On Friday, January 24, CMC faculty voted to recommend that the Board of Trustees reinstate the school’s standardized testing requirements. The Board should listen.
While no metric is perfect, standardized tests are perhaps the best tool admissions offices have to fairly evaluate students’ academic promise. A faculty report at Dartmouth found that SAT scores were “highly predictive” of first-year success. At UT Austin, when controlling for other academic indicators, students who submitted test scores had a 0.86 percentage point increase in their first-year fall GPA relative to those who did not. If CMC’s goal is to accept students of the highest academic caliber, standardized tests offer critical insight.
In addition, data indicate that test scores are very difficult to game, even with the help of expensive tutors. Multiple studies have compared results from tutored and untutored students using a variety of test preparation methods. SAT scores of students who used commercial test prep or a private tutor, compared to those who did not, were only 10 points higher on the verbal section and 15 points higher on the math section. Importantly, these margins are well within the College Board’s reported standard error of 32 points.
Of course, scores aren’t infallible. A professor interviewed for this story remarked that standardized test scores are often influenced by exogenous factors like test formatting and mathematical notation. Also, limited access to testing centers has created issues in the past, although the new adaptive digital SAT will make administering the test easier.
Regardless, tests are certainly more challenging to exploit than most—if not all—other criteria.
High school GPAs, for example, are subject to grade inflation—a problem that, although widespread, is more prevalent at wealthier schools. Not to mention, it is practically impossible to determine the extent to which the new generation of AI-savvy applicants use ChatGPT for their high school courses, further reducing the reliability of grades as a predictor of one’s performance.
Essays are also a subpar tool. In the best of circumstances, they can provide deep insight into the character and prowess of applicants. But they might also reflect applicants' access to high-quality college counselors or adept ChatGPT use. Additionally, multiple studies show that essay topics are highly correlated with income, and disadvantaged students often feel forced to “sell their pain.”
Extracurriculars, too, don’t say much about a person’s academic merit. They might indicate one’s potential for community engagement, but they could also indicate that a student had the leisure time to join clubs and participate in service activities. It’s much easier for someone who doesn’t have to take care of their siblings or bring in extra family income to participate in extracurriculars.
Recommendations are also skewed by wealth. Letters from private and wealthy public schools can have more of an impact than those from lower-income public schools. Additionally, admissions officers from wealthier schools often have connections at universities and can make personal phone calls that may boost an applicant's chances.
That is not to say that these other measures serve no purpose. Qualitative criteria give colleges more discretion in their admissions processes, allowing the college to create a class with balanced, complementary experiences and interests.
Tests, however, are still necessary. Wide discretion is not always directed toward desirable ends. One might hope that schools use this discretion to promote social mobility, increasing low-income students’ chances of admission. Yet, children from families in the top 1% are more than twice as likely to attend an Ivy-plus college as those from middle-class families with comparable SAT/ACT scores. One paper found that one in eight athletes recruited to play at elite colleges comes from the top 1% of the income bracket, whereas only one in 20 comes from the bottom 60%. In practice, admissions offices often leverage discretion in favor of high-income students despite employing it under the guise of equity. Requiring test scores—and properly weighting them over more corruptible metrics—leads to greater transparency and accountability, inhibiting admissions offices from prioritizing wealth over merit.
While income is strongly correlated with SAT scores, this reflects a more fundamental advantage of the wealthy: years of access to better education from the early childhood level, both in and out of the classroom. In the short term, income has a minimal influence, meaning test scores remain valuable for assessing merit at the time of application.
In fact, standardized tests can help even out the playing field between income groups. One study found that scores can mitigate income-based disparities in application, admissions, and matriculation rates, though primarily for middle-class rather than low-income students.
Test scores, when viewed in the context of an applicant’s household income and the quality of their high school, also allow colleges to identify high-performing low-income students. A Princeton study found that these students, when enrolled at elite schools, tend to see larger gains in metrics like post-graduation income than their wealthier peers.
Ultimately, admitted students must match their schools. As the CMC professor interviewed for this story said, “College admissions should be a matching process where students and campuses look for good fits. Policies that reduce transparency or the provision of information to students, parents, and colleges tend to make that matching process harder to navigate and make people less satisfied with their outcomes.” Clear, objective criteria are the best way to support CMC’s mission and future students.
Reason and research show that standardized tests are a useful tool in an undeniably imperfect admissions toolbox. At best, they will ensure high-quality admitted students and promote fairness and economic mobility. At worst, the application and matching process becomes a little more transparent. The choice is clear.
Comments