ERIC Number: ED468754
Record Type: Non-Journal
Publication Date: 1992-Jun
Reference Count: N/A
The Effects of Dimensionality on True Score Conversion Tables for the Law School Admission Test. LSAC Research Report Series.
Camilli, Gregory; Wang, Ming-mei; Fesq, Jaqueline
The Law School Admission Test (LSAT) was examined to see if the items on a form could be divided into different subgroups in which items looked statistically similar within the subgroups but statistically different between subgroups. Of such subgrouping can be detected, it is likely that the subgroups of items measure different abilities, and the test can be described as multidimensional. The multidimensionality of six forms of the LSAT was studied using factor analysis. Two subgroups of items, or factors, were found for each of the six forms. The LSAT thus appears to measure two different reasoning abilities, inductive and deductive. The effect of dimensionality on equating was also examined by calibrating, with item response theory (IRT) methods, all items on a form to obtain a set of estimated item parameters (Set 1). The test was then divided into two homogeneous subgroups of items, each having been determined to represent a different ability. Items within the subgroups were recalibrated separately to obtain item parameter estimates, and these latter estimates were combined into a second set, Set 2. It was found that the equating tables based on Set 1 were highly similar to those based on Set 2 item statistics. Although the IRT model theoretically requires one-dimensional tests, it appears to give satisfactory results with the LSAT. The equating tables appear to be adequate. Two appendixes contain LISREL computer program printouts for the factor solutions. (Contains 7 tables, 5 figures, and 26 references.) (SLD)
Publication Type: Reports - Research
Education Level: N/A
Authoring Institution: Law School Admission Council, Newtown, PA.
Identifiers - Assessments and Surveys: Law School Admission Test