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)

Though item bias statistics are widely recommended for use in test development and analysis, problems arise in their interpretation. This research evaluates logistic test models and computer simulation methods for providing a frame of reference for interpreting item bias statistics. Specifically, the intent was to produce simulated sampling…

It is postulated that mean square fit statistics do not adequately evaluate the quality of test items adequately. The fit statistics arise from the desire to have statistical (computerized) procedures to assess item quality transcending the group on which they are computed. It appears that none of the current formulations meet these criteria. The…

Uses principles of latent trait theory to derive conditions under which mean differences between groups on an observed variable are in fact indicative of mean differences on an underlying construct. Pays particular attention to the logistic curve as a model for describing the relation between the observed variable and the construct. (Author/CB)

In most item response theory models a particular mathematical form is assumed for all item characteristic curves, e.g., a logistic function. It could be desirable, however, to estimate the shape of the item characteristic curves without prior restrictive assumptions about its mathematical form. We have developed a practical method of estimating…

This paper summarizes three studies that illustrate how application of the three-parameter logistic test model helped solve three relatively intractable testing problems. The three problems are: designing a multi-purpose test, evaluating an multi-level test, and equating a test on the basis of pretest statistics. (Author/JKS)

This primer is an introduction to item response theory (also called item characteristic curve theory, or latent trait theory) as it is used most commonly--for scoring multiple choice achievement or aptitude tests. Written for the testing practitioner with minimum training in statistics and psychometrics, it presents and illustrates the basic…

Ways in which the Statistical Package for the Social Sciences (SPSS) can be used to perform some Rasch analyses are described in detail. It is shown how SPSS and a set of item calibrations can be used to estimate person abilities, standard errors of measurement, test characteristic curve, test information curve, classification consistency on a…

Item response theory bias detection procedures were applied to data from Black and White seniors on the High School and Beyond data files. Overall, the sums-of-squares statistics (weighted by the inverse of the variance errors) were the best indices for quantifying item characteristic curve differences between groups (Author/BW)

New ability estimators have been proposed by Wainer and Wright (1980) and Mislevy and Bock (1981) that are resistant against guessing and careless behaviors exhibited by some examinees. This paper presents another class of ability estimators that are resistant to departures from the underlying assumptions concerning guessing and carelessness. In…

Bayesian techniques are adapted to the estimation of stimulus-response curves. Illustrative examples deal with estimation of person characteristic curves and item characteristic curves in the context of mental testing, and with estimation of a stimulus-response curve using data from a psychophysical experiment. (Author/JKS)

Some examinees' test-taking behavior may be so idiosyncratic that their scores are not comparable to the scores of more typical examinees. Appropriateness indices, which provide quantitative measures of response-pattern atypicality, can be viewed as statistics for testing a null hypothesis of normal test-taking behavior against an alternative…

Item response curves for a set of binary responses are studied from a Bayesian viewpoint of estimating the item parameters. For the two-parameter logistic model with normally distributed ability, restricted bivariate beta priors are used to illustrate the computation of the posterior mode via the EM algorithm. (Author/LMO)

The appropriateness of the use of the standardized residual (SR) to assess congruence between sample test item responses and the one parameter latent trait (Rasch) item characteristic curve is investigated. Latent trait theory is reviewed, as well as theory of the SR, the apparent error in calculating the expected distribution of the SR, and…

An elementary survey of item characteristic curve theory, centered around the problems of individualized (tailored) testing, is presented. Following the introduction, discussions are provided of the following: Test Theory for Itemized Tests; The Guttman Scale; Item Characteristic Curve Theory; An Alternative Model; Specialization, Application, and…