This paper studies the general theory of Akaike's Information Criterion (AIC) and provides two analytical extensions. The extensions make AIC asymptotically consistent and penalize overparameterization more stringently to pick only the simplest of the two models. The criteria are applied in two Monte Carlo experiments. (Author/GDC)

Recent research using single-stage procedures to test the equality of the means of J independent normal distributions when variances are unequal have proven unsatisfactory in controlling Type I errors and power. A method for dealing with the problem of unequal sample sizes while implementing two-stage procedures is discussed. (TJH)

We consider the problem of classifying an unknown observation into one of several populations using tree-structured allocation rules. Although many parametric classification procedures are robust to certain assumption violations, there is need for discriminant procedures that can be utilized regardless of the group-conditional distributions that…

The meaningful investigation of many problems in statistics can be solved through Monte Carlo methods. Monte Carlo studies can help solve problems that are mathematically intractable through the analysis of random samples from populations whose characteristics are known to the researcher. Using Monte Carlo simulation, the values of a statistic are…

Eta-Squared (ES) is often used as a measure of strength of association of an effect, a measure often associated with effect size. It is also considered the proportion of total variance accounted for by an independent variable. It is simple to compute and interpret. However, it has one critical weakness cited by several authors (C. Huberty, 1994;…

Monte Carlo methods were used to investigate the effects of removing extreme data points identified by five indices of influence. Multivariate normal data were simulated and observations were removed from samples if they exceeded the criteria suggested in the literature for each influence statistic. Factors included in the design of the Monte…

This study compared the effectiveness of 10 methods of determining the number of factors to retain in exploratory common factor analysis. The 10 methods included the Kaiser rule and a modified Kaiser criterion, 3 variations of parallel analysis, 4 regression-based variations of the scree procedure, and the minimum average partial procedure. The…

Many multivariate statistical methods call upon the assumption of multivariate normality. However, many applied researchers fail to test this assumption. This omission could be due to ignorance of the existence of tests of multivariate normality or confusion about which test to use. Although at least 50 tests of multivariate normality exist,…

Statistical significance and practical significance can be considered jointly through the use of non-nil null hypotheses that are based on values deemed to be practically significant. When examining differences between the means of two groups, researchers can use a randomization test or an independent t test. The issue addressed in this paper is…

This study focused on the issue of measurement reliability and its attenuation on correlation between two composites and two seemingly different approaches for correcting the attenuation. As expected, Monte Carlo simulation results show that correlation coefficients uncorrected for measurement error are systematically biased downward. For the data…

A Monte Carlo study is reported that shows the comparative performance of alternative approaches under deviations from their respective assumptions in the case of structural equation models with latent variables with attention restricted to point estimates of model parameters. The conditional polychoric correlations method is shown most robust…

The distinction between Type 1 and Type 12 sampling in connection with measurement data is discussed, and a method is presented for simulating data arising from Type 12 sampling. A Monte Carlo study is described that shows conditions under which precise confidence level control under Type 12 sampling is maintained. (SLD)

Monte Carlo study results show that recently proposed multiple comparison procedures (MCPs) that are not intended to control the familywise error rate had consistently larger true model rates than did familywise error controlling MCPs. (SLD)

Used Monte Carlo methods to examine the Type I error rates for randomization tests applied to single-case data arising from ABAB designs involving random, systematic, or response-guided assignment of interventions. Discusses conditions under which Type I error rate is controlled or is not. (SLD)

Compared the quality of item parameter estimates for marginal maximum likelihood (MML) and Markov Chain Monte Carlo (MCMC) with the nominal response model using simulation. The quality of item parameter recovery was nearly identical for MML and MCMC, and both methods tended to produce good estimates. (SLD)