Studied the problem of missing data at the second level of a two-level hierarchal data structure using data generated to simulate the 1982 High School and Beyond data set with five different missing data treatments: listwise deletion, overall mean substitution, group mean substitution, the EM algorithm, and multiple imputation. (SLD)

Presents results of a connectionist simulation that modeled the reanalysis of the Latin gender system in its transition to Old French. The network reanalysis was based solely on formal cues and on frequency. Results are in accordance with the historical data, and certain errors in simulations are also amenable to principled explanations.…

Developed a method to determine the required number and position of sensors for human brain electromagnetic source analysis. Studied the method through a simulation study and an empirical study on visual evoked potentials in one adult male. Results indicate the method is fast and reliable and improves source precision. (SLD)

Developed a sandwich-type covariance matrix to evaluate parameters and provide a rescaled statistic to evaluate a hypothesized general multilevel model. A simulation study with a two-level factor model implies that the rescaled statistic is preferable. (SLD)

Extended conventional two-group differential item function (DIF) analysis for dichotomous items to factorial DIF analysis for polytomous items where multiple grouping factors with multiple groups in each are analyzed jointly. Simulation studies and analysis of a real data set with 1,924 subjects show the parameters of the proposed modeling can be…

Describes a writing assignment requiring students to extract information from a simulated telephone conversation and construct a notice of an upcoming meeting. (JAD)

Explores two methods for the analysis of binary data, and presents a proposal for adapting these data to matched-pair data from school intervention studies. Evaluates the performance of the two methods through simulation and discusses conditions under which each method may be used. (SLD)

Evaluated three types of multidimensional item response theory (MIRT) linking methods through two simulation studies. Results indicate that the best MIRT linking method was an unbiased, effective, and consistent estimator that produced accurate estimates of transformation parameters when errors in estimation of item parameters were manipulated…

Provides a framework for viewing local dependency within dichotomous and polytomous items that are clustered by design and presents a testing procedure that allows researchers to identify individual item pairs that exhibit local dependency while controlling for a false positive rate. Simulation results indicate that the proposed method is…

Studied the effectiveness of various analytical formulas for estimating "R" squared shrinkage in multiple regression analysis, focusing on estimators of the squared population multiple correlation coefficient and the squared population cross validity coefficient. Simulation results suggest that the most widely used Wherry (R. Wherry,…

Evaluated multidimensional linking procedures based on a framework recently proposed by T. Davey, T. Oshima, and K. Lee (1966): (1) the Direct method; (2) the Equated Function method; (3) the Test Characteristic Function method; and (4) the Item Characteristic Function method. Simulation results indicate advantages to the last two methods, but all…

Restricted the range of nonnormal distributions by eliminating scores above a designated cutoff value or eliminating scores above or below the mean by a certain distance. Results of a simulation study show that range restriction sometimes increased the correlation between variables having outlier prone distributions. Discusses practical…

Conducted a simulation study to determine how well two models for local item dependency, the constant order interaction and the dimension-dependent interaction models, could be distinguished. Results suggest that when there is good reason to suspect a dimension-dependent interaction, it is possible to detect it if the sample size is at least 500.…

Developed an approximate statistical test for the hypothesis of equality between the Spearman-Brown extrapolations of two independent values of Cronbach's alpha reliability coefficient. Monte Carlo simulations demonstrate that the procedure effectively controls Type I error. (SLD)

Describes an idiosyncracy of the MULTILOG (D. Thissen, 1991) parameter estimation process discovered during a simulation study involving the graded response model. A misordering reflected in boundary function location parameter estimates resulted in a large negative contribution to the true score followed by a large positive contribution. These…