I’m going to begin this post with dSAT-style text completion.

In a 2023 paper, the Harvard economist Raj Chetty, along with colleagues from Brown University, showed that affluent students enjoy a significant advantage in the elite college admissions process: for applicants who received high academic ratings, comprising a rigorous course load and test scores in the 99th percentile (1510+ SAT/34+ ACT), those with family incomes in the top 1% were admitted to Ivy-plus institutions at a 43% higher rate than middle-class applicants, and applicants in the top .1% were admitted at a 250%* higher rate. The researchers found that high-achieving affluent and non-affluent students applied to Ivy-plus schools at relatively similar rates and received comparable academic assessments; however, wealthy students received significantly higher ratings for “personal qualities,” legacy status, and athletics. Based on these findings, Ivy-plus schools that want to improve the economic diversity of their student bodies should ________

(A) place less weight on non-academic factors that disproportionately favor wealthy applicants.
(B) recruit more heavily at high schools that enroll predominantly low- and middle-income applicants.  
(C) give greater consideration to personal qualities such as empathy and curiosity.
(D) adopt policies allowing applicants to choose whether to submit SAT or ACT scores.

If you approached this question from the standpoint of reason and logic, you probably picked (A): if, all things being equal academically, the emphasis on “character,” along with legacy and athletic-recruitment status, disproportionately favors wealthy applicants, then placing less emphasis on those factors would presumably result in a more even distribution of students from across the economic spectrum.

However, as you may have guessed from the title of this post, this is a trick question.

If you’re an elite private college (other than MIT), the answer is of course (D).


Because the response to basically any form of inequity in the admission process is always to remove standardized testing requirements—even when the data directly suggest that doing so would be counterproductive. Indeed, highly ranked public flagships, which rely more on objective metrics, admit students much more evenly across the economic spectrum.

To be clear, the issue is not a dearth of applicants. Academically qualified middle- and lower-income students are already applying to Ivy-plus institutions in large numbers: slightly more than 11,000 per graduating high school cohort, according to Chetty et. al., with another 13,000 likely to attain high academic ratings if they applied. If those students are not being admitted under the current system because of “soft” factors, not because of their test scores, then there is absolutely no reason to think that academically weaker students—as candidates who do not submit SAT or ACT scores tend to be—would be more likely to gain admission (particularly since Affirmative Action has been dismantled).

So, yes, although there is a direct correlation between family income and test scores, although there are gaping disparities in the type of test preparation applicants have access to, the SAT and ACT are still doing more or less what they were introduced to do: provide talented students from outside the traditional prep-school circuit a way to demonstrate that they could compete academically with their more advantaged peers.

In fact, given the disparities in preparation, Chetty and his colleagues mention that standardized tests probably have even more predictive value for middle- and lower-class applicants than they are given credit for.

I have more thoughts about the paper, but I’ll save them for my next post.