A model based on latent trait theory, with maximum likelihood parameter estimates and associated asymptotic tests, is presented. Latent curve analysis is a method for representing development and is an alternative to repeated measures analysis of variance and first-order auto-regressive models. (SLD)

A polychotomous nominal response model-based computerized adaptive test (CAT) was simulated using data from 1,093 University of Texas students. The ability estimation of this model and its overall performance were compared with those of a dichotomous three-parameter logistic model-based CAT. Advantages and drawbacks of nominal response CAT are…

False positive and false negative error rates were studied for competency testing when failing examinees are permitted to retake the test. Formulas are provided for the beta-binomial and Rasch models. Estimates based on these models are compared for six data sets from the South Carolina Basic Skills Assessment Program. (SLD)

Common approaches to examining the relationship between multitrait-multimethod (MTMM) data and variables outside the MTMM data are compared: averaging various means of each trait and estimating LISREL computer program models, and estimating only relationships between MTMM traits and the outside variables. Problems of correlational bias are…

Application of the bootstrap method to complex psychological analysis is illustrated using a simulation experiment with two populations with small and large samples. The method provides variance estimates, allows testing of nested competing models, and gives a preliminary idea about parameter variability. (SLD)

Previous work on the mathematical and implementation details of the marginalized maximum likelihood estimation procedure is extended to encompass the marginalized Bayesian procedure for estimating item parameters of R. J. Mislevy (1986) and to communicate this procedure to users of the BILOG computer program. (SLD)

Monte Carlo methods are used to evaluate marginal maximum likelihood estimation of item parameters and maximum likelihood estimates of theta in the two-parameter logistic model for varying test lengths, sample sizes, and assumed theta distributions. Results with 100 datasets demonstrate the methods' general precision and stability. Exceptions are…

The following models which provide an estimate of the reliability of a rating scale are described: (1) disattenuation; (2) common factor; (3) external criterion; (4) treatment effects; and (5) regression. Such models are especially useful in selecting, evaluating, and training participants in clinical research. (SLD)

A reparameterization of a latent class model is presented to classify and scale nomial and ordered categorical choice data simultaneously. The model extension represents a nonhomogeneous population as a mixture of homogeneous subpopulations. Simulated data and data from a magazine preference survey of 347 college students illustrate the model.…

A method is proposed for calculating the sample size needed to achieve acceptable statistical power with a given test. The minimally important difference significant (MIDS) criterion for sample size is explained and supported with recommendations for determining sample size. The MIDS criterion is computationally simple and easy to explain. (SLD)

W. M. K. Trochim and others defend the record of the regression-discontinuity (RD) design and blur the statistical tests for treatment effect. Their Monte Carlo results show the problematic nature of RD and its potential bias. New testing strategies and restrictions for the application of RD are proposed. (SLD)

The observed probability "p" is the social scientist's primary tool for evaluating the outcome of statistical hypothesis tests. The small-sample accuracy of nonnull asymptotic distributions of several functions of "p" was studied. Implications for use of the approximations are discussed. (SLD)

The reliability of residual scores created in the analysis of covariance in pretest-posttest designs, adjusted for a singe covariate, is compared with that of simple difference scores as relevant parameters vary. Little gain in reliability is created by analysis of covariance, which also suffers from reliability problems. (SLD)

Modifications are proposed to the recently developed method of comparing one-step M-estimators of location corresponding to two independent groups that provides good control over the probability of Type I error even for unequal sample size, unequal variances, and different shaped distributions. Simulation results reveal cautions required. (SLD)

The extent to which log-linear smoothing could improve the accuracy of common-item equating by the chained equipercentile method in small samples of examinees was investigated with responses from a 100-item test and 93,283 examinees. Smoothing reduced the sample size required for a given degree of accuracy. (SLD)