The normality of number distributions generated by various random-number generators were studied, focusing on when the random-number generator reached a normal distribution and at what sample size. Findings suggest the steps that should be followed when using a random-number generator in a Monte Carlo simulation. (SLD)

The problem of chance model modifications under varying levels of sample size, model size, and severity of misspecification in confirmatory factor analysis models was examined through Monte Carlo simulations. Findings suggest that practitioners should exercise caution when interpreting modified models unless sample size is quite large. (SLD)

Used simulated data to test the integrity of orthogonal factor solutions when varying sample size, factor pattern/structure coefficient magnitude, method of extraction, number of variables, number of factors, and degree of overextraction. Discusses implications of results with regard to overextraction. (SLD)

This study assessed the accuracy of parallel analysis, a technique in which observed eigenvalues are compared to eigenvalues from simulated data when no real factors are present. Three studies with manipulated sizes of real factors and sample sizes illustrate the importance of modeling the data more closely when parallel analysis is used. (SLD)

This article illustrates the use of the SAS system for Monte Carlo simulation work in structural equation modeling (SEM). Data generation procedures for both multivariate normal and nonnormal conditions are discussed, and relevant SAS codes for implementing these procedures are presented. A hypothetical example is presented in which Monte Carlo…

Extensions of latent state-trait models for continuous observed variables to mixture latent state-trait models with and without covariates of change are presented that can separate individuals differing in their occasion-specific variability. An empirical application to the repeated measurement of mood states (N = 501) revealed that a model with 2…

In meta-analysis, it is common to have dependent effect sizes, such as several effect sizes from the same sample but measured at different times. Cheung and Chan proposed the adjusted-individual and adjusted-weighted procedures to estimate the degree of dependence and incorporate this estimate in the meta-analysis. The present study extends the…

One of the major problems that a tree-approach to data analysis often encounters is the instability of tree-structures. The instability issue must be dealt with before data can be interpreted by this method. Examining instability at a node of a tree provides insight into the instability of the whole tree, because the same theory of instability…

Whether "more is ever too much" for the number of indicators (p) per factor (p/f) in confirmatory factor analysis (CFA) was studied by varying sample size (N) from 50 to 1,000 and p/f from 2 to 12 items per factor in 30,000 Monte Carlo simulations. For all sample sizes, solution behavior steadily improved (more proper solutions and more…

A sequence of simulations was carried out to aid in the diagnosis and interpretation of equating differences found between random and matched (nonrandom) samples for four commonly used equating procedures: (1) Tucker linear observed-score equating; (2) Levine equally reliable linear observed-score equating; (3) equipercentile curvilinear…

Conventional wisdom suggests the omnibus F-test needs to be significant before conducting post-hoc pairwise multiple comparisons. However, there is little empirical evidence supporting this practice. Protected tests are conducted only after a significant omnibus F-test while unprotected tests are conducted without regard to the significance of the…

One set of approaches to the problem of clustering with dichotomous data in cluster analysis (CA) was studied. The techniques developed for clustering with binary data involve calculating distances between observations based on the variables and then applying one of the standard CA algorithms to these distances. One of the groups of distances that…

Derived a confidence interval for a ratio of correlated mean absolute deviations. Simulation results show that it performs well in small sample sizes across realistically nonnormal distributions and that it is almost as powerful as the most powerful test examined by R. Wilcox (1990). (SLD)

Recent simulations have shown that, for a given sample size, the Mantel-Haenszel (MH) variances tend to be larger when items are administered to randomly selected examinees than when they are administered adaptively. Results suggest that adaptive testing may lead to more efficient application of MH differential item functioning analyses. (SLD)