Op-ed: The GRE is no diversity tool

In February, the University of Arizona College of Law announced that applicants for admission could submit Graduate Record Exam scores, in lieu of scores from the Law School Admission Test. The announcement made waves because for decades LSAT scores have been a requirement for law-school candidates.

The American Bar Association, the primary accreditor of law schools, requires law schools to use a “valid and reliable admission test” to aid in selecting students. However, no particular test is stipulated. Arizona argues that the GRE is as good a predictor for law school success as the LSAT, which may be true. 

But what is interesting, and perhaps disingenuous, is the school’s claim that accepting GRE scores will promote student-body diversity “in all its forms.” This is because if the GRE is misused in the same manner as the LSAT, the admissions process will remain inequitable, at the expense of racial, ethnic and socioeconomic diversity.

The LSAT is designed to be a partial predictor of first-year performance in law school. Data from the Law School Admission Council show that the test predicts about 36 percent of the variance in first-year grades. This means that almost two-thirds of the variance is predicted by other factors. Similarly, in a recent study, a 6-point score difference between two LSAT scores accounted for only a 0.1 difference in law school grade point average. These data demonstrate that the outsized role the LSAT plays in influencing which law school a student attends and how much that student pays is unjustified. 

Overreliance on the LSAT has a tangible and harmful effect on diversity in legal education. The average score for black LSAT-takers is 142 — 11 points lower than the average of 153 among white and Asian test-takers. Latino test-takers score an average of 146. The effects of these scores are evident in the fact that black applicants have by far the lowest rate of receiving at least one offer of admission — 55 percent compared to 72 percent for Latinos, 76 percent for white applicants, and 84 percent for Asian applicants. 

Even more disturbing, in a study I conducted on enrollment trends, I found increasing racial and ethnic stratification in legal education. In 2015, black and Latino law students were less likely to attend schools with the most favorable outcomes than they were in 2010. Asian and white students were more likely.

Moreover, data from the Law School Survey of Student Engagement suggest LSAT-driven scholarship policies are contributing to increasing disparities in student debt. In 2015, respondents with LSAT scores of 155 or below were almost twice as likely to expect more than $120,000 in law school debt than respondents with higher scores. No such disparities were observed 10 years ago.

The GRE shares two troubling similarities with the LSAT: score trends are typified by racial, ethnic, and socioeconomic disparities, and schools often misuse the test in the admissions process. A recent study found that, on average, white test-takers scored about 100 points (roughly 25 percentile points) higher than their black counterparts on the verbal reasoning section, and about 60 points higher than Mexican Americans. Disparities in quantitative reasoning scores were observed as well.

Like the LSAT, the predictive value of the GRE is limited. Data from the Educational Testing Service show that for MBA students, scores on the verbal reasoning section predicted less than 25 percent of GPA variance; quantitative reasoning scores predicted less than 40 percent. Unfortunately, overreliance on the GRE is common, prompting one group of researchers to speculate that the test was “a strong driver of the continuing under-representation of women and minorities in graduate school.”

None of this is surprising. The only surprise is that the GRE is being promoted as a diversity tool, in spite of all the contrary evidence. These diversity-tinged justifications represent echoes of the same lip service that is given when attempts are made to dissociate admissions processes from the harmful and illegitimate outcomes they produce.

If law schools were truly interested in fostering diversity in all its forms, they would fundamentally alter the manner in which they use admissions tests, not merely tinker with the tests they misuse. Holistic assessments of applicants that put test scores into proper context would allow for nuanced consideration of the other information most law schools already require, like transcripts, personal statements, letters of recommendation, and resumes. A de-emphasis on test scores could also spur a refocusing of the admissions process onto lawyering skills as opposed to a limited range of law school skills.

The tool is only as good — and every bit as harmful — as its use. Merely allowing law-school applicants to submit GRE scores might be a boon for some law schools, in the form of more applications. It might be a boon for the Educational Testing Service. It might even be a boon for applicants who score highly. But it will not be a boon for racial, ethnic or socioeconomic diversity.

Aaron N. Taylor is an associate professor at Saint Louis University School of Law and director of the Law School Survey of Student Engagement. Follow him on Twitter at @theedlawprof.