The Mantel-Haenszel delta difference (MH D-DIF) and the standardized proportion difference (STD P-DIF) are two observed-score methods that have been used to assess differential item functioning (DIF) at Educational Testing Service since the early 1990s. Latentvariable approaches to assessing measurement invariance at the item level have been…

Often, important decisions regarding accountability and placement of students in performance categories are made on the basis of test scores generated from tests, therefore, it is important to evaluate the validity of the inferences derived from test results. One of the threats to the validity of such inferences is aberrant responding. Several…

Many standardized tests are now administered via computer rather than paper-and-pencil format. The computer-based delivery mode brings with it certain advantages. One advantage is the ability to adapt the difficulty level of the test to the ability level of the test taker in what has been termed computerized adaptive testing (CAT). A second…

Orlando and Thissen (2000, 2003) proposed an item-fit index, S-X[superscript 2], for dichotomous item response theory (IRT) models, which has performed better than traditional item-fit statistics such as Yen's (1981) Q[subscript 1] and McKinley and Mill's (1985) G[superscript 2]. This study extends the utility of S-X[superscript 2] to polytomous…

This paper focuses on mixtures of two binomials with one known success parameter. It is shown how moment estimators can be obtained for the remaining unknown parameters of such mixtures, and results are presented from a Monte Carlo study carried out to explore the statistical properties of these estimators. (PN)

The Feldt-Gilmer congeneric reliability coefficients make it possible to estimate the reliability of a test composed of parts of unequal, unknown length. The approximate standard errors of the Feldt-Gilmer coefficients are derived via a method using the multivariate Taylor's expansion. Monte Carlo simulation is employed to corroborate the…

Six experimental approaches to the problems of setting cutoff scores and choosing proper test length are briefly mentioned. Most of these methods share the premise that a test is a random sample of items, from a domain associated with a carefully specified objective. Each item is independent and is scored zero or one, with no provision for…

A computer simulation of Bayes' Theorem was conducted in order to determine the probability that an examinee was a master conditional upon his test score. The inputs were: number of mastery states assumed, test length, prior expectation of masters in the examinee population, and conditional probability of a master getting a randomly selected test…

This project is concerned with the development and implementation of some new statistical techniques that will facilitate a continuing input of information about the student to the instructional manager so that individualization of instruction can be managed effectively. The source of this informational input is typically a short…

A new method of analysis was used in the Michigan Educational Assessment Program to test minimum competencies in fourth grade reading achievement. This technique permitted a substantial decrease in testing time and costs. The original test consisted of 95 items measuring 19 objectives; mastery was indicated by correct responses to four out of the…

The Test of English as a Foreign Language (TOEFL) was administered to 26 hearing impaired college students, in order to test the assumption that the English-language deficiencies of hearing impaired students are similar to those of foreign students. The students were attending Gallaudet College's School of Preparatory Studies and were identified…