Digital Commons Network^™

Casual Mediation Analyses With Structural Mean Models, Thomas R. Tenhave, Marshall Joffe, Kevin Lynch, Greg Brown, Stephen Maisto

UPenn Biostatistics Working Papers

We represent a linear structural mean model (SMM)approach for analyzing mediation of a randomized baseline intervention's effect on a univariate follow-up outcome. Unlike standard mediation analyses, our approach does not assume that the mediating factor is randomly assigned to individuals (i.e., sequential ignorability). Hence, a comparison of the results of the proposed and standard approaches in with respect to mediation offers a sensitivity analyses of the sequential ignorability assumption. The G-estimation procedure for the proposed SMM represents an extension of the work on direct effects of randomized treatment effects for survival outcomes by Robins and Greenland (1994) (Section 5.0 and …

Modeling Differentiated Treatment Effects For Multiple Outcomes Data, Hongfei Guo, Karen Bandeen-Roche

Johns Hopkins University, Dept. of Biostatistics Working Papers

Multiple outcomes data are commonly used to characterize treatment effects in medical research, for instance, multiple symptoms to characterize potential remission of a psychiatric disorder. Often either a global, i.e. symptom-invariant, treatment effect is evaluated. Such a treatment effect may over generalize the effect across the outcomes. On the other hand individual treatment effects, varying across all outcomes, are complicated to interpret, and their estimation may lose precision relative to a global summary. An effective compromise to summarize the treatment effect may be through patterns of the treatment effects, i.e. "differentiated effects." In this paper we propose a two-category model …

Restricted Quasi-Independent Model Resolves Paradoxical Behaviors Of Cohen’S Kappa, Vicki Stover Hertzberg, Frank Xu, Michael Haber

Journal of Modern Applied Statistical Methods

Cohen’s kappa, an index of inter-rater agreement, behaves paradoxically in 2×2 tables. λ_A is derived, an index from the restricted quasi-independent model for 2×2 tables. Simulation studies are used to demonstrate λ_A has superior performance compared to Scott’s pi. Moreover, λ_A does not show paradoxical behavior for 2×2 tables.

Inferences About The Components Of A Generalized Additive Model, Rand R. Wilcox

Journal of Modern Applied Statistical Methods

A method for making inferences about the components of a generalized additive model is described. It is found that a variation of the method, based on means, performs well in simulations. Unlike many other inferential methods, switching from a mean to a 20% trimmed mean was found to offer little or no advantage in terms of both power and controlling the probability of a Type I error.

Determining Parallel Analysis Criteria, Marley W. Watkins

Journal of Modern Applied Statistical Methods

Determining the number of factors to extract is a critical decision in exploratory factor analysis. Simulation studies have found the Parallel Analysis criterion to be accurate, but it is computationally intensive. Two freeware programs that implement Parallel Analysis on Macintosh and Windows operating systems are presented.

Supporting And Preparing Future Decision-Makers With The Needed Tools, Michael Wolf-Branigin

Journal of Modern Applied Statistical Methods

Supporting and Preparing Future Decision-makers with the Needed ToolsEducational and social service researchers and evaluators continue to develop advanced statistical methods. To ensure that our students have the essential skills as they enter direct service, the focus must be on assuring that they learn readily understandable methods that are appropriate for small samples and use repeated measures.

Statistical Tests, Tests Of Significance, And Tests Of A Hypothesis Using Excel, David A. Heiser

Journal of Modern Applied Statistical Methods

Microsoft’s spreadsheet program Excel has many statistical functions and routines. Over the years there have been criticisms about the inaccuracies of these functions and routines (see McCullough 1998, 1999). This article reviews some of these statistical methods used to test for differences between two samples. In practice, the analysis is done by a software program and often with the actual method used unknown. The user has to select the method and variations to be used, without full knowledge of just what calculations are used. Usually there is no convenient trace back to textbook explanations. This article describes the Excel algorithm …

Robust Confidence Intervals For Effect Size In The Two-Group Case, H. J. Keselman, James Algina, Katherine Fradette

Journal of Modern Applied Statistical Methods

The probability coverage of intervals involving robust estimates of effect size based on seven procedures was compared for asymmetrically trimming data in an independent two-groups design, and a method that symmetrically trims the data. Four conditions were varied: (a) percentage of trimming, (b) type of nonnormal population distribution, (c) population effect size, and (d) sample size. Results indicated that coverage probabilities were generally well controlled under the conditions of nonnormality. The symmetric trimming method provided excellent probability coverage. Recommendations are provided.

Testing For Aptitude-Treatment Interactions In Analysis Of Covariance And Randomized Block Designs Under Assumption Violations, Tim Moses, Alan Klockars

Journal of Modern Applied Statistical Methods

This study compared the robustness of two analysis strategies designed to detect Aptitude-Treatment Interactions to two of their similarly-held assumptions, normality and residual variance homogeneity. The analysis strategies were the test of slope differences in analysis of covariance and the test of the Block-by- Treatment interaction in randomized block analysis of variance. With equal sample sizes in the treatment groups the results showed that residual variance heterogeneity has little effect on either strategy but nonnormality makes the test of slope differences liberal and the test of the Block-by-Treatment interaction conservative. With unequal sample sizes in the treatment groups the often-reported …

Type I Error Of Four Pairwise Mean Comparison Procedures Conducted As Protected And Unprotected Tests, J. Jackson Barnette, James E. Mclean

Journal of Modern Applied Statistical Methods

Type I error control accuracy of four commonly used pairwise mean comparison procedures, conducted as protected or unprotected tests, is examined. If error control philosophy is experimentwise, Tukey’s HSD, as an unprotected test, is most accurate and if philosophy is per-experiment, Dunn-Bonferroni, conducted as an unprotected test, is most accurate.

Sample Size Selection For Pair-Wise Comparisons Using Information Criteria, Xuemei Pan, C. Mitchell Dayton

Journal of Modern Applied Statistical Methods

This article provides results for rates of correct identifications of paired-comparison information criteria and Tukey HSD as functions of the pattern of mean differences and of sample size. Therefore, the tables provided are useful for selecting sample sizes in real world applications.

Geosciences Newsletter - 2005, Department Of Geosciences

Geological and Environmental Sciences News

Vol. 2, No. 1

• Mohamed Sultan, Chairperson
• Faculty News
• Emeriti News
• Department News
• Graduate Student News
• Awards & Graduations
• Alumni News
• Professional Meetings
• Outstanding Alumni Academy
• New Faces in the Geosciences

Total Syntheses Of (+)-Geldanamycin, (-)-Ragaglitazar, And (+)-Kurasoin A And Phase-Transfer-Catalyzed Asymmetric Alkylation, Erik J. Hicken

Theses and Dissertations

Geldanamycin possesses various biological activities as seen in the NCI 60 cell line panel (13 nM avg., 70 nM SKBr-3 cells). The predominant mode of action providing these unique results arises from the ability of geldanamycin (GA) to bind to the chaperone heat shock protein 90 (Hsp90). Despite its complicated functionality, the first total synthesis of GA was accomplished, which included two new reactions developed specifically to address the stereochemical features. The final step in the synthesis of GA was a demethylation-oxidation sequence to generate the desired para-quinone. This step could only be accomplished with HNO3/AcOH, producing GA in …

A Bayesian Subset Analysis Of Sensory Evaluation Data, Balgobin Nandram

Journal of Modern Applied Statistical Methods

In social sciences it is easy to carry out sensory experiments using say a J-point hedonic scale. One major problem with the J-point hedonic scale is that a conversion from the category scales to numeric scores might not be sensible because the panelists generally view increments on the hedonic scale as psychologically unequal. In the current problem several products are rated by a set of panelists on the J-point hedonic scale. One objective is to select the best subset of products and to assess the quality of the products by estimating the mean and standard deviation response …

Training Statisticians To Be Alert To The Dangers Of Misapplying Statistical Methods, Vance W. Berger

Journal of Modern Applied Statistical Methods

Statisticians are faced with a variety of challenges. Their ability to cope successfully with these challenges depends, in large part, on the quality of their training. It is not the purpose of this article to present a comprehensive training plan that will overhaul the standard curriculum a statistician might follow under current training regimens (i.e., in a degree program). Rather, the objective is to point out important areas that appear to be under-represented in standard curricula and correspondingly overlooked too often in practice. The hope is that these areas might be better integrated into the training of the next generation …

Power Of The T Test For Normal And Mixed Normal Distributions, Marilyn S. Thompson, Samuel B. Green, Yi-Hsin Chen, Shawn Stockford, Wen-Juo Lo

Journal of Modern Applied Statistical Methods

Previous research suggests that the power of the independent-samples t test decreases when population distributions are mixed normal rather than normal, and that robust methods have superior power under these conditions. However, under some conditions, the power for the independent-samples t test can be greater when the population distributions for the independent groups are mixed normal rather than normal. The implications of these results are discussed.

Misconceptions Leading To Choosing The T Test Over The Wilcoxon Mann-Whitney Test For Shift In Location Parameter, Shlomo S. Sawilowsky

Journal of Modern Applied Statistical Methods

There exist many misconceptions in choosing the t over the Wilcoxon Rank-Sum test when testing for shift. Examples are given in the following three groups: (1) false statement, (2) true premise, but false conclusion, and (3) true statement irrelevant in choosing between the t test and the Wilcoxon Rank Sum test.

Jmasm21: Pcic_Sas: Best Subsets Using Information Criteria, C. Mitchell Dayton, Xuemei Pan

Journal of Modern Applied Statistical Methods

PCIC_SAS is a SAS program for identifying optimal subsets of means based on independent groups. All possible configurations of ordered subsets of groups are considered and a best model is identified using both the AIC and BIC information criteria. Results for models with homogeneous variances as well as models with heterogeneity of variance in the same pattern as the means are reported.

End Matter, Jmasm Editors

Journal of Modern Applied Statistical Methods

No abstract provided.

Second-Order Accurate Inference On Simple, Partial, And Multiple Correlations, Robert J. Boik, Ben Haaland

Journal of Modern Applied Statistical Methods

This article develops confidence interval procedures for functions of simple, partial, and squared multiple correlation coefficients. It is assumed that the observed multivariate data represent a random sample from a distribution that possesses infinite moments, but there is no requirement that the distribution be normal. The coverage error of conventional one-sided large sample intervals decreases at rate 1√n as n increases, where n is an index of sample size. The coverage error of the proposed intervals decreases at rate 1/n as n increases. The results of a simulation study that evaluates the performance of the proposed intervals is …

A Method For Analyzing Unreplicated Experiments Using Information On The Intraclass Correlation Coefficient, Jamis J. Perrett

Journal of Modern Applied Statistical Methods

Many studies are performed on units that cannot be replicated; however, there is often an abundance of subsampling. By placing a reasonable upper bound on the intraclass correlation coefficient (ICC), it is possible to carry out classical tests of significance that have conservative levels of significance.

Misconceptions Leading To Choosing The T Test Over The Wilcoxon Mann-Whitney Test For Shift In Location Parameter, Shlomo S. Sawilowsky

Theoretical and Behavioral Foundations of Education Faculty Publications

There exist many misconceptions in choosing the t over the Wilcoxon Rank-Sum test when testing for shift. Examples are given in the following three groups: (1) false statement, (2) true premise, but false conclusion, and (3) true statement irrelevant in choosing between the t test and the Wilcoxon Rank Sum test.

Jmasm24: Numerical Computing For Third-Order Power Method Polynomials (Excel), Todd C. Headrick

Journal of Modern Applied Statistical Methods

The power method polynomial transformation is a popular procedure used for simulating univariate and multivariate non-normal distributions. It requires software that solves simultaneous nonlinear equations. Potential users of the power method may not have access to commercial software packages (e.g., Mathematica, Fortran). Therefore, algorithms are presented in the more commonly available Excel 2003 spreadsheets. The algorithms solve for (1) coefficients for polynomials of order three, (2) intermediate correlations and Cholesky factorizations for multivariate data generation, and (3) the values of skew and kurtosis for determining if a transformation will produce a valid power method probability density function (pdf). The Excel …

The Influence Of Reliability On Four Rules For Determining The Number Of Components To Retain, Gibbs Y. Kanyongo

Journal of Modern Applied Statistical Methods

Imperfectly reliable scores impact the performance of factor analytic procedures. A series of Monte Carlo studies was conducted to generate scores with known component structure from population matrices with varying levels of reliability. The scores were submitted to four procedures: Kaiser rule, scree plot, parallel analysis, and modified Horn’s parallel analysis to find if each procedure accurately determines the number of components at the different reliability levels. The performance of each procedure was judged by the percentage of the number of times that the procedure was correct and the mean components that each procedure extracted in each cell. Generally, the …

Corrections For Type I Error In Social Science Research: A Disconnect Between Theory And Practice, Kenneth Lachlan, Patric R. Spence

Journal of Modern Applied Statistical Methods

Type I errors are a common problem in factorial ANOVA and ANOVA based analyses. Despite decades of literature offering solutions to the Type I error problems associated with multiple significance tests, simple solutions such as Bonferroni corrections have been largely ignored by social scientists. To examine this discontinuity between theory and practice, a content analysis was performed on 5 flagship social science journals. Results indicate that corrections for Type I error are seldom utilized, even in designs so complicated as to almost guarantee erroneous rejection of null hypotheses.

2005 Scholars And Artists Bibliography, Stephen D. Slane Dr., Michael Schwartz Library, Cleveland State University, Friends Of The Michael Schwartz Library

Scholars and Artists Bibliographies

This bibliography was created for the annual Friends of the Michael Schwartz Library Scholars and Artists Reception, recognizing scholarly and creative achievements of Cleveland State University faculty, staff and emeriti. Dr, Steve Slane was the guest speaker.

Student Fact Book, Fall 2005, Twenty-Ninth Annual Edition, Wright State University, Office Of Student Information Systems, Wright State University

Wright State University Student Fact Books

The student fact book has general demographic information on all students enrolled at Wright State University for Fall Quarter, 2005.

Is The Number Of Sick Persons In A Cohort Constant Over Time?, Paula Diehr, Ann Derleth, Anne Newman, Liming Cai

UW Biostatistics Working Paper Series

Objectives: To estimate the number of persons in a cohort who are sick, over time.

Methods: We calculated the number of sick persons in the Cardiovascular Health Study (CHS), a cohort study of older adults followed up to 14 years, using eight definitions of “healthy” and “sick”. We projected the number in each health state over time for a birth cohort.

Results: The number of sick persons in CHS was approximately constant for 14 years, for all definitions of “sick”. The estimated number of sick persons in the birth cohort was approximately constant from ages 55-75, after which it decreased. …

Santa Clara Magazine, Volume 47 Number 2, Fall 2005, Santa Clara University

Santa Clara Magazine

6 - THE DRIVE TO CREATE by Elizabeth Kelley Gillogly '93. Albert Hoagland, an adjunct professor at SCU for more than 20 years, helped to build the first disk drive. Now is he working to preserve the history of magnetic disk storage.

10 - UNFINISHED MESSAGE by Toshio Mori. We share an excerpt from a collection of short stories by Mori, whose work highlights the plight of Japanese immigrants in the U.S. during World War II. The book is a part of the California Legacy Series, a partnership between SCU and Heyday Books.

14 - THE FAMILIAR STRANGER By Cynthia …

Multifractal Structure In Nonrepresentational Art, Jonas R. Mureika, C. C. Dyer, G. C. Cupchik

Physics Faculty Works

Multifractal analysis techniques are applied to patterns in several abstract expressionist artworks, painted by various artists. The analysis is carried out on two distinct types of structures: the physical patterns formed by a specific color (“blobs”) and patterns formed by the luminance gradient between adjacent colors (“edges”). It is found that the multifractal analysis method applied to “blobs” cannot distinguish between artists of the same movement, yielding a multifractal spectrum of dimensions between about 1.5 and 1.8. The method can distinguish between different types of images, however, as demonstrated by studying a radically different type of art. The data suggest …