An important determinant of equating process efficiency is the correlation between the anchor test and components of each form. Use of some monotonic function of this correlation as a measure of equating efficiency is suggested. A model relating anchor test length and test reliability to this measure of efficiency is presented. (Author/DWH)

This paper discusses various issues involved in using the Rasch Model with multiple-choice tests and questions the suitability of this model for multiple-choice items. Results of some past studies supporting the model are shown to be irrelevant. The effects of the model's misfit on test equating are demonstrated. (Author JAZ)

Linear conversions were developed relating scores on recent forms of the Graduate Record Examinations. Conversions based on specially selected subpopulations were compared with total-group conversions and evaluated. Conclusions indicated that the data clearly support the assumption of population independence for homogenoeous tests, but not quite…

The test equating accuracy of content-representative anchors (subsets of items in common) versus nonrepresentative, but substantially longer, anchors was compared for a professional certification examination. Through a chain of equatings, it was found that content representation in anchors was critical. (Author/GDC)

Preceeding through the development of a sequence of paired calibrations determined by a hierarchical clustering algorithm, the proposed method of constructing a universal criterion scale does not rely on covariate information. The procedure is illustrated with data from American law schools. (Author/BW)

A universal criterion scale is applied to the direct estimation of validity for various groups of law school applicants and to the testing of the assumptions underlying the more familiar extrapolation methods. Data from American law schools are used to illustrate the procedures. (Author/BW)

Three methods of test equating (linear, equipercentile, and item response theory) were investigated with respect to the issue of scale drift. Results indicate that all three models work well in limited settings but that the item response theory approach provided the most stable results overall. (JKS)

Discusses, with illustrations, the difficulty of estimating college students' mathematical ability from the results of one or two examinations. (MM)

There is an urgent need for more research in predictive measures to test the ability of applicants for training in occupational therapy. (CK)

Functional-level testing is fundamental to the evaluation models of the Elementary Secondary Education Act Title I Evaluation and Reporting System, although criteria for determining what is a functional-level test are unclear. Reliability and validity of test scores in the range of chance is at issue. (Author/RL)

Three levels of a mathematics computation test were equated using the Rasch model. Sixth, seventh, and eighth graders were administered different levels of the test. Lack of consistency among equatings suggested that the Rasch model did not produce a satisfactory vertical equating of this computation test. (Author/RD)

Five indices are discussed that should logically discriminate between situations in which: (1) the linear equating method (LEM) adequately adjusts for difference between score distributions of two approximately parallel test forms; or (2) a method more complex than the linear equating method is needed. (RL)

Discusses the use of scaling test scores for an algebra class. Provides example data, several equations used in scaling, and graphs. (YP)

In applying I. I. Bejar's method for detecting the dimensionality of achievement tests, researchers should be cautious in interpreting the slope of the principal axis. Other information from the data is needed in conjunction with Bejar's method of addressing item dimensionality. (SLD)

An experiment is reported in which 225 high school students were asked to choose among several multiple-choice items but then were required to answer them all. It is concluded that allowing choice while having fair tests is only possible when choice is irrelevant in terms of difficulty. (SLD)