Estimating confidence intervals (CIs) for the correlation has been a challenge because the correlation sampling distribution changes depending on the correlation magnitude. The Fisher z-transformation was one of the first attempts at estimating correlation CIs but has historically shown to not have acceptable coverage probability if data were…

There is a need for effect sizes that are readily interpretable by a broad audience. One index that might fill this need is [pi], which represents the proportion of scores in one group that exceed the mean of another group. The robustness of estimates of [pi] to violations of normality had not been explored. Using simulated data, three estimates…

Using Monte Carlo simulations, this research examined the performance of four missing data methods in SEM under different multivariate distributional conditions. The effects of four independent variables (sample size, missing proportion, distribution shape, and factor loading magnitude) were investigated on six outcome variables: convergence rate,…

The authors performed a Monte Carlo simulation to empirically investigate the robustness and power of 4 methods in testing mean differences for 2 independent groups under conditions in which 2 populations may not demonstrate the same pattern of nonnormality. The approaches considered were the t test, Wilcoxon rank-sum test, Welch-James test with…

The authors generated exact probability distributions for sample sizes up to 35 in each of three groups ("n" less than or equal to 105) and up to 10 in each of four groups ("n" less than or equal to 40). They compared the exact distributions to the chi-square, gamma, and beta approximations. The beta approximation was best in…

The authors assessed correct model identification rates of Akaike's information criterion (AIC), corrected criterion (AICC), consistent AIC (CAIC), Hannon and Quinn's information criterion (HQIC), and Bayesian information criterion (BIC) for selecting among cross-classified random effects models. Performance of default values for the 5…

Regression discontinuity design is an alternative to randomized experiments to make causal inference when random assignment is not possible. This article first presents the formal identification and estimation of regression discontinuity treatment effects in the framework of Rubin's causal model, followed by a thorough literature review of…

The authors describe and illustrate 6 factors that affect the size of a Pearson correlation: (a) the amount of variability in the data, (b) differences in the shapes of the 2 distributions, (c) lack of linearity, (d) the presence of 1 or more "outliers," (e) characteristics of the sample, and (f) measurement error. Also discussed are ways to…

The power functions of Student t tests performed on initial scores, ordinary ranks, 3 kinds of modular ranks, and dichotomies were investigated for 1 normal and 3 nonnormal distributions using 2 samples of 26 simulated scores each. Advantages of extending the rank transformation concept are discussed. (SLD)

It is argued that outlier-prone distributions reduce the power of nonparametric tests, but power can be restored through procedures usually associated with parametric tests. Computer simulation is used to show how an outlier detection and downweighting procedure augments the power of the t-test and the Wilcoxon-Mann-Whitney test. (SLD)

Assessed the relative merits of the Student "t" test and the Wilcoxon rank sum test under four population distributions and six sample-size pairings through Monte Carlo methods. The Wilcoxon rank sum test demonstrated an advantage in statistical power for nonnormal distributions (but not normal distributions), with fewer Type III errors…

A sampling subspace in hypothesis testing where Type II error is made for completely illogical reasons from the standpoint of probability is described. The case of unequal probabilities of populations or conditions is also considered. (Author/BS)

A computer program randomly sampled ordered pairs of scores from known populations that departed from bivariate normal form and calculated correlation coefficients from sample values. Hypotheses were tested (1) that population correlations are zero using the t statistic; and (2) that population correlations have non-zero values using the r to z…

The meta-analytic method proposed by S. W. Raudenbush (1988) for studying variance heterogeneity was studied. Results of a Monte Carlo study indicate that the Type I error rate of the test is sensitive to even modestly platykurtic score distributions and to the ratio of study sample size to the number of studies. (SLD)

A Monte Carlo evaluation of four test statistics for comparing dependent zero-order correlations was conducted with four sample sizes and three population distributions. Results indicate that choice of optimal test statistic depends on sample size and distribution, and predictor intercorrelation and effect size or magnitude of the…