Causes of local item dependence in a large-scale performance assessment were studied using data from the Maryland School Performance Assessment Program. Contextual characteristics (content and response requirements) were identified to differentiate locally independent and dependent item clusters. Hypothesized explanations are offered for high…

Explores the asymptotic theory of a method of nonparametric item characteristic curve estimation based on kernel smoothing with the theory of obtaining the proper linear ordering of examinees with respect to their true latent abilities. Results support the usefulness of estimates of the type produced by the TESTGRAF program. (SLD)

This study compared the Mantel-Haenszel procedure, the simultaneous item bias approach (SIB), and the logistic regression approach (LR) with respect to Type I error rates and power to detect nonuniform differential item functioning (DIF). The SIB and LR procedures were equally powerful in detecting nonuniform DIF. (SLD)

Compared selected methods of assessing the structure of tests with dichotomous items using real data from a 62-item test of reading ability and computer-generated data for multiple unidimensional and multidimensional cases. All methods performed reasonably well over a relatively wide range of conditions. (SLD)

Examined the case of adaptive testing under a multidimensional response model with large numbers of constraints on the content of the test and items selected using a shadow test approach. Illustrated the procedure with five different cases of multidimensionality that differ in the numbers of ability dimensions and test structure with respect of…

Presents an alternative to measuring student overall satisfaction that uses, rather than a simple "yes/no" question, a multiple-item weighted gap score analysis approach. Results suggest that this approach may have more diagnostic value to both academics and practitioners. (EV)

Studied the results provided by two strategies for fitting data to Latent Trait Theory models, Total-Persons-Items (TPI) and Total-Items-Persons (TIP). To assess these strategies, 30 items measuring religious attitudes were administered to 821 persons. Results show that TPI maximizes the number of persons with good fit, and TIP maximizes the…

Outlines an approach to differential item functioning (DIF) analysis using hierarchical linear regression that makes it possible to combine results of logistic regression analyses across items to identify consistent sources of DIF, to quantify the proportion of explained variation in DIF coefficients, and to compare the predictive accuracy of…

Five approaches to writing test items with computer assistance are described. A model of knowledge using a set of structures and a system for implementing the scheme are outlined. The approaches include the author-supplied approach, replacement-set procedures, computer-supplied prototype items, subject-matter mapping, and discourse analysis. (TJH)

Multiple-choice response and scoring methods that attempt to determine an examinee's degree of knowledge about each item in order to produce a total test score are reviewed. There is apparently little advantage to such schemes; however, they may have secondary benefits such as providing feedback to enhance learning. (SLD)

Two-group classification is discussed when a unidimensional latent trait "theta" is appropriate for explaining data. If data have a monotone likelihood ratio, then optimal allocation rules can be based on its magnitude when allocation must be made to one of the two groups related to the unidimensional latent trait. (SLD)

Examined the degree to which coefficient alpha is affected by including items with different distribution shapes within a unidimensional scale. Computer simulation results indicate that reliability does not increase dramatically as a result of using differentially shaped items within a scale. Discusses implications for test construction. (SLD)

Investigates the structure of item-pair conditional covariances given a unidimensional latent composite using a generalized semiparametric-modeling approach. Conditional covariances are shown to be highly informative about the latent dimensionality required to achieve local independence. Uses the new approach to justify the DIMTEST procedure. (SLD)

Focuses on the practical effects of using different statistical treatments with omitted and not-reached items in an item-response theory application. The strategy selected for scoring such items has considerable impact on the interpretation of results for individual or group-level assessments. (Author/SLD)

Compared three methods for developing a common metric under item response theory through simulation. For smaller numbers of common items, linking using the characteristic curve method yielded smaller root mean square differences for both item discrimination and difficulty parameters. For larger numbers of common items, the three methods were…