (and possible traces of calcium-carbonate content) made minor differences to the laser diffraction particle size distributions compared to no pretreatment. These differences were found to be marginally statistically significant in the Podosol topsoil and Vertosol subsoil. There are well known reasons why sedimentation methods may be considered to ‘overestimate’ plate-like clay particles, while laser diffraction will ‘underestimate’ the proportion of clay particles. In this study we used Lin’s concordance correlation coefficient to determine the equivalence of laser diffraction and sieve plummet balance results. The results suggested that the laser diffraction equivalent thresholds corresponding to the sieve plummet balance cumulative particle sizes of < 2 μm, < 20 μm, and < 200 μm, were < 9 μm, < 26 μm, < 275 μm respectively. The many advantages of laser diffraction for soil particle size analysis, and the empirical results of this study, suggest that deployment of laser diffraction as a standard test procedure can provide reliable results, provided consistent sample preparation is used. PMID:28472043

Effect size measures in a two-independent-samples case with nonnormal and nonhomogeneous data.

PubMed

Li, Johnson Ching-Hong

2016-12-01

In psychological science, the "new statistics" refer to the new statistical practices that focus on effect size (ES) evaluation instead of conventional null-hypothesis significance testing (Cumming, Psychological Science, 25, 7-29, 2014). In a two-independent-samples scenario, Cohen's (1988) standardized mean difference (d) is the most popular ES, but its accuracy relies on two assumptions: normality and homogeneity of variances. Five other ESs-the unscaled robust d (d r * ; Hogarty & Kromrey, 2001), scaled robust d (d r ; Algina, Keselman, & Penfield, Psychological Methods, 10, 317-328, 2005), point-biserial correlation (r pb ; McGrath & Meyer, Psychological Methods, 11, 386-401, 2006), common-language ES (CL; Cliff, Psychological Bulletin, 114, 494-509, 1993), and nonparametric estimator for CL (A w ; Ruscio, Psychological Methods, 13, 19-30, 2008)-may be robust to violations of these assumptions, but no study has systematically evaluated their performance. Thus, in this simulation study the performance of these six ESs was examined across five factors: data distribution, sample, base rate, variance ratio, and sample size. The results showed that A w and d r were generally robust to these violations, and A w slightly outperformed d r . Implications for the use of A w and d r in real-world research are discussed.

Foundational Principles for Large-Scale Inference: Illustrations Through Correlation Mining

DOE Office of Scientific and Technical Information (OSTI.GOV)

Hero, Alfred O.; Rajaratnam, Bala

When can reliable inference be drawn in the ‘‘Big Data’’ context? This article presents a framework for answering this fundamental question in the context of correlation mining, with implications for general large-scale inference. In large-scale data applications like genomics, connectomics, and eco-informatics, the data set is often variable rich but sample starved: a regime where the number n of acquired samples (statistical replicates) is far fewer than the number p of observed variables (genes, neurons, voxels, or chemical constituents). Much of recent work has focused on understanding the computational complexity of proposed methods for ‘‘Big Data.’’ Sample complexity, however, hasmore » received relatively less attention, especially in the setting when the sample size n is fixed, and the dimension p grows without bound. To address this gap, we develop a unified statistical framework that explicitly quantifies the sample complexity of various inferential tasks. Sampling regimes can be divided into several categories: 1) the classical asymptotic regime where the variable dimension is fixed and the sample size goes to infinity; 2) the mixed asymptotic regime where both variable dimension and sample size go to infinity at comparable rates; and 3) the purely high-dimensional asymptotic regime where the variable dimension goes to infinity and the sample size is fixed. Each regime has its niche but only the latter regime applies to exa-scale data dimension. We illustrate this high-dimensional framework for the problem of correlation mining, where it is the matrix of pairwise and partial correlations among the variables that are of interest. Correlation mining arises in numerous applications and subsumes the regression context as a special case. We demonstrate various regimes of correlation mining based on the unifying perspective of high-dimensional learning rates and sample complexity for different structured covariance models and different inference tasks.« less

Foundational Principles for Large-Scale Inference: Illustrations Through Correlation Mining

PubMed Central

Hero, Alfred O.; Rajaratnam, Bala

2015-01-01

When can reliable inference be drawn in fue “Big Data” context? This paper presents a framework for answering this fundamental question in the context of correlation mining, wifu implications for general large scale inference. In large scale data applications like genomics, connectomics, and eco-informatics fue dataset is often variable-rich but sample-starved: a regime where the number n of acquired samples (statistical replicates) is far fewer than fue number p of observed variables (genes, neurons, voxels, or chemical constituents). Much of recent work has focused on understanding the computational complexity of proposed methods for “Big Data”. Sample complexity however has received relatively less attention, especially in the setting when the sample size n is fixed, and the dimension p grows without bound. To address fuis gap, we develop a unified statistical framework that explicitly quantifies the sample complexity of various inferential tasks. Sampling regimes can be divided into several categories: 1) the classical asymptotic regime where fue variable dimension is fixed and fue sample size goes to infinity; 2) the mixed asymptotic regime where both variable dimension and sample size go to infinity at comparable rates; 3) the purely high dimensional asymptotic regime where the variable dimension goes to infinity and the sample size is fixed. Each regime has its niche but only the latter regime applies to exa cale data dimension. We illustrate this high dimensional framework for the problem of correlation mining, where it is the matrix of pairwise and partial correlations among the variables fua t are of interest. Correlation mining arises in numerous applications and subsumes the regression context as a special case. we demonstrate various regimes of correlation mining based on the unifying perspective of high dimensional learning rates and sample complexity for different structured covariance models and different inference tasks. PMID:27087700

Foundational Principles for Large-Scale Inference: Illustrations Through Correlation Mining

DOE PAGES

Hero, Alfred O.; Rajaratnam, Bala

2015-12-09

When can reliable inference be drawn in the ‘‘Big Data’’ context? This article presents a framework for answering this fundamental question in the context of correlation mining, with implications for general large-scale inference. In large-scale data applications like genomics, connectomics, and eco-informatics, the data set is often variable rich but sample starved: a regime where the number n of acquired samples (statistical replicates) is far fewer than the number p of observed variables (genes, neurons, voxels, or chemical constituents). Much of recent work has focused on understanding the computational complexity of proposed methods for ‘‘Big Data.’’ Sample complexity, however, hasmore » received relatively less attention, especially in the setting when the sample size n is fixed, and the dimension p grows without bound. To address this gap, we develop a unified statistical framework that explicitly quantifies the sample complexity of various inferential tasks. Sampling regimes can be divided into several categories: 1) the classical asymptotic regime where the variable dimension is fixed and the sample size goes to infinity; 2) the mixed asymptotic regime where both variable dimension and sample size go to infinity at comparable rates; and 3) the purely high-dimensional asymptotic regime where the variable dimension goes to infinity and the sample size is fixed. Each regime has its niche but only the latter regime applies to exa-scale data dimension. We illustrate this high-dimensional framework for the problem of correlation mining, where it is the matrix of pairwise and partial correlations among the variables that are of interest. Correlation mining arises in numerous applications and subsumes the regression context as a special case. We demonstrate various regimes of correlation mining based on the unifying perspective of high-dimensional learning rates and sample complexity for different structured covariance models and different inference tasks.« less

IGESS: a statistical approach to integrating individual-level genotype data and summary statistics in genome-wide association studies.

PubMed

Dai, Mingwei; Ming, Jingsi; Cai, Mingxuan; Liu, Jin; Yang, Can; Wan, Xiang; Xu, Zongben

2017-09-15

Results from genome-wide association studies (GWAS) suggest that a complex phenotype is often affected by many variants with small effects, known as 'polygenicity'. Tens of thousands of samples are often required to ensure statistical power of identifying these variants with small effects. However, it is often the case that a research group can only get approval for the access to individual-level genotype data with a limited sample size (e.g. a few hundreds or thousands). Meanwhile, summary statistics generated using single-variant-based analysis are becoming publicly available. The sample sizes associated with the summary statistics datasets are usually quite large. How to make the most efficient use of existing abundant data resources largely remains an open question. In this study, we propose a statistical approach, IGESS, to increasing statistical power of identifying risk variants and improving accuracy of risk prediction by i ntegrating individual level ge notype data and s ummary s tatistics. An efficient algorithm based on variational inference is developed to handle the genome-wide analysis. Through comprehensive simulation studies, we demonstrated the advantages of IGESS over the methods which take either individual-level data or summary statistics data as input. We applied IGESS to perform integrative analysis of Crohns Disease from WTCCC and summary statistics from other studies. IGESS was able to significantly increase the statistical power of identifying risk variants and improve the risk prediction accuracy from 63.2% ( ±0.4% ) to 69.4% ( ±0.1% ) using about 240 000 variants. The IGESS software is available at https://github.com/daviddaigithub/IGESS . zbxu@xjtu.edu.cn or xwan@comp.hkbu.edu.hk or eeyang@hkbu.edu.hk. Supplementary data are available at Bioinformatics online. © The Author (2017). Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Statistical differences between relative quantitative molecular fingerprints from microbial communities.

PubMed

Portillo, M C; Gonzalez, J M

2008-08-01

Molecular fingerprints of microbial communities are a common method for the analysis and comparison of environmental samples. The significance of differences between microbial community fingerprints was analyzed considering the presence of different phylotypes and their relative abundance. A method is proposed by simulating coverage of the analyzed communities as a function of sampling size applying a Cramér-von Mises statistic. Comparisons were performed by a Monte Carlo testing procedure. As an example, this procedure was used to compare several sediment samples from freshwater ponds using a relative quantitative PCR-DGGE profiling technique. The method was able to discriminate among different samples based on their molecular fingerprints, and confirmed the lack of differences between aliquots from a single sample.

Two models of the sound-signal frequency dependence on the animal body size as exemplified by the ground squirrels of Eurasia (mammalia, rodentia).

PubMed

Nikol'skii, A A

2017-11-01

Dependence of the sound-signal frequency on the animal body length was studied in 14 ground squirrel species (genus Spermophilus) of Eurasia. Regression analysis of the total sample yielded a low determination coefficient (R 2 = 26%), because the total sample proved to be heterogeneous in terms of signal frequency within the dimension classes of animals. When the total sample was divided into two groups according to signal frequency, two statistically significant models (regression equations) were obtained in which signal frequency depended on the body size at high determination coefficients (R 2 = 73 and 94% versus 26% for the total sample). Thus, the problem of correlation between animal body size and the frequency of their vocal signals does not have a unique solution.

Experimental and environmental factors affect spurious detection of ecological thresholds

USGS Publications Warehouse

Daily, Jonathan P.; Hitt, Nathaniel P.; Smith, David; Snyder, Craig D.

2012-01-01

Threshold detection methods are increasingly popular for assessing nonlinear responses to environmental change, but their statistical performance remains poorly understood. We simulated linear change in stream benthic macroinvertebrate communities and evaluated the performance of commonly used threshold detection methods based on model fitting (piecewise quantile regression [PQR]), data partitioning (nonparametric change point analysis [NCPA]), and a hybrid approach (significant zero crossings [SiZer]). We demonstrated that false detection of ecological thresholds (type I errors) and inferences on threshold locations are influenced by sample size, rate of linear change, and frequency of observations across the environmental gradient (i.e., sample-environment distribution, SED). However, the relative importance of these factors varied among statistical methods and between inference types. False detection rates were influenced primarily by user-selected parameters for PQR (τ) and SiZer (bandwidth) and secondarily by sample size (for PQR) and SED (for SiZer). In contrast, the location of reported thresholds was influenced primarily by SED. Bootstrapped confidence intervals for NCPA threshold locations revealed strong correspondence to SED. We conclude that the choice of statistical methods for threshold detection should be matched to experimental and environmental constraints to minimize false detection rates and avoid spurious inferences regarding threshold location.

Variance Estimation, Design Effects, and Sample Size Calculations for Respondent-Driven Sampling

PubMed Central

2006-01-01

Hidden populations, such as injection drug users and sex workers, are central to a number of public health problems. However, because of the nature of these groups, it is difficult to collect accurate information about them, and this difficulty complicates disease prevention efforts. A recently developed statistical approach called respondent-driven sampling improves our ability to study hidden populations by allowing researchers to make unbiased estimates of the prevalence of certain traits in these populations. Yet, not enough is known about the sample-to-sample variability of these prevalence estimates. In this paper, we present a bootstrap method for constructing confidence intervals around respondent-driven sampling estimates and demonstrate in simulations that it outperforms the naive method currently in use. We also use simulations and real data to estimate the design effects for respondent-driven sampling in a number of situations. We conclude with practical advice about the power calculations that are needed to determine the appropriate sample size for a study using respondent-driven sampling. In general, we recommend a sample size twice as large as would be needed under simple random sampling. PMID:16937083

Unequal cluster sizes in stepped-wedge cluster randomised trials: a systematic review

PubMed Central

Morris, Tom; Gray, Laura

2017-01-01

Objectives To investigate the extent to which cluster sizes vary in stepped-wedge cluster randomised trials (SW-CRT) and whether any variability is accounted for during the sample size calculation and analysis of these trials. Setting Any, not limited to healthcare settings. Participants Any taking part in an SW-CRT published up to March 2016. Primary and secondary outcome measures The primary outcome is the variability in cluster sizes, measured by the coefficient of variation (CV) in cluster size. Secondary outcomes include the difference between the cluster sizes assumed during the sample size calculation and those observed during the trial, any reported variability in cluster sizes and whether the methods of sample size calculation and methods of analysis accounted for any variability in cluster sizes. Results Of the 101 included SW-CRTs, 48% mentioned that the included clusters were known to vary in size, yet only 13% of these accounted for this during the calculation of the sample size. However, 69% of the trials did use a method of analysis appropriate for when clusters vary in size. Full trial reports were available for 53 trials. The CV was calculated for 23 of these: the median CV was 0.41 (IQR: 0.22–0.52). Actual cluster sizes could be compared with those assumed during the sample size calculation for 14 (26%) of the trial reports; the cluster sizes were between 29% and 480% of that which had been assumed. Conclusions Cluster sizes often vary in SW-CRTs. Reporting of SW-CRTs also remains suboptimal. The effect of unequal cluster sizes on the statistical power of SW-CRTs needs further exploration and methods appropriate to studies with unequal cluster sizes need to be employed. PMID:29146637

Accuracy assessment of percent canopy cover, cover type, and size class

Treesearch

H. T. Schreuder; S. Bain; R. C. Czaplewski

2003-01-01

Truth for vegetation cover percent and type is obtained from very large-scale photography (VLSP), stand structure as measured by size classes, and vegetation types from a combination of VLSP and ground sampling. We recommend using the Kappa statistic with bootstrap confidence intervals for overall accuracy, and similarly bootstrap confidence intervals for percent...

"TNOs are Cool": A survey of the trans-Neptunian region. XIII. Statistical analysis of multiple trans-Neptunian objects observed with Herschel Space Observatory

NASA Astrophysics Data System (ADS)

Kovalenko, I. D.; Doressoundiram, A.; Lellouch, E.; Vilenius, E.; Müller, T.; Stansberry, J.

2017-11-01

Context. Gravitationally bound multiple systems provide an opportunity to estimate the mean bulk density of the objects, whereas this characteristic is not available for single objects. Being a primitive population of the outer solar system, binary and multiple trans-Neptunian objects (TNOs) provide unique information about bulk density and internal structure, improving our understanding of their formation and evolution. Aims: The goal of this work is to analyse parameters of multiple trans-Neptunian systems, observed with Herschel and Spitzer space telescopes. Particularly, statistical analysis is done for radiometric size and geometric albedo, obtained from photometric observations, and for estimated bulk density. Methods: We use Monte Carlo simulation to estimate the real size distribution of TNOs. For this purpose, we expand the dataset of diameters by adopting the Minor Planet Center database list with available values of the absolute magnitude therein, and the albedo distribution derived from Herschel radiometric measurements. We use the 2-sample Anderson-Darling non-parametric statistical method for testing whether two samples of diameters, for binary and single TNOs, come from the same distribution. Additionally, we use the Spearman's coefficient as a measure of rank correlations between parameters. Uncertainties of estimated parameters together with lack of data are taken into account. Conclusions about correlations between parameters are based on statistical hypothesis testing. Results: We have found that the difference in size distributions of multiple and single TNOs is biased by small objects. The test on correlations between parameters shows that the effective diameter of binary TNOs strongly correlates with heliocentric orbital inclination and with magnitude difference between components of binary system. The correlation between diameter and magnitude difference implies that small and large binaries are formed by different mechanisms. Furthermore, the statistical test indicates, although not significant with the sample size, that a moderately strong correlation exists between diameter and bulk density. Herschel is an ESA space observatory with science instruments provided by European-led Principal Investigator consortia and with important participation from NASA.

Using the bootstrap to establish statistical significance for relative validity comparisons among patient-reported outcome measures

PubMed Central

2013-01-01

Background Relative validity (RV), a ratio of ANOVA F-statistics, is often used to compare the validity of patient-reported outcome (PRO) measures. We used the bootstrap to establish the statistical significance of the RV and to identify key factors affecting its significance. Methods Based on responses from 453 chronic kidney disease (CKD) patients to 16 CKD-specific and generic PRO measures, RVs were computed to determine how well each measure discriminated across clinically-defined groups of patients compared to the most discriminating (reference) measure. Statistical significance of RV was quantified by the 95% bootstrap confidence interval. Simulations examined the effects of sample size, denominator F-statistic, correlation between comparator and reference measures, and number of bootstrap replicates. Results The statistical significance of the RV increased as the magnitude of denominator F-statistic increased or as the correlation between comparator and reference measures increased. A denominator F-statistic of 57 conveyed sufficient power (80%) to detect an RV of 0.6 for two measures correlated at r = 0.7. Larger denominator F-statistics or higher correlations provided greater power. Larger sample size with a fixed denominator F-statistic or more bootstrap replicates (beyond 500) had minimal impact. Conclusions The bootstrap is valuable for establishing the statistical significance of RV estimates. A reasonably large denominator F-statistic (F > 57) is required for adequate power when using the RV to compare the validity of measures with small or moderate correlations (r < 0.7). Substantially greater power can be achieved when comparing measures of a very high correlation (r > 0.9). PMID:23721463

Statistical analysis of hydrological response in urbanising catchments based on adaptive sampling using inter-amount times

NASA Astrophysics Data System (ADS)

ten Veldhuis, Marie-Claire; Schleiss, Marc

2017-04-01

In this study, we introduced an alternative approach for analysis of hydrological flow time series, using an adaptive sampling framework based on inter-amount times (IATs). The main difference with conventional flow time series is the rate at which low and high flows are sampled: the unit of analysis for IATs is a fixed flow amount, instead of a fixed time window. We analysed statistical distributions of flows and IATs across a wide range of sampling scales to investigate sensitivity of statistical properties such as quantiles, variance, skewness, scaling parameters and flashiness indicators to the sampling scale. We did this based on streamflow time series for 17 (semi)urbanised basins in North Carolina, US, ranging from 13 km2 to 238 km2 in size. Results showed that adaptive sampling of flow time series based on inter-amounts leads to a more balanced representation of low flow and peak flow values in the statistical distribution. While conventional sampling gives a lot of weight to low flows, as these are most ubiquitous in flow time series, IAT sampling gives relatively more weight to high flow values, when given flow amounts are accumulated in shorter time. As a consequence, IAT sampling gives more information about the tail of the distribution associated with high flows, while conventional sampling gives relatively more information about low flow periods. We will present results of statistical analyses across a range of subdaily to seasonal scales and will highlight some interesting insights that can be derived from IAT statistics with respect to basin flashiness and impact urbanisation on hydrological response.

Power and sample-size estimation for microbiome studies using pairwise distances and PERMANOVA.

PubMed

Kelly, Brendan J; Gross, Robert; Bittinger, Kyle; Sherrill-Mix, Scott; Lewis, James D; Collman, Ronald G; Bushman, Frederic D; Li, Hongzhe

2015-08-01

The variation in community composition between microbiome samples, termed beta diversity, can be measured by pairwise distance based on either presence-absence or quantitative species abundance data. PERMANOVA, a permutation-based extension of multivariate analysis of variance to a matrix of pairwise distances, partitions within-group and between-group distances to permit assessment of the effect of an exposure or intervention (grouping factor) upon the sampled microbiome. Within-group distance and exposure/intervention effect size must be accurately modeled to estimate statistical power for a microbiome study that will be analyzed with pairwise distances and PERMANOVA. We present a framework for PERMANOVA power estimation tailored to marker-gene microbiome studies that will be analyzed by pairwise distances, which includes: (i) a novel method for distance matrix simulation that permits modeling of within-group pairwise distances according to pre-specified population parameters; (ii) a method to incorporate effects of different sizes within the simulated distance matrix; (iii) a simulation-based method for estimating PERMANOVA power from simulated distance matrices; and (iv) an R statistical software package that implements the above. Matrices of pairwise distances can be efficiently simulated to satisfy the triangle inequality and incorporate group-level effects, which are quantified by the adjusted coefficient of determination, omega-squared (ω2). From simulated distance matrices, available PERMANOVA power or necessary sample size can be estimated for a planned microbiome study. © The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.

Kidney function endpoints in kidney transplant trials: a struggle for power.

PubMed

Ibrahim, A; Garg, A X; Knoll, G A; Akbari, A; White, C A

2013-03-01

Kidney function endpoints are commonly used in randomized controlled trials (RCTs) in kidney transplantation (KTx). We conducted this study to estimate the proportion of ongoing RCTs with kidney function endpoints in KTx where the proposed sample size is large enough to detect meaningful differences in glomerular filtration rate (GFR) with adequate statistical power. RCTs were retrieved using the key word "kidney transplantation" from the National Institute of Health online clinical trial registry. Included trials had at least one measure of kidney function tracked for at least 1 month after transplant. We determined the proportion of two-arm parallel trials that had sufficient sample sizes to detect a minimum 5, 7.5 and 10 mL/min difference in GFR between arms. Fifty RCTs met inclusion criteria. Only 7% of the trials were above a sample size of 562, the number needed to detect a minimum 5 mL/min difference between the groups should one exist (assumptions: α = 0.05; power = 80%, 10% loss to follow-up, common standard deviation of 20 mL/min). The result increased modestly to 36% of trials when a minimum 10 mL/min difference was considered. Only a minority of ongoing trials have adequate statistical power to detect between-group differences in kidney function using conventional sample size estimating parameters. For this reason, some potentially effective interventions which ultimately could benefit patients may be abandoned from future assessment. © Copyright 2013 The American Society of Transplantation and the American Society of Transplant Surgeons.

Sensitivity and specificity of normality tests and consequences on reference interval accuracy at small sample size: a computer-simulation study.

PubMed

Le Boedec, Kevin

2016-12-01

According to international guidelines, parametric methods must be chosen for RI construction when the sample size is small and the distribution is Gaussian. However, normality tests may not be accurate at small sample size. The purpose of the study was to evaluate normality test performance to properly identify samples extracted from a Gaussian population at small sample sizes, and assess the consequences on RI accuracy of applying parametric methods to samples that falsely identified the parent population as Gaussian. Samples of n = 60 and n = 30 values were randomly selected 100 times from simulated Gaussian, lognormal, and asymmetric populations of 10,000 values. The sensitivity and specificity of 4 normality tests were compared. Reference intervals were calculated using 6 different statistical methods from samples that falsely identified the parent population as Gaussian, and their accuracy was compared. Shapiro-Wilk and D'Agostino-Pearson tests were the best performing normality tests. However, their specificity was poor at sample size n = 30 (specificity for P < .05: .51 and .50, respectively). The best significance levels identified when n = 30 were 0.19 for Shapiro-Wilk test and 0.18 for D'Agostino-Pearson test. Using parametric methods on samples extracted from a lognormal population but falsely identified as Gaussian led to clinically relevant inaccuracies. At small sample size, normality tests may lead to erroneous use of parametric methods to build RI. Using nonparametric methods (or alternatively Box-Cox transformation) on all samples regardless of their distribution or adjusting, the significance level of normality tests depending on sample size would limit the risk of constructing inaccurate RI. © 2016 American Society for Veterinary Clinical Pathology.

Experiments with central-limit properties of spatial samples from locally covariant random fields

USGS Publications Warehouse

Barringer, T.H.; Smith, T.E.

1992-01-01

When spatial samples are statistically dependent, the classical estimator of sample-mean standard deviation is well known to be inconsistent. For locally dependent samples, however, consistent estimators of sample-mean standard deviation can be constructed. The present paper investigates the sampling properties of one such estimator, designated as the tau estimator of sample-mean standard deviation. In particular, the asymptotic normality properties of standardized sample means based on tau estimators are studied in terms of computer experiments with simulated sample-mean distributions. The effects of both sample size and dependency levels among samples are examined for various value of tau (denoting the size of the spatial kernel for the estimator). The results suggest that even for small degrees of spatial dependency, the tau estimator exhibits significantly stronger normality properties than does the classical estimator of standardized sample means. ?? 1992.

In vivo Comet assay--statistical analysis and power calculations of mice testicular cells.

PubMed

Hansen, Merete Kjær; Sharma, Anoop Kumar; Dybdahl, Marianne; Boberg, Julie; Kulahci, Murat

2014-11-01

The in vivo Comet assay is a sensitive method for evaluating DNA damage. A recurrent concern is how to analyze the data appropriately and efficiently. A popular approach is to summarize the raw data into a summary statistic prior to the statistical analysis. However, consensus on which summary statistic to use has yet to be reached. Another important consideration concerns the assessment of proper sample sizes in the design of Comet assay studies. This study aims to identify a statistic suitably summarizing the % tail DNA of mice testicular samples in Comet assay studies. A second aim is to provide curves for this statistic outlining the number of animals and gels to use. The current study was based on 11 compounds administered via oral gavage in three doses to male mice: CAS no. 110-26-9, CAS no. 512-56-1, CAS no. 111873-33-7, CAS no. 79-94-7, CAS no. 115-96-8, CAS no. 598-55-0, CAS no. 636-97-5, CAS no. 85-28-9, CAS no. 13674-87-8, CAS no. 43100-38-5 and CAS no. 60965-26-6. Testicular cells were examined using the alkaline version of the Comet assay and the DNA damage was quantified as % tail DNA using a fully automatic scoring system. From the raw data 23 summary statistics were examined. A linear mixed-effects model was fitted to the summarized data and the estimated variance components were used to generate power curves as a function of sample size. The statistic that most appropriately summarized the within-sample distributions was the median of the log-transformed data, as it most consistently conformed to the assumptions of the statistical model. Power curves for 1.5-, 2-, and 2.5-fold changes of the highest dose group compared to the control group when 50 and 100 cells were scored per gel are provided to aid in the design of future Comet assay studies on testicular cells. Copyright © 2014 Elsevier B.V. All rights reserved.

Sampling errors in the estimation of empirical orthogonal functions. [for climatology studies

NASA Technical Reports Server (NTRS)

North, G. R.; Bell, T. L.; Cahalan, R. F.; Moeng, F. J.

1982-01-01

Empirical Orthogonal Functions (EOF's), eigenvectors of the spatial cross-covariance matrix of a meteorological field, are reviewed with special attention given to the necessary weighting factors for gridded data and the sampling errors incurred when too small a sample is available. The geographical shape of an EOF shows large intersample variability when its associated eigenvalue is 'close' to a neighboring one. A rule of thumb indicating when an EOF is likely to be subject to large sampling fluctuations is presented. An explicit example, based on the statistics of the 500 mb geopotential height field, displays large intersample variability in the EOF's for sample sizes of a few hundred independent realizations, a size seldom exceeded by meteorological data sets.

Designing clinical trials to test disease-modifying agents: application to the treatment trials of Alzheimer's disease.

PubMed

Xiong, Chengjie; van Belle, Gerald; Miller, J Philip; Morris, John C

2011-02-01

Therapeutic trials of disease-modifying agents on Alzheimer's disease (AD) require novel designs and analyses involving switch of treatments for at least a portion of subjects enrolled. Randomized start and randomized withdrawal designs are two examples of such designs. Crucial design parameters such as sample size and the time of treatment switch are important to understand in designing such clinical trials. The purpose of this article is to provide methods to determine sample sizes and time of treatment switch as well as optimum statistical tests of treatment efficacy for clinical trials of disease-modifying agents on AD. A general linear mixed effects model is proposed to test the disease-modifying efficacy of novel therapeutic agents on AD. This model links the longitudinal growth from both the placebo arm and the treatment arm at the time of treatment switch for these in the delayed treatment arm or early withdrawal arm and incorporates the potential correlation on the rate of cognitive change before and after the treatment switch. Sample sizes and the optimum time for treatment switch of such trials as well as optimum test statistic for the treatment efficacy are determined according to the model. Assuming an evenly spaced longitudinal design over a fixed duration, the optimum treatment switching time in a randomized start or a randomized withdrawal trial is half way through the trial. With the optimum test statistic for the treatment efficacy and over a wide spectrum of model parameters, the optimum sample size allocations are fairly close to the simplest design with a sample size ratio of 1:1:1 among the treatment arm, the delayed treatment or early withdrawal arm, and the placebo arm. The application of the proposed methodology to AD provides evidence that much larger sample sizes are required to adequately power disease-modifying trials when compared with those for symptomatic agents, even when the treatment switch time and efficacy test are optimally chosen. The proposed method assumes that the only and immediate effect of treatment switch is on the rate of cognitive change. Crucial design parameters for the clinical trials of disease-modifying agents on AD can be optimally chosen. Government and industry officials as well as academia researchers should consider the optimum use of the clinical trials design for disease-modifying agents on AD in their effort to search for the treatments with the potential to modify the underlying pathophysiology of AD.

Experimental toxicology: Issues of statistics, experimental design, and replication.

PubMed

Briner, Wayne; Kirwan, Jeral

2017-01-01

The difficulty of replicating experiments has drawn considerable attention. Issues with replication occur for a variety of reasons ranging from experimental design to laboratory errors to inappropriate statistical analysis. Here we review a variety of guidelines for statistical analysis, design, and execution of experiments in toxicology. In general, replication can be improved by using hypothesis driven experiments with adequate sample sizes, randomization, and blind data collection techniques. Copyright © 2016 Elsevier B.V. All rights reserved.

Caution regarding the choice of standard deviations to guide sample size calculations in clinical trials.

PubMed

Chen, Henian; Zhang, Nanhua; Lu, Xiaosun; Chen, Sophie

2013-08-01

The method used to determine choice of standard deviation (SD) is inadequately reported in clinical trials. Underestimations of the population SD may result in underpowered clinical trials. This study demonstrates how using the wrong method to determine population SD can lead to inaccurate sample sizes and underpowered studies, and offers recommendations to maximize the likelihood of achieving adequate statistical power. We review the practice of reporting sample size and its effect on the power of trials published in major journals. Simulated clinical trials were used to compare the effects of different methods of determining SD on power and sample size calculations. Prior to 1996, sample size calculations were reported in just 1%-42% of clinical trials. This proportion increased from 38% to 54% after the initial Consolidated Standards of Reporting Trials (CONSORT) was published in 1996, and from 64% to 95% after the revised CONSORT was published in 2001. Nevertheless, underpowered clinical trials are still common. Our simulated data showed that all minimal and 25th-percentile SDs fell below 44 (the population SD), regardless of sample size (from 5 to 50). For sample sizes 5 and 50, the minimum sample SDs underestimated the population SD by 90.7% and 29.3%, respectively. If only one sample was available, there was less than 50% chance that the actual power equaled or exceeded the planned power of 80% for detecting a median effect size (Cohen's d = 0.5) when using the sample SD to calculate the sample size. The proportions of studies with actual power of at least 80% were about 95%, 90%, 85%, and 80% when we used the larger SD, 80% upper confidence limit (UCL) of SD, 70% UCL of SD, and 60% UCL of SD to calculate the sample size, respectively. When more than one sample was available, the weighted average SD resulted in about 50% of trials being underpowered; the proportion of trials with power of 80% increased from 90% to 100% when the 75th percentile and the maximum SD from 10 samples were used. Greater sample size is needed to achieve a higher proportion of studies having actual power of 80%. This study only addressed sample size calculation for continuous outcome variables. We recommend using the 60% UCL of SD, maximum SD, 80th-percentile SD, and 75th-percentile SD to calculate sample size when 1 or 2 samples, 3 samples, 4-5 samples, and more than 5 samples of data are available, respectively. Using the sample SD or average SD to calculate sample size should be avoided.

How accurate is the Pearson r-from-Z approximation? A Monte Carlo simulation study.

PubMed

Hittner, James B; May, Kim

2012-01-01

The Pearson r-from-Z approximation estimates the sample correlation (as an effect size measure) from the ratio of two quantities: the standard normal deviate equivalent (Z-score) corresponding to a one-tailed p-value divided by the square root of the total (pooled) sample size. The formula has utility in meta-analytic work when reports of research contain minimal statistical information. Although simple to implement, the accuracy of the Pearson r-from-Z approximation has not been empirically evaluated. To address this omission, we performed a series of Monte Carlo simulations. Results indicated that in some cases the formula did accurately estimate the sample correlation. However, when sample size was very small (N = 10) and effect sizes were small to small-moderate (ds of 0.1 and 0.3), the Pearson r-from-Z approximation was very inaccurate. Detailed figures that provide guidance as to when the Pearson r-from-Z formula will likely yield valid inferences are presented.

Predicting stellar angular diameters from V, IC, H and K photometry

NASA Astrophysics Data System (ADS)

Adams, Arthur D.; Boyajian, Tabetha S.; von Braun, Kaspar

2018-01-01

Determining the physical properties of microlensing events depends on having accurate angular sizes of the source star. Using long baseline optical interferometry, we are able to measure the angular sizes of nearby stars with uncertainties ≤2 per cent. We present empirically derived relations of angular diameters which are calibrated using both a sample of dwarfs/subgiants and a sample of giant stars. These relations are functions of five colour indices in the visible and near-infrared, and have uncertainties of 1.8-6.5 per cent depending on the colour used. We find that a combined sample of both main-sequence and evolved stars of A-K spectral types is well fitted by a single relation for each colour considered. We find that in the colours considered, metallicity does not play a statistically significant role in predicting stellar size, leading to a means of predicting observed sizes of stars from colour alone.

Statistical considerations in monitoring birds over large areas

USGS Publications Warehouse

Johnson, D.H.

2000-01-01

The proper design of a monitoring effort depends primarily on the objectives desired, constrained by the resources available to conduct the work. Typically, managers have numerous objectives, such as determining abundance of the species, detecting changes in population size, evaluating responses to management activities, and assessing habitat associations. A design that is optimal for one objective will likely not be optimal for others. Careful consideration of the importance of the competing objectives may lead to a design that adequately addresses the priority concerns, although it may not be optimal for any individual objective. Poor design or inadequate sample sizes may result in such weak conclusions that the effort is wasted. Statistical expertise can be used at several stages, such as estimating power of certain hypothesis tests, but is perhaps most useful in fundamental considerations of describing objectives and designing sampling plans.

10 CFR Appendix B to Subpart C of... - Sampling Plan for Enforcement Testing of Covered Equipment and Certain Low-Volume Covered Products

Code of Federal Regulations, 2014 CFR

2014-01-01

... percent, one-sided confidence limit and a sample size of n1. (2) For an energy consumption standard (ECS..., where ECS is the energy consumption standard and t is a statistic based on a 97.5 percent, one-sided...

10 CFR Appendix B to Subpart C of... - Sampling Plan for Enforcement Testing of Covered Equipment and Certain Low-Volume Covered Products

Code of Federal Regulations, 2013 CFR

2013-01-01

... percent, one-sided confidence limit and a sample size of n1. (2) For an energy consumption standard (ECS..., where ECS is the energy consumption standard and t is a statistic based on a 97.5 percent, one-sided...

10 CFR Appendix B to Subpart C of... - Sampling Plan for Enforcement Testing of Covered Equipment and Certain Low-Volume Covered Products

Code of Federal Regulations, 2012 CFR

2012-01-01

... percent, one-sided confidence limit and a sample size of n1. (2) For an energy consumption standard (ECS..., where ECS is the energy consumption standard and t is a statistic based on a 97.5 percent, one-sided...

Bootstrap Estimation of Sample Statistic Bias in Structural Equation Modeling.

ERIC Educational Resources Information Center

Thompson, Bruce; Fan, Xitao

This study empirically investigated bootstrap bias estimation in the area of structural equation modeling (SEM). Three correctly specified SEM models were used under four different sample size conditions. Monte Carlo experiments were carried out to generate the criteria against which bootstrap bias estimation should be judged. For SEM fit indices,…

Correlating sampling and intensity statistics in nanoparticle diffraction experiments

DOE Office of Scientific and Technical Information (OSTI.GOV)

Öztürk, Hande; Yan, Hanfei; Hill, John P.

2015-07-28

It is shown in a previous article [Öztürk, Yan, Hill & Noyan (2014).J. Appl. Cryst.47, 1016–1025] that the sampling statistics of diffracting particle populations within a polycrystalline ensemble depended on the size of the constituent crystallites: broad X-ray peak breadths enabled some nano-sized particles to contribute more than one diffraction spot to Debye–Scherrer rings. Here it is shown that the equations proposed by Alexander, Klug & Kummer [J. Appl. Phys.(1948),19, 742–753] (AKK) to link diffracting particle and diffracted intensity statistics are not applicable if the constituent crystallites of the powder are below 10 nm. In this size range, (i) themore » one-to-one correspondence between diffracting particles and Laue spots assumed in the AKK analysis is not satisfied, and (ii) the crystallographic correlation between Laue spots originating from the same grain invalidates the assumption that all diffracting plane normals are randomly oriented and uncorrelated. Such correlation produces unexpected results in the selection of diffracting grains. For example, three or more Laue spots from a given grain for a particular reflection can only be observed at certain wavelengths. In addition, correcting the diffracted intensity values by the traditional Lorentz term, 1/cos θ, to compensate for the variation of particles sampled within a reflection band does not maintain fidelity to the number of poles contributing to the diffracted signal. A new term, cos θ B/cos θ, corrects this problem.« less

Topical anesthesia for rubber dam clamp placement in sealant placement: comparison of lidocaine/prilocaine gel and benzocaine.

PubMed

Yoon, Richard K; Chussid, Steven

2009-01-01

The purpose of this prospective study was to compare the efficacy of Oraqix to benzocaine while placing a rubber dam clamp during sealant placement on children. A sample size of 45 7- to 12-year-old patients who presented for bilateral sealants on permanent first molars participated in this study. A split-mouth design was implemented with Oraqix applied to one side and 20 percent benzocaine gel applied to the other. After placing the topical anesthetic and the rubber dam clamp, patients completed a Feces Pain Scale (FPS) to rate the level of discomfort after clamp placement. Twenty-seven subjects (60%) were female and 18 subjects (40%) were mole; 15 (33%) were younger than 9 years old and 30 (67%) were at least 9 years old. The overall difference in mean FPS ratings was not statistically significant (P = .27). Regarding gender, there was no statistically significant difference in males (P = .65) or females (P = .26). There was also no difference in mean FPS ratings when looking at age groups younger than 9 years old with P=.77 In the 9 years and older age groups, however there was a statistically significant difference, with P = .04. Application of Oraqix did not reduce discomfort when compared to benzocaine in this small sample size. Oraqix was more effective than benzocaine in the age group 9 and older. A larger sample size is needed to determine its efficacy in children younger than 9years old.

A Statistical Analysis of the Economic Drivers of Battery Energy Storage in Commercial Buildings: Preprint

DOE Office of Scientific and Technical Information (OSTI.GOV)

Long, Matthew; Simpkins, Travis; Cutler, Dylan

There is significant interest in using battery energy storage systems (BESS) to reduce peak demand charges, and therefore the life cycle cost of electricity, in commercial buildings. This paper explores the drivers of economic viability of BESS in commercial buildings through statistical analysis. A sample population of buildings was generated, a techno-economic optimization model was used to size and dispatch the BESS, and the resulting optimal BESS sizes were analyzed for relevant predictor variables. Explanatory regression analyses were used to demonstrate that peak demand charges are the most significant predictor of an economically viable battery, and that the shape ofmore » the load profile is the most significant predictor of the size of the battery.« less

A Statistical Analysis of the Economic Drivers of Battery Energy Storage in Commercial Buildings

DOE Office of Scientific and Technical Information (OSTI.GOV)

Long, Matthew; Simpkins, Travis; Cutler, Dylan

There is significant interest in using battery energy storage systems (BESS) to reduce peak demand charges, and therefore the life cycle cost of electricity, in commercial buildings. This paper explores the drivers of economic viability of BESS in commercial buildings through statistical analysis. A sample population of buildings was generated, a techno-economic optimization model was used to size and dispatch the BESS, and the resulting optimal BESS sizes were analyzed for relevant predictor variables. Explanatory regression analyses were used to demonstrate that peak demand charges are the most significant predictor of an economically viable battery, and that the shape ofmore » the load profile is the most significant predictor of the size of the battery.« less

Simulation of parametric model towards the fixed covariate of right censored lung cancer data

NASA Astrophysics Data System (ADS)

Afiqah Muhamad Jamil, Siti; Asrul Affendi Abdullah, M.; Kek, Sie Long; Ridwan Olaniran, Oyebayo; Enera Amran, Syahila

2017-09-01

In this study, simulation procedure was applied to measure the fixed covariate of right censored data by using parametric survival model. The scale and shape parameter were modified to differentiate the analysis of parametric regression survival model. Statistically, the biases, mean biases and the coverage probability were used in this analysis. Consequently, different sample sizes were employed to distinguish the impact of parametric regression model towards right censored data with 50, 100, 150 and 200 number of sample. R-statistical software was utilised to develop the coding simulation with right censored data. Besides, the final model of right censored simulation was compared with the right censored lung cancer data in Malaysia. It was found that different values of shape and scale parameter with different sample size, help to improve the simulation strategy for right censored data and Weibull regression survival model is suitable fit towards the simulation of survival of lung cancer patients data in Malaysia.

Merging National Forest and National Forest Health Inventories to Obtain an Integrated Forest Resource Inventory – Experiences from Bavaria, Slovenia and Sweden

PubMed Central

Kovač, Marko; Bauer, Arthur; Ståhl, Göran

2014-01-01

Backgrounds, Material and Methods To meet the demands of sustainable forest management and international commitments, European nations have designed a variety of forest-monitoring systems for specific needs. While the majority of countries are committed to independent, single-purpose inventorying, a minority of countries have merged their single-purpose forest inventory systems into integrated forest resource inventories. The statistical efficiencies of the Bavarian, Slovene and Swedish integrated forest resource inventory designs are investigated with the various statistical parameters of the variables of growing stock volume, shares of damaged trees, and deadwood volume. The parameters are derived by using the estimators for the given inventory designs. The required sample sizes are derived via the general formula for non-stratified independent samples and via statistical power analyses. The cost effectiveness of the designs is compared via two simple cost effectiveness ratios. Results In terms of precision, the most illustrative parameters of the variables are relative standard errors; their values range between 1% and 3% if the variables’ variations are low (s%<80%) and are higher in the case of higher variations. A comparison of the actual and required sample sizes shows that the actual sample sizes were deliberately set high to provide precise estimates for the majority of variables and strata. In turn, the successive inventories are statistically efficient, because they allow detecting the mean changes of variables with powers higher than 90%; the highest precision is attained for the changes of growing stock volume and the lowest for the changes of the shares of damaged trees. Two indicators of cost effectiveness also show that the time input spent for measuring one variable decreases with the complexity of inventories. Conclusion There is an increasing need for credible information on forest resources to be used for decision making and national and international policy making. Such information can be cost-efficiently provided through integrated forest resource inventories. PMID:24941120

A Future Moon Mission: Curatorial Statistics on Regolith Fragments Applicable to Sample Collection by Raking

NASA Technical Reports Server (NTRS)

Allton, J. H.; Bevill, T. J.

2003-01-01

The strategy of raking rock fragments from the lunar regolith as a means of acquiring representative samples has wide support due to science return, spacecraft simplicity (reliability) and economy [3, 4, 5]. While there exists widespread agreement that raking or sieving the bulk regolith is good strategy, there is lively discussion about the minimum sample size. Advocates of consor-tium studies desire fragments large enough to support petrologic and isotopic studies. Fragments from 5 to 10 mm are thought adequate [4, 5]. Yet, Jolliff et al. [6] demonstrated use of 2-4 mm fragments as repre-sentative of larger rocks. Here we make use of cura-torial records and sample catalogs to give a different perspective on minimum sample size for a robotic sample collector.

Modeling motor vehicle crashes using Poisson-gamma models: examining the effects of low sample mean values and small sample size on the estimation of the fixed dispersion parameter.

PubMed

Lord, Dominique

2006-07-01

There has been considerable research conducted on the development of statistical models for predicting crashes on highway facilities. Despite numerous advancements made for improving the estimation tools of statistical models, the most common probabilistic structure used for modeling motor vehicle crashes remains the traditional Poisson and Poisson-gamma (or Negative Binomial) distribution; when crash data exhibit over-dispersion, the Poisson-gamma model is usually the model of choice most favored by transportation safety modelers. Crash data collected for safety studies often have the unusual attributes of being characterized by low sample mean values. Studies have shown that the goodness-of-fit of statistical models produced from such datasets can be significantly affected. This issue has been defined as the "low mean problem" (LMP). Despite recent developments on methods to circumvent the LMP and test the goodness-of-fit of models developed using such datasets, no work has so far examined how the LMP affects the fixed dispersion parameter of Poisson-gamma models used for modeling motor vehicle crashes. The dispersion parameter plays an important role in many types of safety studies and should, therefore, be reliably estimated. The primary objective of this research project was to verify whether the LMP affects the estimation of the dispersion parameter and, if it is, to determine the magnitude of the problem. The secondary objective consisted of determining the effects of an unreliably estimated dispersion parameter on common analyses performed in highway safety studies. To accomplish the objectives of the study, a series of Poisson-gamma distributions were simulated using different values describing the mean, the dispersion parameter, and the sample size. Three estimators commonly used by transportation safety modelers for estimating the dispersion parameter of Poisson-gamma models were evaluated: the method of moments, the weighted regression, and the maximum likelihood method. In an attempt to complement the outcome of the simulation study, Poisson-gamma models were fitted to crash data collected in Toronto, Ont. characterized by a low sample mean and small sample size. The study shows that a low sample mean combined with a small sample size can seriously affect the estimation of the dispersion parameter, no matter which estimator is used within the estimation process. The probability the dispersion parameter becomes unreliably estimated increases significantly as the sample mean and sample size decrease. Consequently, the results show that an unreliably estimated dispersion parameter can significantly undermine empirical Bayes (EB) estimates as well as the estimation of confidence intervals for the gamma mean and predicted response. The paper ends with recommendations about minimizing the likelihood of producing Poisson-gamma models with an unreliable dispersion parameter for modeling motor vehicle crashes.

A Monte Carlo Study of Levene's Test of Homogeneity of Variance: Empirical Frequencies of Type I Error in Normal Distributions.

ERIC Educational Resources Information Center

Neel, John H.; Stallings, William M.

An influential statistics test recommends a Levene text for homogeneity of variance. A recent note suggests that Levene's test is upwardly biased for small samples. Another report shows inflated Alpha estimates and low power. Neither study utilized more than two sample sizes. This Monte Carlo study involved sampling from a normal population for…

Anthropometric Characteristics of Columbia, South Carolina, Youth Baseball Players and Dixie Youth World Series Players

ERIC Educational Resources Information Center

French, Karen E.; Spurgeon, John H.; Nevett, Michael E.

2007-01-01

The purpose of this study was to compare measures of body size in two samples of youth baseball players with normative data from the United States National Center for Health Statistics (NCHS) growth charts. One sample of youth baseball players participated in a local little league. The second sample of youth baseball players were members of eight…

A Bayesian sequential design with adaptive randomization for 2-sided hypothesis test.

PubMed

Yu, Qingzhao; Zhu, Lin; Zhu, Han

2017-11-01

Bayesian sequential and adaptive randomization designs are gaining popularity in clinical trials thanks to their potentials to reduce the number of required participants and save resources. We propose a Bayesian sequential design with adaptive randomization rates so as to more efficiently attribute newly recruited patients to different treatment arms. In this paper, we consider 2-arm clinical trials. Patients are allocated to the 2 arms with a randomization rate to achieve minimum variance for the test statistic. Algorithms are presented to calculate the optimal randomization rate, critical values, and power for the proposed design. Sensitivity analysis is implemented to check the influence on design by changing the prior distributions. Simulation studies are applied to compare the proposed method and traditional methods in terms of power and actual sample sizes. Simulations show that, when total sample size is fixed, the proposed design can obtain greater power and/or cost smaller actual sample size than the traditional Bayesian sequential design. Finally, we apply the proposed method to a real data set and compare the results with the Bayesian sequential design without adaptive randomization in terms of sample sizes. The proposed method can further reduce required sample size. Copyright © 2017 John Wiley & Sons, Ltd.

A single test for rejecting the null hypothesis in subgroups and in the overall sample.

PubMed

Lin, Yunzhi; Zhou, Kefei; Ganju, Jitendra

2017-01-01

In clinical trials, some patient subgroups are likely to demonstrate larger effect sizes than other subgroups. For example, the effect size, or informally the benefit with treatment, is often greater in patients with a moderate condition of a disease than in those with a mild condition. A limitation of the usual method of analysis is that it does not incorporate this ordering of effect size by patient subgroup. We propose a test statistic which supplements the conventional test by including this information and simultaneously tests the null hypothesis in pre-specified subgroups and in the overall sample. It results in more power than the conventional test when the differences in effect sizes across subgroups are at least moderately large; otherwise it loses power. The method involves combining p-values from models fit to pre-specified subgroups and the overall sample in a manner that assigns greater weight to subgroups in which a larger effect size is expected. Results are presented for randomized trials with two and three subgroups.

Maximum type 1 error rate inflation in multiarmed clinical trials with adaptive interim sample size modifications.

PubMed

Graf, Alexandra C; Bauer, Peter; Glimm, Ekkehard; Koenig, Franz

2014-07-01

Sample size modifications in the interim analyses of an adaptive design can inflate the type 1 error rate, if test statistics and critical boundaries are used in the final analysis as if no modification had been made. While this is already true for designs with an overall change of the sample size in a balanced treatment-control comparison, the inflation can be much larger if in addition a modification of allocation ratios is allowed as well. In this paper, we investigate adaptive designs with several treatment arms compared to a single common control group. Regarding modifications, we consider treatment arm selection as well as modifications of overall sample size and allocation ratios. The inflation is quantified for two approaches: a naive procedure that ignores not only all modifications, but also the multiplicity issue arising from the many-to-one comparison, and a Dunnett procedure that ignores modifications, but adjusts for the initially started multiple treatments. The maximum inflation of the type 1 error rate for such types of design can be calculated by searching for the "worst case" scenarios, that are sample size adaptation rules in the interim analysis that lead to the largest conditional type 1 error rate in any point of the sample space. To show the most extreme inflation, we initially assume unconstrained second stage sample size modifications leading to a large inflation of the type 1 error rate. Furthermore, we investigate the inflation when putting constraints on the second stage sample sizes. It turns out that, for example fixing the sample size of the control group, leads to designs controlling the type 1 error rate. © 2014 The Author. Biometrical Journal published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Statistical Modelling of Temperature and Moisture Uptake of Biochars Exposed to Selected Relative Humidity of Air.

PubMed

Bastistella, Luciane; Rousset, Patrick; Aviz, Antonio; Caldeira-Pires, Armando; Humbert, Gilles; Nogueira, Manoel

2018-02-09

New experimental techniques, as well as modern variants on known methods, have recently been employed to investigate the fundamental reactions underlying the oxidation of biochar. The purpose of this paper was to experimentally and statistically study how the relative humidity of air, mass, and particle size of four biochars influenced the adsorption of water and the increase in temperature. A random factorial design was employed using the intuitive statistical software Xlstat. A simple linear regression model and an analysis of variance with a pairwise comparison were performed. The experimental study was carried out on the wood of Quercus pubescens , Cyclobalanopsis glauca , Trigonostemon huangmosun , and Bambusa vulgaris , and involved five relative humidity conditions (22, 43, 75, 84, and 90%), two mass samples (0.1 and 1 g), and two particle sizes (powder and piece). Two response variables including water adsorption and temperature increase were analyzed and discussed. The temperature did not increase linearly with the adsorption of water. Temperature was modeled by nine explanatory variables, while water adsorption was modeled by eight. Five variables, including factors and their interactions, were found to be common to the two models. Sample mass and relative humidity influenced the two qualitative variables, while particle size and biochar type only influenced the temperature.

Optimally estimating the sample mean from the sample size, median, mid-range, and/or mid-quartile range.

PubMed

Luo, Dehui; Wan, Xiang; Liu, Jiming; Tong, Tiejun

2018-06-01

The era of big data is coming, and evidence-based medicine is attracting increasing attention to improve decision making in medical practice via integrating evidence from well designed and conducted clinical research. Meta-analysis is a statistical technique widely used in evidence-based medicine for analytically combining the findings from independent clinical trials to provide an overall estimation of a treatment effectiveness. The sample mean and standard deviation are two commonly used statistics in meta-analysis but some trials use the median, the minimum and maximum values, or sometimes the first and third quartiles to report the results. Thus, to pool results in a consistent format, researchers need to transform those information back to the sample mean and standard deviation. In this article, we investigate the optimal estimation of the sample mean for meta-analysis from both theoretical and empirical perspectives. A major drawback in the literature is that the sample size, needless to say its importance, is either ignored or used in a stepwise but somewhat arbitrary manner, e.g. the famous method proposed by Hozo et al. We solve this issue by incorporating the sample size in a smoothly changing weight in the estimators to reach the optimal estimation. Our proposed estimators not only improve the existing ones significantly but also share the same virtue of the simplicity. The real data application indicates that our proposed estimators are capable to serve as "rules of thumb" and will be widely applied in evidence-based medicine.

Small studies may overestimate the effect sizes in critical care meta-analyses: a meta-epidemiological study

PubMed Central

2013-01-01

Introduction Small-study effects refer to the fact that trials with limited sample sizes are more likely to report larger beneficial effects than large trials. However, this has never been investigated in critical care medicine. Thus, the present study aimed to examine the presence and extent of small-study effects in critical care medicine. Methods Critical care meta-analyses involving randomized controlled trials and reported mortality as an outcome measure were considered eligible for the study. Component trials were classified as large (≥100 patients per arm) and small (<100 patients per arm) according to their sample sizes. Ratio of odds ratio (ROR) was calculated for each meta-analysis and then RORs were combined using a meta-analytic approach. ROR<1 indicated larger beneficial effect in small trials. Small and large trials were compared in methodological qualities including sequence generating, blinding, allocation concealment, intention to treat and sample size calculation. Results A total of 27 critical care meta-analyses involving 317 trials were included. Of them, five meta-analyses showed statistically significant RORs <1, and other meta-analyses did not reach a statistical significance. Overall, the pooled ROR was 0.60 (95% CI: 0.53 to 0.68); the heterogeneity was moderate with an I2 of 50.3% (chi-squared = 52.30; P = 0.002). Large trials showed significantly better reporting quality than small trials in terms of sequence generating, allocation concealment, blinding, intention to treat, sample size calculation and incomplete follow-up data. Conclusions Small trials are more likely to report larger beneficial effects than large trials in critical care medicine, which could be partly explained by the lower methodological quality in small trials. Caution should be practiced in the interpretation of meta-analyses involving small trials. PMID:23302257

A Guerilla Guide to Common Problems in ‘Neurostatistics’: Essential Statistical Topics in Neuroscience

PubMed Central

Smith, Paul F.

2017-01-01

Effective inferential statistical analysis is essential for high quality studies in neuroscience. However, recently, neuroscience has been criticised for the poor use of experimental design and statistical analysis. Many of the statistical issues confronting neuroscience are similar to other areas of biology; however, there are some that occur more regularly in neuroscience studies. This review attempts to provide a succinct overview of some of the major issues that arise commonly in the analyses of neuroscience data. These include: the non-normal distribution of the data; inequality of variance between groups; extensive correlation in data for repeated measurements across time or space; excessive multiple testing; inadequate statistical power due to small sample sizes; pseudo-replication; and an over-emphasis on binary conclusions about statistical significance as opposed to effect sizes. Statistical analysis should be viewed as just another neuroscience tool, which is critical to the final outcome of the study. Therefore, it needs to be done well and it is a good idea to be proactive and seek help early, preferably before the study even begins. PMID:29371855

A Guerilla Guide to Common Problems in 'Neurostatistics': Essential Statistical Topics in Neuroscience.

PubMed

Smith, Paul F

2017-01-01

Effective inferential statistical analysis is essential for high quality studies in neuroscience. However, recently, neuroscience has been criticised for the poor use of experimental design and statistical analysis. Many of the statistical issues confronting neuroscience are similar to other areas of biology; however, there are some that occur more regularly in neuroscience studies. This review attempts to provide a succinct overview of some of the major issues that arise commonly in the analyses of neuroscience data. These include: the non-normal distribution of the data; inequality of variance between groups; extensive correlation in data for repeated measurements across time or space; excessive multiple testing; inadequate statistical power due to small sample sizes; pseudo-replication; and an over-emphasis on binary conclusions about statistical significance as opposed to effect sizes. Statistical analysis should be viewed as just another neuroscience tool, which is critical to the final outcome of the study. Therefore, it needs to be done well and it is a good idea to be proactive and seek help early, preferably before the study even begins.

Seabed mapping and characterization of sediment variability using the usSEABED data base

USGS Publications Warehouse

Goff, J.A.; Jenkins, C.J.; Jeffress, Williams S.

2008-01-01

We present a methodology for statistical analysis of randomly located marine sediment point data, and apply it to the US continental shelf portions of usSEABED mean grain size records. The usSEABED database, like many modern, large environmental datasets, is heterogeneous and interdisciplinary. We statistically test the database as a source of mean grain size data, and from it provide a first examination of regional seafloor sediment variability across the entire US continental shelf. Data derived from laboratory analyses ("extracted") and from word-based descriptions ("parsed") are treated separately, and they are compared statistically and deterministically. Data records are selected for spatial analysis by their location within sample regions: polygonal areas defined in ArcGIS chosen by geography, water depth, and data sufficiency. We derive isotropic, binned semivariograms from the data, and invert these for estimates of noise variance, field variance, and decorrelation distance. The highly erratic nature of the semivariograms is a result both of the random locations of the data and of the high level of data uncertainty (noise). This decorrelates the data covariance matrix for the inversion, and largely prevents robust estimation of the fractal dimension. Our comparison of the extracted and parsed mean grain size data demonstrates important differences between the two. In particular, extracted measurements generally produce finer mean grain sizes, lower noise variance, and lower field variance than parsed values. Such relationships can be used to derive a regionally dependent conversion factor between the two. Our analysis of sample regions on the US continental shelf revealed considerable geographic variability in the estimated statistical parameters of field variance and decorrelation distance. Some regional relationships are evident, and overall there is a tendency for field variance to be higher where the average mean grain size is finer grained. Surprisingly, parsed and extracted noise magnitudes correlate with each other, which may indicate that some portion of the data variability that we identify as "noise" is caused by real grain size variability at very short scales. Our analyses demonstrate that by applying a bias-correction proxy, usSEABED data can be used to generate reliable interpolated maps of regional mean grain size and sediment character. 

Digital image processing of nanometer-size metal particles on amorphous substrates

NASA Technical Reports Server (NTRS)

Soria, F.; Artal, P.; Bescos, J.; Heinemann, K.

1989-01-01

The task of differentiating very small metal aggregates supported on amorphous films from the phase contrast image features inherently stemming from the support is extremely difficult in the nanometer particle size range. Digital image processing was employed to overcome some of the ambiguities in evaluating such micrographs. It was demonstrated that such processing allowed positive particle detection and a limited degree of statistical size analysis even for micrographs where by bare eye examination the distribution between particles and erroneous substrate features would seem highly ambiguous. The smallest size class detected for Pd/C samples peaks at 0.8 nm. This size class was found in various samples prepared under different evaporation conditions and it is concluded that these particles consist of 'a magic number' of 13 atoms and have cubooctahedral or icosahedral crystal structure.

Experimental Design in Clinical 'Omics Biomarker Discovery.

PubMed

Forshed, Jenny

2017-11-03

This tutorial highlights some issues in the experimental design of clinical 'omics biomarker discovery, how to avoid bias and get as true quantities as possible from biochemical analyses, and how to select samples to improve the chance of answering the clinical question at issue. This includes the importance of defining clinical aim and end point, knowing the variability in the results, randomization of samples, sample size, statistical power, and how to avoid confounding factors by including clinical data in the sample selection, that is, how to avoid unpleasant surprises at the point of statistical analysis. The aim of this Tutorial is to help translational clinical and preclinical biomarker candidate research and to improve the validity and potential of future biomarker candidate findings.

Characterization of Inclusion Populations in Mn-Si Deoxidized Steel

NASA Astrophysics Data System (ADS)

García-Carbajal, Alfonso; Herrera-Trejo, Martín; Castro-Cedeño, Edgar-Ivan; Castro-Román, Manuel; Martinez-Enriquez, Arturo-Isaias

2017-12-01

Four plant heats of Mn-Si deoxidized steel were conducted to follow the evolution of the inclusion population through ladle furnace (LF) treatment and subsequent vacuum treatment (VT). The liquid steel was sampled, and the chemical composition and size distribution of the inclusion populations were characterized. The Gumbel generalized extreme-value (GEV) and generalized Pareto (GP) distributions were used for the statistical analysis of the inclusion size distributions. The inclusions found at the beginning of the LF treatment were mostly fully liquid SiO2-Al2O3-MnO inclusions, which then evolved into fully liquid SiO2-Al2O3-CaO-MgO and partly liquid SiO2-CaO-MgO-(Al2O3-MgO) inclusions detected at the end of the VT. The final fully liquid inclusions had a desirable chemical composition for plastic behavior in subsequent metallurgical operations. The GP distribution was found to be undesirable for statistical analysis. The GEV distribution approach led to shape parameter values different from the zero value hypothesized from the Gumbel distribution. According to the GEV approach, some of the final inclusion size distributions had statistically significant differences, whereas the Gumbel approach predicted no statistically significant differences. The heats were organized according to indicators of inclusion cleanliness and a statistical comparison of the size distributions.

Estimating numbers of females with cubs-of-the-year in the Yellowstone grizzly bear population

USGS Publications Warehouse

Keating, K.A.; Schwartz, C.C.; Haroldson, M.A.; Moody, D.

2001-01-01

For grizzly bears (Ursus arctos horribilis) in the Greater Yellowstone Ecosystem (GYE), minimum population size and allowable numbers of human-caused mortalities have been calculated as a function of the number of unique females with cubs-of-the-year (FCUB) seen during a 3- year period. This approach underestimates the total number of FCUB, thereby biasing estimates of population size and sustainable mortality. Also, it does not permit calculation of valid confidence bounds. Many statistical methods can resolve or mitigate these problems, but there is no universal best method. Instead, relative performances of different methods can vary with population size, sample size, and degree of heterogeneity among sighting probabilities for individual animals. We compared 7 nonparametric estimators, using Monte Carlo techniques to assess performances over the range of sampling conditions deemed plausible for the Yellowstone population. Our goal was to estimate the number of FCUB present in the population each year. Our evaluation differed from previous comparisons of such estimators by including sample coverage methods and by treating individual sightings, rather than sample periods, as the sample unit. Consequently, our conclusions also differ from earlier studies. Recommendations regarding estimators and necessary sample sizes are presented, together with estimates of annual numbers of FCUB in the Yellowstone population with bootstrap confidence bounds.

Rule-of-thumb adjustment of sample sizes to accommodate dropouts in a two-stage analysis of repeated measurements.

PubMed

Overall, John E; Tonidandel, Scott; Starbuck, Robert R

2006-01-01

Recent contributions to the statistical literature have provided elegant model-based solutions to the problem of estimating sample sizes for testing the significance of differences in mean rates of change across repeated measures in controlled longitudinal studies with differentially correlated error and missing data due to dropouts. However, the mathematical complexity and model specificity of these solutions make them generally inaccessible to most applied researchers who actually design and undertake treatment evaluation research in psychiatry. In contrast, this article relies on a simple two-stage analysis in which dropout-weighted slope coefficients fitted to the available repeated measurements for each subject separately serve as the dependent variable for a familiar ANCOVA test of significance for differences in mean rates of change. This article is about how a sample of size that is estimated or calculated to provide desired power for testing that hypothesis without considering dropouts can be adjusted appropriately to take dropouts into account. Empirical results support the conclusion that, whatever reasonable level of power would be provided by a given sample size in the absence of dropouts, essentially the same power can be realized in the presence of dropouts simply by adding to the original dropout-free sample size the number of subjects who would be expected to drop from a sample of that original size under conditions of the proposed study.

Using sieving and pretreatment to separate plastics during end-of-life vehicle recycling.

PubMed

Stagner, Jacqueline A; Sagan, Barsha; Tam, Edwin Kl

2013-09-01

Plastics continue to be a challenge for recovering materials at the end-of-life for vehicles. However, it may be possible to improve the recovery of plastics by exploiting material characteristics, such as shape, or by altering their behavior, such as through temperature changes, in relation to recovery processes and handling. Samples of a 2009 Dodge Challenger front fascia were shredded in a laboratory-scale hammer mill shredder. A 2 × 2 factorial design study was performed to determine the effect of sample shape (flat versus curved) and sample temperature (room temperature versus cryogenic temperature) on the size of the particles exiting from the shredder. It was determined that sample shape does not affect the particle size; however, sample temperature does affect the particle size. At cryogenic temperatures, the distribution of particle sizes is much narrower than at room temperature. Having a more uniform particle size could make recovery of plastic particles, such as these more efficient during the recycling of end-of-life vehicles. Samples of Chrysler minivan headlights were also shredded at room temperature and at cryogenic temperatures. The size of the particles of the two different plastics in the headlights is statistically different both at room temperature and at cryogenic temperature, and the particles are distributed narrowly. The research suggests that incremental changes in end-of-life vehicle processing could be effective in aiding materials recovery.

You Cannot Step Into the Same River Twice: When Power Analyses Are Optimistic.

PubMed

McShane, Blakeley B; Böckenholt, Ulf

2014-11-01

Statistical power depends on the size of the effect of interest. However, effect sizes are rarely fixed in psychological research: Study design choices, such as the operationalization of the dependent variable or the treatment manipulation, the social context, the subject pool, or the time of day, typically cause systematic variation in the effect size. Ignoring this between-study variation, as standard power formulae do, results in assessments of power that are too optimistic. Consequently, when researchers attempting replication set sample sizes using these formulae, their studies will be underpowered and will thus fail at a greater than expected rate. We illustrate this with both hypothetical examples and data on several well-studied phenomena in psychology. We provide formulae that account for between-study variation and suggest that researchers set sample sizes with respect to our generally more conservative formulae. Our formulae generalize to settings in which there are multiple effects of interest. We also introduce an easy-to-use website that implements our approach to setting sample sizes. Finally, we conclude with recommendations for quantifying between-study variation. © The Author(s) 2014.

Visual Sample Plan Version 7.0 User's Guide

DOE Office of Scientific and Technical Information (OSTI.GOV)

Matzke, Brett D.; Newburn, Lisa LN; Hathaway, John E.

2014-03-01

User's guide for VSP 7.0 This user's guide describes Visual Sample Plan (VSP) Version 7.0 and provides instructions for using the software. VSP selects the appropriate number and location of environmental samples to ensure that the results of statistical tests performed to provide input to risk decisions have the required confidence and performance. VSP Version 7.0 provides sample-size equations or algorithms needed by specific statistical tests appropriate for specific environmental sampling objectives. It also provides data quality assessment and statistical analysis functions to support evaluation of the data and determine whether the data support decisions regarding sites suspected of contamination.more » The easy-to-use program is highly visual and graphic. VSP runs on personal computers with Microsoft Windows operating systems (XP, Vista, Windows 7, and Windows 8). Designed primarily for project managers and users without expertise in statistics, VSP is applicable to two- and three-dimensional populations to be sampled (e.g., rooms and buildings, surface soil, a defined layer of subsurface soil, water bodies, and other similar applications) for studies of environmental quality. VSP is also applicable for designing sampling plans for assessing chem/rad/bio threat and hazard identification within rooms and buildings, and for designing geophysical surveys for unexploded ordnance (UXO) identification.« less

Methodological Issues Related to the Use of P Less than 0.05 in Health Behavior Research

ERIC Educational Resources Information Center

Duryea, Elias; Graner, Stephen P.; Becker, Jeremy

2009-01-01

This paper reviews methodological issues related to the use of P less than 0.05 in health behavior research and suggests how application and presentation of statistical significance may be improved. Assessment of sample size and P less than 0.05, the file drawer problem, the Law of Large Numbers and the statistical significance arguments in…

Directions for new developments on statistical design and analysis of small population group trials.

PubMed

Hilgers, Ralf-Dieter; Roes, Kit; Stallard, Nigel

2016-06-14

Most statistical design and analysis methods for clinical trials have been developed and evaluated where at least several hundreds of patients could be recruited. These methods may not be suitable to evaluate therapies if the sample size is unavoidably small, which is usually termed by small populations. The specific sample size cut off, where the standard methods fail, needs to be investigated. In this paper, the authors present their view on new developments for design and analysis of clinical trials in small population groups, where conventional statistical methods may be inappropriate, e.g., because of lack of power or poor adherence to asymptotic approximations due to sample size restrictions. Following the EMA/CHMP guideline on clinical trials in small populations, we consider directions for new developments in the area of statistical methodology for design and analysis of small population clinical trials. We relate the findings to the research activities of three projects, Asterix, IDeAl, and InSPiRe, which have received funding since 2013 within the FP7-HEALTH-2013-INNOVATION-1 framework of the EU. As not all aspects of the wide research area of small population clinical trials can be addressed, we focus on areas where we feel advances are needed and feasible. The general framework of the EMA/CHMP guideline on small population clinical trials stimulates a number of research areas. These serve as the basis for the three projects, Asterix, IDeAl, and InSPiRe, which use various approaches to develop new statistical methodology for design and analysis of small population clinical trials. Small population clinical trials refer to trials with a limited number of patients. Small populations may result form rare diseases or specific subtypes of more common diseases. New statistical methodology needs to be tailored to these specific situations. The main results from the three projects will constitute a useful toolbox for improved design and analysis of small population clinical trials. They address various challenges presented by the EMA/CHMP guideline as well as recent discussions about extrapolation. There is a need for involvement of the patients' perspective in the planning and conduct of small population clinical trials for a successful therapy evaluation.

Professional School Counseling (PSC) Publication Pattern Review: A Meta-Study of Author and Article Characteristics from the First 15 Years

ERIC Educational Resources Information Center

Erford, Bradley T.; Giguere, Monica; Glenn, Kacie; Ciarlone, Hallie

2015-01-01

Patterns of articles published in "Professional School Counseling" (PSC) from the first 15 volumes were reviewed in this meta-study. Author characteristics (e.g., sex, employment setting, nation of domicile) and article characteristics (e.g., topic, type, design, sample, sample size, participant type, statistical procedures and…

Predictor sort sampling and one-sided confidence bounds on quantiles

Treesearch

Steve Verrill; Victoria L. Herian; David W. Green

2002-01-01

Predictor sort experiments attempt to make use of the correlation between a predictor that can be measured prior to the start of an experiment and the response variable that we are investigating. Properly designed and analyzed, they can reduce necessary sample sizes, increase statistical power, and reduce the lengths of confidence intervals. However, if the non- random...

Pupil size in Jewish theological seminary students.

PubMed

Shemesh, G; Kesler, A; Lazar, M; Rothkoff, L

2004-01-01

To investigate the authors' clinical impression that pupil size among myopic Jewish theological seminary students is different from pupil size of similar secular subjects. This cross-sectional study was conducted on 28 male Jewish theological seminary students and 28 secular students or workers who were matched for age and refraction. All participants were consecutively enrolled. Scotopic and photopic pupil size was measured by means of a Colvard pupillometer. Comparisons of various parameters between the groups were performed using the two-sample t-test, Fisher exact test, a paired-sample t-test, a two-way analysis of variance, and Pearson correlation coefficients as appropriate. The two groups were statistically matched for age, refraction, and visual acuity. The seminary students were undercorrected by an average of 2.35 diopters (D), while the secular subjects were undercorrected by only 0.65 D (p<0.01). The average pupil size was larger in the religious group under both scotopic and photopic luminance. This difference was maintained when the two groups were compared according to iris color under both conditions, reaching a level of statistical significance (p<0.0001). There was a significant difference in photopic pupil size between dark and light irises (p=0.049), but this difference was not maintained under scotopic conditions. The average pupil size of young ultraorthodox seminary students was significantly larger than that of matched secular subjects. Whether this is the result of intensive close-up work or of apparently characteristic undercorrection of the myopia is undetermined.

Sampling studies to estimate the HIV prevalence rate in female commercial sex workers.

PubMed

Pascom, Ana Roberta Pati; Szwarcwald, Célia Landmann; Barbosa Júnior, Aristides

2010-01-01

We investigated sampling methods being used to estimate the HIV prevalence rate among female commercial sex workers. The studies were classified according to the adequacy or not of the sample size to estimate HIV prevalence rate and according to the sampling method (probabilistic or convenience). We identified 75 studies that estimated the HIV prevalence rate among female sex workers. Most of the studies employed convenience samples. The sample size was not adequate to estimate HIV prevalence rate in 35 studies. The use of convenience sample limits statistical inference for the whole group. It was observed that there was an increase in the number of published studies since 2005, as well as in the number of studies that used probabilistic samples. This represents a large advance in the monitoring of risk behavior practices and HIV prevalence rate in this group.

Power and sample-size estimation for microbiome studies using pairwise distances and PERMANOVA

PubMed Central

Kelly, Brendan J.; Gross, Robert; Bittinger, Kyle; Sherrill-Mix, Scott; Lewis, James D.; Collman, Ronald G.; Bushman, Frederic D.; Li, Hongzhe

2015-01-01

Motivation: The variation in community composition between microbiome samples, termed beta diversity, can be measured by pairwise distance based on either presence–absence or quantitative species abundance data. PERMANOVA, a permutation-based extension of multivariate analysis of variance to a matrix of pairwise distances, partitions within-group and between-group distances to permit assessment of the effect of an exposure or intervention (grouping factor) upon the sampled microbiome. Within-group distance and exposure/intervention effect size must be accurately modeled to estimate statistical power for a microbiome study that will be analyzed with pairwise distances and PERMANOVA. Results: We present a framework for PERMANOVA power estimation tailored to marker-gene microbiome studies that will be analyzed by pairwise distances, which includes: (i) a novel method for distance matrix simulation that permits modeling of within-group pairwise distances according to pre-specified population parameters; (ii) a method to incorporate effects of different sizes within the simulated distance matrix; (iii) a simulation-based method for estimating PERMANOVA power from simulated distance matrices; and (iv) an R statistical software package that implements the above. Matrices of pairwise distances can be efficiently simulated to satisfy the triangle inequality and incorporate group-level effects, which are quantified by the adjusted coefficient of determination, omega-squared (ω2). From simulated distance matrices, available PERMANOVA power or necessary sample size can be estimated for a planned microbiome study. Availability and implementation: http://github.com/brendankelly/micropower. Contact: brendank@mail.med.upenn.edu or hongzhe@upenn.edu PMID:25819674

Eye-gaze determination of user intent at the computer interface

DOE Office of Scientific and Technical Information (OSTI.GOV)

Goldberg, J.H.; Schryver, J.C.

1993-12-31

Determination of user intent at the computer interface through eye-gaze monitoring can significantly aid applications for the disabled, as well as telerobotics and process control interfaces. Whereas current eye-gaze control applications are limited to object selection and x/y gazepoint tracking, a methodology was developed here to discriminate a more abstract interface operation: zooming-in or out. This methodology first collects samples of eve-gaze location looking at controlled stimuli, at 30 Hz, just prior to a user`s decision to zoom. The sample is broken into data frames, or temporal snapshots. Within a data frame, all spatial samples are connected into a minimummore » spanning tree, then clustered, according to user defined parameters. Each cluster is mapped to one in the prior data frame, and statistics are computed from each cluster. These characteristics include cluster size, position, and pupil size. A multiple discriminant analysis uses these statistics both within and between data frames to formulate optimal rules for assigning the observations into zooming, zoom-out, or no zoom conditions. The statistical procedure effectively generates heuristics for future assignments, based upon these variables. Future work will enhance the accuracy and precision of the modeling technique, and will empirically test users in controlled experiments.« less

Development and Validation of Pathogen Environmental Monitoring Programs for Small Cheese Processing Facilities.

PubMed

Beno, Sarah M; Stasiewicz, Matthew J; Andrus, Alexis D; Ralyea, Robert D; Kent, David J; Martin, Nicole H; Wiedmann, Martin; Boor, Kathryn J

2016-12-01

Pathogen environmental monitoring programs (EMPs) are essential for food processing facilities of all sizes that produce ready-to-eat food products exposed to the processing environment. We developed, implemented, and evaluated EMPs targeting Listeria spp. and Salmonella in nine small cheese processing facilities, including seven farmstead facilities. Individual EMPs with monthly sample collection protocols were designed specifically for each facility. Salmonella was detected in only one facility, with likely introduction from the adjacent farm indicated by pulsed-field gel electrophoresis data. Listeria spp. were isolated from all nine facilities during routine sampling. The overall Listeria spp. (other than Listeria monocytogenes ) and L. monocytogenes prevalences in the 4,430 environmental samples collected were 6.03 and 1.35%, respectively. Molecular characterization and subtyping data suggested persistence of a given Listeria spp. strain in seven facilities and persistence of L. monocytogenes in four facilities. To assess routine sampling plans, validation sampling for Listeria spp. was performed in seven facilities after at least 6 months of routine sampling. This validation sampling was performed by independent individuals and included collection of 50 to 150 samples per facility, based on statistical sample size calculations. Two of the facilities had a significantly higher frequency of detection of Listeria spp. during the validation sampling than during routine sampling, whereas two other facilities had significantly lower frequencies of detection. This study provides a model for a science- and statistics-based approach to developing and validating pathogen EMPs.

Biostatistical analysis of quantitative immunofluorescence microscopy images.

PubMed

Giles, C; Albrecht, M A; Lam, V; Takechi, R; Mamo, J C

2016-12-01

Semiquantitative immunofluorescence microscopy has become a key methodology in biomedical research. Typical statistical workflows are considered in the context of avoiding pseudo-replication and marginalising experimental error. However, immunofluorescence microscopy naturally generates hierarchically structured data that can be leveraged to improve statistical power and enrich biological interpretation. Herein, we describe a robust distribution fitting procedure and compare several statistical tests, outlining their potential advantages/disadvantages in the context of biological interpretation. Further, we describe tractable procedures for power analysis that incorporates the underlying distribution, sample size and number of images captured per sample. The procedures outlined have significant potential for increasing understanding of biological processes and decreasing both ethical and financial burden through experimental optimization. © 2016 The Authors Journal of Microscopy © 2016 Royal Microscopical Society.

Steep discounting of delayed monetary and food rewards in obesity: a meta-analysis.

PubMed

Amlung, M; Petker, T; Jackson, J; Balodis, I; MacKillop, J

2016-08-01

An increasing number of studies have investigated delay discounting (DD) in relation to obesity, but with mixed findings. This meta-analysis synthesized the literature on the relationship between monetary and food DD and obesity, with three objectives: (1) to characterize the relationship between DD and obesity in both case-control comparisons and continuous designs; (2) to examine potential moderators, including case-control v. continuous design, money v. food rewards, sample sex distribution, and sample age (18 years); and (3) to evaluate publication bias. From 134 candidate articles, 39 independent investigations yielded 29 case-control and 30 continuous comparisons (total n = 10 278). Random-effects meta-analysis was conducted using Cohen's d as the effect size. Publication bias was evaluated using fail-safe N, Begg-Mazumdar and Egger tests, meta-regression of publication year and effect size, and imputation of missing studies. The primary analysis revealed a medium effect size across studies that was highly statistically significant (d = 0.43, p < 10-14). None of the moderators examined yielded statistically significant differences, although notably larger effect sizes were found for studies with case-control designs, food rewards and child/adolescent samples. Limited evidence of publication bias was present, although the Begg-Mazumdar test and meta-regression suggested a slightly diminishing effect size over time. Steep DD of food and money appears to be a robust feature of obesity that is relatively consistent across the DD assessment methodologies and study designs examined. These findings are discussed in the context of research on DD in drug addiction, the neural bases of DD in obesity, and potential clinical applications.

Statistics Clinic

NASA Technical Reports Server (NTRS)

Feiveson, Alan H.; Foy, Millennia; Ploutz-Snyder, Robert; Fiedler, James

2014-01-01

Do you have elevated p-values? Is the data analysis process getting you down? Do you experience anxiety when you need to respond to criticism of statistical methods in your manuscript? You may be suffering from Insufficient Statistical Support Syndrome (ISSS). For symptomatic relief of ISSS, come for a free consultation with JSC biostatisticians at our help desk during the poster sessions at the HRP Investigators Workshop. Get answers to common questions about sample size, missing data, multiple testing, when to trust the results of your analyses and more. Side effects may include sudden loss of statistics anxiety, improved interpretation of your data, and increased confidence in your results.

Statistical Design for Biospecimen Cohort Size in Proteomics-based Biomarker Discovery and Verification Studies

PubMed Central

Skates, Steven J.; Gillette, Michael A.; LaBaer, Joshua; Carr, Steven A.; Anderson, N. Leigh; Liebler, Daniel C.; Ransohoff, David; Rifai, Nader; Kondratovich, Marina; Težak, Živana; Mansfield, Elizabeth; Oberg, Ann L.; Wright, Ian; Barnes, Grady; Gail, Mitchell; Mesri, Mehdi; Kinsinger, Christopher R.; Rodriguez, Henry; Boja, Emily S.

2014-01-01

Protein biomarkers are needed to deepen our understanding of cancer biology and to improve our ability to diagnose, monitor and treat cancers. Important analytical and clinical hurdles must be overcome to allow the most promising protein biomarker candidates to advance into clinical validation studies. Although contemporary proteomics technologies support the measurement of large numbers of proteins in individual clinical specimens, sample throughput remains comparatively low. This problem is amplified in typical clinical proteomics research studies, which routinely suffer from a lack of proper experimental design, resulting in analysis of too few biospecimens to achieve adequate statistical power at each stage of a biomarker pipeline. To address this critical shortcoming, a joint workshop was held by the National Cancer Institute (NCI), National Heart, Lung and Blood Institute (NHLBI), and American Association for Clinical Chemistry (AACC), with participation from the U.S. Food and Drug Administration (FDA). An important output from the workshop was a statistical framework for the design of biomarker discovery and verification studies. Herein, we describe the use of quantitative clinical judgments to set statistical criteria for clinical relevance, and the development of an approach to calculate biospecimen sample size for proteomic studies in discovery and verification stages prior to clinical validation stage. This represents a first step towards building a consensus on quantitative criteria for statistical design of proteomics biomarker discovery and verification research. PMID:24063748

Statistical design for biospecimen cohort size in proteomics-based biomarker discovery and verification studies.

PubMed

Skates, Steven J; Gillette, Michael A; LaBaer, Joshua; Carr, Steven A; Anderson, Leigh; Liebler, Daniel C; Ransohoff, David; Rifai, Nader; Kondratovich, Marina; Težak, Živana; Mansfield, Elizabeth; Oberg, Ann L; Wright, Ian; Barnes, Grady; Gail, Mitchell; Mesri, Mehdi; Kinsinger, Christopher R; Rodriguez, Henry; Boja, Emily S

2013-12-06

Protein biomarkers are needed to deepen our understanding of cancer biology and to improve our ability to diagnose, monitor, and treat cancers. Important analytical and clinical hurdles must be overcome to allow the most promising protein biomarker candidates to advance into clinical validation studies. Although contemporary proteomics technologies support the measurement of large numbers of proteins in individual clinical specimens, sample throughput remains comparatively low. This problem is amplified in typical clinical proteomics research studies, which routinely suffer from a lack of proper experimental design, resulting in analysis of too few biospecimens to achieve adequate statistical power at each stage of a biomarker pipeline. To address this critical shortcoming, a joint workshop was held by the National Cancer Institute (NCI), National Heart, Lung, and Blood Institute (NHLBI), and American Association for Clinical Chemistry (AACC) with participation from the U.S. Food and Drug Administration (FDA). An important output from the workshop was a statistical framework for the design of biomarker discovery and verification studies. Herein, we describe the use of quantitative clinical judgments to set statistical criteria for clinical relevance and the development of an approach to calculate biospecimen sample size for proteomic studies in discovery and verification stages prior to clinical validation stage. This represents a first step toward building a consensus on quantitative criteria for statistical design of proteomics biomarker discovery and verification research.

Empirical assessment of published effect sizes and power in the recent cognitive neuroscience and psychology literature.

PubMed

Szucs, Denes; Ioannidis, John P A

2017-03-01

We have empirically assessed the distribution of published effect sizes and estimated power by analyzing 26,841 statistical records from 3,801 cognitive neuroscience and psychology papers published recently. The reported median effect size was D = 0.93 (interquartile range: 0.64-1.46) for nominally statistically significant results and D = 0.24 (0.11-0.42) for nonsignificant results. Median power to detect small, medium, and large effects was 0.12, 0.44, and 0.73, reflecting no improvement through the past half-century. This is so because sample sizes have remained small. Assuming similar true effect sizes in both disciplines, power was lower in cognitive neuroscience than in psychology. Journal impact factors negatively correlated with power. Assuming a realistic range of prior probabilities for null hypotheses, false report probability is likely to exceed 50% for the whole literature. In light of our findings, the recently reported low replication success in psychology is realistic, and worse performance may be expected for cognitive neuroscience.

Improved population estimates through the use of auxiliary information

USGS Publications Warehouse

Johnson, D.H.; Ralph, C.J.; Scott, J.M.

1981-01-01

When estimating the size of a population of birds, the investigator may have, in addition to an estimator based on a statistical sample, information on one of several auxiliary variables, such as: (1) estimates of the population made on previous occasions, (2) measures of habitat variables associated with the size of the population, and (3) estimates of the population sizes of other species that correlate with the species of interest. Although many studies have described the relationships between each of these kinds of data and the population size to be estimated, very little work has been done to improve the estimator by incorporating such auxiliary information. A statistical methodology termed 'empirical Bayes' seems to be appropriate to these situations. The potential that empirical Bayes methodology has for improved estimation of the population size of the Mallard (Anas platyrhynchos) is explored. In the example considered, three empirical Bayes estimators were found to reduce the error by one-fourth to one-half of that of the usual estimator.

Testing for qualitative heterogeneity: An application to composite endpoints in survival analysis.

PubMed

Oulhaj, Abderrahim; El Ghouch, Anouar; Holman, Rury R

2017-01-01

Composite endpoints are frequently used in clinical outcome trials to provide more endpoints, thereby increasing statistical power. A key requirement for a composite endpoint to be meaningful is the absence of the so-called qualitative heterogeneity to ensure a valid overall interpretation of any treatment effect identified. Qualitative heterogeneity occurs when individual components of a composite endpoint exhibit differences in the direction of a treatment effect. In this paper, we develop a general statistical method to test for qualitative heterogeneity, that is to test whether a given set of parameters share the same sign. This method is based on the intersection-union principle and, provided that the sample size is large, is valid whatever the model used for parameters estimation. We propose two versions of our testing procedure, one based on a random sampling from a Gaussian distribution and another version based on bootstrapping. Our work covers both the case of completely observed data and the case where some observations are censored which is an important issue in many clinical trials. We evaluated the size and power of our proposed tests by carrying out some extensive Monte Carlo simulations in the case of multivariate time to event data. The simulations were designed under a variety of conditions on dimensionality, censoring rate, sample size and correlation structure. Our testing procedure showed very good performances in terms of statistical power and type I error. The proposed test was applied to a data set from a single-center, randomized, double-blind controlled trial in the area of Alzheimer's disease.

Unequal cluster sizes in stepped-wedge cluster randomised trials: a systematic review.

PubMed

Kristunas, Caroline; Morris, Tom; Gray, Laura

2017-11-15

To investigate the extent to which cluster sizes vary in stepped-wedge cluster randomised trials (SW-CRT) and whether any variability is accounted for during the sample size calculation and analysis of these trials. Any, not limited to healthcare settings. Any taking part in an SW-CRT published up to March 2016. The primary outcome is the variability in cluster sizes, measured by the coefficient of variation (CV) in cluster size. Secondary outcomes include the difference between the cluster sizes assumed during the sample size calculation and those observed during the trial, any reported variability in cluster sizes and whether the methods of sample size calculation and methods of analysis accounted for any variability in cluster sizes. Of the 101 included SW-CRTs, 48% mentioned that the included clusters were known to vary in size, yet only 13% of these accounted for this during the calculation of the sample size. However, 69% of the trials did use a method of analysis appropriate for when clusters vary in size. Full trial reports were available for 53 trials. The CV was calculated for 23 of these: the median CV was 0.41 (IQR: 0.22-0.52). Actual cluster sizes could be compared with those assumed during the sample size calculation for 14 (26%) of the trial reports; the cluster sizes were between 29% and 480% of that which had been assumed. Cluster sizes often vary in SW-CRTs. Reporting of SW-CRTs also remains suboptimal. The effect of unequal cluster sizes on the statistical power of SW-CRTs needs further exploration and methods appropriate to studies with unequal cluster sizes need to be employed. © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2017. All rights reserved. No commercial use is permitted unless otherwise expressly granted.

Bayesian evaluation of effect size after replicating an original study

PubMed Central

van Aert, Robbie C. M.; van Assen, Marcel A. L. M.

2017-01-01

The vast majority of published results in the literature is statistically significant, which raises concerns about their reliability. The Reproducibility Project Psychology (RPP) and Experimental Economics Replication Project (EE-RP) both replicated a large number of published studies in psychology and economics. The original study and replication were statistically significant in 36.1% in RPP and 68.8% in EE-RP suggesting many null effects among the replicated studies. However, evidence in favor of the null hypothesis cannot be examined with null hypothesis significance testing. We developed a Bayesian meta-analysis method called snapshot hybrid that is easy to use and understand and quantifies the amount of evidence in favor of a zero, small, medium and large effect. The method computes posterior model probabilities for a zero, small, medium, and large effect and adjusts for publication bias by taking into account that the original study is statistically significant. We first analytically approximate the methods performance, and demonstrate the necessity to control for the original study’s significance to enable the accumulation of evidence for a true zero effect. Then we applied the method to the data of RPP and EE-RP, showing that the underlying effect sizes of the included studies in EE-RP are generally larger than in RPP, but that the sample sizes of especially the included studies in RPP are often too small to draw definite conclusions about the true effect size. We also illustrate how snapshot hybrid can be used to determine the required sample size of the replication akin to power analysis in null hypothesis significance testing and present an easy to use web application (https://rvanaert.shinyapps.io/snapshot/) and R code for applying the method. PMID:28388646

Evolution of sociality by natural selection on variances in reproductive fitness: evidence from a social bee.

PubMed

Stevens, Mark I; Hogendoorn, Katja; Schwarz, Michael P

2007-08-29

The Central Limit Theorem (CLT) is a statistical principle that states that as the number of repeated samples from any population increase, the variance among sample means will decrease and means will become more normally distributed. It has been conjectured that the CLT has the potential to provide benefits for group living in some animals via greater predictability in food acquisition, if the number of foraging bouts increases with group size. The potential existence of benefits for group living derived from a purely statistical principle is highly intriguing and it has implications for the origins of sociality. Here we show that in a social allodapine bee the relationship between cumulative food acquisition (measured as total brood weight) and colony size accords with the CLT. We show that deviations from expected food income decrease with group size, and that brood weights become more normally distributed both over time and with increasing colony size, as predicted by the CLT. Larger colonies are better able to match egg production to expected food intake, and better able to avoid costs associated with producing more brood than can be reared while reducing the risk of under-exploiting the food resources that may be available. These benefits to group living derive from a purely statistical principle, rather than from ecological, ergonomic or genetic factors, and could apply to a wide variety of species. This in turn suggests that the CLT may provide benefits at the early evolutionary stages of sociality and that evolution of group size could result from selection on variances in reproductive fitness. In addition, they may help explain why sociality has evolved in some groups and not others.

Prediction of Active-Region CME Productivity from Magnetograms

NASA Technical Reports Server (NTRS)

Falconer, D. A.; Moore, R. L.; Gary, G. A.

2004-01-01

We report results of an expanded evaluation of whole-active-region magnetic measures as predictors of active-region coronal mass ejection (CME) productivity. Previously, in a sample of 17 vector magnetograms of 12 bipolar active regions observed by the Marshall Space Flight Center (MSFC) vector magnetograph, from each magnetogram we extracted a measure of the size of the active region (the active region s total magnetic flux a) and four measures of the nonpotentiality of the active region: the strong-shear length L(sub SS), the strong-gradient length L(sub SG), the net vertical electric current I(sub N), and the net-current magnetic twist parameter alpha (sub IN). This sample size allowed us to show that each of the four nonpotentiality measures was statistically significantly correlated with active-region CME productivity in time windows of a few days centered on the day of the magnetogram. We have now added a fifth measure of active-region nonpotentiality (the best-constant-alpha magnetic twist parameter (alpha sub BC)), and have expanded the sample to 36 MSFC vector magnetograms of 31 bipolar active regions. This larger sample allows us to demonstrate statistically significant correlations of each of the five nonpotentiality measures with future CME productivity, in time windows of a few days starting from the day of the magnetogram. The two magnetic twist parameters (alpha (sub 1N) and alpha (sub BC)) are normalized measures of an active region s nonpotentially in that they do not depend directly on the size of the active region, while the other three nonpotentiality measures (L(sub SS), L(sub SG), and I(sub N)) are non-normalized measures in that they do depend directly on active-region size. We find (1) Each of the five nonpotentiality measures is statistically significantly correlated (correlation confidence level greater than 95%) with future CME productivity and has a CME prediction success rate of approximately 80%. (2) None of the nonpotentiality measures is a significantly better CME predictor than the others. (3) The active-region phi shows some correlation with CME productivity, but well below a statistically significant level (correlation confidence level less than approximately 80%; CME prediction success rate less than approximately 65%). (4) In addition to depending on magnetic twist, CME productivity appears to have some direct dependence on active-region size (rather than only an indirect dependence through a correlation of magnetic twist with active-region size), but it will take a still larger sample of active regions (50 or more) to certify this. (5) Of the five nonpotentiality measures, L(sub SG) appears to be the best for operational CME forecasting because it is as good or better a CME predictor than the others and it alone does not require a vector magnetogram; L(sub SG) can be measured from a line-of-sight magnetogram such as from the Michelson Doppler Imager (MDI) on the Solar and Heliospheric Observatory (SOHO).

Optical Clock Distribution to VLSI Chips

DTIC Science & Technology

1989-07-01

depletion layer edge represents a lower concentration of minority carriers, minority carriers generated outside the depletion layer statistically tend to...completed and the output voltage levels for the next state to be transfered to the intermediate buffer by means of the second pass transitor . Since 41...a sample size too small to have statistical significance, the results give an indication of the typical operating frequency that is possible with a

Where statistics and molecular microarray experiments biology meet.

PubMed

Kelmansky, Diana M

2013-01-01

This review chapter presents a statistical point of view to microarray experiments with the purpose of understanding the apparent contradictions that often appear in relation to their results. We give a brief introduction of molecular biology for nonspecialists. We describe microarray experiments from their construction and the biological principles the experiments rely on, to data acquisition and analysis. The role of epidemiological approaches and sample size considerations are also discussed.

Prevalence of diseases and statistical power of the Japan Nurses' Health Study.

PubMed

Fujita, Toshiharu; Hayashi, Kunihiko; Katanoda, Kota; Matsumura, Yasuhiro; Lee, Jung Su; Takagi, Hirofumi; Suzuki, Shosuke; Mizunuma, Hideki; Aso, Takeshi

2007-10-01

The Japan Nurses' Health Study (JNHS) is a long-term, large-scale cohort study investigating the effects of various lifestyle factors and healthcare habits on the health of Japanese women. Based on currently limited statistical data regarding the incidence of disease among Japanese women, our initial sample size was tentatively set at 50,000 during the design phase. The actual number of women who agreed to participate in follow-up surveys was approximately 18,000. Taking into account the actual sample size and new information on disease frequency obtained during the baseline component, we established the prevalence of past diagnoses of target diseases, predicted their incidence, and calculated the statistical power for JNHS follow-up surveys. For all diseases except ovarian cancer, the prevalence of a past diagnosis increased markedly with age, and incidence rates could be predicted based on the degree of increase in prevalence between two adjacent 5-yr age groups. The predicted incidence rate for uterine myoma, hypercholesterolemia, and hypertension was > or =3.0 (per 1,000 women, per year), while the rate of thyroid disease, hepatitis, gallstone disease, and benign breast tumor was predicted to be > or =1.0. For these diseases, the statistical power to detect risk factors with a relative risk of 1.5 or more within ten years, was 70% or higher.

VARIABLE SELECTION IN NONPARAMETRIC ADDITIVE MODELS

PubMed Central

Huang, Jian; Horowitz, Joel L.; Wei, Fengrong

2010-01-01

We consider a nonparametric additive model of a conditional mean function in which the number of variables and additive components may be larger than the sample size but the number of nonzero additive components is “small” relative to the sample size. The statistical problem is to determine which additive components are nonzero. The additive components are approximated by truncated series expansions with B-spline bases. With this approximation, the problem of component selection becomes that of selecting the groups of coefficients in the expansion. We apply the adaptive group Lasso to select nonzero components, using the group Lasso to obtain an initial estimator and reduce the dimension of the problem. We give conditions under which the group Lasso selects a model whose number of components is comparable with the underlying model, and the adaptive group Lasso selects the nonzero components correctly with probability approaching one as the sample size increases and achieves the optimal rate of convergence. The results of Monte Carlo experiments show that the adaptive group Lasso procedure works well with samples of moderate size. A data example is used to illustrate the application of the proposed method. PMID:21127739

Sampling benthic macroinvertebrates in a large flood-plain river: Considerations of study design, sample size, and cost

USGS Publications Warehouse

Bartsch, L.A.; Richardson, W.B.; Naimo, T.J.

1998-01-01

Estimation of benthic macroinvertebrate populations over large spatial scales is difficult due to the high variability in abundance and the cost of sample processing and taxonomic analysis. To determine a cost-effective, statistically powerful sample design, we conducted an exploratory study of the spatial variation of benthic macroinvertebrates in a 37 km reach of the Upper Mississippi River. We sampled benthos at 36 sites within each of two strata, contiguous backwater and channel border. Three standard ponar (525 cm(2)) grab samples were obtained at each site ('Original Design'). Analysis of variance and sampling cost of strata-wide estimates for abundance of Oligochaeta, Chironomidae, and total invertebrates showed that only one ponar sample per site ('Reduced Design') yielded essentially the same abundance estimates as the Original Design, while reducing the overall cost by 63%. A posteriori statistical power analysis (alpha = 0.05, beta = 0.20) on the Reduced Design estimated that at least 18 sites per stratum were needed to detect differences in mean abundance between contiguous backwater and channel border areas for Oligochaeta, Chironomidae, and total invertebrates. Statistical power was nearly identical for the three taxonomic groups. The abundances of several taxa of concern (e.g., Hexagenia mayflies and Musculium fingernail clams) were too spatially variable to estimate power with our method. Resampling simulations indicated that to achieve adequate sampling precision for Oligochaeta, at least 36 sample sites per stratum would be required, whereas a sampling precision of 0.2 would not be attained with any sample size for Hexagenia in channel border areas, or Chironomidae and Musculium in both strata given the variance structure of the original samples. Community-wide diversity indices (Brillouin and 1-Simpsons) increased as sample area per site increased. The backwater area had higher diversity than the channel border area. The number of sampling sites required to sample benthic macroinvertebrates during our sampling period depended on the study objective and ranged from 18 to more than 40 sites per stratum. No single sampling regime would efficiently and adequately sample all components of the macroinvertebrate community.

Analyzing thematic maps and mapping for accuracy

USGS Publications Warehouse

Rosenfield, G.H.

1982-01-01

Two problems which exist while attempting to test the accuracy of thematic maps and mapping are: (1) evaluating the accuracy of thematic content, and (2) evaluating the effects of the variables on thematic mapping. Statistical analysis techniques are applicable to both these problems and include techniques for sampling the data and determining their accuracy. In addition, techniques for hypothesis testing, or inferential statistics, are used when comparing the effects of variables. A comprehensive and valid accuracy test of a classification project, such as thematic mapping from remotely sensed data, includes the following components of statistical analysis: (1) sample design, including the sample distribution, sample size, size of the sample unit, and sampling procedure; and (2) accuracy estimation, including estimation of the variance and confidence limits. Careful consideration must be given to the minimum sample size necessary to validate the accuracy of a given. classification category. The results of an accuracy test are presented in a contingency table sometimes called a classification error matrix. Usually the rows represent the interpretation, and the columns represent the verification. The diagonal elements represent the correct classifications. The remaining elements of the rows represent errors by commission, and the remaining elements of the columns represent the errors of omission. For tests of hypothesis that compare variables, the general practice has been to use only the diagonal elements from several related classification error matrices. These data are arranged in the form of another contingency table. The columns of the table represent the different variables being compared, such as different scales of mapping. The rows represent the blocking characteristics, such as the various categories of classification. The values in the cells of the tables might be the counts of correct classification or the binomial proportions of these counts divided by either the row totals or the column totals from the original classification error matrices. In hypothesis testing, when the results of tests of multiple sample cases prove to be significant, some form of statistical test must be used to separate any results that differ significantly from the others. In the past, many analyses of the data in this error matrix were made by comparing the relative magnitudes of the percentage of correct classifications, for either individual categories, the entire map or both. More rigorous analyses have used data transformations and (or) two-way classification analysis of variance. A more sophisticated step of data analysis techniques would be to use the entire classification error matrices using the methods of discrete multivariate analysis or of multiviariate analysis of variance.

Sample size determination for disease prevalence studies with partially validated data.

PubMed

Qiu, Shi-Fang; Poon, Wai-Yin; Tang, Man-Lai

2016-02-01

Disease prevalence is an important topic in medical research, and its study is based on data that are obtained by classifying subjects according to whether a disease has been contracted. Classification can be conducted with high-cost gold standard tests or low-cost screening tests, but the latter are subject to the misclassification of subjects. As a compromise between the two, many research studies use partially validated datasets in which all data points are classified by fallible tests, and some of the data points are validated in the sense that they are also classified by the completely accurate gold-standard test. In this article, we investigate the determination of sample sizes for disease prevalence studies with partially validated data. We use two approaches. The first is to find sample sizes that can achieve a pre-specified power of a statistical test at a chosen significance level, and the second is to find sample sizes that can control the width of a confidence interval with a pre-specified confidence level. Empirical studies have been conducted to demonstrate the performance of various testing procedures with the proposed sample sizes. The applicability of the proposed methods are illustrated by a real-data example. © The Author(s) 2012.

A Scanning Transmission Electron Microscopy Method for Determining Manganese Composition in Welding Fume as a Function of Primary Particle Size

PubMed Central

Richman, Julie D.; Livi, Kenneth J.T.; Geyh, Alison S.

2011-01-01

Increasing evidence suggests that the physicochemical properties of inhaled nanoparticles influence the resulting toxicokinetics and toxicodynamics. This report presents a method using scanning transmission electron microscopy (STEM) to measure the Mn content throughout the primary particle size distribution of welding fume particle samples collected on filters for application in exposure and health research. Dark field images were collected to assess the primary particle size distribution and energy-dispersive X-ray and electron energy loss spectroscopy were performed for measurement of Mn composition as a function of primary particle size. A manual method incorporating imaging software was used to measure the primary particle diameter and to select an integration region for compositional analysis within primary particles throughout the size range. To explore the variation in the developed metric, the method was applied to 10 gas metal arc welding (GMAW) fume particle samples of mild steel that were collected under a variety of conditions. The range of Mn composition by particle size was −0.10 to 0.19 %/nm, where a positive estimate indicates greater relative abundance of Mn increasing with primary particle size and a negative estimate conversely indicates decreasing Mn content with size. However, the estimate was only statistically significant (p<0.05) in half of the samples (n=5), which all had a positive estimate. In the remaining samples, no significant trend was measured. Our findings indicate that the method is reproducible and that differences in the abundance of Mn by primary particle size among welding fume samples can be detected. PMID:21625364

A Scanning Transmission Electron Microscopy Method for Determining Manganese Composition in Welding Fume as a Function of Primary Particle Size.

PubMed

Richman, Julie D; Livi, Kenneth J T; Geyh, Alison S

2011-06-01

Increasing evidence suggests that the physicochemical properties of inhaled nanoparticles influence the resulting toxicokinetics and toxicodynamics. This report presents a method using scanning transmission electron microscopy (STEM) to measure the Mn content throughout the primary particle size distribution of welding fume particle samples collected on filters for application in exposure and health research. Dark field images were collected to assess the primary particle size distribution and energy-dispersive X-ray and electron energy loss spectroscopy were performed for measurement of Mn composition as a function of primary particle size. A manual method incorporating imaging software was used to measure the primary particle diameter and to select an integration region for compositional analysis within primary particles throughout the size range. To explore the variation in the developed metric, the method was applied to 10 gas metal arc welding (GMAW) fume particle samples of mild steel that were collected under a variety of conditions. The range of Mn composition by particle size was -0.10 to 0.19 %/nm, where a positive estimate indicates greater relative abundance of Mn increasing with primary particle size and a negative estimate conversely indicates decreasing Mn content with size. However, the estimate was only statistically significant (p<0.05) in half of the samples (n=5), which all had a positive estimate. In the remaining samples, no significant trend was measured. Our findings indicate that the method is reproducible and that differences in the abundance of Mn by primary particle size among welding fume samples can be detected.

The Effect of Unequal Samples, Heterogeneity of Covariance Matrices, and Number of Variables on Discriminant Analysis Classification Tables and Related Statistics.

ERIC Educational Resources Information Center

Spearing, Debra; Woehlke, Paula

To assess the effect on discriminant analysis in terms of correct classification into two groups, the following parameters were systematically altered using Monte Carlo techniques: sample sizes; proportions of one group to the other; number of independent variables; and covariance matrices. The pairing of the off diagonals (or covariances) with…

The Influence of Experimental Design on the Detection of Performance Differences

ERIC Educational Resources Information Center

Bates, B. T.; Dufek, J. S.; James, C. R.; Harry, J. R.; Eggleston, J. D.

2016-01-01

We demonstrate the effect of sample and trial size on statistical outcomes for single-subject analyses (SSA) and group analyses (GA) for a frequently studied performance activity and common intervention. Fifty strides of walking data collected in two blocks of 25 trials for two shoe conditions were analyzed for samples of five, eight, 10, and 12…

A computational framework for estimating statistical power and planning hypothesis-driven experiments involving one-dimensional biomechanical continua.

PubMed

Pataky, Todd C; Robinson, Mark A; Vanrenterghem, Jos

2018-01-03

Statistical power assessment is an important component of hypothesis-driven research but until relatively recently (mid-1990s) no methods were available for assessing power in experiments involving continuum data and in particular those involving one-dimensional (1D) time series. The purpose of this study was to describe how continuum-level power analyses can be used to plan hypothesis-driven biomechanics experiments involving 1D data. In particular, we demonstrate how theory- and pilot-driven 1D effect modeling can be used for sample-size calculations for both single- and multi-subject experiments. For theory-driven power analysis we use the minimum jerk hypothesis and single-subject experiments involving straight-line, planar reaching. For pilot-driven power analysis we use a previously published knee kinematics dataset. Results show that powers on the order of 0.8 can be achieved with relatively small sample sizes, five and ten for within-subject minimum jerk analysis and between-subject knee kinematics, respectively. However, the appropriate sample size depends on a priori justifications of biomechanical meaning and effect size. The main advantage of the proposed technique is that it encourages a priori justification regarding the clinical and/or scientific meaning of particular 1D effects, thereby robustly structuring subsequent experimental inquiry. In short, it shifts focus from a search for significance to a search for non-rejectable hypotheses. Copyright © 2017 Elsevier Ltd. All rights reserved.

Characterization of stormwater runoff from bridge decks in eastern Massachusetts, 2014–16

USGS Publications Warehouse

Smith, Kirk P.; Sorenson, Jason R.; Granato, Gregory E.

2018-05-02

The quality of stormwater runoff from bridge decks (hereafter referred to as “bridge-deck runoff”) was characterized in a field study from August 2014 through August 2016 in which concentrations of suspended sediment (SS) and total nutrients were monitored. These new data were collected to supplement existing highway-runoff data collected in Massachusetts which were deficient in bridge-deck runoff concentration data. Monitoring stations were installed at three bridges maintained by the Massachusetts Department of Transportation in eastern Massachusetts (State Route 2A in the city of Boston, Interstate 90 in the town of Weston, and State Route 20 near Quinsigamond Village in the city of Worcester). The bridges had annual average daily traffic volumes from 21,200 to 124,000 vehicles per day; the land use surrounding the monitoring stations was 25 to 67 percent impervious.Automatic-monitoring techniques were used to collect more than 160 flow-proportional composite samples of bridge-deck runoff. Samples were analyzed for concentrations of SS, loss on ignition of suspended solids (LOI), particulate carbon (PC), total phosphorus (TP), total dissolved nitrogen (DN), and particulate nitrogen (PN). The distribution of particle size of SS also was determined for composite samples. Samples of bridge-deck runoff were collected year round during rain, mixed precipitation, and snowmelt runoff and with different dry antecedent periods throughout the 2-year sampling period.At the three bridge-deck-monitoring stations, median concentrations of SS in composite samples of bridge-deck runoff ranged from 1,490 to 2,020 milligrams per liter (mg/L); however, the range of SS in individual composites was vast at 44 to 142,000 mg/L. Median concentrations of SS were similar in composite samples collected from the State Route 2A and Interstate 90 bridge (2,010 and 2,020 mg/L, respectively), and lowest at the State Route 20 bridge (1,490 mg/L). Concentrations of coarse sediment (greater than 0.25 millimeters in diameter) dominated the SS matrix by more than an order of magnitude. Concentrations of LOI and PC in composite samples ranged from 15 to 1,740 mg/L and 6.68 to 1,360 mg/L, respectively, and generally represented less than 10 and 3 percent of the median mass of SS, respectively. Concentrations of TP in composite samples ranged from 0.09 to 7.02 mg/L; median concentrations of TP ranged from 0.505 to 0.69 mg/L and were highest on the bridge on State Route 2A in Boston. Concentrations of total nitrogen (TN) (sum DN and PN) in composite samples were variable (0.36 to 29 mg/L). Median DN (0.64 to 0.90 mg/L) concentrations generally represented about 40 percent of the TN concentration at each bridge and were similar to annual volume-weighted mean concentrations of nitrogen in precipitation in Massachusetts.Nonparametric statistical methods were used to test for differences between sample constituent concentrations among the three bridges. These results indicated that there are no statistically significant differences for concentrations of SS, LOI, PC, and TP among the three bridges (one-way analysis of variance test on rank-transformed data, 95-percent confidence level). Test results for concentrations of TN in composite samples indicated that concentrations of TN collected on State Route 20 near Quinsigamond Village were significantly higher than those concentrations collected on State Route 2A in Boston and Interstate 90 near Weston. Median concentrations of TN were about 93 and 55 percent lower at State Route 2A and at Interstate 90, respectively, compared to the median concentrations of TN at State Route 20.Samples of sediment were collected from five fixed locations on each bridge on three occasions during dry weather to calculate semiquantitative distributions of sediment yields on the bridge surface relative to the monitoring location. Mean yields of bridge-deck sediment during this study for State Route 2A in Boston, Interstate 90 near Weston, and State Route 20 near Quinsigamond Village were 1,500, 250, and 5,700 pounds per curb-mile, respectively. Sediment yields at each sampling location varied widely (26 to 25,000 pounds per curb-mile) but were similar to yields reported elsewhere in Massachusetts and the United States. Yields calculated for each sampling location indicated that the sediment was not evenly distributed across each bridge in this study for plausible reasons such as bridge slope, vehicular tracking, and bridge deterioration.Bridge-deck sediment quality was largely affected by the distribution of sediment particle size. Concentrations of TP in the fine sediment-size fraction (less than 0.0625 millimeter in diameter) of samples of bridge-deck sediment were about 6 times greater than in the coarse size fraction. Concentrations for many total-recoverable metals were 2 to 17 times greater in the fine size fraction compared to concentrations in the coarse size fraction (greater than or equal to 0.25 millimeter in diameter), and concentrations of total-recoverable copper and lead in the fine size fraction were 2 to 65 times higher compared to concentrations in the intermediate (greater than or equal to 0.0625 to 0.25 millimeter in diameter) or the coarse size fraction. However, the proportion of sediment particles less than 0.0625 millimeter in diameter in composite samples of bridge-deck runoff was small (median values range from 4 to 8 percent at each bridge) compared to the larger sediment particle-size mass. As a result, more than 50 percent of the sediment-associated TP, aluminum, chromium, manganese, and nickel was estimated to be associated with the coarse size fraction of the SS load. In contrast, about 95 percent of the estimated sediment-associated copper concentration was associated with the fine size fraction of the SS load.Version 1.0.2 of the Stochastic Empirical Loading and Dilution Model was used to simulate long-term (29–30-year) concentrations and annual yields of SS, TP, and TN in bridge-deck runoff and in discharges from a hypothetical stormwater treatment best-management practice structure. Three methods (traditional statistics, robust statistics, and L-moments) were used to calculate statistics for stochastic simulations because the high variability in measured concentration values during the field study resulted in extreme simulated concentrations. Statistics of each dataset, including the average, standard deviation, and skew of the common (base 10) logarithms, for each of the three bridges, and for a lumped dataset, were calculated and used for simulations; statistics representing the median of statistics calculated for the three bridges also were used for simulations. These median statistics were selected for the interpretive simulations so that the simulations could be used to estimate concentrations and yields from other, unmonitored bridges in Massachusetts. Comparisons of the standard and robust statistics indicated that simulation results with either method would be similar, which indicated that the large variability in simulated results was not caused by a few outliers. Comparison to statistics calculated by the L-moments methods indicated that L-moments do not produce extreme concentrations; however, they also do not produce results that represent the bulk of concentration data.The runoff-quality risk analysis indicated that bridge-deck runoff would exceed discharge standards commonly used for large, advanced wastewater treatment plants, but that commonly used stormwater best-management practices may reduce the percentage of exceedances by one-half. Results of simulations indicated that long-term average yields of TN, TP, and SS may be about 21.4, 6.44, and 40,600 pounds per acre per year, respectively. These yields are about 1.3, 3.4, and 16 times simulated ultra-urban highway yields in Massachusetts; however, simulations indicated that use of a best-management practice structure to treat bridge-deck runoff may reduce discharge yields to about 10, 2.8, and 4,300, pounds per acre per year, respectively.

[Application of statistics on chronic-diseases-relating observational research papers].

PubMed

Hong, Zhi-heng; Wang, Ping; Cao, Wei-hua

2012-09-01

To study the application of statistics on Chronic-diseases-relating observational research papers which were recently published in the Chinese Medical Association Magazines, with influential index above 0.5. Using a self-developed criterion, two investigators individually participated in assessing the application of statistics on Chinese Medical Association Magazines, with influential index above 0.5. Different opinions reached an agreement through discussion. A total number of 352 papers from 6 magazines, including the Chinese Journal of Epidemiology, Chinese Journal of Oncology, Chinese Journal of Preventive Medicine, Chinese Journal of Cardiology, Chinese Journal of Internal Medicine and Chinese Journal of Endocrinology and Metabolism, were reviewed. The rate of clear statement on the following contents as: research objectives, t target audience, sample issues, objective inclusion criteria and variable definitions were 99.43%, 98.57%, 95.43%, 92.86% and 96.87%. The correct rates of description on quantitative and qualitative data were 90.94% and 91.46%, respectively. The rates on correctly expressing the results, on statistical inference methods related to quantitative, qualitative data and modeling were 100%, 95.32% and 87.19%, respectively. 89.49% of the conclusions could directly response to the research objectives. However, 69.60% of the papers did not mention the exact names of the study design, statistically, that the papers were using. 11.14% of the papers were in lack of further statement on the exclusion criteria. Percentage of the papers that could clearly explain the sample size estimation only taking up as 5.16%. Only 24.21% of the papers clearly described the variable value assignment. Regarding the introduction on statistical conduction and on database methods, the rate was only 24.15%. 18.75% of the papers did not express the statistical inference methods sufficiently. A quarter of the papers did not use 'standardization' appropriately. As for the aspect of statistical inference, the rate of description on statistical testing prerequisite was only 24.12% while 9.94% papers did not even employ the statistical inferential method that should be used. The main deficiencies on the application of Statistics used in papers related to Chronic-diseases-related observational research were as follows: lack of sample-size determination, variable value assignment description not sufficient, methods on statistics were not introduced clearly or properly, lack of consideration for pre-requisition regarding the use of statistical inferences.

The widespread misuse of effect sizes.

PubMed

Dankel, Scott J; Mouser, J Grant; Mattocks, Kevin T; Counts, Brittany R; Jessee, Matthew B; Buckner, Samuel L; Loprinzi, Paul D; Loenneke, Jeremy P

2017-05-01

Studies comparing multiple groups (i.e., experimental and control) often examine the efficacy of an intervention by calculating within group effect sizes using Cohen's d. This method is inappropriate and largely impacted by the pre-test variability as opposed to the variability in the intervention itself. Furthermore, the percentage change is often analyzed, but this is highly impacted by the baseline values and can be potentially misleading. Thus, the objective of this study was to illustrate the common misuse of the effect size and percent change measures. Here we provide a realistic sample data set comparing two resistance training groups with the same pre-test to post-test change. Statistical tests that are commonly performed within the literature were computed. Analyzing the within group effect size favors the control group, while the percent change favors the experimental group. The most appropriate way to present the data would be to plot the individual responses or, for larger samples, provide the mean change and 95% confidence intervals of the mean change. This details the magnitude and variability within the response to the intervention itself in units that are easily interpretable. This manuscript demonstrates the common misuse of the effect size and details the importance for investigators to always report raw values, even when alternative statistics are performed. Copyright © 2016 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

Internal pilots for a class of linear mixed models with Gaussian and compound symmetric data

PubMed Central

Gurka, Matthew J.; Coffey, Christopher S.; Muller, Keith E.

2015-01-01

SUMMARY An internal pilot design uses interim sample size analysis, without interim data analysis, to adjust the final number of observations. The approach helps to choose a sample size sufficiently large (to achieve the statistical power desired), but not too large (which would waste money and time). We report on recent research in cerebral vascular tortuosity (curvature in three dimensions) which would benefit greatly from internal pilots due to uncertainty in the parameters of the covariance matrix used for study planning. Unfortunately, observations correlated across the four regions of the brain and small sample sizes preclude using existing methods. However, as in a wide range of medical imaging studies, tortuosity data have no missing or mistimed data, a factorial within-subject design, the same between-subject design for all responses, and a Gaussian distribution with compound symmetry. For such restricted models, we extend exact, small sample univariate methods for internal pilots to linear mixed models with any between-subject design (not just two groups). Planning a new tortuosity study illustrates how the new methods help to avoid sample sizes that are too small or too large while still controlling the type I error rate. PMID:17318914

Scale-dependent effect sizes of ecological drivers on biodiversity: why standardised sampling is not enough.

PubMed

Chase, Jonathan M; Knight, Tiffany M

2013-05-01

There is little consensus about how natural (e.g. productivity, disturbance) and anthropogenic (e.g. invasive species, habitat destruction) ecological drivers influence biodiversity. Here, we show that when sampling is standardised by area (species density) or individuals (rarefied species richness), the measured effect sizes depend critically on the spatial grain and extent of sampling, as well as the size of the species pool. This compromises comparisons of effects sizes within studies using standard statistics, as well as among studies using meta-analysis. To derive an unambiguous effect size, we advocate that comparisons need to be made on a scale-independent metric, such as Hurlbert's Probability of Interspecific Encounter. Analyses of this metric can be used to disentangle the relative influence of changes in the absolute and relative abundances of individuals, as well as their intraspecific aggregations, in driving differences in biodiversity among communities. This and related approaches are necessary to achieve generality in understanding how biodiversity responds to ecological drivers and will necessitate a change in the way many ecologists collect and analyse their data. © 2013 John Wiley & Sons Ltd/CNRS.

Failure to Replicate a Genetic Association May Provide Important Clues About Genetic Architecture

PubMed Central

Greene, Casey S.; Penrod, Nadia M.; Williams, Scott M.; Moore, Jason H.

2009-01-01

Replication has become the gold standard for assessing statistical results from genome-wide association studies. Unfortunately this replication requirement may cause real genetic effects to be missed. A real result can fail to replicate for numerous reasons including inadequate sample size or variability in phenotype definitions across independent samples. In genome-wide association studies the allele frequencies of polymorphisms may differ due to sampling error or population differences. We hypothesize that some statistically significant independent genetic effects may fail to replicate in an independent dataset when allele frequencies differ and the functional polymorphism interacts with one or more other functional polymorphisms. To test this hypothesis, we designed a simulation study in which case-control status was determined by two interacting polymorphisms with heritabilities ranging from 0.025 to 0.4 with replication sample sizes ranging from 400 to 1600 individuals. We show that the power to replicate the statistically significant independent main effect of one polymorphism can drop dramatically with a change of allele frequency of less than 0.1 at a second interacting polymorphism. We also show that differences in allele frequency can result in a reversal of allelic effects where a protective allele becomes a risk factor in replication studies. These results suggest that failure to replicate an independent genetic effect may provide important clues about the complexity of the underlying genetic architecture. We recommend that polymorphisms that fail to replicate be checked for interactions with other polymorphisms, particularly when samples are collected from groups with distinct ethnic backgrounds or different geographic regions. PMID:19503614

Accounting for missing data in the estimation of contemporary genetic effective population size (N(e) ).

PubMed

Peel, D; Waples, R S; Macbeth, G M; Do, C; Ovenden, J R

2013-03-01

Theoretical models are often applied to population genetic data sets without fully considering the effect of missing data. Researchers can deal with missing data by removing individuals that have failed to yield genotypes and/or by removing loci that have failed to yield allelic determinations, but despite their best efforts, most data sets still contain some missing data. As a consequence, realized sample size differs among loci, and this poses a problem for unbiased methods that must explicitly account for random sampling error. One commonly used solution for the calculation of contemporary effective population size (N(e) ) is to calculate the effective sample size as an unweighted mean or harmonic mean across loci. This is not ideal because it fails to account for the fact that loci with different numbers of alleles have different information content. Here we consider this problem for genetic estimators of contemporary effective population size (N(e) ). To evaluate bias and precision of several statistical approaches for dealing with missing data, we simulated populations with known N(e) and various degrees of missing data. Across all scenarios, one method of correcting for missing data (fixed-inverse variance-weighted harmonic mean) consistently performed the best for both single-sample and two-sample (temporal) methods of estimating N(e) and outperformed some methods currently in widespread use. The approach adopted here may be a starting point to adjust other population genetics methods that include per-locus sample size components. © 2012 Blackwell Publishing Ltd.

STATISTICAL ANALYSIS OF TANK 18F FLOOR SAMPLE RESULTS

DOE Office of Scientific and Technical Information (OSTI.GOV)

Harris, S.

2010-09-02

Representative sampling has been completed for characterization of the residual material on the floor of Tank 18F as per the statistical sampling plan developed by Shine [1]. Samples from eight locations have been obtained from the tank floor and two of the samples were archived as a contingency. Six samples, referred to in this report as the current scrape samples, have been submitted to and analyzed by SRNL [2]. This report contains the statistical analysis of the floor sample analytical results to determine if further data are needed to reduce uncertainty. Included are comparisons with the prior Mantis samples resultsmore » [3] to determine if they can be pooled with the current scrape samples to estimate the upper 95% confidence limits (UCL{sub 95%}) for concentration. Statistical analysis revealed that the Mantis and current scrape sample results are not compatible. Therefore, the Mantis sample results were not used to support the quantification of analytes in the residual material. Significant spatial variability among the current sample results was not found. Constituent concentrations were similar between the North and South hemispheres as well as between the inner and outer regions of the tank floor. The current scrape sample results from all six samples fall within their 3-sigma limits. In view of the results from numerous statistical tests, the data were pooled from all six current scrape samples. As such, an adequate sample size was provided for quantification of the residual material on the floor of Tank 18F. The uncertainty is quantified in this report by an upper 95% confidence limit (UCL{sub 95%}) on each analyte concentration. The uncertainty in analyte concentration was calculated as a function of the number of samples, the average, and the standard deviation of the analytical results. The UCL{sub 95%} was based entirely on the six current scrape sample results (each averaged across three analytical determinations).« less

STATISTICAL ANALYSIS OF TANK 19F FLOOR SAMPLE RESULTS

DOE Office of Scientific and Technical Information (OSTI.GOV)

Harris, S.

2010-09-02

Representative sampling has been completed for characterization of the residual material on the floor of Tank 19F as per the statistical sampling plan developed by Harris and Shine. Samples from eight locations have been obtained from the tank floor and two of the samples were archived as a contingency. Six samples, referred to in this report as the current scrape samples, have been submitted to and analyzed by SRNL. This report contains the statistical analysis of the floor sample analytical results to determine if further data are needed to reduce uncertainty. Included are comparisons with the prior Mantis samples resultsmore » to determine if they can be pooled with the current scrape samples to estimate the upper 95% confidence limits (UCL95%) for concentration. Statistical analysis revealed that the Mantis and current scrape sample results are not compatible. Therefore, the Mantis sample results were not used to support the quantification of analytes in the residual material. Significant spatial variability among the current scrape sample results was not found. Constituent concentrations were similar between the North and South hemispheres as well as between the inner and outer regions of the tank floor. The current scrape sample results from all six samples fall within their 3-sigma limits. In view of the results from numerous statistical tests, the data were pooled from all six current scrape samples. As such, an adequate sample size was provided for quantification of the residual material on the floor of Tank 19F. The uncertainty is quantified in this report by an UCL95% on each analyte concentration. The uncertainty in analyte concentration was calculated as a function of the number of samples, the average, and the standard deviation of the analytical results. The UCL95% was based entirely on the six current scrape sample results (each averaged across three analytical determinations).« less

Information Entropy Production of Maximum Entropy Markov Chains from Spike Trains

NASA Astrophysics Data System (ADS)

Cofré, Rodrigo; Maldonado, Cesar

2018-01-01

We consider the maximum entropy Markov chain inference approach to characterize the collective statistics of neuronal spike trains, focusing on the statistical properties of the inferred model. We review large deviations techniques useful in this context to describe properties of accuracy and convergence in terms of sampling size. We use these results to study the statistical fluctuation of correlations, distinguishability and irreversibility of maximum entropy Markov chains. We illustrate these applications using simple examples where the large deviation rate function is explicitly obtained for maximum entropy models of relevance in this field.

US Food and Drug Administration survey of methyl mercury in canned tuna

DOE Office of Scientific and Technical Information (OSTI.GOV)

Yess, J.

1993-01-01

Methyl mercury was determined by the US Food and Drug Administration (FDA) in 220 samples of canned tuna collected in 1991. Samples were chosen to represent different styles, colors, and packs as available. Emphasis was placed on water-packed tuna, small can size, and the highest-volume brand names. The average methyl mercury (expressed as Hg) found for the 220 samples was 0.17 ppm; the range was <0.10-0.75 ppm. Statistically, a significantly higher level of methyl mercury was found in solid white and chunk tuna. Methyl mercury level was not related to can size. None of the 220 samples had methyl mercurymore » levels that exceeded the 1 ppm FDA action level. 11 refs., 1 tab.« less

Areal Control Using Generalized Least Squares As An Alternative to Stratification

Treesearch

Raymond L. Czaplewski

2001-01-01

Stratification for both variance reduction and areal control proliferates the number of strata, which causes small sample sizes in many strata. This might compromise statistical efficiency. Generalized least squares can, in principle, replace stratification for areal control.

A roughness-corrected index of relative bed stability for regional stream surveys

EPA Science Inventory

Quantitative regional assessments of streambed sedimentation and its likely causes are hampered because field investigations typically lack the requisite sample size, measurements, or precision for sound geomorphic and statistical interpretation. We adapted an index of relative b...

A Statistical Skull Geometry Model for Children 0-3 Years Old

PubMed Central

Li, Zhigang; Park, Byoung-Keon; Liu, Weiguo; Zhang, Jinhuan; Reed, Matthew P.; Rupp, Jonathan D.; Hoff, Carrie N.; Hu, Jingwen

2015-01-01

Head injury is the leading cause of fatality and long-term disability for children. Pediatric heads change rapidly in both size and shape during growth, especially for children under 3 years old (YO). To accurately assess the head injury risks for children, it is necessary to understand the geometry of the pediatric head and how morphologic features influence injury causation within the 0–3 YO population. In this study, head CT scans from fifty-six 0–3 YO children were used to develop a statistical model of pediatric skull geometry. Geometric features important for injury prediction, including skull size and shape, skull thickness and suture width, along with their variations among the sample population, were quantified through a series of image and statistical analyses. The size and shape of the pediatric skull change significantly with age and head circumference. The skull thickness and suture width vary with age, head circumference and location, which will have important effects on skull stiffness and injury prediction. The statistical geometry model developed in this study can provide a geometrical basis for future development of child anthropomorphic test devices and pediatric head finite element models. PMID:25992998

A statistical skull geometry model for children 0-3 years old.

PubMed

Li, Zhigang; Park, Byoung-Keon; Liu, Weiguo; Zhang, Jinhuan; Reed, Matthew P; Rupp, Jonathan D; Hoff, Carrie N; Hu, Jingwen

2015-01-01

Head injury is the leading cause of fatality and long-term disability for children. Pediatric heads change rapidly in both size and shape during growth, especially for children under 3 years old (YO). To accurately assess the head injury risks for children, it is necessary to understand the geometry of the pediatric head and how morphologic features influence injury causation within the 0-3 YO population. In this study, head CT scans from fifty-six 0-3 YO children were used to develop a statistical model of pediatric skull geometry. Geometric features important for injury prediction, including skull size and shape, skull thickness and suture width, along with their variations among the sample population, were quantified through a series of image and statistical analyses. The size and shape of the pediatric skull change significantly with age and head circumference. The skull thickness and suture width vary with age, head circumference and location, which will have important effects on skull stiffness and injury prediction. The statistical geometry model developed in this study can provide a geometrical basis for future development of child anthropomorphic test devices and pediatric head finite element models.

Hierarchical distance-sampling models to estimate population size and habitat-specific abundance of an island endemic

USGS Publications Warehouse

Sillett, Scott T.; Chandler, Richard B.; Royle, J. Andrew; Kéry, Marc; Morrison, Scott A.

2012-01-01

Population size and habitat-specific abundance estimates are essential for conservation management. A major impediment to obtaining such estimates is that few statistical models are able to simultaneously account for both spatial variation in abundance and heterogeneity in detection probability, and still be amenable to large-scale applications. The hierarchical distance-sampling model of J. A. Royle, D. K. Dawson, and S. Bates provides a practical solution. Here, we extend this model to estimate habitat-specific abundance and rangewide population size of a bird species of management concern, the Island Scrub-Jay (Aphelocoma insularis), which occurs solely on Santa Cruz Island, California, USA. We surveyed 307 randomly selected, 300 m diameter, point locations throughout the 250-km2 island during October 2008 and April 2009. Population size was estimated to be 2267 (95% CI 1613-3007) and 1705 (1212-2369) during the fall and spring respectively, considerably lower than a previously published but statistically problematic estimate of 12 500. This large discrepancy emphasizes the importance of proper survey design and analysis for obtaining reliable information for management decisions. Jays were most abundant in low-elevation chaparral habitat; the detection function depended primarily on the percent cover of chaparral and forest within count circles. Vegetation change on the island has been dramatic in recent decades, due to release from herbivory following the eradication of feral sheep (Ovis aries) from the majority of the island in the mid-1980s. We applied best-fit fall and spring models of habitat-specific jay abundance to a vegetation map from 1985, and estimated the population size of A. insularis was 1400-1500 at that time. The 20-30% increase in the jay population suggests that the species has benefited from the recovery of native vegetation since sheep removal. Nevertheless, this jay's tiny range and small population size make it vulnerable to natural disasters and to habitat alteration related to climate change. Our results demonstrate that hierarchical distance-sampling models hold promise for estimating population size and spatial density variation at large scales. Our statistical methods have been incorporated into the R package unmarked to facilitate their use by animal ecologists, and we provide annotated code in the Supplement.

Sampling errors in the measurement of rain and hail parameters

NASA Technical Reports Server (NTRS)

Gertzman, H. S.; Atlas, D.

1977-01-01

Attention is given to a general derivation of the fractional standard deviation (FSD) of any integrated property X such that X(D) = cD to the n. This work extends that of Joss and Waldvogel (1969). The equation is applicable to measuring integrated properties of cloud, rain or hail populations (such as water content, precipitation rate, kinetic energy, or radar reflectivity) which are subject to statistical sampling errors due to the Poisson distributed fluctuations of particles sampled in each particle size interval and the weighted sum of the associated variances in proportion to their contribution to the integral parameter to be measured. Universal curves are presented which are applicable to the exponential size distribution permitting FSD estimation of any parameters from n = 0 to n = 6. The equations and curves also permit corrections for finite upper limits in the size spectrum and a realistic fall speed law.

Discovery sequence and the nature of low permeability gas accumulations

USGS Publications Warehouse

Attanasi, E.D.

2005-01-01

There is an ongoing discussion regarding the geologic nature of accumulations that host gas in low-permeability sandstone environments. This note examines the discovery sequence of the accumulations in low permeability sandstone plays that were classified as continuous-type by the U.S. Geological Survey for the 1995 National Oil and Gas Assessment. It compares the statistical character of historical discovery sequences of accumulations associated with continuous-type sandstone gas plays to those of conventional plays. The seven sandstone plays with sufficient data exhibit declining size with sequence order, on average, and in three of the seven the trend is statistically significant. Simulation experiments show that both a skewed endowment size distribution and a discovery process that mimics sampling proportional to size are necessary to generate a discovery sequence that consistently produces a statistically significant negative size order relationship. The empirical findings suggest that discovery sequence could be used to constrain assessed gas in untested areas. The plays examined represent 134 of the 265 trillion cubic feet of recoverable gas assessed in undeveloped areas of continuous-type gas plays in low permeability sandstone environments reported in the 1995 National Assessment. ?? 2005 International Association for Mathematical Geology.

Particle size distribution of main-channel-bed sediments along the upper Mississippi River, USA

USGS Publications Warehouse

Remo, Jonathan; Heine, Ruben A.; Ickes, Brian

2016-01-01

In this study, we compared pre-lock-and-dam (ca. 1925) with a modern longitudinal survey of main-channel-bed sediments along a 740-km segment of the upper Mississippi River (UMR) between Davenport, IA, and Cairo, IL. This comparison was undertaken to gain a better understanding of how bed sediments are distributed longitudinally and to assess change since the completion of the UMR lock and dam navigation system and Missouri River dams (i.e., mid-twentieth century). The comparison of the historic and modern longitudinal bed sediment surveys showed similar bed sediment sizes and distributions along the study segment with the majority (> 90%) of bed sediment samples having a median diameter (D50) of fine to coarse sand. The fine tail (≤ D10) of the sediment size distributions was very fine to medium sand, and the coarse tail (≥ D90) of sediment-size distribution was coarse sand to gravel. Coarsest sediments in both surveys were found within or immediately downstream of bedrock-floored reaches. Statistical analysis revealed that the particle-size distributions between the survey samples were statistically identical, suggesting no overall difference in main-channel-bed sediment-size distribution between 1925 and present. This was a surprising result given the magnitude of river engineering undertaken along the study segment over the past ~ 90 years. The absence of substantial differences in main-channel-bed-sediment size suggests that flow competencies within the highly engineered navigation channel today are similar to conditions within the less-engineered historic channel.

Statistical considerations in evaluating pharmacogenomics-based clinical effect for confirmatory trials.

PubMed

Wang, Sue-Jane; O'Neill, Robert T; Hung, Hm James

2010-10-01

The current practice for seeking genomically favorable patients in randomized controlled clinical trials using genomic convenience samples. To discuss the extent of imbalance, confounding, bias, design efficiency loss, type I error, and type II error that can occur in the evaluation of the convenience samples, particularly when they are small samples. To articulate statistical considerations for a reasonable sample size to minimize the chance of imbalance, and, to highlight the importance of replicating the subgroup finding in independent studies. Four case examples reflecting recent regulatory experiences are used to underscore the problems with convenience samples. Probability of imbalance for a pre-specified subgroup is provided to elucidate sample size needed to minimize the chance of imbalance. We use an example drug development to highlight the level of scientific rigor needed, with evidence replicated for a pre-specified subgroup claim. The convenience samples evaluated ranged from 18% to 38% of the intent-to-treat samples with sample size ranging from 100 to 5000 patients per arm. The baseline imbalance can occur with probability higher than 25%. Mild to moderate multiple confounders yielding the same directional bias in favor of the treated group can make treatment group incomparable at baseline and result in a false positive conclusion that there is a treatment difference. Conversely, if the same directional bias favors the placebo group or there is loss in design efficiency, the type II error can increase substantially. Pre-specification of a genomic subgroup hypothesis is useful only for some degree of type I error control. Complete ascertainment of genomic samples in a randomized controlled trial should be the first step to explore if a favorable genomic patient subgroup suggests a treatment effect when there is no clear prior knowledge and understanding about how the mechanism of a drug target affects the clinical outcome of interest. When stratified randomization based on genomic biomarker status cannot be implemented in designing a pharmacogenomics confirmatory clinical trial, if there is one genomic biomarker prognostic for clinical response, as a general rule of thumb, a sample size of at least 100 patients may be needed to be considered for the lower prevalence genomic subgroup to minimize the chance of an imbalance of 20% or more difference in the prevalence of the genomic marker. The sample size may need to be at least 150, 350, and 1350, respectively, if an imbalance of 15%, 10% and 5% difference is of concern.

A weighted generalized score statistic for comparison of predictive values of diagnostic tests.

PubMed

Kosinski, Andrzej S

2013-03-15

Positive and negative predictive values are important measures of a medical diagnostic test performance. We consider testing equality of two positive or two negative predictive values within a paired design in which all patients receive two diagnostic tests. The existing statistical tests for testing equality of predictive values are either Wald tests based on the multinomial distribution or the empirical Wald and generalized score tests within the generalized estimating equations (GEE) framework. As presented in the literature, these test statistics have considerably complex formulas without clear intuitive insight. We propose their re-formulations that are mathematically equivalent but algebraically simple and intuitive. As is clearly seen with a new re-formulation we presented, the generalized score statistic does not always reduce to the commonly used score statistic in the independent samples case. To alleviate this, we introduce a weighted generalized score (WGS) test statistic that incorporates empirical covariance matrix with newly proposed weights. This statistic is simple to compute, always reduces to the score statistic in the independent samples situation, and preserves type I error better than the other statistics as demonstrated by simulations. Thus, we believe that the proposed WGS statistic is the preferred statistic for testing equality of two predictive values and for corresponding sample size computations. The new formulas of the Wald statistics may be useful for easy computation of confidence intervals for difference of predictive values. The introduced concepts have potential to lead to development of the WGS test statistic in a general GEE setting. Copyright © 2012 John Wiley & Sons, Ltd.

A weighted generalized score statistic for comparison of predictive values of diagnostic tests

PubMed Central

Kosinski, Andrzej S.

2013-01-01

Positive and negative predictive values are important measures of a medical diagnostic test performance. We consider testing equality of two positive or two negative predictive values within a paired design in which all patients receive two diagnostic tests. The existing statistical tests for testing equality of predictive values are either Wald tests based on the multinomial distribution or the empirical Wald and generalized score tests within the generalized estimating equations (GEE) framework. As presented in the literature, these test statistics have considerably complex formulas without clear intuitive insight. We propose their re-formulations which are mathematically equivalent but algebraically simple and intuitive. As is clearly seen with a new re-formulation we present, the generalized score statistic does not always reduce to the commonly used score statistic in the independent samples case. To alleviate this, we introduce a weighted generalized score (WGS) test statistic which incorporates empirical covariance matrix with newly proposed weights. This statistic is simple to compute, it always reduces to the score statistic in the independent samples situation, and it preserves type I error better than the other statistics as demonstrated by simulations. Thus, we believe the proposed WGS statistic is the preferred statistic for testing equality of two predictive values and for corresponding sample size computations. The new formulas of the Wald statistics may be useful for easy computation of confidence intervals for difference of predictive values. The introduced concepts have potential to lead to development of the weighted generalized score test statistic in a general GEE setting. PMID:22912343

Landsat image and sample design for water reservoirs (Rapel dam Central Chile).

PubMed

Lavanderos, L; Pozo, M E; Pattillo, C; Miranda, H

1990-01-01

Spatial heterogeneity of the Rapel reservoir surface waters is analyzed through Landsat images. The image digital counts are used with the aim or developing an aprioristic quantitative sample design.Natural horizontal stratification of the Rapel Reservoir (Central Chile) is produced mainly by suspended solids. The spatial heterogeneity conditions of the reservoir for the Spring 86-Summer 87 period were determined by qualitative analysis and image processing of the MSS Landsat, bands 1 and 3. The space-time variations of the different observed strata obtained with multitemporal image analysis.A random stratified sample design (r.s.s.d) was developed, based on the digital counts statistical analysis. Strata population size as well as the average, variance and sampling size of the digital counts were obtained by the r.s.s.d method.Stratification determined by analysis of satellite images were later correlated with ground data. Though the stratification of the reservoir is constant over time, the shape and size of the strata varys.

Will Outer Tropical Cyclone Size Change due to Anthropogenic Warming?

NASA Astrophysics Data System (ADS)

Schenkel, B. A.; Lin, N.; Chavas, D. R.; Vecchi, G. A.; Knutson, T. R.; Oppenheimer, M.

2017-12-01

Prior research has shown significant interbasin and intrabasin variability in outer tropical cyclone (TC) size. Moreover, outer TC size has even been shown to vary substantially over the lifetime of the majority of TCs. However, the factors responsible for both setting initial outer TC size and determining its evolution throughout the TC lifetime remain uncertain. Given these gaps in our physical understanding, there remains uncertainty in how outer TC size will change, if at all, due to anthropogenic warming. The present study seeks to quantify whether outer TC size will change significantly in response to anthropogenic warming using data from a high-resolution global climate model and a regional hurricane model. Similar to prior work, the outer TC size metric used in this study is the radius in which the azimuthal-mean surface azimuthal wind equals 8 m/s. The initial results from the high-resolution global climate model data suggest that the distribution of outer TC size shifts significantly towards larger values in each global TC basin during future climates, as revealed by 1) statistically significant increase of the median outer TC size by 5-10% (p<0.05) according to a 1,000-sample bootstrap resampling approach with replacement and 2) statistically significant differences between distributions of outer TC size from current and future climate simulations as shown using two-sample Kolmogorov Smirnov testing (p<<0.01). Additional analysis of the high-resolution global climate model data reveals that outer TC size does not uniformly increase within each basin in future climates, but rather shows substantial locational dependence. Future work will incorporate the regional mesoscale hurricane model data to help focus on identifying the source of the spatial variability in outer TC size increases within each basin during future climates and, more importantly, why outer TC size changes in response to anthropogenic warming.

Validation of fixed sample size plans for monitoring lepidopteran pests of Brassica oleracea crops in North Korea.

PubMed

Hamilton, A J; Waters, E K; Kim, H J; Pak, W S; Furlong, M J

2009-06-01

The combined action of two lepidoteran pests, Plutella xylostella L. (Plutellidae) and Pieris rapae L. (Pieridae),causes significant yield losses in cabbage (Brassica oleracea variety capitata) crops in the Democratic People's Republic of Korea. Integrated pest management (IPM) strategies for these cropping systems are in their infancy, and sampling plans have not yet been developed. We used statistical resampling to assess the performance of fixed sample size plans (ranging from 10 to 50 plants). First, the precision (D = SE/mean) of the plans in estimating the population mean was assessed. There was substantial variation in achieved D for all sample sizes, and sample sizes of at least 20 and 45 plants were required to achieve the acceptable precision level of D < or = 0.3 at least 50 and 75% of the time, respectively. Second, the performance of the plans in classifying the population density relative to an economic threshold (ET) was assessed. To account for the different damage potentials of the two species the ETs were defined in terms of standard insects (SIs), where 1 SI = 1 P. rapae = 5 P. xylostella larvae. The plans were implemented using different economic thresholds (ETs) for the three growth stages of the crop: precupping (1 SI/plant), cupping (0.5 SI/plant), and heading (4 SI/plant). Improvement in the classification certainty with increasing sample sizes could be seen through the increasing steepness of operating characteristic curves. Rather than prescribe a particular plan, we suggest that the results of these analyses be used to inform practitioners of the relative merits of the different sample sizes.

Methodological reporting of randomized trials in five leading Chinese nursing journals.

PubMed

Shi, Chunhu; Tian, Jinhui; Ren, Dan; Wei, Hongli; Zhang, Lihuan; Wang, Quan; Yang, Kehu

2014-01-01

Randomized controlled trials (RCTs) are not always well reported, especially in terms of their methodological descriptions. This study aimed to investigate the adherence of methodological reporting complying with CONSORT and explore associated trial level variables in the Chinese nursing care field. In June 2012, we identified RCTs published in five leading Chinese nursing journals and included trials with details of randomized methods. The quality of methodological reporting was measured through the methods section of the CONSORT checklist and the overall CONSORT methodological items score was calculated and expressed as a percentage. Meanwhile, we hypothesized that some general and methodological characteristics were associated with reporting quality and conducted a regression with these data to explore the correlation. The descriptive and regression statistics were calculated via SPSS 13.0. In total, 680 RCTs were included. The overall CONSORT methodological items score was 6.34 ± 0.97 (Mean ± SD). No RCT reported descriptions and changes in "trial design," changes in "outcomes" and "implementation," or descriptions of the similarity of interventions for "blinding." Poor reporting was found in detailing the "settings of participants" (13.1%), "type of randomization sequence generation" (1.8%), calculation methods of "sample size" (0.4%), explanation of any interim analyses and stopping guidelines for "sample size" (0.3%), "allocation concealment mechanism" (0.3%), additional analyses in "statistical methods" (2.1%), and targeted subjects and methods of "blinding" (5.9%). More than 50% of trials described randomization sequence generation, the eligibility criteria of "participants," "interventions," and definitions of the "outcomes" and "statistical methods." The regression analysis found that publication year and ITT analysis were weakly associated with CONSORT score. The completeness of methodological reporting of RCTs in the Chinese nursing care field is poor, especially with regard to the reporting of trial design, changes in outcomes, sample size calculation, allocation concealment, blinding, and statistical methods.

Ranking metrics in gene set enrichment analysis: do they matter?

PubMed

Zyla, Joanna; Marczyk, Michal; Weiner, January; Polanska, Joanna

2017-05-12

There exist many methods for describing the complex relation between changes of gene expression in molecular pathways or gene ontologies under different experimental conditions. Among them, Gene Set Enrichment Analysis seems to be one of the most commonly used (over 10,000 citations). An important parameter, which could affect the final result, is the choice of a metric for the ranking of genes. Applying a default ranking metric may lead to poor results. In this work 28 benchmark data sets were used to evaluate the sensitivity and false positive rate of gene set analysis for 16 different ranking metrics including new proposals. Furthermore, the robustness of the chosen methods to sample size was tested. Using k-means clustering algorithm a group of four metrics with the highest performance in terms of overall sensitivity, overall false positive rate and computational load was established i.e. absolute value of Moderated Welch Test statistic, Minimum Significant Difference, absolute value of Signal-To-Noise ratio and Baumgartner-Weiss-Schindler test statistic. In case of false positive rate estimation, all selected ranking metrics were robust with respect to sample size. In case of sensitivity, the absolute value of Moderated Welch Test statistic and absolute value of Signal-To-Noise ratio gave stable results, while Baumgartner-Weiss-Schindler and Minimum Significant Difference showed better results for larger sample size. Finally, the Gene Set Enrichment Analysis method with all tested ranking metrics was parallelised and implemented in MATLAB, and is available at https://github.com/ZAEDPolSl/MrGSEA . Choosing a ranking metric in Gene Set Enrichment Analysis has critical impact on results of pathway enrichment analysis. The absolute value of Moderated Welch Test has the best overall sensitivity and Minimum Significant Difference has the best overall specificity of gene set analysis. When the number of non-normally distributed genes is high, using Baumgartner-Weiss-Schindler test statistic gives better outcomes. Also, it finds more enriched pathways than other tested metrics, which may induce new biological discoveries.

Bed-sediment grain-size and morphologic data from Suisun, Grizzly, and Honker Bays, CA, 1998-2002

USGS Publications Warehouse

Hampton, Margaret A.; Snyder, Noah P.; Chin, John L.; Allison, Dan W.; Rubin, David M.

2003-01-01

The USGS Place Based Studies Program for San Francisco Bay investigates this sensitive estuarine system to aid in resource management. As part of the inter-disciplinary research program, the USGS collected side-scan sonar data and bed-sediment samples from north San Francisco Bay to characterize bed-sediment texture and investigate temporal trends in sedimentation. The study area is located in central California and consists of Suisun Bay, and Grizzly and Honker Bays, sub-embayments of Suisun Bay. During the study (1998-2002), the USGS collected three side-scan sonar data sets and approximately 300 sediment samples. The side-scan data revealed predominantly fine-grained material on the bayfloor. We also mapped five different bottom types from the data set, categorized as featureless, furrows, sand waves, machine-made, and miscellaneous. We performed detailed grain-size and statistical analyses on the sediment samples. Overall, we found that grain size ranged from clay to fine sand, with the coarsest material in the channels and finer material located in the shallow bays. Grain-size analyses revealed high spatial variability in size distributions in the channel areas. In contrast, the shallow regions exhibited low spatial variability and consistent sediment size over time.

The analysis of morphometric data on rocky mountain wolves and artic wolves using statistical method

NASA Astrophysics Data System (ADS)

Ammar Shafi, Muhammad; Saifullah Rusiman, Mohd; Hamzah, Nor Shamsidah Amir; Nor, Maria Elena; Ahmad, Noor’ani; Azia Hazida Mohamad Azmi, Nur; Latip, Muhammad Faez Ab; Hilmi Azman, Ahmad

2018-04-01

Morphometrics is a quantitative analysis depending on the shape and size of several specimens. Morphometric quantitative analyses are commonly used to analyse fossil record, shape and size of specimens and others. The aim of the study is to find the differences between rocky mountain wolves and arctic wolves based on gender. The sample utilised secondary data which included seven variables as independent variables and two dependent variables. Statistical modelling was used in the analysis such was the analysis of variance (ANOVA) and multivariate analysis of variance (MANOVA). The results showed there exist differentiating results between arctic wolves and rocky mountain wolves based on independent factors and gender.

Evaluating test-retest reliability in patient-reported outcome measures for older people: A systematic review.

PubMed

Park, Myung Sook; Kang, Kyung Ja; Jang, Sun Joo; Lee, Joo Yun; Chang, Sun Ju

2018-03-01

This study aimed to evaluate the components of test-retest reliability including time interval, sample size, and statistical methods used in patient-reported outcome measures in older people and to provide suggestions on the methodology for calculating test-retest reliability for patient-reported outcomes in older people. This was a systematic literature review. MEDLINE, Embase, CINAHL, and PsycINFO were searched from January 1, 2000 to August 10, 2017 by an information specialist. This systematic review was guided by both the Preferred Reporting Items for Systematic Reviews and Meta-Analyses checklist and the guideline for systematic review published by the National Evidence-based Healthcare Collaborating Agency in Korea. The methodological quality was assessed by the Consensus-based Standards for the selection of health Measurement Instruments checklist box B. Ninety-five out of 12,641 studies were selected for the analysis. The median time interval for test-retest reliability was 14days, and the ratio of sample size for test-retest reliability to the number of items in each measure ranged from 1:1 to 1:4. The most frequently used statistical methods for continuous scores was intraclass correlation coefficients (ICCs). Among the 63 studies that used ICCs, 21 studies presented models for ICC calculations and 30 studies reported 95% confidence intervals of the ICCs. Additional analyses using 17 studies that reported a strong ICC (>0.09) showed that the mean time interval was 12.88days and the mean ratio of the number of items to sample size was 1:5.37. When researchers plan to assess the test-retest reliability of patient-reported outcome measures for older people, they need to consider an adequate time interval of approximately 13days and the sample size of about 5 times the number of items. Particularly, statistical methods should not only be selected based on the types of scores of the patient-reported outcome measures, but should also be described clearly in the studies that report the results of test-retest reliability. Copyright © 2017 Elsevier Ltd. All rights reserved.

Coalescent: an open-science framework for importance sampling in coalescent theory.

PubMed

Tewari, Susanta; Spouge, John L

2015-01-01

Background. In coalescent theory, computer programs often use importance sampling to calculate likelihoods and other statistical quantities. An importance sampling scheme can exploit human intuition to improve statistical efficiency of computations, but unfortunately, in the absence of general computer frameworks on importance sampling, researchers often struggle to translate new sampling schemes computationally or benchmark against different schemes, in a manner that is reliable and maintainable. Moreover, most studies use computer programs lacking a convenient user interface or the flexibility to meet the current demands of open science. In particular, current computer frameworks can only evaluate the efficiency of a single importance sampling scheme or compare the efficiencies of different schemes in an ad hoc manner. Results. We have designed a general framework (http://coalescent.sourceforge.net; language: Java; License: GPLv3) for importance sampling that computes likelihoods under the standard neutral coalescent model of a single, well-mixed population of constant size over time following infinite sites model of mutation. The framework models the necessary core concepts, comes integrated with several data sets of varying size, implements the standard competing proposals, and integrates tightly with our previous framework for calculating exact probabilities. For a given dataset, it computes the likelihood and provides the maximum likelihood estimate of the mutation parameter. Well-known benchmarks in the coalescent literature validate the accuracy of the framework. The framework provides an intuitive user interface with minimal clutter. For performance, the framework switches automatically to modern multicore hardware, if available. It runs on three major platforms (Windows, Mac and Linux). Extensive tests and coverage make the framework reliable and maintainable. Conclusions. In coalescent theory, many studies of computational efficiency consider only effective sample size. Here, we evaluate proposals in the coalescent literature, to discover that the order of efficiency among the three importance sampling schemes changes when one considers running time as well as effective sample size. We also describe a computational technique called "just-in-time delegation" available to improve the trade-off between running time and precision by constructing improved importance sampling schemes from existing ones. Thus, our systems approach is a potential solution to the "2(8) programs problem" highlighted by Felsenstein, because it provides the flexibility to include or exclude various features of similar coalescent models or importance sampling schemes.

R2 effect-size measures for mediation analysis

PubMed Central

Fairchild, Amanda J.; MacKinnon, David P.; Taborga, Marcia P.; Taylor, Aaron B.

2010-01-01

R2 effect-size measures are presented to assess variance accounted for in mediation models. The measures offer a means to evaluate both component paths and the overall mediated effect in mediation models. Statistical simulation results indicate acceptable bias across varying parameter and sample-size combinations. The measures are applied to a real-world example using data from a team-based health promotion program to improve the nutrition and exercise habits of firefighters. SAS and SPSS computer code are also provided for researchers to compute the measures in their own data. PMID:19363189

R2 effect-size measures for mediation analysis.

PubMed

Fairchild, Amanda J; Mackinnon, David P; Taborga, Marcia P; Taylor, Aaron B

2009-05-01

R(2) effect-size measures are presented to assess variance accounted for in mediation models. The measures offer a means to evaluate both component paths and the overall mediated effect in mediation models. Statistical simulation results indicate acceptable bias across varying parameter and sample-size combinations. The measures are applied to a real-world example using data from a team-based health promotion program to improve the nutrition and exercise habits of firefighters. SAS and SPSS computer code are also provided for researchers to compute the measures in their own data.

A Naturalistic Study of Driving Behavior in Older Adults and Preclinical Alzheimer Disease.

PubMed

Babulal, Ganesh M; Stout, Sarah H; Benzinger, Tammie L S; Ott, Brian R; Carr, David B; Webb, Mollie; Traub, Cindy M; Addison, Aaron; Morris, John C; Warren, David K; Roe, Catherine M

2017-01-01

A clinical consequence of symptomatic Alzheimer's disease (AD) is impaired driving performance. However, decline in driving performance may begin in the preclinical stage of AD. We used a naturalistic driving methodology to examine differences in driving behavior over one year in a small sample of cognitively normal older adults with ( n = 10) and without ( n = 10) preclinical AD. As expected with a small sample size, there were no statistically significant differences between the two groups, but older adults with preclinical AD drove less often, were less likely to drive at night, and had fewer aggressive behaviors such as hard braking, speeding, and sudden acceleration. The sample size required to power a larger study to determine differences was calculated.

Development and Validation of the Caring Loneliness Scale.

PubMed

Karhe, Liisa; Kaunonen, Marja; Koivisto, Anna-Maija

2016-12-01

The Caring Loneliness Scale (CARLOS) includes 5 categories derived from earlier qualitative research. This article assesses the reliability and construct validity of a scale designed to measure patient experiences of loneliness in a professional caring relationship. Statistical analysis with 4 different sample sizes included Cronbach's alpha and exploratory factor analysis with principal axis factoring extraction. The sample size of 250 gave the most useful and comprehensible structure, but all 4 samples yielded underlying content of loneliness experiences. The initial 5 categories were reduced to 4 factors with 24 items and Cronbach's alpha ranging from .77 to .90. The findings support the reliability and validity of CARLOS for the assessment of Finnish breast cancer and heart surgery patients' experiences but as all instruments, further validation is needed.

Practical Advice on Calculating Confidence Intervals for Radioprotection Effects and Reducing Animal Numbers in Radiation Countermeasure Experiments

PubMed Central

Landes, Reid D.; Lensing, Shelly Y.; Kodell, Ralph L.; Hauer-Jensen, Martin

2014-01-01

The dose of a substance that causes death in P% of a population is called an LDP, where LD stands for lethal dose. In radiation research, a common LDP of interest is the radiation dose that kills 50% of the population by a specified time, i.e., lethal dose 50 or LD50. When comparing LD50 between two populations, relative potency is the parameter of interest. In radiation research, this is commonly known as the dose reduction factor (DRF). Unfortunately, statistical inference on dose reduction factor is seldom reported. We illustrate how to calculate confidence intervals for dose reduction factor, which may then be used for statistical inference. Further, most dose reduction factor experiments use hundreds, rather than tens of animals. Through better dosing strategies and the use of a recently available sample size formula, we also show how animal numbers may be reduced while maintaining high statistical power. The illustrations center on realistic examples comparing LD50 values between a radiation countermeasure group and a radiation-only control. We also provide easy-to-use spreadsheets for sample size calculations and confidence interval calculations, as well as SAS® and R code for the latter. PMID:24164553

Statistical analysis of hydrological response in urbanising catchments based on adaptive sampling using inter-amount times

NASA Astrophysics Data System (ADS)

ten Veldhuis, Marie-Claire; Schleiss, Marc

2017-04-01

Urban catchments are typically characterised by a more flashy nature of the hydrological response compared to natural catchments. Predicting flow changes associated with urbanisation is not straightforward, as they are influenced by interactions between impervious cover, basin size, drainage connectivity and stormwater management infrastructure. In this study, we present an alternative approach to statistical analysis of hydrological response variability and basin flashiness, based on the distribution of inter-amount times. We analyse inter-amount time distributions of high-resolution streamflow time series for 17 (semi-)urbanised basins in North Carolina, USA, ranging from 13 to 238 km2 in size. We show that in the inter-amount-time framework, sampling frequency is tuned to the local variability of the flow pattern, resulting in a different representation and weighting of high and low flow periods in the statistical distribution. This leads to important differences in the way the distribution quantiles, mean, coefficient of variation and skewness vary across scales and results in lower mean intermittency and improved scaling. Moreover, we show that inter-amount-time distributions can be used to detect regulation effects on flow patterns, identify critical sampling scales and characterise flashiness of hydrological response. The possibility to use both the classical approach and the inter-amount-time framework to identify minimum observable scales and analyse flow data opens up interesting areas for future research.

An empirical determination of the minimum number of measurements needed to estimate the mean random vitrinite reflectance of disseminated organic matter

USGS Publications Warehouse

Barker, C.E.; Pawlewicz, M.J.

1993-01-01

In coal samples, published recommendations based on statistical methods suggest 100 measurements are needed to estimate the mean random vitrinite reflectance (Rv-r) to within ??2%. Our survey of published thermal maturation studies indicates that those using dispersed organic matter (DOM) mostly have an objective of acquiring 50 reflectance measurements. This smaller objective size in DOM versus that for coal samples poses a statistical contradiction because the standard deviations of DOM reflectance distributions are typically larger indicating a greater sample size is needed to accurately estimate Rv-r in DOM. However, in studies of thermal maturation using DOM, even 50 measurements can be an unrealistic requirement given the small amount of vitrinite often found in such samples. Furthermore, there is generally a reduced need for assuring precision like that needed for coal applications. Therefore, a key question in thermal maturation studies using DOM is how many measurements of Rv-r are needed to adequately estimate the mean. Our empirical approach to this problem is to compute the reflectance distribution statistics: mean, standard deviation, skewness, and kurtosis in increments of 10 measurements. This study compares these intermediate computations of Rv-r statistics with a final one computed using all measurements for that sample. Vitrinite reflectance was measured on mudstone and sandstone samples taken from borehole M-25 in the Cerro Prieto, Mexico geothermal system which was selected because the rocks have a wide range of thermal maturation and a comparable humic DOM with depth. The results of this study suggest that after only 20-30 measurements the mean Rv-r is generally known to within 5% and always to within 12% of the mean Rv-r calculated using all of the measured particles. Thus, even in the worst case, the precision after measuring only 20-30 particles is in good agreement with the general precision of one decimal place recommended for mean Rv-r measurements on DOM. The coefficient of variation (V = standard deviation/mean) is proposed as a statistic to indicate the reliability of the mean Rv-r estimates made at n ??? 20. This preliminary study suggests a V 0.2 suggests an unreliable mean in such small samples. ?? 1993.

Small sample mediation testing: misplaced confidence in bootstrapped confidence intervals.

PubMed

Koopman, Joel; Howe, Michael; Hollenbeck, John R; Sin, Hock-Peng

2015-01-01

Bootstrapping is an analytical tool commonly used in psychology to test the statistical significance of the indirect effect in mediation models. Bootstrapping proponents have particularly advocated for its use for samples of 20-80 cases. This advocacy has been heeded, especially in the Journal of Applied Psychology, as researchers are increasingly utilizing bootstrapping to test mediation with samples in this range. We discuss reasons to be concerned with this escalation, and in a simulation study focused specifically on this range of sample sizes, we demonstrate not only that bootstrapping has insufficient statistical power to provide a rigorous hypothesis test in most conditions but also that bootstrapping has a tendency to exhibit an inflated Type I error rate. We then extend our simulations to investigate an alternative empirical resampling method as well as a Bayesian approach and demonstrate that they exhibit comparable statistical power to bootstrapping in small samples without the associated inflated Type I error. Implications for researchers testing mediation hypotheses in small samples are presented. For researchers wishing to use these methods in their own research, we have provided R syntax in the online supplemental materials. (c) 2015 APA, all rights reserved.

Calculating Confidence, Uncertainty, and Numbers of Samples When Using Statistical Sampling Approaches to Characterize and Clear Contaminated Areas

DOE Office of Scientific and Technical Information (OSTI.GOV)

Piepel, Gregory F.; Matzke, Brett D.; Sego, Landon H.

2013-04-27

This report discusses the methodology, formulas, and inputs needed to make characterization and clearance decisions for Bacillus anthracis-contaminated and uncontaminated (or decontaminated) areas using a statistical sampling approach. Specifically, the report includes the methods and formulas for calculating the • number of samples required to achieve a specified confidence in characterization and clearance decisions • confidence in making characterization and clearance decisions for a specified number of samples for two common statistically based environmental sampling approaches. In particular, the report addresses an issue raised by the Government Accountability Office by providing methods and formulas to calculate the confidence that amore » decision area is uncontaminated (or successfully decontaminated) if all samples collected according to a statistical sampling approach have negative results. Key to addressing this topic is the probability that an individual sample result is a false negative, which is commonly referred to as the false negative rate (FNR). The two statistical sampling approaches currently discussed in this report are 1) hotspot sampling to detect small isolated contaminated locations during the characterization phase, and 2) combined judgment and random (CJR) sampling during the clearance phase. Typically if contamination is widely distributed in a decision area, it will be detectable via judgment sampling during the characterization phrase. Hotspot sampling is appropriate for characterization situations where contamination is not widely distributed and may not be detected by judgment sampling. CJR sampling is appropriate during the clearance phase when it is desired to augment judgment samples with statistical (random) samples. The hotspot and CJR statistical sampling approaches are discussed in the report for four situations: 1. qualitative data (detect and non-detect) when the FNR = 0 or when using statistical sampling methods that account for FNR > 0 2. qualitative data when the FNR > 0 but statistical sampling methods are used that assume the FNR = 0 3. quantitative data (e.g., contaminant concentrations expressed as CFU/cm2) when the FNR = 0 or when using statistical sampling methods that account for FNR > 0 4. quantitative data when the FNR > 0 but statistical sampling methods are used that assume the FNR = 0. For Situation 2, the hotspot sampling approach provides for stating with Z% confidence that a hotspot of specified shape and size with detectable contamination will be found. Also for Situation 2, the CJR approach provides for stating with X% confidence that at least Y% of the decision area does not contain detectable contamination. Forms of these statements for the other three situations are discussed in Section 2.2. Statistical methods that account for FNR > 0 currently only exist for the hotspot sampling approach with qualitative data (or quantitative data converted to qualitative data). This report documents the current status of methods and formulas for the hotspot and CJR sampling approaches. Limitations of these methods are identified. Extensions of the methods that are applicable when FNR = 0 to account for FNR > 0, or to address other limitations, will be documented in future revisions of this report if future funding supports the development of such extensions. For quantitative data, this report also presents statistical methods and formulas for 1. quantifying the uncertainty in measured sample results 2. estimating the true surface concentration corresponding to a surface sample 3. quantifying the uncertainty of the estimate of the true surface concentration. All of the methods and formulas discussed in the report were applied to example situations to illustrate application of the methods and interpretation of the results.« less

Estimating the mean and standard deviation of environmental data with below detection limit observations: Considering highly skewed data and model misspecification.

PubMed

Shoari, Niloofar; Dubé, Jean-Sébastien; Chenouri, Shoja'eddin

2015-11-01

In environmental studies, concentration measurements frequently fall below detection limits of measuring instruments, resulting in left-censored data. Some studies employ parametric methods such as the maximum likelihood estimator (MLE), robust regression on order statistic (rROS), and gamma regression on order statistic (GROS), while others suggest a non-parametric approach, the Kaplan-Meier method (KM). Using examples of real data from a soil characterization study in Montreal, we highlight the need for additional investigations that aim at unifying the existing literature. A number of studies have examined this issue; however, those considering data skewness and model misspecification are rare. These aspects are investigated in this paper through simulations. Among other findings, results show that for low skewed data, the performance of different statistical methods is comparable, regardless of the censoring percentage and sample size. For highly skewed data, the performance of the MLE method under lognormal and Weibull distributions is questionable; particularly, when the sample size is small or censoring percentage is high. In such conditions, MLE under gamma distribution, rROS, GROS, and KM are less sensitive to skewness. Related to model misspecification, MLE based on lognormal and Weibull distributions provides poor estimates when the true distribution of data is misspecified. However, the methods of rROS, GROS, and MLE under gamma distribution are generally robust to model misspecifications regardless of skewness, sample size, and censoring percentage. Since the characteristics of environmental data (e.g., type of distribution and skewness) are unknown a priori, we suggest using MLE based on gamma distribution, rROS and GROS. Copyright © 2015 Elsevier Ltd. All rights reserved.

Quality of statistical reporting in developmental disability journals.

PubMed

Namasivayam, Aravind K; Yan, Tina; Wong, Wing Yiu Stephanie; van Lieshout, Pascal

2015-12-01

Null hypothesis significance testing (NHST) dominates quantitative data analysis, but its use is controversial and has been heavily criticized. The American Psychological Association has advocated the reporting of effect sizes (ES), confidence intervals (CIs), and statistical power analysis to complement NHST results to provide a more comprehensive understanding of research findings. The aim of this paper is to carry out a sample survey of statistical reporting practices in two journals with the highest h5-index scores in the areas of developmental disability and rehabilitation. Using a checklist that includes critical recommendations by American Psychological Association, we examined 100 randomly selected articles out of 456 articles reporting inferential statistics in the year 2013 in the Journal of Autism and Developmental Disorders (JADD) and Research in Developmental Disabilities (RDD). The results showed that for both journals, ES were reported only half the time (JADD 59.3%; RDD 55.87%). These findings are similar to psychology journals, but are in stark contrast to ES reporting in educational journals (73%). Furthermore, a priori power and sample size determination (JADD 10%; RDD 6%), along with reporting and interpreting precision measures (CI: JADD 13.33%; RDD 16.67%), were the least reported metrics in these journals, but not dissimilar to journals in other disciplines. To advance the science in developmental disability and rehabilitation and to bridge the research-to-practice divide, reforms in statistical reporting, such as providing supplemental measures to NHST, are clearly needed.

Sampling Errors in Monthly Rainfall Totals for TRMM and SSM/I, Based on Statistics of Retrieved Rain Rates and Simple Models

NASA Technical Reports Server (NTRS)

Bell, Thomas L.; Kundu, Prasun K.; Einaudi, Franco (Technical Monitor)

2000-01-01

Estimates from TRMM satellite data of monthly total rainfall over an area are subject to substantial sampling errors due to the limited number of visits to the area by the satellite during the month. Quantitative comparisons of TRMM averages with data collected by other satellites and by ground-based systems require some estimate of the size of this sampling error. A method of estimating this sampling error based on the actual statistics of the TRMM observations and on some modeling work has been developed. "Sampling error" in TRMM monthly averages is defined here relative to the monthly total a hypothetical satellite permanently stationed above the area would have reported. "Sampling error" therefore includes contributions from the random and systematic errors introduced by the satellite remote sensing system. As part of our long-term goal of providing error estimates for each grid point accessible to the TRMM instruments, sampling error estimates for TRMM based on rain retrievals from TRMM microwave (TMI) data are compared for different times of the year and different oceanic areas (to minimize changes in the statistics due to algorithmic differences over land and ocean). Changes in sampling error estimates due to changes in rain statistics due 1) to evolution of the official algorithms used to process the data, and 2) differences from other remote sensing systems such as the Defense Meteorological Satellite Program (DMSP) Special Sensor Microwave/Imager (SSM/I), are analyzed.

Relationships between media use, body fatness and physical activity in children and youth: a meta-analysis.

PubMed

Marshall, S J; Biddle, S J H; Gorely, T; Cameron, N; Murdey, I

2004-10-01

To review the empirical evidence of associations between television (TV) viewing, video/computer game use and (a) body fatness, and (b) physical activity. Meta-analysis. Published English-language studies were located from computerized literature searches, bibliographies of primary studies and narrative reviews, and manual searches of personal archives. Included studies presented at least one empirical association between TV viewing, video/computer game use and body fatness or physical activity among samples of children and youth aged 3-18 y. The mean sample-weighted corrected effect size (Pearson r). Based on data from 52 independent samples, the mean sample-weighted effect size between TV viewing and body fatness was 0.066 (95% CI=0.056-0.078; total N=44,707). The sample-weighted fully corrected effect size was 0.084. Based on data from six independent samples, the mean sample-weighted effect size between video/computer game use and body fatness was 0.070 (95% CI=-0.048 to 0.188; total N=1,722). The sample-weighted fully corrected effect size was 0.128. Based on data from 39 independent samples, the mean sample-weighted effect size between TV viewing and physical activity was -0.096 (95% CI=-0.080 to -0.112; total N=141,505). The sample-weighted fully corrected effect size was -0.129. Based on data from 10 independent samples, the mean sample-weighted effect size between video/computer game use and physical activity was -0.104 (95% CI=-0.080 to -0.128; total N=119,942). The sample-weighted fully corrected effect size was -0.141. A statistically significant relationship exists between TV viewing and body fatness among children and youth although it is likely to be too small to be of substantial clinical relevance. The relationship between TV viewing and physical activity is small but negative. The strength of these relationships remains virtually unchanged even after correcting for common sources of bias known to impact study outcomes. While the total amount of time per day engaged in sedentary behavior is inevitably prohibitive of physical activity, media-based inactivity may be unfairly implicated in recent epidemiologic trends of overweight and obesity among children and youth. Relationships between sedentary behavior and health are unlikely to be explained using single markers of inactivity, such as TV viewing or video/computer game use.

Evaluation of multiple-frequency, active and passive acoustics as surrogates for bedload transport

USGS Publications Warehouse

Wood, Molly S.; Fosness, Ryan L.; Pachman, Gregory; Lorang, Mark; Tonolla, Diego

2015-01-01

The use of multiple-frequency, active acoustics through deployment of acoustic Doppler current profilers (ADCPs) shows potential for estimating bedload in selected grain size categories. The U.S. Geological Survey (USGS), in cooperation with the University of Montana (UM), evaluated the use of multiple-frequency, active and passive acoustics as surrogates for bedload transport during a pilot study on the Kootenai River, Idaho, May 17-18, 2012. Four ADCPs with frequencies ranging from 600 to 2000 kHz were used to measure apparent moving bed velocities at 20 stations across the river in conjunction with physical bedload samples. Additionally, UM scientists measured the sound frequencies of moving particles with two hydrophones, considered passive acoustics, along longitudinal transects in the study reach. Some patterns emerged in the preliminary analysis which show promise for future studies. Statistically significant relations were successfully developed between apparent moving bed velocities measured by ADCPs with frequencies 1000 and 1200 kHz and bedload in 0.5 to 2.0 mm grain size categories. The 600 kHz ADCP seemed somewhat sensitive to the movement of gravel bedload in the size range 8.0 to 31.5 mm, but the relation was not statistically significant. The passive hydrophone surveys corroborated the sample results and could be used to map spatial variability in bedload transport and to select a measurement cross-section with moving bedload for active acoustic surveys and physical samples.

Improving risk classification of critical illness with biomarkers: a simulation study

PubMed Central

Seymour, Christopher W.; Cooke, Colin R.; Wang, Zheyu; Kerr, Kathleen F.; Yealy, Donald M.; Angus, Derek C.; Rea, Thomas D.; Kahn, Jeremy M.; Pepe, Margaret S.

2012-01-01

Purpose Optimal triage of patients at risk of critical illness requires accurate risk prediction, yet little data exists on the performance criteria required of a potential biomarker to be clinically useful. Materials and Methods We studied an adult cohort of non-arrest, non-trauma emergency medical services encounters transported to a hospital from 2002–2006. We simulated hypothetical biomarkers increasingly associated with critical illness during hospitalization, and determined the biomarker strength and sample size necessary to improve risk classification beyond a best clinical model. Results Of 57,647 encounters, 3,121 (5.4%) were hospitalized with critical illness and 54,526 (94.6%) without critical illness. The addition of a moderate strength biomarker (odds ratio=3.0 for critical illness) to a clinical model improved discrimination (c-statistic 0.85 vs. 0.8, p<0.01), reclassification (net reclassification improvement=0.15, 95%CI: 0.13,0.18), and increased the proportion of cases in the highest risk categoryby+8.6% (95%CI: 7.5,10.8%). Introducing correlation between the biomarker and physiological variables in the clinical risk score did not modify the results. Statistically significant changes in net reclassification required a sample size of at least 1000 subjects. Conclusions Clinical models for triage of critical illness could be significantly improved by incorporating biomarkers, yet, substantial sample sizes and biomarker strength may be required. PMID:23566734

Acute Respiratory Distress Syndrome Measurement Error. Potential Effect on Clinical Study Results

PubMed Central

Cooke, Colin R.; Iwashyna, Theodore J.; Hofer, Timothy P.

2016-01-01

Rationale: Identifying patients with acute respiratory distress syndrome (ARDS) is a recognized challenge. Experts often have only moderate agreement when applying the clinical definition of ARDS to patients. However, no study has fully examined the implications of low reliability measurement of ARDS on clinical studies. Objectives: To investigate how the degree of variability in ARDS measurement commonly reported in clinical studies affects study power, the accuracy of treatment effect estimates, and the measured strength of risk factor associations. Methods: We examined the effect of ARDS measurement error in randomized clinical trials (RCTs) of ARDS-specific treatments and cohort studies using simulations. We varied the reliability of ARDS diagnosis, quantified as the interobserver reliability (κ-statistic) between two reviewers. In RCT simulations, patients identified as having ARDS were enrolled, and when measurement error was present, patients without ARDS could be enrolled. In cohort studies, risk factors as potential predictors were analyzed using reviewer-identified ARDS as the outcome variable. Measurements and Main Results: Lower reliability measurement of ARDS during patient enrollment in RCTs seriously degraded study power. Holding effect size constant, the sample size necessary to attain adequate statistical power increased by more than 50% as reliability declined, although the result was sensitive to ARDS prevalence. In a 1,400-patient clinical trial, the sample size necessary to maintain similar statistical power increased to over 1,900 when reliability declined from perfect to substantial (κ = 0.72). Lower reliability measurement diminished the apparent effectiveness of an ARDS-specific treatment from a 15.2% (95% confidence interval, 9.4–20.9%) absolute risk reduction in mortality to 10.9% (95% confidence interval, 4.7–16.2%) when reliability declined to moderate (κ = 0.51). In cohort studies, the effect on risk factor associations was similar. Conclusions: ARDS measurement error can seriously degrade statistical power and effect size estimates of clinical studies. The reliability of ARDS measurement warrants careful attention in future ARDS clinical studies. PMID:27159648

Sample size re-estimation and other midcourse adjustments with sequential parallel comparison design.

PubMed

Silverman, Rachel K; Ivanova, Anastasia

2017-01-01

Sequential parallel comparison design (SPCD) was proposed to reduce placebo response in a randomized trial with placebo comparator. Subjects are randomized between placebo and drug in stage 1 of the trial, and then, placebo non-responders are re-randomized in stage 2. Efficacy analysis includes all data from stage 1 and all placebo non-responding subjects from stage 2. This article investigates the possibility to re-estimate the sample size and adjust the design parameters, allocation proportion to placebo in stage 1 of SPCD, and weight of stage 1 data in the overall efficacy test statistic during an interim analysis.

Single-arm phase II trial design under parametric cure models.

PubMed

Wu, Jianrong

2015-01-01

The current practice of designing single-arm phase II survival trials is limited under the exponential model. Trial design under the exponential model may not be appropriate when a portion of patients are cured. There is no literature available for designing single-arm phase II trials under the parametric cure model. In this paper, a test statistic is proposed, and a sample size formula is derived for designing single-arm phase II trials under a class of parametric cure models. Extensive simulations showed that the proposed test and sample size formula perform very well under different scenarios. Copyright © 2015 John Wiley & Sons, Ltd.

Estimating size and scope economies in the Portuguese water sector using the Bayesian stochastic frontier analysis.

PubMed

Carvalho, Pedro; Marques, Rui Cunha

2016-02-15

This study aims to search for economies of size and scope in the Portuguese water sector applying Bayesian and classical statistics to make inference in stochastic frontier analysis (SFA). This study proves the usefulness and advantages of the application of Bayesian statistics for making inference in SFA over traditional SFA which just uses classical statistics. The resulting Bayesian methods allow overcoming some problems that arise in the application of the traditional SFA, such as the bias in small samples and skewness of residuals. In the present case study of the water sector in Portugal, these Bayesian methods provide more plausible and acceptable results. Based on the results obtained we found that there are important economies of output density, economies of size, economies of vertical integration and economies of scope in the Portuguese water sector, pointing out to the huge advantages in undertaking mergers by joining the retail and wholesale components and by joining the drinking water and wastewater services. Copyright © 2015 Elsevier B.V. All rights reserved.

Texture Classification by Texton: Statistical versus Binary

PubMed Central

Guo, Zhenhua; Zhang, Zhongcheng; Li, Xiu; Li, Qin; You, Jane

2014-01-01

Using statistical textons for texture classification has shown great success recently. The maximal response 8 (Statistical_MR8), image patch (Statistical_Joint) and locally invariant fractal (Statistical_Fractal) are typical statistical texton algorithms and state-of-the-art texture classification methods. However, there are two limitations when using these methods. First, it needs a training stage to build a texton library, thus the recognition accuracy will be highly depended on the training samples; second, during feature extraction, local feature is assigned to a texton by searching for the nearest texton in the whole library, which is time consuming when the library size is big and the dimension of feature is high. To address the above two issues, in this paper, three binary texton counterpart methods were proposed, Binary_MR8, Binary_Joint, and Binary_Fractal. These methods do not require any training step but encode local feature into binary representation directly. The experimental results on the CUReT, UIUC and KTH-TIPS databases show that binary texton could get sound results with fast feature extraction, especially when the image size is not big and the quality of image is not poor. PMID:24520346

Logical and Methodological Issues Affecting Genetic Studies of Humans Reported in Top Neuroscience Journals.

PubMed

Grabitz, Clara R; Button, Katherine S; Munafò, Marcus R; Newbury, Dianne F; Pernet, Cyril R; Thompson, Paul A; Bishop, Dorothy V M

2018-01-01

Genetics and neuroscience are two areas of science that pose particular methodological problems because they involve detecting weak signals (i.e., small effects) in noisy data. In recent years, increasing numbers of studies have attempted to bridge these disciplines by looking for genetic factors associated with individual differences in behavior, cognition, and brain structure or function. However, different methodological approaches to guarding against false positives have evolved in the two disciplines. To explore methodological issues affecting neurogenetic studies, we conducted an in-depth analysis of 30 consecutive articles in 12 top neuroscience journals that reported on genetic associations in nonclinical human samples. It was often difficult to estimate effect sizes in neuroimaging paradigms. Where effect sizes could be calculated, the studies reporting the largest effect sizes tended to have two features: (i) they had the smallest samples and were generally underpowered to detect genetic effects, and (ii) they did not fully correct for multiple comparisons. Furthermore, only a minority of studies used statistical methods for multiple comparisons that took into account correlations between phenotypes or genotypes, and only nine studies included a replication sample or explicitly set out to replicate a prior finding. Finally, presentation of methodological information was not standardized and was often distributed across Methods sections and Supplementary Material, making it challenging to assemble basic information from many studies. Space limits imposed by journals could mean that highly complex statistical methods were described in only a superficial fashion. In summary, methods that have become standard in the genetics literature-stringent statistical standards, use of large samples, and replication of findings-are not always adopted when behavioral, cognitive, or neuroimaging phenotypes are used, leading to an increased risk of false-positive findings. Studies need to correct not just for the number of phenotypes collected but also for the number of genotypes examined, genetic models tested, and subsamples investigated. The field would benefit from more widespread use of methods that take into account correlations between the factors corrected for, such as spectral decomposition, or permutation approaches. Replication should become standard practice; this, together with the need for larger sample sizes, will entail greater emphasis on collaboration between research groups. We conclude with some specific suggestions for standardized reporting in this area.

Developing GIS-based eastern equine encephalitis vector-host models in Tuskegee, Alabama.

PubMed

Jacob, Benjamin G; Burkett-Cadena, Nathan D; Luvall, Jeffrey C; Parcak, Sarah H; McClure, Christopher J W; Estep, Laura K; Hill, Geoffrey E; Cupp, Eddie W; Novak, Robert J; Unnasch, Thomas R

2010-02-24

A site near Tuskegee, Alabama was examined for vector-host activities of eastern equine encephalomyelitis virus (EEEV). Land cover maps of the study site were created in ArcInfo 9.2 from QuickBird data encompassing visible and near-infrared (NIR) band information (0.45 to 0.72 microm) acquired July 15, 2008. Georeferenced mosquito and bird sampling sites, and their associated land cover attributes from the study site, were overlaid onto the satellite data. SAS 9.1.4 was used to explore univariate statistics and to generate regression models using the field and remote-sampled mosquito and bird data. Regression models indicated that Culex erracticus and Northern Cardinals were the most abundant mosquito and bird species, respectively. Spatial linear prediction models were then generated in Geostatistical Analyst Extension of ArcGIS 9.2. Additionally, a model of the study site was generated, based on a Digital Elevation Model (DEM), using ArcScene extension of ArcGIS 9.2. For total mosquito count data, a first-order trend ordinary kriging process was fitted to the semivariogram at a partial sill of 5.041 km, nugget of 6.325 km, lag size of 7.076 km, and range of 31.43 km, using 12 lags. For total adult Cx. erracticus count, a first-order trend ordinary kriging process was fitted to the semivariogram at a partial sill of 5.764 km, nugget of 6.114 km, lag size of 7.472 km, and range of 32.62 km, using 12 lags. For the total bird count data, a first-order trend ordinary kriging process was fitted to the semivariogram at a partial sill of 4.998 km, nugget of 5.413 km, lag size of 7.549 km and range of 35.27 km, using 12 lags. For the Northern Cardinal count data, a first-order trend ordinary kriging process was fitted to the semivariogram at a partial sill of 6.387 km, nugget of 5.935 km, lag size of 8.549 km and a range of 41.38 km, using 12 lags. Results of the DEM analyses indicated a statistically significant inverse linear relationship between total sampled mosquito data and elevation (R2 = -.4262; p < .0001), with a standard deviation (SD) of 10.46, and total sampled bird data and elevation (R2 = -.5111; p < .0001), with a SD of 22.97. DEM statistics also indicated a significant inverse linear relationship between total sampled Cx. erracticus data and elevation (R2 = -.4711; p < .0001), with a SD of 11.16, and the total sampled Northern Cardinal data and elevation (R2 = -.5831; p < .0001), SD of 11.42. These data demonstrate that GIS/remote sensing models and spatial statistics can capture space-varying functional relationships between field-sampled mosquito and bird parameters for determining risk for EEEV transmission.

Sampling design and required sample size for evaluating contamination levels of 137Cs in Japanese fir needles in a mixed deciduous forest stand in Fukushima, Japan.

PubMed

Oba, Yurika; Yamada, Toshihiro

2017-05-01

We estimated the sample size (the number of samples) required to evaluate the concentration of radiocesium ( 137 Cs) in Japanese fir (Abies firma Sieb. & Zucc.), 5 years after the outbreak of the Fukushima Daiichi Nuclear Power Plant accident. We investigated the spatial structure of the contamination levels in this species growing in a mixed deciduous broadleaf and evergreen coniferous forest stand. We sampled 40 saplings with a tree height of 150 cm-250 cm in a Fukushima forest community. The results showed that: (1) there was no correlation between the 137 Cs concentration in needles and soil, and (2) the difference in the spatial distribution pattern of 137 Cs concentration between needles and soil suggest that the contribution of root uptake to 137 Cs in new needles of this species may be minor in the 5 years after the radionuclides were released into the atmosphere. The concentration of 137 Cs in needles showed a strong positive spatial autocorrelation in the distance class from 0 to 2.5 m, suggesting that the statistical analysis of data should consider spatial autocorrelation in the case of an assessment of the radioactive contamination of forest trees. According to our sample size analysis, a sample size of seven trees was required to determine the mean contamination level within an error in the means of no more than 10%. This required sample size may be feasible for most sites. Copyright © 2017 Elsevier Ltd. All rights reserved.

On the repeated measures designs and sample sizes for randomized controlled trials.

PubMed

Tango, Toshiro

2016-04-01

For the analysis of longitudinal or repeated measures data, generalized linear mixed-effects models provide a flexible and powerful tool to deal with heterogeneity among subject response profiles. However, the typical statistical design adopted in usual randomized controlled trials is an analysis of covariance type analysis using a pre-defined pair of "pre-post" data, in which pre-(baseline) data are used as a covariate for adjustment together with other covariates. Then, the major design issue is to calculate the sample size or the number of subjects allocated to each treatment group. In this paper, we propose a new repeated measures design and sample size calculations combined with generalized linear mixed-effects models that depend not only on the number of subjects but on the number of repeated measures before and after randomization per subject used for the analysis. The main advantages of the proposed design combined with the generalized linear mixed-effects models are (1) it can easily handle missing data by applying the likelihood-based ignorable analyses under the missing at random assumption and (2) it may lead to a reduction in sample size, compared with the simple pre-post design. The proposed designs and the sample size calculations are illustrated with real data arising from randomized controlled trials. © The Author 2015. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Effect of dislocation pile-up on size-dependent yield strength in finite single-crystal micro-samples

DOE Office of Scientific and Technical Information (OSTI.GOV)

Pan, Bo; Shibutani, Yoji, E-mail: sibutani@mech.eng.osaka-u.ac.jp; Zhang, Xu

2015-07-07

Recent research has explained that the steeply increasing yield strength in metals depends on decreasing sample size. In this work, we derive a statistical physical model of the yield strength of finite single-crystal micro-pillars that depends on single-ended dislocation pile-up inside the micro-pillars. We show that this size effect can be explained almost completely by considering the stochastic lengths of the dislocation source and the dislocation pile-up length in the single-crystal micro-pillars. The Hall–Petch-type relation holds even in a microscale single-crystal, which is characterized by its dislocation source lengths. Our quantitative conclusions suggest that the number of dislocation sources andmore » pile-ups are significant factors for the size effect. They also indicate that starvation of dislocation sources is another reason for the size effect. Moreover, we investigated the explicit relationship between the stacking fault energy and the dislocation “pile-up” effect inside the sample: materials with low stacking fault energy exhibit an obvious dislocation pile-up effect. Our proposed physical model predicts a sample strength that agrees well with experimental data, and our model can give a more precise prediction than the current single arm source model, especially for materials with low stacking fault energy.« less

Impacts of Industrial Wind Turbine Noise on Sleep Quality: Results From a Field Study of Rural Residents in Ontario, Canada.

PubMed

Lane, James D; Bigelow, Philip L; Majowicz, Shannon E; McColl, R Stephen

2016-07-01

The objectives of this study were to determine whether grid-connected industrial wind turbines (IWTs) are a risk factor for poor sleep quality, and if IWT noise is associated with sleep parameters in rural Ontarians. A daily sleep diary and actigraphy-derived measures of sleep were obtained from 12 participants from an IWT community and 10 participants from a comparison community with no wind power installations. The equivalent and maximum sound pressure levels within the bedroom were also assessed. No statistically significant differences were observed between IWT residents and non-IWT residents for any of the parameters measured in this study. Actigraphy and sleep diaries are feasible tools to understand the impact of IWTs on the quality of sleep for nearby residents. Further studies with larger sample sizes should be conducted to determine whether the lack of statistical significance observed here is a result of sample size, or reflects a true lack of association.

Accounting for response misclassification and covariate measurement error improves power and reduces bias in epidemiologic studies.

PubMed

Cheng, Dunlei; Branscum, Adam J; Stamey, James D

2010-07-01

To quantify the impact of ignoring misclassification of a response variable and measurement error in a covariate on statistical power, and to develop software for sample size and power analysis that accounts for these flaws in epidemiologic data. A Monte Carlo simulation-based procedure is developed to illustrate the differences in design requirements and inferences between analytic methods that properly account for misclassification and measurement error to those that do not in regression models for cross-sectional and cohort data. We found that failure to account for these flaws in epidemiologic data can lead to a substantial reduction in statistical power, over 25% in some cases. The proposed method substantially reduced bias by up to a ten-fold margin compared to naive estimates obtained by ignoring misclassification and mismeasurement. We recommend as routine practice that researchers account for errors in measurement of both response and covariate data when determining sample size, performing power calculations, or analyzing data from epidemiological studies. 2010 Elsevier Inc. All rights reserved.

Partitioning heritability by functional annotation using genome-wide association summary statistics.

PubMed

Finucane, Hilary K; Bulik-Sullivan, Brendan; Gusev, Alexander; Trynka, Gosia; Reshef, Yakir; Loh, Po-Ru; Anttila, Verneri; Xu, Han; Zang, Chongzhi; Farh, Kyle; Ripke, Stephan; Day, Felix R; Purcell, Shaun; Stahl, Eli; Lindstrom, Sara; Perry, John R B; Okada, Yukinori; Raychaudhuri, Soumya; Daly, Mark J; Patterson, Nick; Neale, Benjamin M; Price, Alkes L

2015-11-01

Recent work has demonstrated that some functional categories of the genome contribute disproportionately to the heritability of complex diseases. Here we analyze a broad set of functional elements, including cell type-specific elements, to estimate their polygenic contributions to heritability in genome-wide association studies (GWAS) of 17 complex diseases and traits with an average sample size of 73,599. To enable this analysis, we introduce a new method, stratified LD score regression, for partitioning heritability from GWAS summary statistics while accounting for linked markers. This new method is computationally tractable at very large sample sizes and leverages genome-wide information. Our findings include a large enrichment of heritability in conserved regions across many traits, a very large immunological disease-specific enrichment of heritability in FANTOM5 enhancers and many cell type-specific enrichments, including significant enrichment of central nervous system cell types in the heritability of body mass index, age at menarche, educational attainment and smoking behavior.

Empirical assessment of published effect sizes and power in the recent cognitive neuroscience and psychology literature

PubMed Central

Szucs, Denes; Ioannidis, John P. A.

2017-01-01

We have empirically assessed the distribution of published effect sizes and estimated power by analyzing 26,841 statistical records from 3,801 cognitive neuroscience and psychology papers published recently. The reported median effect size was D = 0.93 (interquartile range: 0.64–1.46) for nominally statistically significant results and D = 0.24 (0.11–0.42) for nonsignificant results. Median power to detect small, medium, and large effects was 0.12, 0.44, and 0.73, reflecting no improvement through the past half-century. This is so because sample sizes have remained small. Assuming similar true effect sizes in both disciplines, power was lower in cognitive neuroscience than in psychology. Journal impact factors negatively correlated with power. Assuming a realistic range of prior probabilities for null hypotheses, false report probability is likely to exceed 50% for the whole literature. In light of our findings, the recently reported low replication success in psychology is realistic, and worse performance may be expected for cognitive neuroscience. PMID:28253258

Statistical theory and methodology for remote sensing data analysis

NASA Technical Reports Server (NTRS)

Odell, P. L.

1974-01-01

A model is developed for the evaluation of acreages (proportions) of different crop-types over a geographical area using a classification approach and methods for estimating the crop acreages are given. In estimating the acreages of a specific croptype such as wheat, it is suggested to treat the problem as a two-crop problem: wheat vs. nonwheat, since this simplifies the estimation problem considerably. The error analysis and the sample size problem is investigated for the two-crop approach. Certain numerical results for sample sizes are given for a JSC-ERTS-1 data example on wheat identification performance in Hill County, Montana and Burke County, North Dakota. Lastly, for a large area crop acreages inventory a sampling scheme is suggested for acquiring sample data and the problem of crop acreage estimation and the error analysis is discussed.

Forest inventory using multistage sampling with probability proportional to size. [Brazil

NASA Technical Reports Server (NTRS)

Parada, N. D. J. (Principal Investigator); Lee, D. C. L.; Hernandezfilho, P.; Shimabukuro, Y. E.; Deassis, O. R.; Demedeiros, J. S.

1984-01-01

A multistage sampling technique, with probability proportional to size, for forest volume inventory using remote sensing data is developed and evaluated. The study area is located in the Southeastern Brazil. The LANDSAT 4 digital data of the study area are used in the first stage for automatic classification of reforested areas. Four classes of pine and eucalypt with different tree volumes are classified utilizing a maximum likelihood classification algorithm. Color infrared aerial photographs are utilized in the second stage of sampling. In the third state (ground level) the time volume of each class is determined. The total time volume of each class is expanded through a statistical procedure taking into account all the three stages of sampling. This procedure results in an accurate time volume estimate with a smaller number of aerial photographs and reduced time in field work.

Estimating and comparing microbial diversity in the presence of sequencing errors

PubMed Central

Chiu, Chun-Huo

2016-01-01

Estimating and comparing microbial diversity are statistically challenging due to limited sampling and possible sequencing errors for low-frequency counts, producing spurious singletons. The inflated singleton count seriously affects statistical analysis and inferences about microbial diversity. Previous statistical approaches to tackle the sequencing errors generally require different parametric assumptions about the sampling model or about the functional form of frequency counts. Different parametric assumptions may lead to drastically different diversity estimates. We focus on nonparametric methods which are universally valid for all parametric assumptions and can be used to compare diversity across communities. We develop here a nonparametric estimator of the true singleton count to replace the spurious singleton count in all methods/approaches. Our estimator of the true singleton count is in terms of the frequency counts of doubletons, tripletons and quadrupletons, provided these three frequency counts are reliable. To quantify microbial alpha diversity for an individual community, we adopt the measure of Hill numbers (effective number of taxa) under a nonparametric framework. Hill numbers, parameterized by an order q that determines the measures’ emphasis on rare or common species, include taxa richness (q = 0), Shannon diversity (q = 1, the exponential of Shannon entropy), and Simpson diversity (q = 2, the inverse of Simpson index). A diversity profile which depicts the Hill number as a function of order q conveys all information contained in a taxa abundance distribution. Based on the estimated singleton count and the original non-singleton frequency counts, two statistical approaches (non-asymptotic and asymptotic) are developed to compare microbial diversity for multiple communities. (1) A non-asymptotic approach refers to the comparison of estimated diversities of standardized samples with a common finite sample size or sample completeness. This approach aims to compare diversity estimates for equally-large or equally-complete samples; it is based on the seamless rarefaction and extrapolation sampling curves of Hill numbers, specifically for q = 0, 1 and 2. (2) An asymptotic approach refers to the comparison of the estimated asymptotic diversity profiles. That is, this approach compares the estimated profiles for complete samples or samples whose size tends to be sufficiently large. It is based on statistical estimation of the true Hill number of any order q ≥ 0. In the two approaches, replacing the spurious singleton count by our estimated count, we can greatly remove the positive biases associated with diversity estimates due to spurious singletons and also make fair comparisons across microbial communities, as illustrated in our simulation results and in applying our method to analyze sequencing data from viral metagenomes. PMID:26855872

An analysis of Apollo lunar soil samples 12070,889, 12030,187, and 12070,891: Basaltic diversity at the Apollo 12 landing site and implications for classification of small-sized lunar samples

NASA Astrophysics Data System (ADS)

Alexander, Louise; Snape, Joshua F.; Joy, Katherine H.; Downes, Hilary; Crawford, Ian A.

2016-09-01

Lunar mare basalts provide insights into the compositional diversity of the Moon's interior. Basalt fragments from the lunar regolith can potentially sample lava flows from regions of the Moon not previously visited, thus, increasing our understanding of lunar geological evolution. As part of a study of basaltic diversity at the Apollo 12 landing site, detailed petrological and geochemical data are provided here for 13 basaltic chips. In addition to bulk chemistry, we have analyzed the major, minor, and trace element chemistry of mineral phases which highlight differences between basalt groups. Where samples contain olivine, the equilibrium parent melt magnesium number (Mg#; atomic Mg/[Mg + Fe]) can be calculated to estimate parent melt composition. Ilmenite and plagioclase chemistry can also determine differences between basalt groups. We conclude that samples of approximately 1-2 mm in size can be categorized provided that appropriate mineral phases (olivine, plagioclase, and ilmenite) are present. Where samples are fine-grained (grain size <0.3 mm), a "paired samples t-test" can provide a statistical comparison between a particular sample and known lunar basalts. Of the fragments analyzed here, three are found to belong to each of the previously identified olivine and ilmenite basalt suites, four to the pigeonite basalt suite, one is an olivine cumulate, and two could not be categorized because of their coarse grain sizes and lack of appropriate mineral phases. Our approach introduces methods that can be used to investigate small sample sizes (i.e., fines) from future sample return missions to investigate lava flow diversity and petrological significance.

Static Scene Statistical Non-Uniformity Correction

DTIC Science & Technology

2015-03-01

Error NUC Non-Uniformity Correction RMSE Root Mean Squared Error RSD Relative Standard Deviation S3NUC Static Scene Statistical Non-Uniformity...Deviation ( RSD ) which normalizes the standard deviation, σ, to the mean estimated value, µ using the equation RS D = σ µ × 100. The RSD plot of the gain...estimates is shown in Figure 4.1(b). The RSD plot shows that after a sample size of approximately 10, the different photocount values and the inclusion

Researchers’ Intuitions About Power in Psychological Research

PubMed Central

Bakker, Marjan; Hartgerink, Chris H. J.; Wicherts, Jelte M.; van der Maas, Han L. J.

2016-01-01

Many psychology studies are statistically underpowered. In part, this may be because many researchers rely on intuition, rules of thumb, and prior practice (along with practical considerations) to determine the number of subjects to test. In Study 1, we surveyed 291 published research psychologists and found large discrepancies between their reports of their preferred amount of power and the actual power of their studies (calculated from their reported typical cell size, typical effect size, and acceptable alpha). Furthermore, in Study 2, 89% of the 214 respondents overestimated the power of specific research designs with a small expected effect size, and 95% underestimated the sample size needed to obtain .80 power for detecting a small effect. Neither researchers’ experience nor their knowledge predicted the bias in their self-reported power intuitions. Because many respondents reported that they based their sample sizes on rules of thumb or common practice in the field, we recommend that researchers conduct and report formal power analyses for their studies. PMID:27354203

Researchers' Intuitions About Power in Psychological Research.

PubMed

Bakker, Marjan; Hartgerink, Chris H J; Wicherts, Jelte M; van der Maas, Han L J

2016-08-01

Many psychology studies are statistically underpowered. In part, this may be because many researchers rely on intuition, rules of thumb, and prior practice (along with practical considerations) to determine the number of subjects to test. In Study 1, we surveyed 291 published research psychologists and found large discrepancies between their reports of their preferred amount of power and the actual power of their studies (calculated from their reported typical cell size, typical effect size, and acceptable alpha). Furthermore, in Study 2, 89% of the 214 respondents overestimated the power of specific research designs with a small expected effect size, and 95% underestimated the sample size needed to obtain .80 power for detecting a small effect. Neither researchers' experience nor their knowledge predicted the bias in their self-reported power intuitions. Because many respondents reported that they based their sample sizes on rules of thumb or common practice in the field, we recommend that researchers conduct and report formal power analyses for their studies. © The Author(s) 2016.

Some challenges with statistical inference in adaptive designs.

PubMed

Hung, H M James; Wang, Sue-Jane; Yang, Peiling

2014-01-01

Adaptive designs have generated a great deal of attention to clinical trial communities. The literature contains many statistical methods to deal with added statistical uncertainties concerning the adaptations. Increasingly encountered in regulatory applications are adaptive statistical information designs that allow modification of sample size or related statistical information and adaptive selection designs that allow selection of doses or patient populations during the course of a clinical trial. For adaptive statistical information designs, a few statistical testing methods are mathematically equivalent, as a number of articles have stipulated, but arguably there are large differences in their practical ramifications. We pinpoint some undesirable features of these methods in this work. For adaptive selection designs, the selection based on biomarker data for testing the correlated clinical endpoints may increase statistical uncertainty in terms of type I error probability, and most importantly the increased statistical uncertainty may be impossible to assess.

On the comparison of the strength of morphological integration across morphometric datasets.

PubMed

Adams, Dean C; Collyer, Michael L

2016-11-01

Evolutionary morphologists frequently wish to understand the extent to which organisms are integrated, and whether the strength of morphological integration among subsets of phenotypic variables differ among taxa or other groups. However, comparisons of the strength of integration across datasets are difficult, in part because the summary measures that characterize these patterns (RV coefficient and r PLS ) are dependent both on sample size and on the number of variables. As a solution to this issue, we propose a standardized test statistic (a z-score) for measuring the degree of morphological integration between sets of variables. The approach is based on a partial least squares analysis of trait covariation, and its permutation-based sampling distribution. Under the null hypothesis of a random association of variables, the method displays a constant expected value and confidence intervals for datasets of differing sample sizes and variable number, thereby providing a consistent measure of integration suitable for comparisons across datasets. A two-sample test is also proposed to statistically determine whether levels of integration differ between datasets, and an empirical example examining cranial shape integration in Mediterranean wall lizards illustrates its use. Some extensions of the procedure are also discussed. © 2016 The Author(s). Evolution © 2016 The Society for the Study of Evolution.

Statistical methods for identifying and bounding a UXO target area or minefield

DOE Office of Scientific and Technical Information (OSTI.GOV)

McKinstry, Craig A.; Pulsipher, Brent A.; Gilbert, Richard O.

2003-09-18

The sampling unit for minefield or UXO area characterization is typically represented by a geographical block or transect swath that lends itself to characterization by geophysical instrumentation such as mobile sensor arrays. New spatially based statistical survey methods and tools, more appropriate for these unique sampling units have been developed and implemented at PNNL (Visual Sample Plan software, ver. 2.0) with support from the US Department of Defense. Though originally developed to support UXO detection and removal efforts, these tools may also be used in current form or adapted to support demining efforts and aid in the development of newmore » sensors and detection technologies by explicitly incorporating both sampling and detection error in performance assessments. These tools may be used to (1) determine transect designs for detecting and bounding target areas of critical size, shape, and density of detectable items of interest with a specified confidence probability, (2) evaluate the probability that target areas of a specified size, shape and density have not been missed by a systematic or meandering transect survey, and (3) support post-removal verification by calculating the number of transects required to achieve a specified confidence probability that no UXO or mines have been missed.« less

The local environment of ice particles in arctic mixed-phase clouds

NASA Astrophysics Data System (ADS)

Schlenczek, Oliver; Fugal, Jacob P.; Schledewitz, Waldemar; Borrmann, Stephan

2015-04-01

During the RACEPAC field campaign in April and May 2014, research flights were made with the Polar 5 and Polar 6 aircraft from the Alfred Wegener Institute in Arctic clouds near Inuvik, Northwest Territories, Canada. One flight with the Polar 6 aircraft, done on May 16, 2014, flew under precipitating, stratiform, mid-level clouds with several penetrations through cloud base. Measurements with HALOHolo, an airborne digital in-line holographic instrument for cloud particles, show ice particles in a field of other cloud particles in a local three-dimensional sample volume (~14x19x130 mm3 or ~35 cm^3). Each holographic sample volume is a snapshot of a 3-dimensional piece of cloud at the cm-scale with typically thousands of cloud droplets per sample volume, so each sample volume yields a statistically significant droplet size distribution. Holograms are recorded at a rate of six times per second, which provides one volume sample approx. every 12 meters along the flight path. The size resolution limit for cloud droplets is better than 1 µm due to advanced sizing algorithms. Shown are preliminary results of, (1) the ice/liquid water partitioning at the cloud base and the distribution of water droplets around each ice particle, and (2) spatial and temporal variability of the cloud droplet size distributions at cloud base.

Global Sensitivity Analysis with Small Sample Sizes: Ordinary Least Squares Approach

DOE Office of Scientific and Technical Information (OSTI.GOV)

Davis, Michael J.; Liu, Wei; Sivaramakrishnan, Raghu

2016-12-21

A new version of global sensitivity analysis is developed in this paper. This new version coupled with tools from statistics, machine learning, and optimization can devise small sample sizes that allow for the accurate ordering of sensitivity coefficients for the first 10-30 most sensitive chemical reactions in complex chemical-kinetic mechanisms, and is particularly useful for studying the chemistry in realistic devices. A key part of the paper is calibration of these small samples. Because these small sample sizes are developed for use in realistic combustion devices, the calibration is done over the ranges of conditions in such devices, with amore » test case being the operating conditions of a compression ignition engine studied earlier. Compression ignition engines operate under low-temperature combustion conditions with quite complicated chemistry making this calibration difficult, leading to the possibility of false positives and false negatives in the ordering of the reactions. So an important aspect of the paper is showing how to handle the trade-off between false positives and false negatives using ideas from the multiobjective optimization literature. The combination of the new global sensitivity method and the calibration are sample sizes a factor of approximately 10 times smaller than were available with our previous algorithm.« less

Design and analysis of three-arm trials with negative binomially distributed endpoints.

PubMed

Mütze, Tobias; Munk, Axel; Friede, Tim

2016-02-20

A three-arm clinical trial design with an experimental treatment, an active control, and a placebo control, commonly referred to as the gold standard design, enables testing of non-inferiority or superiority of the experimental treatment compared with the active control. In this paper, we propose methods for designing and analyzing three-arm trials with negative binomially distributed endpoints. In particular, we develop a Wald-type test with a restricted maximum-likelihood variance estimator for testing non-inferiority or superiority. For this test, sample size and power formulas as well as optimal sample size allocations will be derived. The performance of the proposed test will be assessed in an extensive simulation study with regard to type I error rate, power, sample size, and sample size allocation. For the purpose of comparison, Wald-type statistics with a sample variance estimator and an unrestricted maximum-likelihood estimator are included in the simulation study. We found that the proposed Wald-type test with a restricted variance estimator performed well across the considered scenarios and is therefore recommended for application in clinical trials. The methods proposed are motivated and illustrated by a recent clinical trial in multiple sclerosis. The R package ThreeArmedTrials, which implements the methods discussed in this paper, is available on CRAN. Copyright © 2015 John Wiley & Sons, Ltd.

Observational studies of patients in the emergency department: a comparison of 4 sampling methods.

PubMed

Valley, Morgan A; Heard, Kennon J; Ginde, Adit A; Lezotte, Dennis C; Lowenstein, Steven R

2012-08-01

We evaluate the ability of 4 sampling methods to generate representative samples of the emergency department (ED) population. We analyzed the electronic records of 21,662 consecutive patient visits at an urban, academic ED. From this population, we simulated different models of study recruitment in the ED by using 2 sample sizes (n=200 and n=400) and 4 sampling methods: true random, random 4-hour time blocks by exact sample size, random 4-hour time blocks by a predetermined number of blocks, and convenience or "business hours." For each method and sample size, we obtained 1,000 samples from the population. Using χ(2) tests, we measured the number of statistically significant differences between the sample and the population for 8 variables (age, sex, race/ethnicity, language, triage acuity, arrival mode, disposition, and payer source). Then, for each variable, method, and sample size, we compared the proportion of the 1,000 samples that differed from the overall ED population to the expected proportion (5%). Only the true random samples represented the population with respect to sex, race/ethnicity, triage acuity, mode of arrival, language, and payer source in at least 95% of the samples. Patient samples obtained using random 4-hour time blocks and business hours sampling systematically differed from the overall ED patient population for several important demographic and clinical variables. However, the magnitude of these differences was not large. Common sampling strategies selected for ED-based studies may affect parameter estimates for several representative population variables. However, the potential for bias for these variables appears small. Copyright © 2012. Published by Mosby, Inc.

Determination of Minimum Training Sample Size for Microarray-Based Cancer Outcome Prediction–An Empirical Assessment

PubMed Central

Cheng, Ningtao; Wu, Leihong; Cheng, Yiyu

2013-01-01

The promise of microarray technology in providing prediction classifiers for cancer outcome estimation has been confirmed by a number of demonstrable successes. However, the reliability of prediction results relies heavily on the accuracy of statistical parameters involved in classifiers. It cannot be reliably estimated with only a small number of training samples. Therefore, it is of vital importance to determine the minimum number of training samples and to ensure the clinical value of microarrays in cancer outcome prediction. We evaluated the impact of training sample size on model performance extensively based on 3 large-scale cancer microarray datasets provided by the second phase of MicroArray Quality Control project (MAQC-II). An SSNR-based (scale of signal-to-noise ratio) protocol was proposed in this study for minimum training sample size determination. External validation results based on another 3 cancer datasets confirmed that the SSNR-based approach could not only determine the minimum number of training samples efficiently, but also provide a valuable strategy for estimating the underlying performance of classifiers in advance. Once translated into clinical routine applications, the SSNR-based protocol would provide great convenience in microarray-based cancer outcome prediction in improving classifier reliability. PMID:23861920

Malaria prevalence metrics in low- and middle-income countries: an assessment of precision in nationally-representative surveys.

PubMed

Alegana, Victor A; Wright, Jim; Bosco, Claudio; Okiro, Emelda A; Atkinson, Peter M; Snow, Robert W; Tatem, Andrew J; Noor, Abdisalan M

2017-11-21

One pillar to monitoring progress towards the Sustainable Development Goals is the investment in high quality data to strengthen the scientific basis for decision-making. At present, nationally-representative surveys are the main source of data for establishing a scientific evidence base, monitoring, and evaluation of health metrics. However, little is known about the optimal precisions of various population-level health and development indicators that remains unquantified in nationally-representative household surveys. Here, a retrospective analysis of the precision of prevalence from these surveys was conducted. Using malaria indicators, data were assembled in nine sub-Saharan African countries with at least two nationally-representative surveys. A Bayesian statistical model was used to estimate between- and within-cluster variability for fever and malaria prevalence, and insecticide-treated bed nets (ITNs) use in children under the age of 5 years. The intra-class correlation coefficient was estimated along with the optimal sample size for each indicator with associated uncertainty. Results suggest that the estimated sample sizes for the current nationally-representative surveys increases with declining malaria prevalence. Comparison between the actual sample size and the modelled estimate showed a requirement to increase the sample size for parasite prevalence by up to 77.7% (95% Bayesian credible intervals 74.7-79.4) for the 2015 Kenya MIS (estimated sample size of children 0-4 years 7218 [7099-7288]), and 54.1% [50.1-56.5] for the 2014-2015 Rwanda DHS (12,220 [11,950-12,410]). This study highlights the importance of defining indicator-relevant sample sizes to achieve the required precision in the current national surveys. While expanding the current surveys would need additional investment, the study highlights the need for improved approaches to cost effective sampling.

Assessment of variations in thermal cycle life data of thermal barrier coated rods

NASA Astrophysics Data System (ADS)

Hendricks, R. C.; McDonald, G.

An analysis of thermal cycle life data for 22 thermal barrier coated (TBC) specimens was conducted. The Zr02-8Y203/NiCrAlY plasma spray coated Rene 41 rods were tested in a Mach 0.3 Jet A/air burner flame. All specimens were subjected to the same coating and subsequent test procedures in an effort to control three parametric groups; material properties, geometry and heat flux. Statistically, the data sample space had a mean of 1330 cycles with a standard deviation of 520 cycles. The data were described by normal or log-normal distributions, but other models could also apply; the sample size must be increased to clearly delineate a statistical failure model. The statistical methods were also applied to adhesive/cohesive strength data for 20 TBC discs of the same composition, with similar results. The sample space had a mean of 9 MPa with a standard deviation of 4.2 MPa.

Assessment of variations in thermal cycle life data of thermal barrier coated rods

NASA Technical Reports Server (NTRS)

Hendricks, R. C.; Mcdonald, G.

1981-01-01

An analysis of thermal cycle life data for 22 thermal barrier coated (TBC) specimens was conducted. The Zr02-8Y203/NiCrAlY plasma spray coated Rene 41 rods were tested in a Mach 0.3 Jet A/air burner flame. All specimens were subjected to the same coating and subsequent test procedures in an effort to control three parametric groups; material properties, geometry and heat flux. Statistically, the data sample space had a mean of 1330 cycles with a standard deviation of 520 cycles. The data were described by normal or log-normal distributions, but other models could also apply; the sample size must be increased to clearly delineate a statistical failure model. The statistical methods were also applied to adhesive/cohesive strength data for 20 TBC discs of the same composition, with similar results. The sample space had a mean of 9 MPa with a standard deviation of 4.2 MPa.

Comparative forensic soil analysis of New Jersey state parks using a combination of simple techniques with multivariate statistics.

PubMed

Bonetti, Jennifer; Quarino, Lawrence

2014-05-01

This study has shown that the combination of simple techniques with the use of multivariate statistics offers the potential for the comparative analysis of soil samples. Five samples were obtained from each of twelve state parks across New Jersey in both the summer and fall seasons. Each sample was examined using particle-size distribution, pH analysis in both water and 1 M CaCl2 , and a loss on ignition technique. Data from each of the techniques were combined, and principal component analysis (PCA) and canonical discriminant analysis (CDA) were used for multivariate data transformation. Samples from different locations could be visually differentiated from one another using these multivariate plots. Hold-one-out cross-validation analysis showed error rates as low as 3.33%. Ten blind study samples were analyzed resulting in no misclassifications using Mahalanobis distance calculations and visual examinations of multivariate plots. Seasonal variation was minimal between corresponding samples, suggesting potential success in forensic applications. © 2014 American Academy of Forensic Sciences.

Analysis of Statistical Methods Currently used in Toxicology Journals

PubMed Central

Na, Jihye; Yang, Hyeri

2014-01-01

Statistical methods are frequently used in toxicology, yet it is not clear whether the methods employed by the studies are used consistently and conducted based on sound statistical grounds. The purpose of this paper is to describe statistical methods used in top toxicology journals. More specifically, we sampled 30 papers published in 2014 from Toxicology and Applied Pharmacology, Archives of Toxicology, and Toxicological Science and described methodologies used to provide descriptive and inferential statistics. One hundred thirteen endpoints were observed in those 30 papers, and most studies had sample size less than 10, with the median and the mode being 6 and 3 & 6, respectively. Mean (105/113, 93%) was dominantly used to measure central tendency, and standard error of the mean (64/113, 57%) and standard deviation (39/113, 34%) were used to measure dispersion, while few studies provide justifications regarding why the methods being selected. Inferential statistics were frequently conducted (93/113, 82%), with one-way ANOVA being most popular (52/93, 56%), yet few studies conducted either normality or equal variance test. These results suggest that more consistent and appropriate use of statistical method is necessary which may enhance the role of toxicology in public health. PMID:25343012

Analysis of Statistical Methods Currently used in Toxicology Journals.

PubMed

Na, Jihye; Yang, Hyeri; Bae, SeungJin; Lim, Kyung-Min

2014-09-01

Statistical methods are frequently used in toxicology, yet it is not clear whether the methods employed by the studies are used consistently and conducted based on sound statistical grounds. The purpose of this paper is to describe statistical methods used in top toxicology journals. More specifically, we sampled 30 papers published in 2014 from Toxicology and Applied Pharmacology, Archives of Toxicology, and Toxicological Science and described methodologies used to provide descriptive and inferential statistics. One hundred thirteen endpoints were observed in those 30 papers, and most studies had sample size less than 10, with the median and the mode being 6 and 3 & 6, respectively. Mean (105/113, 93%) was dominantly used to measure central tendency, and standard error of the mean (64/113, 57%) and standard deviation (39/113, 34%) were used to measure dispersion, while few studies provide justifications regarding why the methods being selected. Inferential statistics were frequently conducted (93/113, 82%), with one-way ANOVA being most popular (52/93, 56%), yet few studies conducted either normality or equal variance test. These results suggest that more consistent and appropriate use of statistical method is necessary which may enhance the role of toxicology in public health.

Diagnostic test accuracy and prevalence inferences based on joint and sequential testing with finite population sampling.

PubMed

Su, Chun-Lung; Gardner, Ian A; Johnson, Wesley O

2004-07-30

The two-test two-population model, originally formulated by Hui and Walter, for estimation of test accuracy and prevalence estimation assumes conditionally independent tests, constant accuracy across populations and binomial sampling. The binomial assumption is incorrect if all individuals in a population e.g. child-care centre, village in Africa, or a cattle herd are sampled or if the sample size is large relative to population size. In this paper, we develop statistical methods for evaluating diagnostic test accuracy and prevalence estimation based on finite sample data in the absence of a gold standard. Moreover, two tests are often applied simultaneously for the purpose of obtaining a 'joint' testing strategy that has either higher overall sensitivity or specificity than either of the two tests considered singly. Sequential versions of such strategies are often applied in order to reduce the cost of testing. We thus discuss joint (simultaneous and sequential) testing strategies and inference for them. Using the developed methods, we analyse two real and one simulated data sets, and we compare 'hypergeometric' and 'binomial-based' inferences. Our findings indicate that the posterior standard deviations for prevalence (but not sensitivity and specificity) based on finite population sampling tend to be smaller than their counterparts for infinite population sampling. Finally, we make recommendations about how small the sample size should be relative to the population size to warrant use of the binomial model for prevalence estimation. Copyright 2004 John Wiley & Sons, Ltd.

Alteration of histological gastritis after cure of Helicobacter pylori infection.

PubMed

Hojo, M; Miwa, H; Ohkusa, T; Ohkura, R; Kurosawa, A; Sato, N

2002-11-01

It is still disputed whether gastric atrophy or intestinal metaplasia improves after the cure of Helicobacter pylori infection. To clarify the histological changes after the cure of H. pylori infection through a literature survey. Fifty-one selected reports from 1066 relevant articles were reviewed. The extracted data were pooled according to histological parameters of gastritis based on the (updated) Sydney system. Activity improved more rapidly than inflammation. Eleven of 25 reports described significant improvement of atrophy. Atrophy was not improved in one of four studies with a large sample size (> 100 samples) and in two of five studies with a long follow-up period (> 12 months), suggesting that disagreement between the studies was not totally due to sample size or follow-up period. Methodological flaws, such as patient selection, and statistical analysis based on the assumption that atrophy improves continuously and generally in all patients might be responsible for the inconsistent results. Four of 28 studies described significant improvement of intestinal metaplasia [corrected]. Activity and inflammation were improved after the cure of H. pylori infection. Atrophy did not improve generally among all patients, but improved in certain patients. Improvement of intestinal metaplasia was difficult to analyse due to methodological problems including statistical power.

Metrological characterization of X-ray diffraction methods at different acquisition geometries for determination of crystallite size in nano-scale materials

DOE Office of Scientific and Technical Information (OSTI.GOV)

Uvarov, Vladimir, E-mail: vladimiru@savion.huji.ac.il; Popov, Inna

2013-11-15

Crystallite size values were determined by X-ray diffraction methods for 183 powder samples. The tested size range was from a few to about several hundred nanometers. Crystallite size was calculated with direct use of the Scherrer equation, the Williamson–Hall method and the Rietveld procedure via the application of a series of commercial and free software. The results were statistically treated to estimate the significance of the difference in size resulting from these methods. We also estimated effect of acquisition conditions (Bragg–Brentano, parallel-beam geometry, step size, counting time) and data processing on the calculated crystallite size values. On the basis ofmore » the obtained results it is possible to conclude that direct use of the Scherrer equation, Williamson–Hall method and the Rietveld refinement employed by a series of software (EVA, PCW and TOPAS respectively) yield very close results for crystallite sizes less than 60 nm for parallel beam geometry and less than 100 nm for Bragg–Brentano geometry. However, we found that despite the fact that the differences between the crystallite sizes, which were calculated by various methods, are small by absolute values, they are statistically significant in some cases. The values of crystallite size determined from XRD were compared with those obtained by imaging in a transmission (TEM) and scanning electron microscopes (SEM). It was found that there was a good correlation in size only for crystallites smaller than 50 – 60 nm. Highlights: • The crystallite sizes for 183 nanopowders were calculated using different XRD methods • Obtained results were subject to statistical treatment • Results obtained with Bragg-Brentano and parallel beam geometries were compared • Influence of conditions of XRD pattern acquisition on results was estimated • Calculated by XRD crystallite sizes were compared with same obtained by TEM and SEM.« less

Optimizing Integrated Terminal Airspace Operations Under Uncertainty

NASA Technical Reports Server (NTRS)

Bosson, Christabelle; Xue, Min; Zelinski, Shannon

2014-01-01

In the terminal airspace, integrated departures and arrivals have the potential to increase operations efficiency. Recent research has developed geneticalgorithm- based schedulers for integrated arrival and departure operations under uncertainty. This paper presents an alternate method using a machine jobshop scheduling formulation to model the integrated airspace operations. A multistage stochastic programming approach is chosen to formulate the problem and candidate solutions are obtained by solving sample average approximation problems with finite sample size. Because approximate solutions are computed, the proposed algorithm incorporates the computation of statistical bounds to estimate the optimality of the candidate solutions. A proof-ofconcept study is conducted on a baseline implementation of a simple problem considering a fleet mix of 14 aircraft evolving in a model of the Los Angeles terminal airspace. A more thorough statistical analysis is also performed to evaluate the impact of the number of scenarios considered in the sampled problem. To handle extensive sampling computations, a multithreading technique is introduced.

Geochemistry of sediments in the Northern and Central Adriatic Sea

NASA Astrophysics Data System (ADS)

De Lazzari, A.; Rampazzo, G.; Pavoni, B.

2004-03-01

Major, minor and trace elements, loss of ignition, specific surface area, quantities of calcite and dolomite, qualitative mineralogical composition, grain-size distribution and organic micropollutants (PAH, PCB, DDT) were determined on surficial marine sediments sampled during the 1990 ASCOP (Adriatic Scientific Cooperative Program) cruise. Mineralogical composition and carbonate content of the samples were found to be comparable with data previously reported in the literature, whereas geochemical composition and distribution of major, minor and trace elements for samples in international waters and in the central basin have never been reported before. The large amount of information contained in the variables of different origin has been processed by means of a comprehensive approach which establishes the relations among the components through the mathematical-statistical calculation of principal components (factors). These account for the major part of data variance loosing only marginal parts of information and are independent from the units of measure. The sample descriptors concerning natural components and contamination load are discussed by means of a statistical model based on an R-mode Factor analysis calculating four significant factors which explain 86.8% of the total variance, and represent important relationships between grain size, mineralogy, geochemistry and organic micropollutants. A description and an interpretation of factor composition is discussed on the basis of pollution inputs, basin geology and hydrodynamics. The areal distribution of the factors showed that it is the fine grain-size fraction, with oxides and hydroxides of colloidal origin, which are the main means of transport and thus the principal link between chemical, physical and granulometric elements in the Adriatic.

Imaging Extended Emission-Line Regions of Obscured AGN with the Subaru Hyper Suprime-Cam Survey

NASA Astrophysics Data System (ADS)

Sun, Ai-Lei; Greene, Jenny E.; Zakamska, Nadia L.; Goulding, Andy; Strauss, Michael A.; Huang, Song; Johnson, Sean; Kawaguchi, Toshihiro; Matsuoka, Yoshiki; Marsteller, Alisabeth A.; Nagao, Tohru; Toba, Yoshiki

2018-05-01

Narrow-line regions excited by active galactic nuclei (AGN) are important for studying AGN photoionization and feedback. Their strong [O III] lines can be detected with broadband images, allowing morphological studies of these systems with large-area imaging surveys. We develop a new broad-band imaging technique to reconstruct the images of the [O III] line using the Subaru Hyper Suprime-Cam (HSC) Survey aided with spectra from the Sloan Digital Sky Survey (SDSS). The technique involves a careful subtraction of the galactic continuum to isolate emission from the [O III]λ5007 and [O III]λ4959 lines. Compared to traditional targeted observations, this technique is more efficient at covering larger samples without dedicated observational resources. We apply this technique to an SDSS spectroscopically selected sample of 300 obscured AGN at redshifts 0.1 - 0.7, uncovering extended emission-line region candidates with sizes up to tens of kpc. With the largest sample of uniformly derived narrow-line region sizes, we revisit the narrow-line region size - luminosity relation. The area and radii of the [O III] emission-line regions are strongly correlated with the AGN luminosity inferred from the mid-infrared (15 μm rest-frame) with a power-law slope of 0.62^{+0.05}_{-0.06}± 0.10 (statistical and systematic errors), consistent with previous spectroscopic findings. We discuss the implications for the physics of AGN emission-line regions and future applications of this technique, which should be useful for current and next-generation imaging surveys to study AGN photoionization and feedback with large statistical samples.

The SDSS-IV MaNGA Sample: Design, Optimization, and Usage Considerations

NASA Astrophysics Data System (ADS)

Wake, David A.; Bundy, Kevin; Diamond-Stanic, Aleksandar M.; Yan, Renbin; Blanton, Michael R.; Bershady, Matthew A.; Sánchez-Gallego, José R.; Drory, Niv; Jones, Amy; Kauffmann, Guinevere; Law, David R.; Li, Cheng; MacDonald, Nicholas; Masters, Karen; Thomas, Daniel; Tinker, Jeremy; Weijmans, Anne-Marie; Brownstein, Joel R.

2017-09-01

We describe the sample design for the SDSS-IV MaNGA survey and present the final properties of the main samples along with important considerations for using these samples for science. Our target selection criteria were developed while simultaneously optimizing the size distribution of the MaNGA integral field units (IFUs), the IFU allocation strategy, and the target density to produce a survey defined in terms of maximizing signal-to-noise ratio, spatial resolution, and sample size. Our selection strategy makes use of redshift limits that only depend on I-band absolute magnitude (M I ), or, for a small subset of our sample, M I and color (NUV - I). Such a strategy ensures that all galaxies span the same range in angular size irrespective of luminosity and are therefore covered evenly by the adopted range of IFU sizes. We define three samples: the Primary and Secondary samples are selected to have a flat number density with respect to M I and are targeted to have spectroscopic coverage to 1.5 and 2.5 effective radii (R e ), respectively. The Color-Enhanced supplement increases the number of galaxies in the low-density regions of color-magnitude space by extending the redshift limits of the Primary sample in the appropriate color bins. The samples cover the stellar mass range 5× {10}8≤slant {M}* ≤slant 3× {10}11 {M}⊙ {h}-2 and are sampled at median physical resolutions of 1.37 and 2.5 kpc for the Primary and Secondary samples, respectively. We provide weights that will statistically correct for our luminosity and color-dependent selection function and IFU allocation strategy, thus correcting the observed sample to a volume-limited sample.

A novel measure of effect size for mediation analysis.

PubMed

Lachowicz, Mark J; Preacher, Kristopher J; Kelley, Ken

2018-06-01

Mediation analysis has become one of the most popular statistical methods in the social sciences. However, many currently available effect size measures for mediation have limitations that restrict their use to specific mediation models. In this article, we develop a measure of effect size that addresses these limitations. We show how modification of a currently existing effect size measure results in a novel effect size measure with many desirable properties. We also derive an expression for the bias of the sample estimator for the proposed effect size measure and propose an adjusted version of the estimator. We present a Monte Carlo simulation study conducted to examine the finite sampling properties of the adjusted and unadjusted estimators, which shows that the adjusted estimator is effective at recovering the true value it estimates. Finally, we demonstrate the use of the effect size measure with an empirical example. We provide freely available software so that researchers can immediately implement the methods we discuss. Our developments here extend the existing literature on effect sizes and mediation by developing a potentially useful method of communicating the magnitude of mediation. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Improving the analysis of composite endpoints in rare disease trials.

PubMed

McMenamin, Martina; Berglind, Anna; Wason, James M S

2018-05-22

Composite endpoints are recommended in rare diseases to increase power and/or to sufficiently capture complexity. Often, they are in the form of responder indices which contain a mixture of continuous and binary components. Analyses of these outcomes typically treat them as binary, thus only using the dichotomisations of continuous components. The augmented binary method offers a more efficient alternative and is therefore especially useful for rare diseases. Previous work has indicated the method may have poorer statistical properties when the sample size is small. Here we investigate small sample properties and implement small sample corrections. We re-sample from a previous trial with sample sizes varying from 30 to 80. We apply the standard binary and augmented binary methods and determine the power, type I error rate, coverage and average confidence interval width for each of the estimators. We implement Firth's adjustment for the binary component models and a small sample variance correction for the generalized estimating equations, applying the small sample adjusted methods to each sub-sample as before for comparison. For the log-odds treatment effect the power of the augmented binary method is 20-55% compared to 12-20% for the standard binary method. Both methods have approximately nominal type I error rates. The difference in response probabilities exhibit similar power but both unadjusted methods demonstrate type I error rates of 6-8%. The small sample corrected methods have approximately nominal type I error rates. On both scales, the reduction in average confidence interval width when using the adjusted augmented binary method is 17-18%. This is equivalent to requiring a 32% smaller sample size to achieve the same statistical power. The augmented binary method with small sample corrections provides a substantial improvement for rare disease trials using composite endpoints. We recommend the use of the method for the primary analysis in relevant rare disease trials. We emphasise that the method should be used alongside other efforts in improving the quality of evidence generated from rare disease trials rather than replace them.

Effects of Al(OH)O nanoparticle agglomerate size in epoxy resin on tension, bending, and fracture properties

NASA Astrophysics Data System (ADS)

Jux, Maximilian; Finke, Benedikt; Mahrholz, Thorsten; Sinapius, Michael; Kwade, Arno; Schilde, Carsten

2017-04-01

Several epoxy Al(OH)O (boehmite) dispersions in an epoxy resin are produced in a kneader to study the mechanistic correlation between the nanoparticle size and mechanical properties of the prepared nanocomposites. The agglomerate size is set by a targeted variation in solid content and temperature during dispersion, resulting in a different level of stress intensity and thus a different final agglomerate size during the process. The suspension viscosity was used for the estimation of stress energy in laminar shear flow. Agglomerate size measurements are executed via dynamic light scattering to ensure the quality of the produced dispersions. Furthermore, various nanocomposite samples are prepared for three-point bending, tension, and fracture toughness tests. The screening of the size effect is executed with at least seven samples per agglomerate size and test method. The variation of solid content is found to be a reliable method to adjust the agglomerate size between 138-354 nm during dispersion. The size effect on the Young's modulus and the critical stress intensity is only marginal. Nevertheless, there is a statistically relevant trend showing a linear increase with a decrease in agglomerate size. In contrast, the size effect is more dominant to the sample's strain and stress at failure. Unlike microscaled agglomerates or particles, which lead to embrittlement of the composite material, nanoscaled agglomerates or particles cause the composite elongation to be nearly of the same level as the base material. The observed effect is valid for agglomerate sizes between 138-354 nm and a particle mass fraction of 10 wt%.

The potential of composite cognitive scores for tracking progression in Huntington's disease.

PubMed

Jones, Rebecca; Stout, Julie C; Labuschagne, Izelle; Say, Miranda; Justo, Damian; Coleman, Allison; Dumas, Eve M; Hart, Ellen; Owen, Gail; Durr, Alexandra; Leavitt, Blair R; Roos, Raymund; O'Regan, Alison; Langbehn, Doug; Tabrizi, Sarah J; Frost, Chris

2014-01-01

Composite scores derived from joint statistical modelling of individual risk factors are widely used to identify individuals who are at increased risk of developing disease or of faster disease progression. We investigated the ability of composite measures developed using statistical models to differentiate progressive cognitive deterioration in Huntington's disease (HD) from natural decline in healthy controls. Using longitudinal data from TRACK-HD, the optimal combinations of quantitative cognitive measures to differentiate premanifest and early stage HD individuals respectively from controls was determined using logistic regression. Composite scores were calculated from the parameters of each statistical model. Linear regression models were used to calculate effect sizes (ES) quantifying the difference in longitudinal change over 24 months between premanifest and early stage HD groups respectively and controls. ES for the composites were compared with ES for individual cognitive outcomes and other measures used in HD research. The 0.632 bootstrap was used to eliminate biases which result from developing and testing models in the same sample. In early HD, the composite score from the HD change prediction model produced an ES for difference in rate of 24-month change relative to controls of 1.14 (95% CI: 0.90 to 1.39), larger than the ES for any individual cognitive outcome and UHDRS Total Motor Score and Total Functional Capacity. In addition, this composite gave a statistically significant difference in rate of change in premanifest HD compared to controls over 24-months (ES: 0.24; 95% CI: 0.04 to 0.44), even though none of the individual cognitive outcomes produced statistically significant ES over this period. Composite scores developed using appropriate statistical modelling techniques have the potential to materially reduce required sample sizes for randomised controlled trials.

Evaluation of Primary Immunization Coverage of Infants Under Universal Immunization Programme in an Urban Area of Bangalore City Using Cluster Sampling and Lot Quality Assurance Sampling Techniques

PubMed Central

K, Punith; K, Lalitha; G, Suman; BS, Pradeep; Kumar K, Jayanth

2008-01-01

Research Question: Is LQAS technique better than cluster sampling technique in terms of resources to evaluate the immunization coverage in an urban area? Objective: To assess and compare the lot quality assurance sampling against cluster sampling in the evaluation of primary immunization coverage. Study Design: Population-based cross-sectional study. Study Setting: Areas under Mathikere Urban Health Center. Study Subjects: Children aged 12 months to 23 months. Sample Size: 220 in cluster sampling, 76 in lot quality assurance sampling. Statistical Analysis: Percentages and Proportions, Chi square Test. Results: (1) Using cluster sampling, the percentage of completely immunized, partially immunized and unimmunized children were 84.09%, 14.09% and 1.82%, respectively. With lot quality assurance sampling, it was 92.11%, 6.58% and 1.31%, respectively. (2) Immunization coverage levels as evaluated by cluster sampling technique were not statistically different from the coverage value as obtained by lot quality assurance sampling techniques. Considering the time and resources required, it was found that lot quality assurance sampling is a better technique in evaluating the primary immunization coverage in urban area. PMID:19876474

Determining Sample Sizes for Precise Contrast Analysis with Heterogeneous Variances

ERIC Educational Resources Information Center

Jan, Show-Li; Shieh, Gwowen

2014-01-01

The analysis of variance (ANOVA) is one of the most frequently used statistical analyses in practical applications. Accordingly, the single and multiple comparison procedures are frequently applied to assess the differences among mean effects. However, the underlying assumption of homogeneous variances may not always be tenable. This study…

American Samoa's forest resources, 2001.

Treesearch

Joseph A. Donnegan; Sheri S. Mann; Sarah L. Butler; Bruce A. Hiserote

2004-01-01

The Forest Inventory and Analysis Program of the Pacific Northwest Research Station collected, analyzed, and summarized data from field plots, and mapped land cover on four islands in American Samoa. This statistical sample provides estimates of forest area, stem volume, biomass, numbers of trees, damages to trees, and tree size distribution. The summary provides...

Intraclass Correlation Values for Planning Group-Randomized Trials in Education

ERIC Educational Resources Information Center

Hedges, Larry V.; Hedberg, E. C.

2007-01-01

Experiments that assign intact groups to treatment conditions are increasingly common in social research. In educational research, the groups assigned are often schools. The design of group-randomized experiments requires knowledge of the intraclass correlation structure to compute statistical power and sample sizes required to achieve adequate…

A Comparison of Normal and Elliptical Estimation Methods in Structural Equation Models.

ERIC Educational Resources Information Center

Schumacker, Randall E.; Cheevatanarak, Suchittra

Monte Carlo simulation compared chi-square statistics, parameter estimates, and root mean square error of approximation values using normal and elliptical estimation methods. Three research conditions were imposed on the simulated data: sample size, population contamination percent, and kurtosis. A Bentler-Weeks structural model established the…

From the Knowledge of Understanding to Military Deception

DTIC Science & Technology

2008-05-21

and decision-making. Experimental research could lead to a validation of the theory. In the experiment, I tried to cause a decrease in ambiguity and...strong evidence that indicate a relationship. Further research with a larger sample size might show a significant statistical relationship. In order...2 Research question

Pearson-type goodness-of-fit test with bootstrap maximum likelihood estimation.

PubMed

Yin, Guosheng; Ma, Yanyuan

2013-01-01

The Pearson test statistic is constructed by partitioning the data into bins and computing the difference between the observed and expected counts in these bins. If the maximum likelihood estimator (MLE) of the original data is used, the statistic generally does not follow a chi-squared distribution or any explicit distribution. We propose a bootstrap-based modification of the Pearson test statistic to recover the chi-squared distribution. We compute the observed and expected counts in the partitioned bins by using the MLE obtained from a bootstrap sample. This bootstrap-sample MLE adjusts exactly the right amount of randomness to the test statistic, and recovers the chi-squared distribution. The bootstrap chi-squared test is easy to implement, as it only requires fitting exactly the same model to the bootstrap data to obtain the corresponding MLE, and then constructs the bin counts based on the original data. We examine the test size and power of the new model diagnostic procedure using simulation studies and illustrate it with a real data set.

Computed Tomography to Estimate the Representative Elementary Area for Soil Porosity Measurements

PubMed Central

Borges, Jaqueline Aparecida Ribaski; Pires, Luiz Fernando; Belmont Pereira, André

2012-01-01

Computed tomography (CT) is a technique that provides images of different solid and porous materials. CT could be an ideal tool to study representative sizes of soil samples because of the noninvasive characteristic of this technique. The scrutiny of such representative elementary sizes (RESs) has been the target of attention of many researchers related to soil physics field owing to the strong relationship between physical properties and size of the soil sample. In the current work, data from gamma-ray CT were used to assess RES in measurements of soil porosity (ϕ). For statistical analysis, a study on the full width at a half maximum (FWHM) of the adjustment of distribution of ϕ at different areas (1.2 to 1162.8 mm2) selected inside of tomographic images was proposed herein. The results obtained point out that samples with a section area corresponding to at least 882.1 mm2 were the ones that provided representative values of ϕ for the studied Brazilian tropical soil. PMID:22666133

A Review of ETS Differential Item Functioning Assessment Procedures: Flagging Rules, Minimum Sample Size Requirements, and Criterion Refinement. Research Report. ETS RR-12-08

ERIC Educational Resources Information Center

Zwick, Rebecca

2012-01-01

Differential item functioning (DIF) analysis is a key component in the evaluation of the fairness and validity of educational tests. The goal of this project was to review the status of ETS DIF analysis procedures, focusing on three aspects: (a) the nature and stringency of the statistical rules used to flag items, (b) the minimum sample size…

Correlating Sampling and Intensity Statistics in Nanoparticle Diffraction Experiments

DOE Office of Scientific and Technical Information (OSTI.GOV)

Ozturk, Hande; Yan, Hanfei; Hill, John P.

2015-08-01

In this article, [Öztürk, Yan, Hill & Noyan (2014). J. Appl. Cryst. 47, 1016-1025] it was shown that the sampling statistics of diffracting particle populations within a polycrystalline ensemble depended on the size of the constituent crystallites: broad X-ray peak breadths enabled some nano-sized particles to contribute more than one diffraction spot to Debye-Scherrer rings. Here it is shown that the equations proposed by Alexander, Klug & Kummer [J. Appl. Phys. (1948), 19, 742-753] (AKK) to link diffracting particle and diffracted intensity statistics are not applicable if the constituent crystallites of the powder are below 10 nm. In this sizemore » range, (i) the one-to-one correspondence between diffracting particles and Laue spots assumed in the AKK analysis is not satisfied, and (ii) the crystallographic correlation between Laue spots originating from the same grain invalidates the assumption that all diffracting plane normals are randomly oriented and uncorrelated. Such correlation produces unexpected results in the selection of diffracting grains. Three or more Laue spots from a given grain for a particular reflection can only be observed at certain wavelengths. In addition, correcting the diffracted intensity values by the traditional Lorentz term, 1/cos [theta], to compensate for the variation of particles sampled within a reflection band does not maintain fidelity to the number of poles contributing to the diffracted signal. A new term, cos [theta]B/cos [theta], corrects this problem.« less

Maturation of the Middle Phalanx of the Third Finger: A Comparative Study between Right and Left Hand.

PubMed

Gracco, Antonio; Bruno, Giovanni; De Stefani, Alberto; Siviero, Laura; Perri, Alessandro; Stellini, Edoardo

Recently a classification of patient's skeletal age based on the phalanx maturation, The Middle Phalanx Maturation of the third finger (MPM) method, was suggested. The aim of this study is to evaluate if there is a difference in MPM between the right and left hand. Two hundred fifty-four patients were obtained from the Complex Operating Unit of Orthodontics of Padua University Hospital. The total sample size has been selected by appropriate statistical calculations resulting in 130 patients. It was decided to further double the sample size of a previous study to ensure a robust statistical analysis. Radiographs of the right and left were obtained using the MPM method. Stages were compared using the right hand as a reference. The statistical analysis (Fisher exact test) was performed for the entire sample and related to gender in order to compare the right and the left hand stages. In MPS2, 6 out 49 (12.2%) males and 7 out 27 females (25.9%) showed MPS3 in the left hand (p-value < 0.05). In all other stages, a total agreement (100%) was found. The authors confirm the use of the right hand as reference. In patients with MPS2 an additional radiograph on the left hand can be taken in order to increase the diagnostic accuracy. In all other stages other radiographs are not needed as a total agreement between the right and left hand was found.

Supercritical Fluid Extraction and Analysis of Tropospheric Aerosol Particles

NASA Astrophysics Data System (ADS)

Hansen, Kristen J.

An integrated sampling and supercritical fluid extraction (SFE) cell has been designed for whole-sample analysis of organic compounds on tropospheric aerosol particles. The low-volume extraction cell has been interfaced with a sampling manifold for aerosol particle collection in the field. After sample collection, the entire SFE cell was coupled to a gas chromatograph; after on-line extraction, the cryogenically -focused sample was separated and the volatile compounds detected with either a mass spectrometer or a flame ionization detector. A 20-minute extraction at 450 atm and 90 ^circC with pure supercritical CO _2 is sufficient for quantitative extraction of most volatile compounds in aerosol particle samples. A comparison between SFE and thermal desorption, the traditional whole-sample technique for analyses of this type, was performed using ambient aerosol particle samples, as well as samples containing known amounts of standard analytes. The results of these studies indicate that SFE of atmospheric aerosol particles provides quantitative measurement of several classes of organic compounds. SFE provides information that is complementary to that gained by the thermal desorption analysis. The results also indicate that SFE with CO _2 can be validated as an alternative to thermal desorption for quantitative recovery of several organic compounds. In 1989, the organic constituents of atmospheric aerosol particles collected at Niwot Ridge, Colorado, along with various physical and meteorological data, were measured during a collaborative field study. Temporal changes in the composition of samples collected during summertime at the rural site were studied. Thermal desorption-GC/FID was used to quantify selected compounds in samples collected during the field study. The statistical analysis of the 1989 Niwot Ridge data set is presented in this work. Principal component analysis was performed on thirty-one variables selected from the data set in order to ascertain different source and process components, and to examine concentration changes in groups of variables with respect to time of day and meteorological conditions. Seven orthogonal groups of variables resulted from the statistical analysis; the groups serve as molecular markers for different biologic and anthropogenic emission sources. In addition, the results of the statistical analysis were used to investigate how several emission source contributions vary with respect to local atmospheric dynamics. Field studies were conducted in the urban environment in and around Boulder, CO. to characterize the dynamics, chemistry, and emission sources which affect the composition and concentration of different size-fractions of aerosol particles in the Boulder air mass. Relationships between different size fractions of particles and some gas-phase pollutants were elucidated. These field studies included an investigation of seasonal variations in the organic content and concentration of aerosol particles, and how these characteristics are related to local meteorology and to the concentration of some gas-phase pollutants. The elemental and organic composition of aerosol particles was investigated according to particle size in preliminary studies of size-differentiated samples of aerosol particles. In order to aid in future studies of urban aerosol particles, samples were collected at a forest fire near Boulder. Molecular markers specific to wood burning processes will be useful indicators of residential wood burning activities in future field studies.

A statistical model and national data set for partioning fish-tissue mercury concentration variation between spatiotemporal and sample characteristic effects

USGS Publications Warehouse

Wente, Stephen P.

2004-01-01

Many Federal, Tribal, State, and local agencies monitor mercury in fish-tissue samples to identify sites with elevated fish-tissue mercury (fish-mercury) concentrations, track changes in fish-mercury concentrations over time, and produce fish-consumption advisories. Interpretation of such monitoring data commonly is impeded by difficulties in separating the effects of sample characteristics (species, tissues sampled, and sizes of fish) from the effects of spatial and temporal trends on fish-mercury concentrations. Without such a separation, variation in fish-mercury concentrations due to differences in the characteristics of samples collected over time or across space can be misattributed to temporal or spatial trends; and/or actual trends in fish-mercury concentration can be misattributed to differences in sample characteristics. This report describes a statistical model and national data set (31,813 samples) for calibrating the aforementioned statistical model that can separate spatiotemporal and sample characteristic effects in fish-mercury concentration data. This model could be useful for evaluating spatial and temporal trends in fishmercury concentrations and developing fish-consumption advisories. The observed fish-mercury concentration data and model predictions can be accessed, displayed geospatially, and downloaded via the World Wide Web (http://emmma.usgs.gov). This report and the associated web site may assist in the interpretation of large amounts of data from widespread fishmercury monitoring efforts.

A comparison of confidence/credible interval methods for the area under the ROC curve for continuous diagnostic tests with small sample size.

PubMed

Feng, Dai; Cortese, Giuliana; Baumgartner, Richard

2017-12-01

The receiver operating characteristic (ROC) curve is frequently used as a measure of accuracy of continuous markers in diagnostic tests. The area under the ROC curve (AUC) is arguably the most widely used summary index for the ROC curve. Although the small sample size scenario is common in medical tests, a comprehensive study of small sample size properties of various methods for the construction of the confidence/credible interval (CI) for the AUC has been by and large missing in the literature. In this paper, we describe and compare 29 non-parametric and parametric methods for the construction of the CI for the AUC when the number of available observations is small. The methods considered include not only those that have been widely adopted, but also those that have been less frequently mentioned or, to our knowledge, never applied to the AUC context. To compare different methods, we carried out a simulation study with data generated from binormal models with equal and unequal variances and from exponential models with various parameters and with equal and unequal small sample sizes. We found that the larger the true AUC value and the smaller the sample size, the larger the discrepancy among the results of different approaches. When the model is correctly specified, the parametric approaches tend to outperform the non-parametric ones. Moreover, in the non-parametric domain, we found that a method based on the Mann-Whitney statistic is in general superior to the others. We further elucidate potential issues and provide possible solutions to along with general guidance on the CI construction for the AUC when the sample size is small. Finally, we illustrate the utility of different methods through real life examples.

Estimating sample size for landscape-scale mark-recapture studies of North American migratory tree bats

USGS Publications Warehouse

Ellison, Laura E.; Lukacs, Paul M.

2014-01-01

Concern for migratory tree-roosting bats in North America has grown because of possible population declines from wind energy development. This concern has driven interest in estimating population-level changes. Mark-recapture methodology is one possible analytical framework for assessing bat population changes, but sample size requirements to produce reliable estimates have not been estimated. To illustrate the sample sizes necessary for a mark-recapture-based monitoring program we conducted power analyses using a statistical model that allows reencounters of live and dead marked individuals. We ran 1,000 simulations for each of five broad sample size categories in a Burnham joint model, and then compared the proportion of simulations in which 95% confidence intervals overlapped between and among years for a 4-year study. Additionally, we conducted sensitivity analyses of sample size to various capture probabilities and recovery probabilities. More than 50,000 individuals per year would need to be captured and released to accurately determine 10% and 15% declines in annual survival. To detect more dramatic declines of 33% or 50% survival over four years, then sample sizes of 25,000 or 10,000 per year, respectively, would be sufficient. Sensitivity analyses reveal that increasing recovery of dead marked individuals may be more valuable than increasing capture probability of marked individuals. Because of the extraordinary effort that would be required, we advise caution should such a mark-recapture effort be initiated because of the difficulty in attaining reliable estimates. We make recommendations for what techniques show the most promise for mark-recapture studies of bats because some techniques violate the assumptions of mark-recapture methodology when used to mark bats.

Is psychology suffering from a replication crisis? What does "failure to replicate" really mean?

PubMed

Maxwell, Scott E; Lau, Michael Y; Howard, George S

2015-09-01

Psychology has recently been viewed as facing a replication crisis because efforts to replicate past study findings frequently do not show the same result. Often, the first study showed a statistically significant result but the replication does not. Questions then arise about whether the first study results were false positives, and whether the replication study correctly indicates that there is truly no effect after all. This article suggests these so-called failures to replicate may not be failures at all, but rather are the result of low statistical power in single replication studies, and the result of failure to appreciate the need for multiple replications in order to have enough power to identify true effects. We provide examples of these power problems and suggest some solutions using Bayesian statistics and meta-analysis. Although the need for multiple replication studies may frustrate those who would prefer quick answers to psychology's alleged crisis, the large sample sizes typically needed to provide firm evidence will almost always require concerted efforts from multiple investigators. As a result, it remains to be seen how many of the recently claimed failures to replicate will be supported or instead may turn out to be artifacts of inadequate sample sizes and single study replications. (PsycINFO Database Record (c) 2015 APA, all rights reserved).

Test Statistics and Confidence Intervals to Establish Noninferiority between Treatments with Ordinal Categorical Data.

PubMed

Zhang, Fanghong; Miyaoka, Etsuo; Huang, Fuping; Tanaka, Yutaka

2015-01-01

The problem for establishing noninferiority is discussed between a new treatment and a standard (control) treatment with ordinal categorical data. A measure of treatment effect is used and a method of specifying noninferiority margin for the measure is provided. Two Z-type test statistics are proposed where the estimation of variance is constructed under the shifted null hypothesis using U-statistics. Furthermore, the confidence interval and the sample size formula are given based on the proposed test statistics. The proposed procedure is applied to a dataset from a clinical trial. A simulation study is conducted to compare the performance of the proposed test statistics with that of the existing ones, and the results show that the proposed test statistics are better in terms of the deviation from nominal level and the power.

Statistical methods and errors in family medicine articles between 2010 and 2014-Suez Canal University, Egypt: A cross-sectional study.

PubMed

Nour-Eldein, Hebatallah

2016-01-01

With limited statistical knowledge of most physicians it is not uncommon to find statistical errors in research articles. To determine the statistical methods and to assess the statistical errors in family medicine (FM) research articles that were published between 2010 and 2014. This was a cross-sectional study. All 66 FM research articles that were published over 5 years by FM authors with affiliation to Suez Canal University were screened by the researcher between May and August 2015. Types and frequencies of statistical methods were reviewed in all 66 FM articles. All 60 articles with identified inferential statistics were examined for statistical errors and deficiencies. A comprehensive 58-item checklist based on statistical guidelines was used to evaluate the statistical quality of FM articles. Inferential methods were recorded in 62/66 (93.9%) of FM articles. Advanced analyses were used in 29/66 (43.9%). Contingency tables 38/66 (57.6%), regression (logistic, linear) 26/66 (39.4%), and t-test 17/66 (25.8%) were the most commonly used inferential tests. Within 60 FM articles with identified inferential statistics, no prior sample size 19/60 (31.7%), application of wrong statistical tests 17/60 (28.3%), incomplete documentation of statistics 59/60 (98.3%), reporting P value without test statistics 32/60 (53.3%), no reporting confidence interval with effect size measures 12/60 (20.0%), use of mean (standard deviation) to describe ordinal/nonnormal data 8/60 (13.3%), and errors related to interpretation were mainly for conclusions without support by the study data 5/60 (8.3%). Inferential statistics were used in the majority of FM articles. Data analysis and reporting statistics are areas for improvement in FM research articles.

Statistical methods and errors in family medicine articles between 2010 and 2014-Suez Canal University, Egypt: A cross-sectional study

PubMed Central

Nour-Eldein, Hebatallah

2016-01-01

Background: With limited statistical knowledge of most physicians it is not uncommon to find statistical errors in research articles. Objectives: To determine the statistical methods and to assess the statistical errors in family medicine (FM) research articles that were published between 2010 and 2014. Methods: This was a cross-sectional study. All 66 FM research articles that were published over 5 years by FM authors with affiliation to Suez Canal University were screened by the researcher between May and August 2015. Types and frequencies of statistical methods were reviewed in all 66 FM articles. All 60 articles with identified inferential statistics were examined for statistical errors and deficiencies. A comprehensive 58-item checklist based on statistical guidelines was used to evaluate the statistical quality of FM articles. Results: Inferential methods were recorded in 62/66 (93.9%) of FM articles. Advanced analyses were used in 29/66 (43.9%). Contingency tables 38/66 (57.6%), regression (logistic, linear) 26/66 (39.4%), and t-test 17/66 (25.8%) were the most commonly used inferential tests. Within 60 FM articles with identified inferential statistics, no prior sample size 19/60 (31.7%), application of wrong statistical tests 17/60 (28.3%), incomplete documentation of statistics 59/60 (98.3%), reporting P value without test statistics 32/60 (53.3%), no reporting confidence interval with effect size measures 12/60 (20.0%), use of mean (standard deviation) to describe ordinal/nonnormal data 8/60 (13.3%), and errors related to interpretation were mainly for conclusions without support by the study data 5/60 (8.3%). Conclusion: Inferential statistics were used in the majority of FM articles. Data analysis and reporting statistics are areas for improvement in FM research articles. PMID:27453839

Efficient computation of the joint sample frequency spectra for multiple populations.

PubMed

Kamm, John A; Terhorst, Jonathan; Song, Yun S

2017-01-01

A wide range of studies in population genetics have employed the sample frequency spectrum (SFS), a summary statistic which describes the distribution of mutant alleles at a polymorphic site in a sample of DNA sequences and provides a highly efficient dimensional reduction of large-scale population genomic variation data. Recently, there has been much interest in analyzing the joint SFS data from multiple populations to infer parameters of complex demographic histories, including variable population sizes, population split times, migration rates, admixture proportions, and so on. SFS-based inference methods require accurate computation of the expected SFS under a given demographic model. Although much methodological progress has been made, existing methods suffer from numerical instability and high computational complexity when multiple populations are involved and the sample size is large. In this paper, we present new analytic formulas and algorithms that enable accurate, efficient computation of the expected joint SFS for thousands of individuals sampled from hundreds of populations related by a complex demographic model with arbitrary population size histories (including piecewise-exponential growth). Our results are implemented in a new software package called momi (MOran Models for Inference). Through an empirical study we demonstrate our improvements to numerical stability and computational complexity.

Efficient computation of the joint sample frequency spectra for multiple populations

PubMed Central

Kamm, John A.; Terhorst, Jonathan; Song, Yun S.

2016-01-01

A wide range of studies in population genetics have employed the sample frequency spectrum (SFS), a summary statistic which describes the distribution of mutant alleles at a polymorphic site in a sample of DNA sequences and provides a highly efficient dimensional reduction of large-scale population genomic variation data. Recently, there has been much interest in analyzing the joint SFS data from multiple populations to infer parameters of complex demographic histories, including variable population sizes, population split times, migration rates, admixture proportions, and so on. SFS-based inference methods require accurate computation of the expected SFS under a given demographic model. Although much methodological progress has been made, existing methods suffer from numerical instability and high computational complexity when multiple populations are involved and the sample size is large. In this paper, we present new analytic formulas and algorithms that enable accurate, efficient computation of the expected joint SFS for thousands of individuals sampled from hundreds of populations related by a complex demographic model with arbitrary population size histories (including piecewise-exponential growth). Our results are implemented in a new software package called momi (MOran Models for Inference). Through an empirical study we demonstrate our improvements to numerical stability and computational complexity. PMID:28239248

A Guideline to Univariate Statistical Analysis for LC/MS-Based Untargeted Metabolomics-Derived Data

PubMed Central

Vinaixa, Maria; Samino, Sara; Saez, Isabel; Duran, Jordi; Guinovart, Joan J.; Yanes, Oscar

2012-01-01

Several metabolomic software programs provide methods for peak picking, retention time alignment and quantification of metabolite features in LC/MS-based metabolomics. Statistical analysis, however, is needed in order to discover those features significantly altered between samples. By comparing the retention time and MS/MS data of a model compound to that from the altered feature of interest in the research sample, metabolites can be then unequivocally identified. This paper reports on a comprehensive overview of a workflow for statistical analysis to rank relevant metabolite features that will be selected for further MS/MS experiments. We focus on univariate data analysis applied in parallel on all detected features. Characteristics and challenges of this analysis are discussed and illustrated using four different real LC/MS untargeted metabolomic datasets. We demonstrate the influence of considering or violating mathematical assumptions on which univariate statistical test rely, using high-dimensional LC/MS datasets. Issues in data analysis such as determination of sample size, analytical variation, assumption of normality and homocedasticity, or correction for multiple testing are discussed and illustrated in the context of our four untargeted LC/MS working examples. PMID:24957762

A Guideline to Univariate Statistical Analysis for LC/MS-Based Untargeted Metabolomics-Derived Data.

PubMed

Vinaixa, Maria; Samino, Sara; Saez, Isabel; Duran, Jordi; Guinovart, Joan J; Yanes, Oscar

2012-10-18

Several metabolomic software programs provide methods for peak picking, retention time alignment and quantification of metabolite features in LC/MS-based metabolomics. Statistical analysis, however, is needed in order to discover those features significantly altered between samples. By comparing the retention time and MS/MS data of a model compound to that from the altered feature of interest in the research sample, metabolites can be then unequivocally identified. This paper reports on a comprehensive overview of a workflow for statistical analysis to rank relevant metabolite features that will be selected for further MS/MS experiments. We focus on univariate data analysis applied in parallel on all detected features. Characteristics and challenges of this analysis are discussed and illustrated using four different real LC/MS untargeted metabolomic datasets. We demonstrate the influence of considering or violating mathematical assumptions on which univariate statistical test rely, using high-dimensional LC/MS datasets. Issues in data analysis such as determination of sample size, analytical variation, assumption of normality and homocedasticity, or correction for multiple testing are discussed and illustrated in the context of our four untargeted LC/MS working examples.

A more powerful test based on ratio distribution for retention noninferiority hypothesis.

PubMed

Deng, Ling; Chen, Gang

2013-03-11

Rothmann et al. ( 2003 ) proposed a method for the statistical inference of fraction retention noninferiority (NI) hypothesis. A fraction retention hypothesis is defined as a ratio of the new treatment effect verse the control effect in the context of a time to event endpoint. One of the major concerns using this method in the design of an NI trial is that with a limited sample size, the power of the study is usually very low. This makes an NI trial not applicable particularly when using time to event endpoint. To improve power, Wang et al. ( 2006 ) proposed a ratio test based on asymptotic normality theory. Under a strong assumption (equal variance of the NI test statistic under null and alternative hypotheses), the sample size using Wang's test was much smaller than that using Rothmann's test. However, in practice, the assumption of equal variance is generally questionable for an NI trial design. This assumption is removed in the ratio test proposed in this article, which is derived directly from a Cauchy-like ratio distribution. In addition, using this method, the fundamental assumption used in Rothmann's test, that the observed control effect is always positive, that is, the observed hazard ratio for placebo over the control is greater than 1, is no longer necessary. Without assuming equal variance under null and alternative hypotheses, the sample size required for an NI trial can be significantly reduced if using the proposed ratio test for a fraction retention NI hypothesis.

Sub-sampling genetic data to estimate black bear population size: A case study

USGS Publications Warehouse

Tredick, C.A.; Vaughan, M.R.; Stauffer, D.F.; Simek, S.L.; Eason, T.

2007-01-01

Costs for genetic analysis of hair samples collected for individual identification of bears average approximately US$50 [2004] per sample. This can easily exceed budgetary allowances for large-scale studies or studies of high-density bear populations. We used 2 genetic datasets from 2 areas in the southeastern United States to explore how reducing costs of analysis by sub-sampling affected precision and accuracy of resulting population estimates. We used several sub-sampling scenarios to create subsets of the full datasets and compared summary statistics, population estimates, and precision of estimates generated from these subsets to estimates generated from the complete datasets. Our results suggested that bias and precision of estimates improved as the proportion of total samples used increased, and heterogeneity models (e.g., Mh[CHAO]) were more robust to reduced sample sizes than other models (e.g., behavior models). We recommend that only high-quality samples (>5 hair follicles) be used when budgets are constrained, and efforts should be made to maximize capture and recapture rates in the field.

Using variance components to estimate power in a hierarchically nested sampling design improving monitoring of larval Devils Hole pupfish

USGS Publications Warehouse

Dzul, Maria C.; Dixon, Philip M.; Quist, Michael C.; Dinsomore, Stephen J.; Bower, Michael R.; Wilson, Kevin P.; Gaines, D. Bailey

2013-01-01

We used variance components to assess allocation of sampling effort in a hierarchically nested sampling design for ongoing monitoring of early life history stages of the federally endangered Devils Hole pupfish (DHP) (Cyprinodon diabolis). Sampling design for larval DHP included surveys (5 days each spring 2007–2009), events, and plots. Each survey was comprised of three counting events, where DHP larvae on nine plots were counted plot by plot. Statistical analysis of larval abundance included three components: (1) evaluation of power from various sample size combinations, (2) comparison of power in fixed and random plot designs, and (3) assessment of yearly differences in the power of the survey. Results indicated that increasing the sample size at the lowest level of sampling represented the most realistic option to increase the survey's power, fixed plot designs had greater power than random plot designs, and the power of the larval survey varied by year. This study provides an example of how monitoring efforts may benefit from coupling variance components estimation with power analysis to assess sampling design.

Statistical power comparisons at 3T and 7T with a GO / NOGO task.

PubMed

Torrisi, Salvatore; Chen, Gang; Glen, Daniel; Bandettini, Peter A; Baker, Chris I; Reynolds, Richard; Yen-Ting Liu, Jeffrey; Leshin, Joseph; Balderston, Nicholas; Grillon, Christian; Ernst, Monique

2018-07-15

The field of cognitive neuroscience is weighing evidence about whether to move from standard field strength to ultra-high field (UHF). The present study contributes to the evidence by comparing a cognitive neuroscience paradigm at 3 Tesla (3T) and 7 Tesla (7T). The goal was to test and demonstrate the practical effects of field strength on a standard GO/NOGO task using accessible preprocessing and analysis tools. Two independent matched healthy samples (N = 31 each) were analyzed at 3T and 7T. Results show gains at 7T in statistical strength, the detection of smaller effects and group-level power. With an increased availability of UHF scanners, these gains may be exploited by cognitive neuroscientists and other neuroimaging researchers to develop more efficient or comprehensive experimental designs and, given the same sample size, achieve greater statistical power at 7T. Published by Elsevier Inc.

Statistical methods for efficient design of community surveys of response to noise: Random coefficients regression models

NASA Technical Reports Server (NTRS)

Tomberlin, T. J.

1985-01-01

Research studies of residents' responses to noise consist of interviews with samples of individuals who are drawn from a number of different compact study areas. The statistical techniques developed provide a basis for those sample design decisions. These techniques are suitable for a wide range of sample survey applications. A sample may consist of a random sample of residents selected from a sample of compact study areas, or in a more complex design, of a sample of residents selected from a sample of larger areas (e.g., cities). The techniques may be applied to estimates of the effects on annoyance of noise level, numbers of noise events, the time-of-day of the events, ambient noise levels, or other factors. Methods are provided for determining, in advance, how accurately these effects can be estimated for different sample sizes and study designs. Using a simple cost function, they also provide for optimum allocation of the sample across the stages of the design for estimating these effects. These techniques are developed via a regression model in which the regression coefficients are assumed to be random, with components of variance associated with the various stages of a multi-stage sample design.

Testing the equivalence of modern human cranial covariance structure: Implications for bioarchaeological applications.

PubMed

von Cramon-Taubadel, Noreen; Schroeder, Lauren

2016-10-01

Estimation of the variance-covariance (V/CV) structure of fragmentary bioarchaeological populations requires the use of proxy extant V/CV parameters. However, it is currently unclear whether extant human populations exhibit equivalent V/CV structures. Random skewers (RS) and hierarchical analyses of common principal components (CPC) were applied to a modern human cranial dataset. Cranial V/CV similarity was assessed globally for samples of individual populations (jackknifed method) and for pairwise population sample contrasts. The results were examined in light of potential explanatory factors for covariance difference, such as geographic region, among-group distance, and sample size. RS analyses showed that population samples exhibited highly correlated multivariate responses to selection, and that differences in RS results were primarily a consequence of differences in sample size. The CPC method yielded mixed results, depending upon the statistical criterion used to evaluate the hierarchy. The hypothesis-testing (step-up) approach was deemed problematic due to sensitivity to low statistical power and elevated Type I errors. In contrast, the model-fitting (lowest AIC) approach suggested that V/CV matrices were proportional and/or shared a large number of CPCs. Pairwise population sample CPC results were correlated with cranial distance, suggesting that population history explains some of the variability in V/CV structure among groups. The results indicate that patterns of covariance in human craniometric samples are broadly similar but not identical. These findings have important implications for choosing extant covariance matrices to use as proxy V/CV parameters in evolutionary analyses of past populations. © 2016 Wiley Periodicals, Inc.

Improved Statistical Sampling and Accuracy with Accelerated Molecular Dynamics on Rotatable Torsions.

PubMed

Doshi, Urmi; Hamelberg, Donald

2012-11-13

In enhanced sampling techniques, the precision of the reweighted ensemble properties is often decreased due to large variation in statistical weights and reduction in the effective sampling size. To abate this reweighting problem, here, we propose a general accelerated molecular dynamics (aMD) approach in which only the rotatable dihedrals are subjected to aMD (RaMD), unlike the typical implementation wherein all dihedrals are boosted (all-aMD). Nonrotatable and improper dihedrals are marginally important to conformational changes or the different rotameric states. Not accelerating them avoids the sharp increases in the potential energies due to small deviations from their minimum energy conformations and leads to improvement in the precision of RaMD. We present benchmark studies on two model dipeptides, Ace-Ala-Nme and Ace-Trp-Nme, simulated with normal MD, all-aMD, and RaMD. We carry out a systematic comparison between the performances of both forms of aMD using a theory that allows quantitative estimation of the effective number of sampled points and the associated uncertainty. Our results indicate that, for the same level of acceleration and simulation length, as used in all-aMD, RaMD results in significantly less loss in the effective sample size and, hence, increased accuracy in the sampling of φ-ψ space. RaMD yields an accuracy comparable to that of all-aMD, from simulation lengths 5 to 1000 times shorter, depending on the peptide and the acceleration level. Such improvement in speed and accuracy over all-aMD is highly remarkable, suggesting RaMD as a promising method for sampling larger biomolecules.

Discovering human germ cell mutagens with whole genome sequencing: Insights from power calculations reveal the importance of controlling for between-family variability.

PubMed

Webster, R J; Williams, A; Marchetti, F; Yauk, C L

2018-07-01

Mutations in germ cells pose potential genetic risks to offspring. However, de novo mutations are rare events that are spread across the genome and are difficult to detect. Thus, studies in this area have generally been under-powered, and no human germ cell mutagen has been identified. Whole Genome Sequencing (WGS) of human pedigrees has been proposed as an approach to overcome these technical and statistical challenges. WGS enables analysis of a much wider breadth of the genome than traditional approaches. Here, we performed power analyses to determine the feasibility of using WGS in human families to identify germ cell mutagens. Different statistical models were compared in the power analyses (ANOVA and multiple regression for one-child families, and mixed effect model sampling between two to four siblings per family). Assumptions were made based on parameters from the existing literature, such as the mutation-by-paternal age effect. We explored two scenarios: a constant effect due to an exposure that occurred in the past, and an accumulating effect where the exposure is continuing. Our analysis revealed the importance of modeling inter-family variability of the mutation-by-paternal age effect. Statistical power was improved by models accounting for the family-to-family variability. Our power analyses suggest that sufficient statistical power can be attained with 4-28 four-sibling families per treatment group, when the increase in mutations ranges from 40 to 10% respectively. Modeling family variability using mixed effect models provided a reduction in sample size compared to a multiple regression approach. Much larger sample sizes were required to detect an interaction effect between environmental exposures and paternal age. These findings inform study design and statistical modeling approaches to improve power and reduce sequencing costs for future studies in this area. Crown Copyright © 2018. Published by Elsevier B.V. All rights reserved.

On the assessment of the added value of new predictive biomarkers.

PubMed

Chen, Weijie; Samuelson, Frank W; Gallas, Brandon D; Kang, Le; Sahiner, Berkman; Petrick, Nicholas

2013-07-29

The surge in biomarker development calls for research on statistical evaluation methodology to rigorously assess emerging biomarkers and classification models. Recently, several authors reported the puzzling observation that, in assessing the added value of new biomarkers to existing ones in a logistic regression model, statistical significance of new predictor variables does not necessarily translate into a statistically significant increase in the area under the ROC curve (AUC). Vickers et al. concluded that this inconsistency is because AUC "has vastly inferior statistical properties," i.e., it is extremely conservative. This statement is based on simulations that misuse the DeLong et al. method. Our purpose is to provide a fair comparison of the likelihood ratio (LR) test and the Wald test versus diagnostic accuracy (AUC) tests. We present a test to compare ideal AUCs of nested linear discriminant functions via an F test. We compare it with the LR test and the Wald test for the logistic regression model. The null hypotheses of these three tests are equivalent; however, the F test is an exact test whereas the LR test and the Wald test are asymptotic tests. Our simulation shows that the F test has the nominal type I error even with a small sample size. Our results also indicate that the LR test and the Wald test have inflated type I errors when the sample size is small, while the type I error converges to the nominal value asymptotically with increasing sample size as expected. We further show that the DeLong et al. method tests a different hypothesis and has the nominal type I error when it is used within its designed scope. Finally, we summarize the pros and cons of all four methods we consider in this paper. We show that there is nothing inherently less powerful or disagreeable about ROC analysis for showing the usefulness of new biomarkers or characterizing the performance of classification models. Each statistical method for assessing biomarkers and classification models has its own strengths and weaknesses. Investigators need to choose methods based on the assessment purpose, the biomarker development phase at which the assessment is being performed, the available patient data, and the validity of assumptions behind the methodologies.

Raindrop Size Distribution in Different Climatic Regimes from Disdrometer and Dual-Polarized Radar Analysis.

NASA Astrophysics Data System (ADS)

Bringi, V. N.; Chandrasekar, V.; Hubbert, J.; Gorgucci, E.; Randeu, W. L.; Schoenhuber, M.

2003-01-01

The application of polarimetric radar data to the retrieval of raindrop size distribution parameters and rain rate in samples of convective and stratiform rain types is presented. Data from the Colorado State University (CSU), CHILL, NCAR S-band polarimetric (S-Pol), and NASA Kwajalein radars are analyzed for the statistics and functional relation of these parameters with rain rate. Surface drop size distribution measurements using two different disdrometers (2D video and RD-69) from a number of climatic regimes are analyzed and compared with the radar retrievals in a statistical and functional approach. The composite statistics based on disdrometer and radar retrievals suggest that, on average, the two parameters (generalized intercept and median volume diameter) for stratiform rain distributions lie on a straight line with negative slope, which appears to be consistent with variations in the microphysics of stratiform precipitation (melting of larger, dry snow particles versus smaller, rimed ice particles). In convective rain, `maritime-like' and `continental-like' clusters could be identified in the same two-parameter space that are consistent with the different multiplicative coefficients in the Z = aR1.5 relations quoted in the literature for maritime and continental regimes.

Sampling through time and phylodynamic inference with coalescent and birth–death models

PubMed Central

Volz, Erik M.; Frost, Simon D. W.

2014-01-01

Many population genetic models have been developed for the purpose of inferring population size and growth rates from random samples of genetic data. We examine two popular approaches to this problem, the coalescent and the birth–death-sampling model (BDM), in the context of estimating population size and birth rates in a population growing exponentially according to the birth–death branching process. For sequences sampled at a single time, we found the coalescent and the BDM gave virtually indistinguishable results in terms of the growth rates and fraction of the population sampled, even when sampling from a small population. For sequences sampled at multiple time points, we find that the birth–death model estimators are subject to large bias if the sampling process is misspecified. Since BDMs incorporate a model of the sampling process, we show how much of the statistical power of BDMs arises from the sequence of sample times and not from the genealogical tree. This motivates the development of a new coalescent estimator, which is augmented with a model of the known sampling process and is potentially more precise than the coalescent that does not use sample time information. PMID:25401173

Multi-Reader ROC studies with Split-Plot Designs: A Comparison of Statistical Methods

PubMed Central

Obuchowski, Nancy A.; Gallas, Brandon D.; Hillis, Stephen L.

2012-01-01

Rationale and Objectives Multi-reader imaging trials often use a factorial design, where study patients undergo testing with all imaging modalities and readers interpret the results of all tests for all patients. A drawback of the design is the large number of interpretations required of each reader. Split-plot designs have been proposed as an alternative, in which one or a subset of readers interprets all images of a sample of patients, while other readers interpret the images of other samples of patients. In this paper we compare three methods of analysis for the split-plot design. Materials and Methods Three statistical methods are presented: Obuchowski-Rockette method modified for the split-plot design, a newly proposed marginal-mean ANOVA approach, and an extension of the three-sample U-statistic method. A simulation study using the Roe-Metz model was performed to compare the type I error rate, power and confidence interval coverage of the three test statistics. Results The type I error rates for all three methods are close to the nominal level but tend to be slightly conservative. The statistical power is nearly identical for the three methods. The coverage of 95% CIs fall close to the nominal coverage for small and large sample sizes. Conclusions The split-plot MRMC study design can be statistically efficient compared with the factorial design, reducing the number of interpretations required per reader. Three methods of analysis, shown to have nominal type I error rate, similar power, and nominal CI coverage, are available for this study design. PMID:23122570

Multi-reader ROC studies with split-plot designs: a comparison of statistical methods.

PubMed

Obuchowski, Nancy A; Gallas, Brandon D; Hillis, Stephen L

2012-12-01

Multireader imaging trials often use a factorial design, in which study patients undergo testing with all imaging modalities and readers interpret the results of all tests for all patients. A drawback of this design is the large number of interpretations required of each reader. Split-plot designs have been proposed as an alternative, in which one or a subset of readers interprets all images of a sample of patients, while other readers interpret the images of other samples of patients. In this paper, the authors compare three methods of analysis for the split-plot design. Three statistical methods are presented: the Obuchowski-Rockette method modified for the split-plot design, a newly proposed marginal-mean analysis-of-variance approach, and an extension of the three-sample U-statistic method. A simulation study using the Roe-Metz model was performed to compare the type I error rate, power, and confidence interval coverage of the three test statistics. The type I error rates for all three methods are close to the nominal level but tend to be slightly conservative. The statistical power is nearly identical for the three methods. The coverage of 95% confidence intervals falls close to the nominal coverage for small and large sample sizes. The split-plot multireader, multicase study design can be statistically efficient compared to the factorial design, reducing the number of interpretations required per reader. Three methods of analysis, shown to have nominal type I error rates, similar power, and nominal confidence interval coverage, are available for this study design. Copyright © 2012 AUR. All rights reserved.

Test Population Selection from Weibull-Based, Monte Carlo Simulations of Fatigue Life

NASA Technical Reports Server (NTRS)

Vlcek, Brian L.; Zaretsky, Erwin V.; Hendricks, Robert C.

2008-01-01

Fatigue life is probabilistic and not deterministic. Experimentally establishing the fatigue life of materials, components, and systems is both time consuming and costly. As a result, conclusions regarding fatigue life are often inferred from a statistically insufficient number of physical tests. A proposed methodology for comparing life results as a function of variability due to Weibull parameters, variability between successive trials, and variability due to size of the experimental population is presented. Using Monte Carlo simulation of randomly selected lives from a large Weibull distribution, the variation in the L10 fatigue life of aluminum alloy AL6061 rotating rod fatigue tests was determined as a function of population size. These results were compared to the L10 fatigue lives of small (10 each) populations from AL2024, AL7075 and AL6061. For aluminum alloy AL6061, a simple algebraic relationship was established for the upper and lower L10 fatigue life limits as a function of the number of specimens failed. For most engineering applications where less than 30 percent variability can be tolerated in the maximum and minimum values, at least 30 to 35 test samples are necessary. The variability of test results based on small sample sizes can be greater than actual differences, if any, that exists between materials and can result in erroneous conclusions. The fatigue life of AL2024 is statistically longer than AL6061 and AL7075. However, there is no statistical difference between the fatigue lives of AL6061 and AL7075 even though AL7075 had a fatigue life 30 percent greater than AL6061.

Test Population Selection from Weibull-Based, Monte Carlo Simulations of Fatigue Life

NASA Technical Reports Server (NTRS)

Vlcek, Brian L.; Zaretsky, Erwin V.; Hendricks, Robert C.

2012-01-01

Fatigue life is probabilistic and not deterministic. Experimentally establishing the fatigue life of materials, components, and systems is both time consuming and costly. As a result, conclusions regarding fatigue life are often inferred from a statistically insufficient number of physical tests. A proposed methodology for comparing life results as a function of variability due to Weibull parameters, variability between successive trials, and variability due to size of the experimental population is presented. Using Monte Carlo simulation of randomly selected lives from a large Weibull distribution, the variation in the L10 fatigue life of aluminum alloy AL6061 rotating rod fatigue tests was determined as a function of population size. These results were compared to the L10 fatigue lives of small (10 each) populations from AL2024, AL7075 and AL6061. For aluminum alloy AL6061, a simple algebraic relationship was established for the upper and lower L10 fatigue life limits as a function of the number of specimens failed. For most engineering applications where less than 30 percent variability can be tolerated in the maximum and minimum values, at least 30 to 35 test samples are necessary. The variability of test results based on small sample sizes can be greater than actual differences, if any, that exists between materials and can result in erroneous conclusions. The fatigue life of AL2024 is statistically longer than AL6061 and AL7075. However, there is no statistical difference between the fatigue lives of AL6061 and AL7075 even though AL7075 had a fatigue life 30 percent greater than AL6061.

Potential Reporting Bias in Neuroimaging Studies of Sex Differences.

PubMed

David, Sean P; Naudet, Florian; Laude, Jennifer; Radua, Joaquim; Fusar-Poli, Paolo; Chu, Isabella; Stefanick, Marcia L; Ioannidis, John P A

2018-04-17

Numerous functional magnetic resonance imaging (fMRI) studies have reported sex differences. To empirically evaluate for evidence of excessive significance bias in this literature, we searched for published fMRI studies of human brain to evaluate sex differences, regardless of the topic investigated, in Medline and Scopus over 10 years. We analyzed the prevalence of conclusions in favor of sex differences and the correlation between study sample sizes and number of significant foci identified. In the absence of bias, larger studies (better powered) should identify a larger number of significant foci. Across 179 papers, median sample size was n = 32 (interquartile range 23-47.5). A median of 5 foci related to sex differences were reported (interquartile range, 2-9.5). Few articles (n = 2) had titles focused on no differences or on similarities (n = 3) between sexes. Overall, 158 papers (88%) reached "positive" conclusions in their abstract and presented some foci related to sex differences. There was no statistically significant relationship between sample size and the number of foci (-0.048% increase for every 10 participants, p = 0.63). The extremely high prevalence of "positive" results and the lack of the expected relationship between sample size and the number of discovered foci reflect probable reporting bias and excess significance bias in this literature.

Sample size determinations for group-based randomized clinical trials with different levels of data hierarchy between experimental and control arms.

PubMed

Heo, Moonseong; Litwin, Alain H; Blackstock, Oni; Kim, Namhee; Arnsten, Julia H

2017-02-01

We derived sample size formulae for detecting main effects in group-based randomized clinical trials with different levels of data hierarchy between experimental and control arms. Such designs are necessary when experimental interventions need to be administered to groups of subjects whereas control conditions need to be administered to individual subjects. This type of trial, often referred to as a partially nested or partially clustered design, has been implemented for management of chronic diseases such as diabetes and is beginning to emerge more commonly in wider clinical settings. Depending on the research setting, the level of hierarchy of data structure for the experimental arm can be three or two, whereas that for the control arm is two or one. Such different levels of data hierarchy assume correlation structures of outcomes that are different between arms, regardless of whether research settings require two or three level data structure for the experimental arm. Therefore, the different correlations should be taken into account for statistical modeling and for sample size determinations. To this end, we considered mixed-effects linear models with different correlation structures between experimental and control arms to theoretically derive and empirically validate the sample size formulae with simulation studies.

Effect of field view size and lighting on unique-hue selection using Natural Color System object colors.

PubMed

Shamey, Renzo; Zubair, Muhammad; Cheema, Hammad

2015-08-01

The aim of this study was twofold, first to determine the effect of field view size and second of illumination conditions on the selection of unique hue samples (UHs: R, Y, G and B) from two rotatable trays, each containing forty highly chromatic Natural Color System (NCS) samples, on one tray corresponding to 1.4° and on the other to 5.7° field of view size. UH selections were made by 25 color-normal observers who repeated assessments three times with a gap of at least 24h between trials. Observers separately assessed UHs under four illumination conditions simulating illuminants D65, A, F2 and F11. An apparent hue shift (statistically significant for UR) was noted for UH selections at 5.7° field of view compared to those at 1.4°. Observers' overall variability was found to be higher for UH stimuli selections at the larger field of view. Intra-observer variability was found to be approximately 18.7% of inter-observer variability in selection of samples for both sample sizes. The highest intra-observer variability was under simulated illuminant D65, followed by A, F11, and F2. Copyright © 2015 Elsevier Ltd. All rights reserved.

Likert scales, levels of measurement and the "laws" of statistics.

PubMed

Norman, Geoff

2010-12-01

Reviewers of research reports frequently criticize the choice of statistical methods. While some of these criticisms are well-founded, frequently the use of various parametric methods such as analysis of variance, regression, correlation are faulted because: (a) the sample size is too small, (b) the data may not be normally distributed, or (c) The data are from Likert scales, which are ordinal, so parametric statistics cannot be used. In this paper, I dissect these arguments, and show that many studies, dating back to the 1930s consistently show that parametric statistics are robust with respect to violations of these assumptions. Hence, challenges like those above are unfounded, and parametric methods can be utilized without concern for "getting the wrong answer".

Statistical methods for conducting agreement (comparison of clinical tests) and precision (repeatability or reproducibility) studies in optometry and ophthalmology.

PubMed

McAlinden, Colm; Khadka, Jyoti; Pesudovs, Konrad

2011-07-01

The ever-expanding choice of ocular metrology and imaging equipment has driven research into the validity of their measurements. Consequently, studies of the agreement between two instruments or clinical tests have proliferated in the ophthalmic literature. It is important that researchers apply the appropriate statistical tests in agreement studies. Correlation coefficients are hazardous and should be avoided. The 'limits of agreement' method originally proposed by Altman and Bland in 1983 is the statistical procedure of choice. Its step-by-step use and practical considerations in relation to optometry and ophthalmology are detailed in addition to sample size considerations and statistical approaches to precision (repeatability or reproducibility) estimates. Ophthalmic & Physiological Optics © 2011 The College of Optometrists.

Testing non-inferiority of a new treatment in three-arm clinical trials with binary endpoints.

PubMed

Tang, Nian-Sheng; Yu, Bin; Tang, Man-Lai

2014-12-18

A two-arm non-inferiority trial without a placebo is usually adopted to demonstrate that an experimental treatment is not worse than a reference treatment by a small pre-specified non-inferiority margin due to ethical concerns. Selection of the non-inferiority margin and establishment of assay sensitivity are two major issues in the design, analysis and interpretation for two-arm non-inferiority trials. Alternatively, a three-arm non-inferiority clinical trial including a placebo is usually conducted to assess the assay sensitivity and internal validity of a trial. Recently, some large-sample approaches have been developed to assess the non-inferiority of a new treatment based on the three-arm trial design. However, these methods behave badly with small sample sizes in the three arms. This manuscript aims to develop some reliable small-sample methods to test three-arm non-inferiority. Saddlepoint approximation, exact and approximate unconditional, and bootstrap-resampling methods are developed to calculate p-values of the Wald-type, score and likelihood ratio tests. Simulation studies are conducted to evaluate their performance in terms of type I error rate and power. Our empirical results show that the saddlepoint approximation method generally behaves better than the asymptotic method based on the Wald-type test statistic. For small sample sizes, approximate unconditional and bootstrap-resampling methods based on the score test statistic perform better in the sense that their corresponding type I error rates are generally closer to the prespecified nominal level than those of other test procedures. Both approximate unconditional and bootstrap-resampling test procedures based on the score test statistic are generally recommended for three-arm non-inferiority trials with binary outcomes.

Variability in group size and the evolution of collective action.

PubMed

Peña, Jorge; Nöldeke, Georg

2016-01-21

Models of the evolution of collective action typically assume that interactions occur in groups of identical size. In contrast, social interactions between animals occur in groups of widely dispersed size. This paper models collective action problems as two-strategy multiplayer games and studies the effect of variability in group size on the evolution of cooperative behavior under the replicator dynamics. The analysis identifies elementary conditions on the payoff structure of the game implying that the evolution of cooperative behavior is promoted or inhibited when the group size experienced by a focal player is more or less variable. Similar but more stringent conditions are applicable when the confounding effect of size-biased sampling, which causes the group-size distribution experienced by a focal player to differ from the statistical distribution of group sizes, is taken into account. Copyright © 2015 Elsevier Ltd. All rights reserved.

Estuarine sediment toxicity tests on diatoms: Sensitivity comparison for three species

NASA Astrophysics Data System (ADS)

Moreno-Garrido, Ignacio; Lubián, Luis M.; Jiménez, Begoña; Soares, Amadeu M. V. M.; Blasco, Julián

2007-01-01

Experimental populations of three marine and estuarine diatoms were exposed to sediments with different levels of pollutants, collected from the Aveiro Lagoon (NW of Portugal). The species selected were Cylindrotheca closterium, Phaeodactylum tricornutum and Navicula sp. Previous experiments were designed to determine the influence of the sediment particle size distribution on growth of the species assayed. Percentage of silt-sized sediment affect to growth of the selected species in the experimental conditions: the higher percentage of silt-sized sediment, the lower growth. In any case, percentages of silt-sized sediment less than 10% did not affect growth. In general, C. closterium seems to be slightly more sensitive to the selected sediments than the other two species. Two groups of sediment samples were determined as a function of the general response of the exposed microalgal populations: three of the six samples used were more toxic than the other three. Chemical analysis of the samples was carried out in order to determine the specific cause of differences in toxicity. After a statistical analysis, concentrations of Sn, Zn, Hg, Cu and Cr (among all physico-chemical analyzed parameters), in order of importance, were the most important factors that divided the two groups of samples (more and less toxic samples). Benthic diatoms seem to be sensitive organisms in sediment toxicity tests. Toxicity data from bioassays involving microphytobentos should be taken into account when environmental risks are calculated.

Inverse Statistics and Asset Allocation Efficiency

NASA Astrophysics Data System (ADS)

Bolgorian, Meysam

In this paper using inverse statistics analysis, the effect of investment horizon on the efficiency of portfolio selection is examined. Inverse statistics analysis is a general tool also known as probability distribution of exit time that is used for detecting the distribution of the time in which a stochastic process exits from a zone. This analysis was used in Refs. 1 and 2 for studying the financial returns time series. This distribution provides an optimal investment horizon which determines the most likely horizon for gaining a specific return. Using samples of stocks from Tehran Stock Exchange (TSE) as an emerging market and S&P 500 as a developed market, effect of optimal investment horizon in asset allocation is assessed. It is found that taking into account the optimal investment horizon in TSE leads to more efficiency for large size portfolios while for stocks selected from S&P 500, regardless of portfolio size, this strategy does not only not produce more efficient portfolios, but also longer investment horizons provides more efficiency.

Statistical analyses support power law distributions found in neuronal avalanches.

PubMed

Klaus, Andreas; Yu, Shan; Plenz, Dietmar

2011-01-01

The size distribution of neuronal avalanches in cortical networks has been reported to follow a power law distribution with exponent close to -1.5, which is a reflection of long-range spatial correlations in spontaneous neuronal activity. However, identifying power law scaling in empirical data can be difficult and sometimes controversial. In the present study, we tested the power law hypothesis for neuronal avalanches by using more stringent statistical analyses. In particular, we performed the following steps: (i) analysis of finite-size scaling to identify scale-free dynamics in neuronal avalanches, (ii) model parameter estimation to determine the specific exponent of the power law, and (iii) comparison of the power law to alternative model distributions. Consistent with critical state dynamics, avalanche size distributions exhibited robust scaling behavior in which the maximum avalanche size was limited only by the spatial extent of sampling ("finite size" effect). This scale-free dynamics suggests the power law as a model for the distribution of avalanche sizes. Using both the Kolmogorov-Smirnov statistic and a maximum likelihood approach, we found the slope to be close to -1.5, which is in line with previous reports. Finally, the power law model for neuronal avalanches was compared to the exponential and to various heavy-tail distributions based on the Kolmogorov-Smirnov distance and by using a log-likelihood ratio test. Both the power law distribution without and with exponential cut-off provided significantly better fits to the cluster size distributions in neuronal avalanches than the exponential, the lognormal and the gamma distribution. In summary, our findings strongly support the power law scaling in neuronal avalanches, providing further evidence for critical state dynamics in superficial layers of cortex.

Guam's forest resources, 2002.

Treesearch

Joseph A. Donnegan; Sarah L. Butler; Walter Grabowiecki; Bruce A. Hiserote; David. Limtiaco

2004-01-01

The Forest Inventory and Analysis Program collected, analyzed, and summarized field data on 46 forested plots on the island of Guam. Estimates of forest area, tree stem volume and biomass, the numbers of trees, tree damages, and the distribution of tree sizes were summarized for this statistical sample. Detailed tables and graphical highlights provide a summary of Guam...

On the Power of Multivariate Latent Growth Curve Models to Detect Correlated Change

ERIC Educational Resources Information Center

Hertzog, Christopher; Lindenberger, Ulman; Ghisletta, Paolo; Oertzen, Timo von

2006-01-01

We evaluated the statistical power of single-indicator latent growth curve models (LGCMs) to detect correlated change between two variables (covariance of slopes) as a function of sample size, number of longitudinal measurement occasions, and reliability (measurement error variance). Power approximations following the method of Satorra and Saris…

Precipitation Under Cyclic Strain in Solution-Treated Al-4wt%Cu I: Mechanical Behavior

DTIC Science & Technology

2009-02-01

minutes and quenched into ice water immediately prior to mechanical testing. Orientation Imaging Microscopy (OIM) was performed on a FEI XL30 SEM...sampled in order to gain statistical significance with a grain size of 350 µm, it was necessary to condense the data from multiple low-magnification

Some Exact Conditional Tests of Independence for R X C Cross-Classification Tables

ERIC Educational Resources Information Center

Agresti, Alan; Wackerly, Dennis

1977-01-01

Exact conditional tests of independence in cross-classification tables are formulated based on chi square and other statistics with stronger operational interpretations, such as some nominal and ordinal measures of association. Guidelines for table dimensions and sample sizes for which the tests are economically implemented on a computer are…

Krypton and xenon in lunar fines

NASA Technical Reports Server (NTRS)

Basford, J. R.; Dragon, J. C.; Pepin, R. O.; Coscio, M. R., Jr.; Murthy, V. R.

1973-01-01

Data from grain-size separates, stepwise-heated fractions, and bulk analyses of 20 samples of fines and breccias from five lunar sites are used to define three-isotope and ordinate intercept correlations in an attempt to resolve the lunar heavy rare gas system in a statistically valid approach. Tables of concentrations and isotope compositions are given.

The Kindness of Strangers Revisited: A Comparison of 24 US Cities

ERIC Educational Resources Information Center

Levine, Robert V.; Reysen, Stephen; Ganz, Ellen

2008-01-01

Three field studies compared helping behavior across a sample of 24 small, medium and large cities across the United States. The relationship of helping to statistics reflecting the demographic, social, and economic characteristics of these communities was then examined. The strongest predictors of city differences in helping were population size,…

A Monte Carlo Approach to Unidimensionality Testing in Polytomous Rasch Models

ERIC Educational Resources Information Center

Christensen, Karl Bang; Kreiner, Svend

2007-01-01

Many statistical tests are designed to test the different assumptions of the Rasch model, but only few are directed at detecting multidimensionality. The Martin-Lof test is an attractive approach, the disadvantage being that its null distribution deviates strongly from the asymptotic chi-square distribution for most realistic sample sizes. A Monte…

Palau's forest resources, 2003.

Treesearch

Joseph A. Donnegan; Sarah L. Butler; Olaf Kuegler; Brent J. Stroud; Bruce A. Hiserote; Kashgar. Rengulbai

2007-01-01

The Forest Inventory and Analysis Program collected, analyzed, and summarized field data on 54 forested plots on the islands in the Republic of Palau. Estimates of forest area, tree stem volume and biomass, the numbers of trees, tree damages, and the distribution of tree sizes were summarized for this statistical sample. Detailed tables and graphical highlights provide...

Electronic Resource Expenditure and the Decline in Reference Transaction Statistics in Academic Libraries

ERIC Educational Resources Information Center

Dubnjakovic, Ana

2012-01-01

The current study investigates factors influencing increase in reference transactions in a typical week in academic libraries across the United States of America. Employing multiple regression analysis and general linear modeling, variables of interest from the "Academic Library Survey (ALS) 2006" survey (sample size 3960 academic libraries) were…

The Effect of the Multivariate Box-Cox Transformation on the Power of MANOVA.

ERIC Educational Resources Information Center

Kirisci, Levent; Hsu, Tse-Chi

Most of the multivariate statistical techniques rely on the assumption of multivariate normality. The effects of non-normality on multivariate tests are assumed to be negligible when variance-covariance matrices and sample sizes are equal. Therefore, in practice, investigators do not usually attempt to remove non-normality. In this simulation…

Basic School Teachers' Perceptions about Curriculum Design in Ghana

ERIC Educational Resources Information Center

Abudu, Amadu Musah; Mensah, Mary Afi

2016-01-01

This study focused on teachers' perceptions about curriculum design and barriers to their participation. The sample size was 130 teachers who responded to a questionnaire. The analyses made use of descriptive statistics and descriptions. The study found that the level of teachers' participation in curriculum design is low. The results further…

A Structural Equation Model for Predicting Business Student Performance

ERIC Educational Resources Information Center

Pomykalski, James J.; Dion, Paul; Brock, James L.

2008-01-01

In this study, the authors developed a structural equation model that accounted for 79% of the variability of a student's final grade point average by using a sample size of 147 students. The model is based on student grades in 4 foundational business courses: introduction to business, macroeconomics, statistics, and using databases. Educators and…

Threshold-dependent sample sizes for selenium assessment with stream fish tissue

USGS Publications Warehouse

Hitt, Nathaniel P.; Smith, David R.

2015-01-01

Natural resource managers are developing assessments of selenium (Se) contamination in freshwater ecosystems based on fish tissue concentrations. We evaluated the effects of sample size (i.e., number of fish per site) on the probability of correctly detecting mean whole-body Se values above a range of potential management thresholds. We modeled Se concentrations as gamma distributions with shape and scale parameters fitting an empirical mean-to-variance relationship in data from southwestern West Virginia, USA (63 collections, 382 individuals). We used parametric bootstrapping techniques to calculate statistical power as the probability of detecting true mean concentrations up to 3 mg Se/kg above management thresholds ranging from 4 to 8 mg Se/kg. Sample sizes required to achieve 80% power varied as a function of management thresholds and Type I error tolerance (α). Higher thresholds required more samples than lower thresholds because populations were more heterogeneous at higher mean Se levels. For instance, to assess a management threshold of 4 mg Se/kg, a sample of eight fish could detect an increase of approximately 1 mg Se/kg with 80% power (given α = 0.05), but this sample size would be unable to detect such an increase from a management threshold of 8 mg Se/kg with more than a coin-flip probability. Increasing α decreased sample size requirements to detect above-threshold mean Se concentrations with 80% power. For instance, at an α-level of 0.05, an 8-fish sample could detect an increase of approximately 2 units above a threshold of 8 mg Se/kg with 80% power, but when α was relaxed to 0.2, this sample size was more sensitive to increasing mean Se concentrations, allowing detection of an increase of approximately 1.2 units with equivalent power. Combining individuals into 2- and 4-fish composite samples for laboratory analysis did not decrease power because the reduced number of laboratory samples was compensated for by increased precision of composites for estimating mean conditions. However, low sample sizes (<5 fish) did not achieve 80% power to detect near-threshold values (i.e., <1 mg Se/kg) under any scenario we evaluated. This analysis can assist the sampling design and interpretation of Se assessments from fish tissue by accounting for natural variation in stream fish populations.

The Peroxidation of Leukocytes Index Ratio Reveals the Prooxidant Effect of Green Tea Extract

PubMed Central

Manafikhi, Husseen; Raguzzini, Anna; Longhitano, Yaroslava; Reggi, Raffaella; Zanza, Christian

2016-01-01

Despite tea increased plasma nonenzymatic antioxidant capacity, the European Food Safety Administration (EFSA) denied claims related to tea and its protection from oxidative damage. Furthermore, the Supplement Information Expert Committee (DSI EC) expressed some doubts on the safety of green tea extract (GTE). We performed a pilot study in order to evaluate the effect of a single dose of two capsules of a GTE supplement (200 mg × 2) on the peroxidation of leukocytes index ratio (PLIR) in relation to uric acid (UA) and ferric reducing antioxidant potential (FRAP), as well as the sample size to reach statistical significance. GTE induced a prooxidant effect on leukocytes, whereas FRAP did not change, in agreement with the EFSA and the DSI EC conclusions. Besides, our results confirm the primary role of UA in the antioxidant defences. The ratio based calculation of the PLIR reduced the sample size to reach statistical significance, compared to the resistance to an exogenous oxidative stress and to the functional capacity of oxidative burst. Therefore, PLIR could be a sensitive marker of redox status. PMID:28101300

Power analysis to detect treatment effect in longitudinal studies with heterogeneous errors and incomplete data.

PubMed

Vallejo, Guillermo; Ato, Manuel; Fernández García, Paula; Livacic Rojas, Pablo E; Tuero Herrero, Ellián

2016-08-01

 S. Usami (2014) describes a method to realistically determine sample size in longitudinal research using a multilevel model. The present research extends the aforementioned work to situations where it is likely that the assumption of homogeneity of the errors across groups is not met and the error term does not follow a scaled identity covariance structure.   For this purpose, we followed a procedure based on transforming the variance components of the linear growth model and the parameter related to the treatment effect into specific and easily understandable indices. At the same time, we provide the appropriate statistical machinery for researchers to use when data loss is unavoidable, and changes in the expected value of the observed responses are not linear.   The empirical powers based on unknown variance components were virtually the same as the theoretical powers derived from the use of statistically processed indexes.   The main conclusion of the study is the accuracy of the proposed method to calculate sample size in the described situations with the stipulated power criteria.

The Peroxidation of Leukocytes Index Ratio Reveals the Prooxidant Effect of Green Tea Extract.

PubMed

Peluso, Ilaria; Manafikhi, Husseen; Raguzzini, Anna; Longhitano, Yaroslava; Reggi, Raffaella; Zanza, Christian; Palmery, Maura

2016-01-01

Despite tea increased plasma nonenzymatic antioxidant capacity, the European Food Safety Administration (EFSA) denied claims related to tea and its protection from oxidative damage. Furthermore, the Supplement Information Expert Committee (DSI EC) expressed some doubts on the safety of green tea extract (GTE). We performed a pilot study in order to evaluate the effect of a single dose of two capsules of a GTE supplement (200 mg × 2) on the peroxidation of leukocytes index ratio (PLIR) in relation to uric acid (UA) and ferric reducing antioxidant potential (FRAP), as well as the sample size to reach statistical significance. GTE induced a prooxidant effect on leukocytes, whereas FRAP did not change, in agreement with the EFSA and the DSI EC conclusions. Besides, our results confirm the primary role of UA in the antioxidant defences. The ratio based calculation of the PLIR reduced the sample size to reach statistical significance, compared to the resistance to an exogenous oxidative stress and to the functional capacity of oxidative burst. Therefore, PLIR could be a sensitive marker of redox status.

Sampling of prenatal and postnatal offspring from individual rat dams enhances animal use without compromising development

NASA Technical Reports Server (NTRS)

Alberts, J. R.; Burden, H. W.; Hawes, N.; Ronca, A. E.

1996-01-01

To assess prenatal and postnatal developmental status in the offspring of a group of animals, it is typical to examine fetuses from some of the dams as well as infants born to the remaining dams. Statistical limitations often arise, particularly when the animals are rare or especially precious, because all offspring of the dam represent only a single statistical observation; littermates are not independent observations (biologically or statistically). We describe a study in which pregnant laboratory rats were laparotomized on day 7 of gestation (GD7) to ascertain the number and distribution of uterine implantation sites and were subjected to a simulated experience on a 10-day space shuttle flight. After the simulated landing on GD18, rats were unilaterally hysterectomized, thus providing a sample of fetuses from 10 independent uteruses, followed by successful vaginal delivery on GD22, yielding postnatal samples from 10 uteruses. A broad profile of maternal and offspring morphologic and physiologic measures indicated that these novel sampling procedures did not compromise maternal well-being and maintained normal offspring development and function. Measures included maternal organ weights and hormone concentrations, offspring body size, growth, organ weights, sexual differentiation, and catecholamine concentrations.

Sampling errors for satellite-derived tropical rainfall - Monte Carlo study using a space-time stochastic model

NASA Technical Reports Server (NTRS)

Bell, Thomas L.; Abdullah, A.; Martin, Russell L.; North, Gerald R.

1990-01-01

Estimates of monthly average rainfall based on satellite observations from a low earth orbit will differ from the true monthly average because the satellite observes a given area only intermittently. This sampling error inherent in satellite monitoring of rainfall would occur even if the satellite instruments could measure rainfall perfectly. The size of this error is estimated for a satellite system being studied at NASA, the Tropical Rainfall Measuring Mission (TRMM). First, the statistical description of rainfall on scales from 1 to 1000 km is examined in detail, based on rainfall data from the Global Atmospheric Research Project Atlantic Tropical Experiment (GATE). A TRMM-like satellite is flown over a two-dimensional time-evolving simulation of rainfall using a stochastic model with statistics tuned to agree with GATE statistics. The distribution of sampling errors found from many months of simulated observations is found to be nearly normal, even though the distribution of area-averaged rainfall is far from normal. For a range of orbits likely to be employed in TRMM, sampling error is found to be less than 10 percent of the mean for rainfall averaged over a 500 x 500 sq km area.

Addressing the "Replication Crisis": Using Original Studies to Design Replication Studies with Appropriate Statistical Power.

PubMed

Anderson, Samantha F; Maxwell, Scott E

2017-01-01

Psychology is undergoing a replication crisis. The discussion surrounding this crisis has centered on mistrust of previous findings. Researchers planning replication studies often use the original study sample effect size as the basis for sample size planning. However, this strategy ignores uncertainty and publication bias in estimated effect sizes, resulting in overly optimistic calculations. A psychologist who intends to obtain power of .80 in the replication study, and performs calculations accordingly, may have an actual power lower than .80. We performed simulations to reveal the magnitude of the difference between actual and intended power based on common sample size planning strategies and assessed the performance of methods that aim to correct for effect size uncertainty and/or bias. Our results imply that even if original studies reflect actual phenomena and were conducted in the absence of questionable research practices, popular approaches to designing replication studies may result in a low success rate, especially if the original study is underpowered. Methods correcting for bias and/or uncertainty generally had higher actual power, but were not a panacea for an underpowered original study. Thus, it becomes imperative that 1) original studies are adequately powered and 2) replication studies are designed with methods that are more likely to yield the intended level of power.

Mapping cell populations in flow cytometry data for cross‐sample comparison using the Friedman–Rafsky test statistic as a distance measure

PubMed Central

Hsiao, Chiaowen; Liu, Mengya; Stanton, Rick; McGee, Monnie; Qian, Yu

2015-01-01

Abstract Flow cytometry (FCM) is a fluorescence‐based single‐cell experimental technology that is routinely applied in biomedical research for identifying cellular biomarkers of normal physiological responses and abnormal disease states. While many computational methods have been developed that focus on identifying cell populations in individual FCM samples, very few have addressed how the identified cell populations can be matched across samples for comparative analysis. This article presents FlowMap‐FR, a novel method for cell population mapping across FCM samples. FlowMap‐FR is based on the Friedman–Rafsky nonparametric test statistic (FR statistic), which quantifies the equivalence of multivariate distributions. As applied to FCM data by FlowMap‐FR, the FR statistic objectively quantifies the similarity between cell populations based on the shapes, sizes, and positions of fluorescence data distributions in the multidimensional feature space. To test and evaluate the performance of FlowMap‐FR, we simulated the kinds of biological and technical sample variations that are commonly observed in FCM data. The results show that FlowMap‐FR is able to effectively identify equivalent cell populations between samples under scenarios of proportion differences and modest position shifts. As a statistical test, FlowMap‐FR can be used to determine whether the expression of a cellular marker is statistically different between two cell populations, suggesting candidates for new cellular phenotypes by providing an objective statistical measure. In addition, FlowMap‐FR can indicate situations in which inappropriate splitting or merging of cell populations has occurred during gating procedures. We compared the FR statistic with the symmetric version of Kullback–Leibler divergence measure used in a previous population matching method with both simulated and real data. The FR statistic outperforms the symmetric version of KL‐distance in distinguishing equivalent from nonequivalent cell populations. FlowMap‐FR was also employed as a distance metric to match cell populations delineated by manual gating across 30 FCM samples from a benchmark FlowCAP data set. An F‐measure of 0.88 was obtained, indicating high precision and recall of the FR‐based population matching results. FlowMap‐FR has been implemented as a standalone R/Bioconductor package so that it can be easily incorporated into current FCM data analytical workflows. © 2015 International Society for Advancement of Cytometry PMID:26274018

Mapping cell populations in flow cytometry data for cross-sample comparison using the Friedman-Rafsky test statistic as a distance measure.

PubMed

Hsiao, Chiaowen; Liu, Mengya; Stanton, Rick; McGee, Monnie; Qian, Yu; Scheuermann, Richard H

2016-01-01

Flow cytometry (FCM) is a fluorescence-based single-cell experimental technology that is routinely applied in biomedical research for identifying cellular biomarkers of normal physiological responses and abnormal disease states. While many computational methods have been developed that focus on identifying cell populations in individual FCM samples, very few have addressed how the identified cell populations can be matched across samples for comparative analysis. This article presents FlowMap-FR, a novel method for cell population mapping across FCM samples. FlowMap-FR is based on the Friedman-Rafsky nonparametric test statistic (FR statistic), which quantifies the equivalence of multivariate distributions. As applied to FCM data by FlowMap-FR, the FR statistic objectively quantifies the similarity between cell populations based on the shapes, sizes, and positions of fluorescence data distributions in the multidimensional feature space. To test and evaluate the performance of FlowMap-FR, we simulated the kinds of biological and technical sample variations that are commonly observed in FCM data. The results show that FlowMap-FR is able to effectively identify equivalent cell populations between samples under scenarios of proportion differences and modest position shifts. As a statistical test, FlowMap-FR can be used to determine whether the expression of a cellular marker is statistically different between two cell populations, suggesting candidates for new cellular phenotypes by providing an objective statistical measure. In addition, FlowMap-FR can indicate situations in which inappropriate splitting or merging of cell populations has occurred during gating procedures. We compared the FR statistic with the symmetric version of Kullback-Leibler divergence measure used in a previous population matching method with both simulated and real data. The FR statistic outperforms the symmetric version of KL-distance in distinguishing equivalent from nonequivalent cell populations. FlowMap-FR was also employed as a distance metric to match cell populations delineated by manual gating across 30 FCM samples from a benchmark FlowCAP data set. An F-measure of 0.88 was obtained, indicating high precision and recall of the FR-based population matching results. FlowMap-FR has been implemented as a standalone R/Bioconductor package so that it can be easily incorporated into current FCM data analytical workflows. © The Authors. Published by Wiley Periodicals, Inc. on behalf of ISAC.

Precision, Reliability, and Effect Size of Slope Variance in Latent Growth Curve Models: Implications for Statistical Power Analysis

PubMed Central

Brandmaier, Andreas M.; von Oertzen, Timo; Ghisletta, Paolo; Lindenberger, Ulman; Hertzog, Christopher

2018-01-01

Latent Growth Curve Models (LGCM) have become a standard technique to model change over time. Prediction and explanation of inter-individual differences in change are major goals in lifespan research. The major determinants of statistical power to detect individual differences in change are the magnitude of true inter-individual differences in linear change (LGCM slope variance), design precision, alpha level, and sample size. Here, we show that design precision can be expressed as the inverse of effective error. Effective error is determined by instrument reliability and the temporal arrangement of measurement occasions. However, it also depends on another central LGCM component, the variance of the latent intercept and its covariance with the latent slope. We derive a new reliability index for LGCM slope variance—effective curve reliability (ECR)—by scaling slope variance against effective error. ECR is interpretable as a standardized effect size index. We demonstrate how effective error, ECR, and statistical power for a likelihood ratio test of zero slope variance formally relate to each other and how they function as indices of statistical power. We also provide a computational approach to derive ECR for arbitrary intercept-slope covariance. With practical use cases, we argue for the complementary utility of the proposed indices of a study's sensitivity to detect slope variance when making a priori longitudinal design decisions or communicating study designs. PMID:29755377

Size variation in early human mandibles and molars from Klasies River, South Africa: comparison with other middle and late Pleistocene assemblages and with modern humans.

PubMed

Royer, Danielle F; Lockwood, Charles A; Scott, Jeremiah E; Grine, Frederick E

2009-10-01

Previous studies of the Middle Stone Age human remains from Klasies River have concluded that they exhibited more sexual dimorphism than extant populations, but these claims have not been assessed statistically. We evaluate these claims by comparing size variation in the best-represented elements at the site, namely the mandibular corpora and M(2)s, to that in samples from three recent human populations using resampling methods. We also examine size variation in these same elements from seven additional middle and late Pleistocene sites: Skhūl, Dolní Vestonice, Sima de los Huesos, Arago, Krapina, Shanidar, and Vindija. Our results demonstrate that size variation in the Klasies assemblage was greater than in recent humans, consistent with arguments that the Klasies people were more dimorphic than living humans. Variation in the Skhūl, Dolní Vestonice, and Sima de los Huesos mandibular samples is also higher than in the recent human samples, indicating that the Klasies sample was not unusual among middle and late Pleistocene hominins. In contrast, the Neandertal samples (Krapina, Shanidar, and Vindija) do not evince relatively high mandibular and molar variation, which may indicate that the level of dimorphism in Neandertals was similar to that observed in extant humans. These results suggest that the reduced levels of dimorphism in Neandertals and living humans may have developed independently, though larger fossil samples are needed to test this hypothesis.

The impact of hypnotic suggestibility in clinical care settings.

PubMed

Montgomery, Guy H; Schnur, Julie B; David, Daniel

2011-07-01

Hypnotic suggestibility has been described as a powerful predictor of outcomes associated with hypnotic interventions. However, there have been no systematic approaches to quantifying this effect across the literature. This meta-analysis evaluates the magnitude of the effect of hypnotic suggestibility on hypnotic outcomes in clinical settings. PsycINFO and PubMed were searched from their inception through July 2009. Thirty-four effects from 10 studies and 283 participants are reported. Results revealed a statistically significant overall effect size in the small to medium range (r = .24; 95% Confidence Interval = -0.28 to 0.75), indicating that greater hypnotic suggestibility led to greater effects of hypnosis interventions. Hypnotic suggestibility accounted for 6% of the variance in outcomes. Smaller sample size studies, use of the SHCS, and pediatric samples tended to result in larger effect sizes. The authors question the usefulness of assessing hypnotic suggestibility in clinical contexts.

The impact of hypnotic suggestibility in clinical care settings

PubMed Central

Montgomery, Guy H.; Schnur, Julie B.; David, Daniel

2013-01-01

Hypnotic suggestibility has been described as a powerful predictor of outcomes associated with hypnotic interventions. However, there have been no systematic approaches to quantifying this effect across the literature. The present meta-analysis evaluates the magnitude of the effect of hypnotic suggestibility on hypnotic outcomes in clinical settings. PsycINFO and PubMed were searched from their inception through July 2009. Thirty-four effects from ten studies and 283 participants are reported. Results revealed a statistically significant overall effect size in the small to medium range (r = 0.24; 95% Confidence Interval = −0.28 to 0.75), indicating that greater hypnotic suggestibility led to greater effects of hypnosis interventions. Hypnotic suggestibility accounted for 6% of the variance in outcomes. Smaller sample size studies, use of the SHCS, and pediatric samples tended to result in larger effect sizes. Results question the usefulness of assessing hypnotic suggestibility in clinical contexts. PMID:21644122

Trends in study design and the statistical methods employed in a leading general medicine journal.

PubMed

Gosho, M; Sato, Y; Nagashima, K; Takahashi, S

2018-02-01

Study design and statistical methods have become core components of medical research, and the methodology has become more multifaceted and complicated over time. The study of the comprehensive details and current trends of study design and statistical methods is required to support the future implementation of well-planned clinical studies providing information about evidence-based medicine. Our purpose was to illustrate study design and statistical methods employed in recent medical literature. This was an extension study of Sato et al. (N Engl J Med 2017; 376: 1086-1087), which reviewed 238 articles published in 2015 in the New England Journal of Medicine (NEJM) and briefly summarized the statistical methods employed in NEJM. Using the same database, we performed a new investigation of the detailed trends in study design and individual statistical methods that were not reported in the Sato study. Due to the CONSORT statement, prespecification and justification of sample size are obligatory in planning intervention studies. Although standard survival methods (eg Kaplan-Meier estimator and Cox regression model) were most frequently applied, the Gray test and Fine-Gray proportional hazard model for considering competing risks were sometimes used for a more valid statistical inference. With respect to handling missing data, model-based methods, which are valid for missing-at-random data, were more frequently used than single imputation methods. These methods are not recommended as a primary analysis, but they have been applied in many clinical trials. Group sequential design with interim analyses was one of the standard designs, and novel design, such as adaptive dose selection and sample size re-estimation, was sometimes employed in NEJM. Model-based approaches for handling missing data should replace single imputation methods for primary analysis in the light of the information found in some publications. Use of adaptive design with interim analyses is increasing after the presentation of the FDA guidance for adaptive design. © 2017 John Wiley & Sons Ltd.