The interaction of subjects with test items is investigated. It is suggested that psychometricians rely too heavily on inferences about the nature of the interaction. An approach to data analysis is proposed which is more directly related to this interaction. (JKS)

An approach to partitioning the variance in student ratings not previously reported in the literature is described. The new approach provides an alternative basis for interpreting faculty evaluations that overcome objections to current practices. Student responses on evaluation forms were cluster-analyzed to establish homogeneous subgroups of…

The MERMAC computer program is offered to the University of Wisconsin faculty for use in scoring and analyzing classroom tests. The characteristics of a good test are discussed; examples are given of the output of the MERMAC program; and the results are used to show how the quality of a test may be improved. Although the MERMAC Program is for…

A study examined the appropriateness of two methods for relating the enhanced American College Testing Program (ACT) Assessment Composite and Scholastic Aptitude Test (SAT) total scores: equipercentile concordance and regression. Study results, potential inaccuracies, and the usefulness of institutional concordances based on their own data are…

Confirmatory multidimensional item response theory (CMIRT) was used to assess the structure of the Graduate Record Examination General Test, about which much information about factorial structure exists, using a sample of 1,001 psychology majors taking the test in 1984 or 1985. Results supported previous findings that, for this population, there…

Because change is fundamental to education and the measurement of change assesses the quality and effectiveness of postsecondary education, this study examined three methods of measuring change: (1) gain scores; (2) residual scores; and (3) repeated measures. Data for the study was obtained from transcripts of 722 graduating seniors at the…

Since all commonly used latent trait models assume a unidimensional test, the applicability of the procedure to obviously multidimensional tests is questionable. This paper presents the results of the application of latent trait, traditional, and factor analyses to a series of actual and hypothetical tests that vary in factoral complexity.…

An approach is described that reveals the hierarchical test structure (HTS) based on the cognitive demands of the test items, and conducts a linear trait modeling by using the HST elements as item difficulty components. This approach, referred to as the Hierarchical Latent Trait Approach (HLTA), employs an algorithm that allows all test items to…

A basic structure approach is proposed for obtaining multidimensional scale values for attitude, achievement, or personality items from response data. The technique permits the unconfounding of scale values due to response bias and content and partitions item indices of popularity or difficulty among a number of relevant dimensions. (Author/BH)

Discusses the use of the WORDS computer program in scoring writing assignments. Explains that computerized analysis has identified factors in experienced writers' essays that graders favor but has failed to distinguish clearly between basic and average writers' works. Concludes that holistic scoring can work effectively for student placement, but…

Scores representing number of items right and self-perceptions were analyzed for a nonquantitative examination that was assembled into three forms. Multivariate ANCOVA revealed no significant effects for the cognitive measure. However, significant sex and sex x order effects were found for perceptions scores not parallel to those reported…

Based upon prior research in the College of Business and Public Administration and the first year of the university testing program at an upper division, midwestern university, implications for testing of older transfer students are drawn. Students should be tested before, or during, their first term of academic history. If remediation is not…

Effect sizes have been increasingly emphasized in research as more researchers have recognized that: (1) all parametric analyses (t-tests, analyses of variance, etc.) are correlational; (2) effect sizes have played an important role in meta-analytic work; and (3) statistical significance testing is limited in its capacity to inform scientific…

Development of new scales from test items usually requires several hundred cases in each of the defined groups. Application of multivariate techniques to existing test scales also encounters such problems as multicolinearity, combining ipsative with normative measures and ordinal with equal interval scales, differential prediction, and…

This study used the residual gain statistic in reading comprehension research in English as a second language to assess individual differences in improvement accruing from learning and training. The subjects were 34 foreign college students with a variety of native languages. A 35-item cloze test with every eighth word eliminated, but with the…