Exploratory graph analysis (EGA) is a commonly applied technique intended to help social scientists discover latent variables. Yet, the results can be influenced by the methodological decisions the researcher makes along the way. In this article, we focus on the choice regarding the number of factors to retain: We compare the performance of the…

The measurement error in principal components extracted from a set of fallible measures is discussed and evaluated. It is shown that as long as one or more measures in a given set of observed variables contains error of measurement, so also does any principal component obtained from the set. The error variance in any principal component is shown…

A common question in test evaluation is whether an a priori assignment of items to subtests is supported by empirical data. If the analysis results indicate the assignment of items to subtests under study is not supported by data, the assignment is often adjusted. In this study the authors compare two methods on the quality of their suggestions to…

This article reports results of a meta-analysis of survey response rates in published research in counseling and clinical psychology over a 20-year span and describes reported survey administration procedures in those fields. Results of 308 survey administrations showed a weighted average response rate of 49.6%. Among possible moderators, response…

In exploratory common factor analysis, extension analysis refers to computing the relationship of the common factors to variables that were not included in the factor analysis. A new extension procedure is presented that gives correlations without using estimated factor scores. Advantages of the new method are illustrated. (SLD)

Differences between per experiment (PE) and experimentwise (EW) error rates were studied through simulation for several multiple-comparison procedures for both pairwise comparisons and planned contrasts. Results suggest ways to control PE rates through new multiple-comparison procedures that maximize experimental power while controlling Type I…

Consider the nonparametric regression model Y = m(X)+ [tau](X)[epsilon], where X and [epsilon] are independent random variables, [epsilon] has a median of zero and variance [sigma][squared], [tau] is some unknown function used to model heteroscedasticity, and m(X) is an unknown function reflecting some conditional measure of location associated…

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…

Using a factorially complex test on a highly homogeneous population is explored, and a procedure is demonstrated that can statistically justify using a cognitive test on which Cronbach's alpha is low. The example involves 649 accountants (auditors). Results indicate that unbundling test and sample provides justification for using the data. (SLD)

META-RAND, a Statistical Analysis System macro that generates random values used for calculating correlations in a meta analysis, is described. Output consists of uncorrected and corrected mean and variances of "r" and several measures of the degree of homogeneity or heterogeneity in the data. (SLD)

This article describes the theoretical distribution of optima arising from two-category univariable optimal discriminant analysis (UniODA) of continuous random data for a one-tailed (directional) hypothesis. Directional UniODA is illustrated through an investigation of the relationship between depression and brain monoamine turnover. (SLD)

Presents a procedure for determining the optimal number of conditions to use in multifaceted measurement designs when resource constraints are imposed. The procedure is illustrated for the case in which the costs per condition vary within the same facet. (Author)

Three problems with stepwise research methods are explored. Computer packages may use incorrect degrees of freedom in stepwise computations. In addition, stepwise methods do not identify correctly the best variable set of a given size. A third problem is that stepwise methods tend to capitalize on sampling error. (SLD)

Two quasi-experimental field studies were conducted to evaluate the psychometric equivalence of computerized and paper-and-pencil job satisfaction measures. The present research extends previous work in the area by providing better control of common threats to validity in quasi-experimental research on test mode effects and by evaluating a more…

While Spearman's rho and Kendall's tau are equally powerful rank-order correlation coefficients under conditions of normality, they have different metrics. Applied to the same data, tau is smaller in absolute value, often no more than two-thirds of the size of rho. This difference in correlational metric appears to need emphasis. (Author/CM)