The process by which a school might use a parameter estimation procedure (such as LOGIST) based on item response theory (IRT) to build a "customized test" from an "on shelf" standardized achievement test was examined. The Iowa Tests of Educational Development was used. Results indicate that estimates from standardized versus…

Analysis of data from standard affective measures showed both an orthogonally rotated and an obliquely rotated factor solution yielded psychological dimensions that correspond to instrument's scales. Study Attitudes and Methods Survey was studied, using data on a total of 483 eleventh graders. Both solutions afforded comparable inferences about…

Test Pac, a test scoring and analysis computer program for moderate-sized sample designs using dichotomous response items, performs comprehensive item analyses and multiple reliability estimates. It also performs single-facet generalizability analysis of variance, single-parameter item response theory analyses, test score reporting, and computer…

Lord's method for detecting subgroup bias at the item level uses a three-parameter item characteristic curve model. A chi square statistic is computed on the multivariate differences between the parameter estimates of item discrimination and difficulty. The LOGIST program and additional programs written in BASIC are used. (Author/GDC)

When the parameters of item response data are estimated by a latent trait model, some variation will remain unaccounted. If baseline graphical configurations of the residual variation can be constructed, measurement irregularities may be exposed. An analytic strategy for the graphical representation and analysis of Rasch model residuals is…

This study examined three statistical methods for selecting items for mastery tests: (1) pretest-posttest; (2) latent trait; and (3) agreement statistics. The correlation between the latent trait method and agreement statistics, proposed here as an alternative, was substantial. Results for the pretest-posttest method confirmed its reputed…

Standard maximum likeliheed estimation was compared using two forms of robust estimation, BIWEIGHT (based on Tukey's Biweight) and AMTJACK (AMT-Robustified Jackknife), and Rasch model person analysis. The two procedures recovered the generating parameters, but Rasch person analysis also helped to identify the nature of a response disturbance. (GDC)

Nine indices of response pattern appropriateness were compared, based on item response theory, with respect to their relation to the total test score and their effectiveness in detecting unusual response patterns. The ability to differentiate between three groups was compared: cooperative examinees, uncooperative examinees, and randomly generated…

Sato's Caution Index is a method of analyzing item response patterns independently of total test score to provide diagnostic information about students. This research supports the validity of the diagnostic interpretations of the Index when compared to graduate students' self evaluation. (Author/BS)

The stability was evaluated of a partial correlation index, comparisons of item characteristic curves, and comparisions of item difficulties in assessing race and sex differences in the performance of verbal items on the Graduate Record Examination Aptitude Test. All three indexes exhibited consistency in identifying the same items in different…

The purpose of this study was to compare the relative robustness of the one-, two-, and modified two-parameter latent trait logistic models for the Iowa Tests of Educational Development. Results suggest that the modified two-parameter model may provide the best representation of the data. (Author/BW)

A unidimensional latent trait model was used to test a single-factor hypothesis of the Lawson Classroom Test of Formal Reasoning. The test failed to provide a valid measure of formal reasoning. This was a result of test format which neglected aspects of formal reasoning emphasized by Inhelder and Piaget. (Author/DWH)

Response data from 3,000 subjects to the SRA Attitude Survey were analyzed using the one-parameter and two-parameter logistic latent trait models. The effects of varying sample size on the accuracy of person and item parameters were investigated. (Author/BS)

A computerized adaptive testing procedure for measuring mathematical skills of 60 deaf college students was evaluated. Ability estimates were compared for stability to those obtained earlier from conventional testing. Adaptive testing increased measurement efficiency by reducing test length. Student attitudes toward computerized testing were…

The relationship between item parameter values obtained from independent Birnbaum (l968) model calibrations of the same item set from two different samples is examined. (Author)