The LSAT as a Predictor of Success in Law School
General LSAC Validity Studies of the LSAT
LSAC has been conducting validity studies of the LSAT since the test was first offered. These studies consider correlations between undergraduate GPA (UGPA), LSAT scores, and first-year average law school grades (FYA). LSAC has repeatedly asserted that LSAT scores are more closely linked to FYA than UGPA, although the organization admits that LSAT scores and UGPA together form an even stronger predictor. One of these annual studies, for example, claimed that LSAT scores alone and FYA produced a correlation coefficient of 0.41, while the FYA correlation coefficient with UGPA alone was much lower at 0.26 (the correlation coefficient scale ranges from –1.0 to 1.0; –1.0, 0, and 1.0 represent perfect negative correlation, no correlation, and perfect positive correlation respectively). These numbers approximate the long-term trend, although there can be considerable variation among individual law schools. LSAC has also studied LSAT/FYA correlations among repeat test-takers, and concluded that average LSAT scores (as opposed to initial, highest, and most recent scores) were mostly strongly connected to FYA.
LSAC Studies of Test-Taker Subgroups
LSAC has examined the effect of race, gender, and disability on the predictive validity of the exam. LSAC data has consistently shown that whites and Asians score higher on average than African-Americans and Hispanics, and that the mean scores of male test-takers are higher than those of females. LSAC has nonetheless found that these discrepancies do not substantially undermine the LSAT's validity as a predictor of FYA. One comparative analysis of LSAT/FYA data divided by race concluded that, while there was some "slight overprediction" (lower FYA than suggested by LSAT scores) for Blacks, Asians, and Hispanics, the results "do not suggest that the use of LSAT score alone or the combination of LSAT score and UGPA contributes to unfair admission decisions for the racial/ethnic subgroups studied here." Another study reviewed the LSAT/FYA correlation for 158 law schools, with scores divided by gender, and found that "the validity coefficients were very similar to each other." LSAC has conceded that at least some of its research on the LSAT and disability has not produced the same results. In a paper exploring the outcomes of disabled students who received testing accommodations, LSAC found that the LSAT "overpredicted" FYA for this particular group of students.
Criticisms of LSAC Validity Research
LSAC validity research has been criticized on a number of fronts. First of all, some researchers question the value of using FYA only for LSAC correlation studies, claiming that FYA provides an incomplete picture of law school success by failing to account for grades in subsequent years. Second, these LSAT correlation studies are not easily or widely available to the public via the organization's website, although "executive summaries" are published and interested persons may email LSAC to request copies of full reports. These critics wonder why LSAC doesn't fully publicize studies that validate the organization's flagship product. Finally, LSAC validity studies do not consider other factors in the backgrounds of law school applicants, such as work experience. It's certainly possible that one of these additional factors is a better predictor of law school success than the LSAT. Unfortunately, these criticisms do not seem to have inspired a large body of independent (non-LSAC) validity research.
LSAT Scores and Bar Passage Rates
According to an article published in U.S. News & World Report, there is a notable connection between law schools' average LSAT scores and their bar passage rates. Because each state administers its own bar exam, the author compared the passage rate for each institution with the state average. Alumni of the 10 law schools with the highest average LSAT scores outperformed overall state bar passage rates by an average of 29 percent. Graduates of the 10 programs with the lowest mean LSAT scores underperformed state rates by 17 percent on average. Additionally, an LSAT score/bar passage rate analysis conducted at the University of Denver Law School found a "medium positive relationship" between the two. This research suggests an additional benefit for high-quality LSAT prep, but stakeholders should remember that correlation is not necessarily causation.
The LSAT and Other Metrics of Success
As suggested by the previous paragraph, LSAT scores are linked to metrics of success (graduation rates, employment rates, starting salaries, etc.) only in terms of their association with elite programs. To put it simply, the best law schools produce the best outcomes, and these institutions also expect the highest LSAT scores from their applicants. It would be extremely difficult to isolate the effects of the LSAT on these outcomes, and to the best of our knowledge, no researchers have yet attempted to do so.