To clarify the findings of R. Gillett (1991) about the inequality of the means of test scores of minority and majority examinees, the standard errors of the quota-selected sample means and the sampling distribution of these means were studied through Monte Carlo simulation. Results explain that the quota selection inequality results from…

The use of p-values in combining results of studies often involves studies that are potentially aberrant. This paper proposes a combined test that permits trimming some of the extreme p-values. The trimmed statistic is based on an inverse cumulative normal transformation of the ordered p-values. (SLD)

The derivation of the variance of similarity judgments is made from the 3-D process in nonlinear psychophysics. The idea of separability of dimensions in metric space theories of similarity is replaced by one parameter that represents the degree of a form of interdimensional cross-sampling. (SLD)

A Monte Carlo study was conducted using simulated dichotomous data to determine the effects of guessing on Rasch item fit statistics and the Logit Residual Index. Results indicate that no significant differences were found between the mean Rasch item fit statistics for each distribution type as the probability of guessing the correct answer…

The main goal of this study was to assess how sensitive unidimensional parameter estimates derived from BILOG were when the unidimensionality assumption was violated and the underlying ability distribution was not multivariate normal. A multidimensional three-parameter logistic distribution that was a straightforward generalization of the…

Several person-fit statistics have been proposed to detect item score patterns that do not fit an item response theory model. To classify response patterns as not fitting a model, a distribution of a person-fit statistic is needed. The null distributions of several fit statistics have been investigated using conventionally administered tests, but…

The functioning of two population-based Mantel-Haenszel (MH) common-odds ratios was compared. One ratio is conditioned on the observed test score, while the other is conditioned on a latent trait or true ability score. When the comparison group distributions are incongruent or nonoverlapping to some degree, the observed score represents different…

Type I error rate and power for the t test, Wilcoxon-Mann-Whitney test, van der Waerden Normal Scores, and Welch-Aspin-Satterthwaite (W) test are compared for two simulated independent random samples from nonnormal distributions. Conditions under which the t test and W test are best to use are discussed. (SLD)

Intraclass correlation and Cronbach's alpha are widely used to describe reliability of tests and measurements. Even with Gaussian data, exact distributions are known only for compound symmetric covariance (equal variances and equal correlations). Recently, large sample Gaussian approximations were derived for the distribution functions. New exact…

This paper describes a study, now in progress, of new methods for representing the sampling variability of Mantel-Haenszel differential item functioning (DIF) results, based on the system for categorizing the severity of DIF that is now in place at the Educational Testing Service. The methods, which involve a Bayesian elaboration of procedures…

Validity generalization meta-analysis (VG) examines the extent to which the validity of an instrument can be transported across settings. VG offers correction and summarization procedures designed in part to remove the effects of statistical artifacts on estimates of association between criterion and predictor. By employing a random effects model,…

This study compared the model of J. E. Grizzle, C. F. Starmer, and G. G. Koch (GSK, 1969) and log-linear model-based approaches for testing hypotheses in r x c contingency tables. Tables were simulated under various conditions of table, sample, row-effect size, and column-effect size. Test statistics for column (main) and interaction effects were…

The effects of sample size, number of variables, and population value of the congruence coefficient on the sampling distribution of the congruence coefficient were examined. Sample data were generated on the basis of the common factor model, and principal axes factor analyses were performed. (Author/LMO)

This paper is concerned with the construction of effect size standard errors in situations where the effect sizes are independent but the data have likely been sampled from non-normal distributions, and possibly for different studies, from different families of non-normal distributions. Asymptotic distribution-free estimators are provided for two…

A method of item pool design is proposed that uses an optimal blueprint for the item pool calculated from the test specifications. The blueprint is a document that specifies the attributes that the items in the computerized adaptive test (CAT) pool should have. The blueprint can be a starting point for the item writing process, and it can be used…