The LSAT vs. the GRE: May They Both Lose
The monopoly that the LSAT has enjoyed for decades in law school admissions appears to be eroding. What does that mean for Native American law school aspirants, and for the law schools that desire to increase their number of Native students?
Up until yesterday, the fight to be able to use the GRE in law admissions instead of the LSAT had been led by the University of Arizona Law School, which had accepted a small number of students with GRE scores. A few other law schools had been studying the possibility, but no other law school joined Arizona until yesterday, when Harvard Law announced that they too would consider the GRE from applicants.
For the foreseeable future, the impact on law admissions will be symbolic only. Arizona and Harvard will each only accept a small number of applicants based upon their GRE scores, and until a few dozen more law schools join them, the total number of applicants affected may be a few hundred out of tens of thousands. So, while the current discussion may have substantial implications years down the road, today’s applicants will find this a niche play at best.
The LSAT vs. GRE discussion will likely generate more heat than light. Folks will look at their differences, which are worth noting:
- The LSAT is a pencil and paper test, while the GRE is delivered on computer;
- The LSAT is offered only 4 times per year, while the GRE is offered almost continuously throughout the year;
- ¾ of the LSAT’s bubble sections, generating its score, are verbal, and only ¼ (one section, informally called “Games”) involves some math-related sensibilities, while fully ½ of the GRE bubble sections are straightforward math, and 1/2 are verbal; and,
- Under the current rules, a student must report an LSAT score to a law school if he/she has taken it, so only students who have taken the GRE and not the LSAT will have their GRE solely considered.
What most LSAT vs. GRE comparisons will miss are the profound and important ways that these two tests are similar, and both deeply problematic. Both are created by psychometricians using the same methods, so they share these significant characteristics:
- They both generate significantly disparate results by gender, race and ethnicity, with Native Americans, Latinos and African Americans scoring much lower on both, on average, than whites and Asian Americans, and females scoring lower than males;
- The foundation for these disparities could be revealed by item level data, which are statistics on individual test questions that test developers routinely refuse to release because they would expose the way that tests are designed to solidify and maintain those disparities;
- Scores for both are affected by inequitable access to high-quality, often expensive, test prep, with groups like Native Americans penalized because Native students often can’t afford to pay for test prep courses; and,
- What equity activists need to do is try to reduce, and even eliminate, the weight that any bubble test is given in any competitive admissions (or financial aid, etc.) decision.
While the last concept may seem fanciful in the law school admissions world, we now have many highly selective undergraduate schools, like Wake Forest, Wesleyan, Mount Holyoke, etc., that have years of successful experience with test-optional admissions policies that serve to reduce the bubble-test burden faced by URM students.
Any Native educators supporting either the LSAT or the GRE are, I contend, missing the forest for the trees. Bubble tests are designed in a way that Native students are placed at yet another disadvantage in admissions, and these tests should be made optional until they are eliminated entirely.
It should be noted that multistate bar scores will tend to correlate with LSAT (or GRE) scores, since all bubble tests have high correlations with other bubble tests. That’s not a reason to keep the LSAT; that’s a reason to make sure that all Native bar takers have the benefit of high-quality bar-prep courses to leverage their ability to get their best score on the multistate.