Wait, I thought SATs made admissions MORE biased

Swinging back to standardized tests at elite colleges?

Not that we don’t have bigger worries on the planet, but the recent messages from Dartmouth, then Yale, leave me wondering about the state of our meritocracy. Hyper competitive applicant pools, expanded in a combination of appeasing rankings and opening doors to historically absent swaths of the population, put ever greater pressure on the keepers of the sorting hats for college admissions. Scrutiny of their policies and procedure understandably grows as acceptance rates dwindle into the single digits.

Against that backdrop and the Supreme Court’s interest in the topic, resulting in dismissing affirmative action as a period piece that’s no longer justified (except at service academies)— if it ever actually was, per their decision about Harvard and UNC— we now hear that requiring standardized test scores actually levels the playing field more than leaving them out. Huh? Didn’t we collectively walk away from those metrics during COVID times, largely for reasons of access but also as a move toward fundamental fairness?

By 2022, nearly post-pandemic, 4 out of 5 institutions granting 4 year degrees in this country had gone test optional— finally following Bowdoin and Bates’ example from decades ago, in the name of equity and exigency, given questions of access to opportunities to even sit for those exams, much less be tutored for them in advance. The collective wisdom, long in coming, suggested that the tests themselves created a barrier to entry based on family background and educational privilege, one that distorted the view of an applicant’s capacity to succeed in higher education— and we could and should be doing better.

I remember when the College Board was on record saying that test prep didn’t help, back in the 80’s and 90’s. They went to the mat on how carefully the instruments had been created, that the writers had taken all the questions out about yachts, such that everyone had the same opportunity to excel, in the absence of any cultural bias. Then came Khan Academy’s online prep courses as an option to the $150/hr cottage tutoring industry, and then came practice exercises from the testers themselves, a clear admission that it mattered if you prepped— something that had always seemed self-evident to anyone who ever made time to think.

Despite these efforts, racial and income disparities persist, pervasively (remember the UC lawsuit), so how could it make sense to turn back to such imperfect indicators? Here’s the argument, per Dartmouth: it made sense to “pause” during the chaos of the pandemic, but using a rigorous empirical approach, it turns out that including standardized test scores leads to better outcomes for the admission process and better outcomes for students, at least in their first year academics.

If you’re wondering how that can be, consider the conclusions of the study in question that landed on the new President’s desk up there in Hanover, NH. First, that there are no better predictors of academic success than their test scores. Next, that other indicators, like scales constructed by admissions offices themselves don’t tend to predict that well, and college counselor letters arrive in wildly different levels of usefulness. Past that, test-optional years did not provide much of an uptick in applications from underrepresented groups. Ultimately, and perhaps counterintuitively, test scores can expand access since they provide a yardstick to measure schools far afield and beyond the traditional zone of familiarity for even the most well staffed admission offices.

Looking at the question from a behavioral standpoint, we’re left to wonder, as Brown has, to what extent those underrepresented group potential applicants are discouraged from applying by the need to provide test scores, perhaps as a function of Stereotype Threat. Beyond that, by contrast, how likely is it that otherwise privileged applicants hide weak scores behind test-optional policies? And what will be the effect of the affirmative action decision on applicant choices more broadly?

Add to these considerations the very real phenomenon of grade inflation in high schools, mirroring what’s happened in universities. Transcripts look more and more alike, indistinguishable from one another, leaving admissions offices to read between the lines, use proxies like AP enrollments, or compare across regions and years in scattergram form. Put yourself in the position of those admissions staffers, with the clock ticking and inboxes overflowing, and ask how you could really differentiate in a way that ultimately made for the fairest, most societally optimal, most beneficial for the institution?

We sure are asking a whole lot from this process. Maybe better would be to ask what it’s telling us in return. Here are my answers, born of experience in the thin air of super-competitive admissions:

1. SAT/ACT/AP/IB scores measure societal inequity right alongside individual potential. They measure the quality of our national effort to educate all children, if we care to look.

2. Absent the development of a genuine national exam system, like those present in most of the rest of the globe, these are our tests and they are better than nothing.

3. Non-test-score considerations, like legacy status, philanthropic potential, specialty talent (largely athletics) are likely to get a very long and public look in the wake of looking at demographic diversity priorities.

4. And we are still left wondering what to do with international applications, where all these issues get even more hard to sort through.

5. We need to get past a fixation on a short list of colleges as “the best.” That finite list of big names may serve our egos but it doesn’t hold up to scrutiny, and it’s fair to see what they do as part meritocracy, part lottery.

The whole system creaks under its own weight. My last question is whether, in the oligopoly of super-selective American college admissions, these two recent powerhouse announcements, following MIT a couple years ago, will pull the rest of the cohort along? Remember when Harvard and Virginia were going to get rid of early decision, for fairness reasons? That didn’t exactly stick, never reaching critical mass.

I read a great piece, in that light, from Derek Bok, famed Crimson President, that addressed the underlying resentment about the way the elites function- the title says it all- Why Americans Love to Hate Harvard. This SAT question invokes many of the same arguments Bok puts forward. Picking on the tests is kind of blaming the messenger. Yes they have baggage, especially in their origin as IQ metrics, but if we acknowledge them as a tool, especially as one of use in measuring relative achievement given the massive differences in educational access K-12, there’s room for those scores as imperfect parts of a an inherently imperfect process.

For those insiders from independent schools, let’s close by wondering what the implications are for the use of SSAT, ERB and ISEE scores for gatekeeping? Three cheers for the full file review. And we haven’t even talked about the relevance of the price tag once someone gets in…