The Mismatch Effect: A Danger for Students of All Races

A very interesting article from Prof. Robert Steinbuch (Arkansas / Little Rock).

Say that you apply to many schools, including some where other students have much better predictors (chiefly test scores and grades from earlier institutions) than you. And you get in! Maybe it was affirmative action (whether based on race, socioeconomic status, or something else), maybe it was a preference because your relatives had gone to the same school, maybe it was something in your admission essay, maybe it was just luck. Either way, you're thrilled.

Might you have reason to be less thrilled, and actually not want to go to this highly ranked school? (Conversely, might the school reconsider its policy that led you to be let in?) If the predictors are actually reasonable predictors (and apparently grades and test scores tend to be reasonable predictors), then you can expect you'll end up lower in the class at the new school than you might be at some other school, precisely because the other students are better than you. The advantage of having a more prestigious degree might be counteracted by the disadvantage of having a less prestigious class rank. But might you end up getting a worse education, being less likely to graduate, and being less likely to pass any professional licensing exams (such as the bar) that you might be expecting to take?

That's the debate about the "mismatch effect," which I've followed over the years (though from a distance); it has mostly focused on whether race-based affirmative action causes problems (such as lower black bar passage rates) as a result of this effect, but it can also be relevant to many students of all races. I was first exposed to it because of the work of my UCLA School of Law colleague Rick Sander, and Robert Steinbuch at Arkansas / Little Rock has been working in it as well; Rob has been kind enough to pass along these thoughts on the subject:

As much as we have discussed the "mismatch effect" -- what takes place when students with academic credentials noticeably lower than their peers learn less as a consequence of that intellectual estrangement -- I believe that we still haven't fully appreciated its negative consequences. We know that mismatched law-school graduates, on average, fare far worse on the bar exam than their non-mismatched peers. And we've generally thought of the harmful mismatch effect as impacting racial minorities, because they're often admitted with dramatically lower standardized-test scores than their peers. But the negative effects of mismatch on non-minorities have been overlooked: this second set of harms often remains hidden because these students are only a small percentage of a relatively large demographic group. Nonetheless, we should consider them as well when we analyze the scope of the mismatch problem.

[I.] Measuring "Low" LSAT Scores

No measurement of intelligence and skill is perfect, but standardized exams increasingly do a good job. So, in 2018, it should be unsurprising that law-school graduates with low Law School Admission Test (LSAT) scores generally perform far more poorly on bar exams than those with high scores. What constitutes a "low" score, though, requires some discussion.

LSAT scores range from 120 to 180. The consumer-advocacy group Law School Transparency says that law-school applicants with LSAT scores below 150 are engaging in a perilous endeavor: a score of 147 to 149 presents a high risk for bar failure; a score of 145 to 146, a very high risk; and a score under 144, an extreme risk. Therefore, students with an LSAT score of, say, 149 will on average perform less well on the bar exam than students with, say, 165 -- all else being equal.

Beyond this easy to apply static view is a more dynamic model that posits that students' relative LSAT scores as compared with those of their law-school classmates is a reliable predictor of bar passage -- perhaps even more reliable than absolute LSAT scores. This mismatch theory posits that students with, say, a 155 will perform differently depending on where in the class they fall (i.e., how selective a law school they attend). In particular, the further students are below the bulk of the class, the more they suffer from "fish out of water" syndrome -- i.e., the more they're "mismatched." And that estrangement itself negatively affects bar performance.

[II.] Mismatch at UALR

Analysis of a large dataset containing information on graduates from the law school at which I teach, the University of Arkansas at Little Rock, Bowen School of Law, demonstrates that LSAT scores of students enrolled at the school (1) solidly predicted bar passage, and (2) varied significantly in relation to ethnicity.

Although color-blind admissions should produce roughly 25 percent of both Whites and African Americans in each LSAT-score quartile, over two-thirds of graduating African Americans were admitted with LSAT scores in the bottom quartile, as contrasted with only 16 percent for White students. (For more details, see the recent article I coauthored: Steinbuch and Love, Color-Blind-Spot: The Intersection of Freedom of Information Law and Affirmative Action in Law School Admissions, 20 Tex. Rev. L. & Pol. 1 (2016)). Although almost exactly a quarter of White students were admitted in the top quartile of LSAT scores (as expected), remarkably, only one percent of enrolled African Americans fell into the top quartile of LSAT scores. Predictably, this led to dramatic differences in bar passage: 80 percent of Whites passed the bar (the first time), while only 60 percent of African Americans did.

Given that the African-American cohort in our dataset on average had much lower LSAT scores than the bulk of the student body, it's fair to conclude that this cohort overall was mismatched. This profile dominated because affirmative-action considerations are designed to consider factors beyond traditional credentials and explains why debates on how to deal with poor bar-passage rates often focus on race-based admissions. However, the ensuing discussion often misses that, while on average Whites will not be mismatched because they have such a large population -- putting many at or above the mean of the class, the number of Whites who are mismatched could easily equal or exceed that of any other racial group.

[III.] A Closer Look at Mismatch's Consequences

Since mismatch effects likely impact students of all races -- indeed, maybe more White students than those in any individual minority cohort -- we should think much more carefully about how to discuss the process of law school admissions and bar passage in the context of race. Mismatch certainly says something important about race-based admissions, but it has much more to tell us.

Indeed, for schools that are not at the top echelon, the absence of race-based admissions might not eliminate the mismatch effect, because at least some of those schools will likely still admit mismatched students, irrespective of race, to fill otherwise empty tuition-producing seats. (It is possible, though, that some White mismatched students are admitted in some instances as cover for low-scoring minorities. If true, then mismatched admissions would drop in a color-blind process. Further study is necessary to determine this.)

A closer look at the UALR data highlights the dramatic increase in negative outcomes for graduates who were mismatched on LSAT scores. Logistic regression produced an odds ratio for LSAT scores that explained that the odds of a UALR graduate failing the bar exam increased by nearly seven percent (p = 0.0023) for each single-point decrease in LSAT score.

We can further see in the chart below just how heavily African Americans are mismatched in the dataset when we break down the bottom quartile in four roughly equal sized sub-groups. As was the case when looking at all quartiles, even within the bottom quartile, the number of African Americans dramatically grows as the LSAT scores drop. Even though there are nine times as many Whites as Blacks in the UALR dataset, the very lowest end of the LSAT-score band has twice the absolute number of African Americans as it does Whites. That remarkable divergence is overwhelmingly attributable to race-based admissions decisions, which explains the White-Black bar-pass disparity even within the lowest quartile.

Segmented LSAT Scores for Bottom Quartile of UALR Data

	LSAT 148	LSAT 147	LSAT 146-45	LSAT < 145
African American	n = 3 100% pass	n = 2 100% pass	n = 10 50% pass	n = 20 35% pass
White	n = 32 75% pass	n = 24 66.67% pass	n = 26 69.23% pass	n = 10 40% pass

However, even though Whites in the bottom quartile had a far higher pass rate than Blacks, there are almost 50 percent more Whites in this category who failed the bar exam than Blacks. Thus, although Whites viewed as one group across quartiles weren't mismatched -- given their right-sided distribution overall -- far more Whites than Blacks in the bottom quartile failed the bar exam. For these White graduates, the negative consequences that flowed from being mismatched is likely not ameliorated by the realization that their cohort is generally not mismatched or that they didn't receive race-based admission preferences.

[IV.] How Disclosure Might Shrink Mismatch's Negative Consequences

So, what's the solution to poor bar-passage rates that result from schools consciously admitting mismatched students of all races? Some suggest that newly conceived remedial programs can offset the considerable headwinds faced by LSAT-score-impaired students. Others recommend cabining the now-overwhelming influence of race in admissions, so that it returns to its once-envisioned role as a tie-breaker.

As advocates battle over these competing approaches, I offer a modest proposal: informing students of their individual likelihood of success, as predicted with some strength by LSAT scores. With little effort, law schools could disclose their individual bar-passage rates broken down by LSAT scores on an annual basis. This would permit far more meaningful decision-making by students: it would allow prospective students to have a much better sense of where they stand before they attend law school.

Of course, prospective students may seek this information today -- but, sadly, institutions have been less than transparent about this kind of admissions data. Indeed, I had to sue my school under the Arkansas Freedom of Information Act to get some of it. When circumstances or individuals conspire to prevent the market from providing material information to consumers, we have the best case for requiring disclosure. Such disclosure is the most benign form of regulation, because it simply affords consumers the opportunity to make better informed decisions.