The multiple-choice science exercises used by the National Assessment of Educational Progress include an "I Don't Know" (IDK) alternative to estimate more accurately knowledge of groups of respondents. Group percentages of IDK responses were examined and compared with correct responses to see if the IDK introduces bias. Variance common…

In order to determine if Rasch Model procedures have any utility for equating pre-existing tests, this study reanalyzed the data from the equating phase of the Anchor Test Study which used a variety of equipercentile and linear model methods. The tests involved included seven reading test batteries, each having from one to three levels and two…

Two approaches--Harnisch/Linn and Rasch-based--to comparing district-level student performance to system-level student performance were examined, using results for 3,607 students on a curriculum-referenced mathematics test. Both approaches addressed relative mean performance on categories of common items. A modification of one approach and an…

A Monte Carlo study was conducted to compare the performance of three statistical indices of test item bias in small samples of examinees. The statistical indices compared were the Delta method, the Mantel-Haenszel (MH) method, and the Standardization method. Sample sizes of 50, 100, and 200 were examined. One thousand samples of each size were…

A predictive adaptive testing (PAT) strategy was developed based on statistical predictive analysis, and its feasibility was studied by comparing PAT performance to those of the Flexilevel, Bayesian modal, and expected a posteriori (EAP) strategies in a simulated environment. The proposed adaptive test is based on the idea of using item difficulty…

This study examined whether an expert system approach involving intelligent selection of items (EXSPRT-I) is as efficient as item response theory (IRT) based three-parameter adaptive mastery testing (AMT) when there are enough subjects to estimate the three IRT item parameters for all items in the test and when subjects in the item parameter…

Two methods of using collateral information from similar institutions to predict college freshman grade average were investigated. One central prediction model, referred to as pooled least squares with adjusted intercepts, assumes that slopes and residual variances are homogeneous across selected colleges. The second model, referred to as Bayesian…

This paper explores various procedures of panel data in the estimation of causal models. The reported analyses are from the Monitoring the Future study, a nationwide questionnaire survey of 16,000 to 17,000 high school seniors conducted annually since 1975. First, the parameters of causal models are estimated in which the dependent variables are…

This study investigated seven methods for analyzing multivariate group differences. Bonferroni t statistics, multivariate analysis of variance (MANOVA) followed by analysis of variance (ANOVA), and five other methods were studied using Monte Carlo methods. Methods were compared with respect to (1) experimentwise error rate; (2) power; (3) number…

Latent trait model calibration procedures were used on data obtained from a group testing program. The one-parameter model of Wright and Panchapakesan and the three-parameter logistic model of Wingersky, Wood, and Lord were selected for comparison. These models and their corresponding estimation procedures were compared, using actual and simulated…

This study attempts to determine whether the pattern of human abilities varies or remains constant over a range of ability levels. Scores on 12 aptitude and achievement tests for 11,743 subjects, subdivided into four groups according to intelligence and socioeconomic status, were used. A technique, developed by Joreskog, for simultaneously factor…

The philosophical underpinning of the purposes of the community college, an examination of the basis for state financial support from a philosophical viewpoint, and an analysis of the ways in which the various states have implemented that philosophy in financial support programs are the central focus of this report. The four current patterns of…

Two studies were carried out to evaluate the quality of multidimensional item response theory (MIRT) model parameter estimates obtained from the computer program NOHARM. The purpose of the first study was to compute empirical estimates of the standard errors of the parameters. In addition, the parameter estimates were evaluated for bias and the…

This paper is the second report on the purpose, methodology, and results of a large-scale statistical analysis of standardized test scores from Chicago public elementary schools for 1988 and 1989. Major purposes were to develop methods for measuring student improvement (change) at the individual and school levels, to find ways of identifying…

"Postdictions" of college grade point averages (GPAs) from high school GPAs and/or American College Testing Program (ACT) composite scores were compared by gender and race for 1,121 freshmen and sophomore students at a West Virginia College. The sample included 945 white and 176 black students, and 553 males and 568 females. Significant…