Item response theory (IRT) refers to a family of mathematical models which describe the relationship between latent continuous variables (attributes or characteristics) and their manifestations (dichotomous/polytomous observed outcomes or responses) with regard to a set of item characteristics. Researchers typically use parametric IRT (PIRT)…

A model integrating latent trait and latent class theories in characterizing individual performance on the basis of qualitative understanding is presented. This HYBRID model is illustrated through experiments with 119 Air Force technicians taking a paper-and-pencil test and 136 Air Force technicians taking a computerized test. (SLD)

This paper focuses on the questions which arise as test practitioners monitor score scales derived from latent trait theory. Large scale assessment programs are dynamic and constantly challenge the assumptions and limits of latent trait models. Even though testing programs evolve, test scores must remain reliable indicators of progress.…

The most familiar models of item response theory are defined at the level of individual subjects. It is possible, however, to define such models for groups of subjects. This paper discusses group-level item response models, their uses, and their relationships to subject-level models. (Author/JKS)

A statistical model is proposed that describes the determination of an educational outcome variable as a nonlinear function of explanatory variables defined at different levels of a survey data hierarchy, such as students and classes. The theoretical and practical derivation of the model is discussed, and an example is given. (Author/BW)

The 14 chapters of this book focus on the technical advances, advances in applied settings, and emerging topics in the testing field. Part 1 discusses methodological advances, Part 2 considers developments in applied settings, and Part 3 reviews emerging topics in the field of testing. Part 1 papers include: (1) "Advances in…

The theoretical differences between the traditional definition of dimensionality and the more recently defined notion of essential dimensionality are presented. Monte Carlo simulations are used to demonstrate the utility of W. F. Stout's procedure to assess the essential unidimensionality of the latent space underlying a set of terms. The…

The National Assessment of Educational Progress (NAEP) makes possible comparison of groups of students and provides information about what these groups know and can do. The scale anchoring techniques described in this chapter address the latter purpose. The direct method and the smoothing method of scale anchoring are discussed. (SLD)

In this paper, algorithms are described for obtaining the maximum likelihood estimates of the parameters in log-linear models. Modified versions of the iterative proportional fitting and Newton-Raphson algorithms are described that work on the minimal sufficient statistics rather than on the usual counts in the full contingency table. This is…

It is suggested that some of the technology applied to state Scholastic Aptitude Test scores to measure states' educational performance (particularly use of a truncated Gaussian model) may make it possible to adjust National Assessment of Educational Progress (NAEP) scores to make inferences about state educational progress possible. (SLD)

The potential value of developing item banks that would both guarantee the quality of teacher-made examinations and allow for national comparison of student achievement is described. Use of Rasch model methods, rather than classical item analysis techniques, to calibrate the items for these banks is illustrated. (BS)

The need for scale linking in the National Assessment of Educational Progress (NAEP) is discussed, and the specific procedures used to carry out the linking in the context of the major analyses of the 1990 NAEP mathematics assessment are described. Issues remaining to be addressed are outlined. (SLD)

Across the field of educational assessment the case has been made for alternatives to the multiple-choice item type. Most of the alternative types of items require a subjective evaluation by a rater. The reliability of this subjective rating is a key component of these types of alternative items. In this paper, measures of reliability are…

Title I evaluations using the RMC Model C design depend for their interpretation on the assumption that the regression of posttest on pretest is linear across the cut score level when there is no treatment; but there are many instances where nonlinearities may occur. If one applies the analysis of covariance, or model C analysis, large errors may…

Results from the Maryland School Performance Assessment Program for 5,392 elementary school students and from the Comprehensive Tests of Basic Skills (multiple choice) for a national sample are used to explore local item independence (LID) of test items. Some strategies are suggested for measuring LID in performance assessments. (SLD)