Merit, Mission, and a Memorandum
A presidential memorandum issued on August 7 asserts the following:
Although the Supreme Court of the United States has definitively held that consideration of race in higher education admissions violates students’ civil rights, the persistent lack of available data… continues to raise concerns about whether race is actually used in practice.
The Secretary of Education is accordingly directed to “expand the scope of required reporting to provide adequate transparency into admissions” and to “make public the enhanced reporting requirements” within 120 days.
It is likely that the new reporting requirements will closely track those delineated in recent agreements with Columbia and Brown. In Columbia’s case, a resolution monitor must be provided with “admissions data… showing both rejected and admitted students broken down by race, color, grade point average, and performance on standardized tests” and “statistical information regarding enrolled students… including the composition of the class broken down by race, color, national origin, grade point average, and performance on standardized tests.”
How might this information be used? A great deal depends on how merit and mission are understood.
After they lost their only child at the age of 15, Leland and Jane Stanford founded a university that would eventually become one of the world’s finest. In a 1902 address, Jane Stanford observed that their goal was “the betterment of mankind morally, spiritually, intellectually, physically, and materially. The public at large, and not alone the comparatively few students who can attend the University, are the chief and ultimate beneficiaries of the foundation.”
Stanford’s mission is not unusual. As Bowen and Bok note in their influential book, highly selective institutions seek to “attract students who seem especially likely to utilize their education to make valuable or distinctive contributions to their professions and to the welfare of society.” And if you look closely at the following image of Columbia’s Low Library, you’ll notice an inscribed statement of purpose that has been on display since 1897: For the Advancement of the Public Good and the Glory of Almighty God:
It is in relation to mission that merit must be assessed. The admissions decision is a bet on the future, not a reward for accomplishments in the past. At the same time, prior achievements provide us with our main source of evidence regarding future potential. How should this evidence be evaluated?
Simple and transparent rules for selection have one very important virtue—their lawfulness can be easily ascertained. One such rule is to base admissions exclusively on an academic index, such as a weighted average of a grade point average and a standardized test score. Under this procedure, there can be no differences across groups (defined by race, color, or national origin) in the likelihood of admission conditional on the academic index.
Such a selection rule is obviously lawful, and may be precisely what the federal government would like universities to adopt. But it is not meritocratic even with respect to post-enrollment classroom performance. For one thing, prior achievement depends in part on human and material resource inputs, and these are not evenly distributed across the population.1 Of two students with a given academic index, one who has enjoyed fewer advantages and faced more serious personal challenges is likely to have greater latent ability. If these obstacles can be mitigated following enrollment, such a student would have higher expected performance in college and (especially) in later life.
There are indeed some spectacular examples of this phenomenon, including a remarkable hand-picked cohort of students at the College of the Holy Cross in 1968 that included Clarence Thomas, Edward P. Jones, Theodore Wells, and Stanley Grayson.
This kind of reasoning has led many institutions to take into account not just an applicant’s raw scores, but also the distribution of scores in their high school cohort. The same logic underlies the Texas top ten percent plan (and others like it), which guarantee enrollment to students based on class rank rather than absolute performance. Such practices are lawful, but would fail a naive test of merit that looked only on the likelihood of admission conditional on an applicant’s academic index.
I spent some time with Glenn Loury on his podcast last week discussing these and related issues, in the shadow of recent deals struck by Columbia and Brown.2 We covered a lot of ground, but I want to briefly mention two important arguments that Glenn made during our conversation.
First, he pointed out that universities could try to fine-tune the weights given to contextual factors rather than absolute levels of prior achievement (in a race-neutral manner) in order to further diversity goals. This is true, they could certainly try. Doing so would be lawful under the Supreme Court ruling as I understand it, and indeed was part of the original rationale for the top percent plans. Notice, however, that universities would not be able to selectively recruit students from elite high schools who happen to belong to underrepresented groups, as was the case prior to the Court ruling.
Glenn’s second point was this: the federal government could look at post-enrollment performance across groups to detect unlawful admissions practices, on the grounds that a group that was discriminated against would secure higher grades (in any given course) once on campus. This is the so-called outcome test originally proposed by Gary Becker, and is based on the following logic: if a group is discriminated against in the sense of facing a higher academic threshold for admission, only its most qualified members will receive offers, and so their post-enrollment performance will be higher on average than that of other groups.3
Since the federal government is demanding that universities submit data on post-enrollment performance, they are presumably planning to implement tests of this kind. But one needs to be very careful with this because the population of enrolled students is not randomly drawn from those who recieve offers of admission. Suppose, for example, that the most highly qualified members of a heavily underrepresented group decline offers and choose to enroll in somewhat less selective instututions where they may not be quite so underrepresented. In that case differences across groups in post-enrollment performance can arise even if a random draw from the admitted population would not exhibit any such disparity.
This is the dilemma that universities now face—they will be judged on data that may not be adequate for the task of determining whether or not they are behaving in a manner that is merit-based and lawful. This may sound a bit paradoxical, but having to report to a resolution monitor may turn out to be an advantage for Columbia relative to Brown and other institutions that must report directly to the Department of Education.4 It may be easier to persuade a single individual (chosen by mutual consent) of the legitimacy of an admissions policy than to convince political appointees who face different kinds of pressures.
But either way, I think that universities will have to make their selection processes much more transparent, publish the (suitably anonymized) data on which they base their decisions, explain carefully the contextual factors that lead them to offer admission to those who may not appear qualified based on standard metrics, and think deeply about how systematic differences across groups in the decision to enroll affects the inferences one can draw from post-enrollment data. Their most immediate concern is federal oversight, but they also need to rebuild confidence among the public at large. This has started to recover a bit after a sharp decline over the past decade, but there is still a long way to go. If universities want to be left in peace to pursue their respective missions as they see fit, public trust is every bit as important as a large endowment or global prestige.
Some of the implications of this fact are explored in a relatively recent paper that also surveys the literature on meritocracy dating back to a very interesting exchange of letters between Thomas Jefferson and John Adams.
I mentioned the outcome test in my last post, and subsequently recieved an email from Johann Gaebler about a recent paper (co-authored with Sharad Goel) that claims to improve upon it. I haven’t read the paper yet but am posting the link in case anyone wants to take a look.
That is undoubtedly true when played out across the full range of colleges and universities that admit across the full range of academic indices.
I wonder, however, how it plays out at the few most exclusive and selective of colleges and universities where essentially everyone selected has about the same academic index -- hovering around the very top of what is achievable. Would that not, in effect, be wholly to the detriment of those who -- for no reason over which they had any control -- had not faced those "serious personal challenges"?
The point is that your observation about differentiation only happens when there is room above the current academic index to achieve even more that you have. If a selective university is effectively selecting from a complete cohort already at the top of the academic index, there is actually very little (or no) room for demonstrating differentiation; and a built-in bias toward those who have endured hardship would, then, be completely equivalent to a built-in bias against those who have been lucky enough to have avoided hardship.
In some cosmic sense, that might seem like justice. On a personal level, I can assure you that it would not.
Perhaps that's why the public fights over such things are so vehement and so hard: because they are nearly always about what happens at the Harvards and Yales and MITs and Stanfords of the world -- and among potential students (and their parents) with sky high expectations -- and not about what happens at the second and third-tier level...
Another critique of the Becker "outcome test" is that it can conflate intergroup differences in acceptance criteria (i.e. discrimination) with analogous differences in the distribution of "merit" (e.g., test scores, grades, etc.). In particular, suppose that group A's distribution of merit is simply a leftward shift of group B's distribution and that both groups are subjected to the same acceptance rule. This will of course imply that fewer members of Group A will be accepted, but it also implies that, *conditional on acceptance*, Group A members will have less merit and also worse post-graduation outcomes.
I wryly note that a) Becker was one of my thesis advisors and that b) Augustus Lowell and I were Exeter classmates:)