Predictive Validity of Accommodated LSAT Scores for the 2002–2006 Entering Law School Classes (TR 09-01)
by Arlene Amodeo, Laura A. Marcus, Andrea E. Thornton, and Peter J. Pashley
Executive Summary
This study was undertaken to evaluate the predictive validity of Law School Admission Test (LSAT) scores earned under accommodated testing conditions. Of special interest was the validity of scores obtained by test takers who were accommodated under extra-time conditions. Within this group, separate predictive validity analyses were also conducted for test takers who were classified within the following categories: Attention Deficit/Hyperactivity Disorder (ADHD), Learning Disorder (LD), Neurological Impairment, and Visual Impairment. Of secondary interest was the validity of LSAT scores obtained by test takers who were accommodated under standard-time conditions (i.e., accommodations that did not include extra testing time).
The measure used to assess the predictive validity of the LSAT for the above groups was law school first-year average (FYA). The predictive validity of undergraduate grade point average (UGPA) was also evaluated, as was the combination of LSAT score and UGPA, commonly referred to as the Index. The sample used in this study was drawn from 549 entering law school classes from 176 distinct law schools collected over 5 years.
The results from this study strongly suggest that
- LSAT scores earned under accommodated testing conditions that included extra testing time are not comparable to LSAT scores earned under standard-time conditions, as evidenced by a tendency of the former to overpredict FYAs (i.e., estimated FYAs based on scores from test takers accommodated with extra testing time tend to be higher than their actual FYAs).
- Results for test takers categorized within the largest categories of ADHD and LD, and within the accommodated with extra testing time subgroup, were consistent with the finding stated above for the overall extra-time group.
These findings are consistent with those found in earlier reports published by the Law School Admission Council (Thornton, Reese, Pashley, & Dalessandro, 2001; Wightman, 1993a).
The current study also indicates that
- LSAT scores earned under standard-time conditions are comparable, regardless of whether other (non–timing-related) accommodations are given.
While considering the results of this study, the reader should keep in mind that they refer only to subgroup results and not to individuals. For example, while results may suggest that FYAs tend to be overpredicted for those accommodated with extra testing time, the performance of an individual who received an extra-time accommodation may actually be underpredicted.
The results from this study indicate that LSAT scores obtained under accommodated conditions that include extra testing time should be evaluated with care. However, more data and further analyses will be required before more specific statements can be made, such as whether UGPA is a better predictor of FYA than LSAT scores for test takers accommodated with extra testing time.