A recent multidimensional scaling analysis of the Test of English-as-a-Foreign-Language (TOEFL) item response data identified clusters of items in the test sections that, being more homogeneous than their parent sections, might be better for diagnostic use. The analysis was repeated using different scoring techniques. Results diverged only for…

Item and factor analyses examined the psychometric characteristics of the White Racial Identity Attitude Scale based on data from 308 white college students. Then five trained judges assigned scale items to appropriate subscales. Results did not support the psychometric adequacy of the scale. (SK)

The capability of the DIMTEST statistical test in assessing essential unidimensionality of item responses to real tests was investigated for 22 real tests of at least 25 items and 700 or more examinees. DIMTEST results on real tests were able to discriminate between essentially unidimensional and multidimensional tests. (SLD)

Examined five Rasch-model-based item-fit indices in terms of their distributional properties and the power of detecting item bias or differential item functioning. Results indicate that, although these five standardized item-fit indices did not depart significantly from a normal distribution, the Type I error rates were not reasonable. (Author/SLD)

Reviews developments in nonparametric item-response theory (NIRT), from its historic origins in item-response theory (IRT) and scale analysis to new theoretical results for practical test construction. Discusses theoretical results from NIRT often relevant to IRT. Contains 134 references. (SLD)

Proposes two new methods for item-exposure control, the Progressive method and the Restricted Maximum Information method. Compares both methods with six other item-selection methods. Discusses advantages of the two new methods and the usefulness of combining them. (SLD)

This book provides a historical overview of item-response theory (IRT) and contains a compendium of item-response-theory research. Each of its six sections deals with a unique application of a model or a family of models. Requires a strong psychometric background to understand some of the discussions. (SLD)

Nurses' ability to think critically in clinical situations cannot be assessed by multiple-choice tests. Context-dependent test items that assess critical thinking are useful for orientation of new staff, competency testing, and clinical staff development. (SK)

Summarizes the benefits of computerized assessment and provides a review of some practical issues concerning measurement, item and examinee characteristics, hardware, and software. Adequate measures of reliability and validity have been established for many computer-based tests, and the benefits of computer testing have been realized in applied…

This brief history traces the development of item response theory (IRT) from concepts originating in 19th-century mathematics and psychology to present-day principles drawn from statistical estimation theory. Connections to other fields and current trends in IRT are outlined. (SLD)

To test the validity of an achievement test, 77 Virginia business computer applications students answered items on vocabulary, access software, data/text entry, editing, and formatting. Teachers said that only 45% of the items were being taught; 59 of 60 word processing items were deemed instructionally valid. (SK)

This special issue is devoted to person-fit analysis, which is also referred to as appropriateness measurement. An introduction to person-fit research is given. Several types of aberrant response behavior on a test are discussed; and whether person-fit statistics can be used to detect dominant score patterns is explored. (SLD)

Some specific person-fit results for the Rasch model are presented, followed by an application to a test measuring knowledge of reasoning with logical quantors. Some issues are relevant to all attempts to use person-fit statistics in research, but the special role of the Rasch model is highlighted. (SLD)

The separate calibration t-test approach of B. Wright and M. Stone (1979) and the common calibration between-fit approach of B. Wright, R. Mead, and R. Draba (1976) appeared to have similar Type I error rates and similar power to detect item bias within a Rasch framework. (SLD)

Studied the appropriateness and implications of incorporating a testlet definition into the estimation of procedures of the conditional standard error of measurement (SEM) for tests composed of testlets. Simulation results for several methods show that an item-based method using a generalizability theory model provided good estimates of the…