Simulation results show that the power to detect small mean differences when fitting a model with free residual variances across groups decreases as the difference in R squared increases. This decrease is more pronounced in the presence of correlated errors and if group sample sizes differ. (SLD)

Ethical conflicts may arise when health care professionals control researchers' access to vulnerable populations. Collaboration and dialogue among researchers, health care providers, and potential subjects are essential in order to recruit enough subjects to maintain research integrity while ensuring their protection. (Contains 41 references.) (SK)

Studied the use of different weighting techniques in structural equation modeling and found, through simulation, that the use of an effective sample size weight provides unbiased estimates of key parameters and their sampling variances. Also discusses use of a popular normalization technique of scaling weights. (SLD)

Demonstrates how substantive researchers can use a Monte Carlo study to decide on sample size and determine power. Presents confirmatory factor analysis and growth models as examples, conducting these analyses with the Mplus program (B. Muthen and L. Muthen 1998). (SLD)

Sixteen tables are presented for critical values of the larger of two sample correlation coefficients from two independent samples, given the sample size and the value of the smaller correlation. These tables allow quick assessment of significance without requiring calculation of the test statistic. (SLD)

Analysis of data of two cooperative extension studies suggests that research costs can be lowered by reducing sample size without compromising results, descriptive studies can have a larger margin of error in calculating sample size without compromise, and comparative studies' results are unaffected by a 3% margin of error but are dramatically…

A multiobjective programming method is proposed for determining samples of examinees needed for estimating the parameters of a group of items. This approach maximizes the information functions of each of three parameters. A numerical verification of the procedure is presented. (SLD)

Describes MINSIZE, an MS-DOS computer program that permits the user to determine the minimum sample size needed for the results of a given analysis to be statistically significant. Program applications for statistical significance tests are presented and illustrated. (SLD)

Type I error rates and rejection rates for three-dimensionality assessment procedures were studied with data sets simulated to reflect short tests and small samples. Results show that the G-squared difference test (D. Bock, R. Gibbons, and E. Muraki, 1988) suffered from a severely inflated Type I error rate at all conditions simulated. (SLD)

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)

Discusses determining sample size for estimation of the squared multiple correlation coefficient and presents regression equations that permit determination of the sample size for estimating this parameter for up to 20 predictor variables. (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)

Whether "more is ever too much" for the number of indicators per factor in confirmatory factor analysis was studied by varying sample size and indicators per factor in 35,000 Monte Carlo solutions. Results suggest that traditional rules calling for fewer indicators for smaller sample size may be inappropriate. (SLD)