A comparison of measures of association for 2x2 data was carried out by computer analysis. For each of 1,539 tables, 14 measures of association were calculated and evaluated. A measure based on the odds-ratio (Chambers, 1982) was most accurate in capturing the rho underlying a majority of the tables. (Author/BW)

In the present study Monte Carlo methods were employed to evaluate the degree to which canonical function and structure coefficients may be differentially sensitive to sampling error. Sampling error influences were investigated across variations in variable and sample (n) sizes, and across variations in average within-set correlation sizes and in…

This article proposes a method for testing equality of variances which exploits Pitman's idea and the computational power of simulations. Several advantages to this method are illustrated. A Monte Carlo study for several combinations of sample sizes and number of variables is presented. (Author/LMO)

A computer simulation compared significance tests of correlation coefficients calculated from initial scores, from ranks assigned by the Spearman method, and from three kinds of modified ranks. Implications of findings for the idea that rank correlation is a nonparametric correlation method are discussed. (SLD)

Computer simulations indicate that the correlation between baseline and change, by itself, does not invalidate the use of gain scores to measure change, but when the negative correlation is accompanied by decrease in variance from pretest to posttest, covariance is a superior measure of change. (SLD)

This study examined three statistical methods for selecting items for mastery tests: (1) pretest-posttest; (2) latent trait; and (3) agreement statistics. The correlation between the latent trait method and agreement statistics, proposed here as an alternative, was substantial. Results for the pretest-posttest method confirmed its reputed…

Four methods for determining the dimensionality of a set of test items were compared: (1) linear factor analysis; (2) residual analysis; (3) nonlinear factor analysis; and (4) Bejar's method. Five artificial test data sets (for 40 items and 1500 examinees) were generated, consistent with the three-parameter logistic model and the assumption of…

Multidimensional item response data were created. The strength of a general factor, the number of common factors, the distribution of items loadingon common factors, and the number of items in simulated tests were manipulated. LOGIST effectively recovered both item and trait parameters in nearly all of the experimental conditions. (Author/JAZ)

Simulated data typical of those from meta analyses are used to evaluate the reliability, Type I and Type II errors, bias, and standard error of the meta-analytic procedures of Schmidt and Hunter (1977). Concerns about power, reliability, and Type I errors are presented. (SLD)

The TESTFACT computer program and full-information factor analysis of test items were used in a computer simulation conducted to correct for the guessing effect. Full-information factor analysis also corrects for omitted items. The present version of TESTFACT handles up to five factors and 150 items. A preliminary smoothing of the tetrachoric…