The power of the nonparametric person-fit statistic, U3, is investigated through simulations as a function of item characteristics, test characteristics, person characteristics, and the group to which examinees belong. Results suggest conditions under which relatively short tests can be used for person-fit analysis. (SLD)

A study of the ability of different models--including the classification techniques of discriminant analysis, logistic regression, and neural networks--to predict the academic success of master's degree students in business administration suggests that prediction is difficult, but that classification and nonparametric techniques may be…

Physical and chemical indicators of water quality monitored by Environmental Canada between 1977 and 1987 in the Niagara River at Niagara-on-the-Lake and in the Saint Lawrence River at Wolfe Island are analyzed for seasonal and annual variations. Results indicate that specific conductivity, sodium, and chloride have decreased significantly over…

Use of parallel analysis (PA), a selection rule for the number-of-factors problem, is investigated from the viewpoint of permutation assessment through a Monte Carlo simulation. Results reveal advantages and limitations of PA. Tables of sample eigenvalues are included. (SLD)

The asymptotic posterior normality of latent variable distributions is established under very general and appropriate hypotheses, providing a probabilistic basis for assessing ability estimation/prediction accuracy in the long test case, as well as a first step in making the Dutch Identity conjecture rigorous. (SLD)

Using an infinite item test framework, it is argued that the usual assumption of local independence should be replaced by a weaker assumption--essential independence. The usual assumption of unidimensionality is replaced by a weaker and more appropriate statistically testable assumption of essential unidimensionality. (TJH)

The advantages and disadvantages of new software for market segmentation analysis are discussed, and the application of this new, chi-square based procedure (CHAID), is illustrated. A comparison is presented of an earlier, binary segmentation technique (THAID) and a multiple discriminant analysis. It is suggested that CHAID is superior to earlier…

Two bootstrapping or resampling strategies were investigated to determine their applicability to estimating standard errors and ensuing confidence intervals on variance components in two-factor random analysis of variance models. In light of prior negative findings regarding the application of bootstrapping to this particular problem, a…

Comparisons of corresponding values of phi and kappa coefficients were made for 270 instances of data generated by a Monte Carlo technique to simulate a test-retest situation. Data were generated for distributions with the same mean but three different levels of standard deviation, standard error of measurement and cutting score. Ten samples of…

A computer program named Program STANDARD is presented and demonstrated. This program calculates the statistical significance of the overall agreement of the categorical assignments. The program is based on Light's statistic, G, for describing the conjoint agreement of many observers with correct or standard set of classifications on nominal…

A method of factor extraction specific to a binary matrix, illustrated here as a person-by-item response matrix, is presented. The extraction procedure, termed ERGO, differs from the more commonly implemented dimensionalizing techniques, factor analysis and multidimensional scaling, by taking into consideration item difficulty. Utilized in the…

Five distribution-free alternatives to parametric analysis of covariance are presented and demonstrated: Quade's distribution-free test, Puri and Sen's solution, McSweeney and Porter's rank transformation, Burnett and Barr's rank difference scores, and Shirley's general linear model solution. The results of simulation studies regarding Type I…

The use of nonparametric quartile regression in examining the impacts of social programs on the distributions of noncategorical outcome measures is demonstrated through the study of earnings effects and income effects in four evaluations of training and employment programs for welfare recipients. Distributional impact estimates yield estimates…

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)