Fisher v. University of Texas at Austin has wound its way back to the Supreme Court, which will once again examine whether consideration of race in undergraduate admissions is constitutional. One striking development during the oral arguments on December 9th was Justice Scalia’s invocation of the disputed mismatch hypothesis: that affirmative action hoists black students into schools that are “too fast” for them, leading to a mismatch between their true qualifications and the schools that they attend. Justice Scalia, along with the conservative think tank The Heritage Foundation and others, continue to present this hypothesis as fact even though the primary evidence for this argument has been called into serious question by many scientists.
At the core of this argument is the idea that schools with race-conscious admissions policies give preference to students of color over white students even though they are objectively less qualified, as evidenced by lower average pre-college test scores and grades. That achievement gaps persist in college is often cited as proof that this idea is correct. In other words, if black students with lower SAT scores than their white peers also receive worse grades in college, or more often switch into “easier” majors, then surely they were less college-ready to begin with. However, the underlying assumption that poorer academic outcomes indicate lower ability ignores a large piece of the puzzle: the taxing psychological environment faced by students of groups historically marginalized in academics.
In the abstract, it can be challenging to separate these two explanations—psychological environment versus lower ability—because they both result in the same outcomes: lower scores and greater attrition from challenging majors. However, social psychology offers a window into this otherwise black box through the use of experimental design. In the randomized controlled experiments that we (and many other social psychologists across the country) conduct in schools, we can directly measure the impact of the taxing psychological environment on students’ academic achievement by observing what happens when we remove psychological strain as a factor.
One psychological strain faced by students of color that we study in our experiments is stereotype threat. Decades of science has shown that pervasive stereotypes of certain identity groups as less intelligent or less capable in academics often lead students of these groups to worry that they could be judged through the lens of these negative stereotypes. This threat is stressful and can undermine these students’ learning, performance, and engagement. Few of us can perform at our best in an environment we perceive as judgmental or skeptical—this is no different, except that the judgmental environment is chronic.
When we give students exercises designed to reduce stereotype threat, we see their school performance and engagement increase, and the racial achievement gap shrinks. They earn better test scores and grades. Meta-analyses comparing the performance of the thousands of students who have received these relatively simple threat-reducing exercises over the past ten years to that of their peers in control groups estimate that reducing stereotype threat alone accounts for the equivalent of about a 62 point increase on the SAT—and this is almost certainly a conservative estimate, because no one experimental exercise can remove stereotype threat or its effects entirely. Moreover, recent research in our laboratory shows that college students who participate in these exercises are also more likely to continue taking classes in “hard” STEM majors. These results suggest that students’ psychological environment plays a large role in both their grades and major selections.
Another aspect of the psychological environment we target with experimental exercises is the expectation of many black and Latino students that they will be treated differently in academic environments than white students. This expectation is not unfounded; research consistently shows that bias still exists today in the education system. For example: professors are less likely to respond to emails from students of color requesting research opportunities or mentorship than emails from white students; readers told that an essay writer is black evaluate that essay as poorer quality than when they are told the writer is white; and black grade school students face harsher disciplinary action and less warmth from teachers and administrators than their white peers. When students observe differences such as these, they may, understandably, come to expect rejection both from their professors and peers, which can undermine their trust, feelings of belonging, and motivation in school, and—in turn—their performance.
To combat these effects, we and other researchers have given black students exercises that promote trust in educators or foster feelings of belonging at school. In these exercises, participants read brief testimonies ostensibly written by older students that normalize adversity and reduce the expectation of race-based rejection. For example, students may read testimonies explaining that instructors' critical feedback is motivated by high expectations or that social adversity in college is common and transient, instead of stemming from racial bias. We find that these exercises reduce the academic achievement gap between treated black students and their white peers by 40-50%. Moreover, these effects tend to strengthen over time; three years after the belonging exercise, the grade gap between treated black and white students had closed by 79%.
If these modest exercises to reduce psychological strain in the academic environment can have such a powerful effect on the performance and engagement of students of color, without changing anything about their knowledge or credentials, it cannot be simply a lack of ability or preparation holding back these students’ achievement. Instead, this extensive body of research suggests that academic outcome measures such as test scores and grades habitually underestimate the ability of students of color, and instead often reflect the presence of threat in their environment—both before and during college.
To put it another way: Let’s imagine that we strap a 15-pound weight to a runner’s back during her qualifying race. She will run slower than she otherwise would have. The lazy solution—the “mismatch” solution—would be to keep the weight on her back and assign the runner to a slower heat where she might be able to win, despite the extra weight. Most would probably agree, however, that the real solution is to remove the weight and assign the runner to the faster heat that matches her true ability level.
Affirmative action takes one very important step: It increases the chances that students of color will be accepted to the schools that match their true ability level, despite test scores that are often depressed by taxing psychological environments. But this is only part of the solution. We also need to remove the weight, so that racial academic achievement gaps don’t persist even after highly qualified minority students are admitted to top colleges and universities. To make progress in racial equality in higher education, the Supreme Court must uphold affirmative action; we as scientists and educators must continue to find ways to build inclusive psychological environments in academia, so that students of color can reach their full potential.