In 1978, the Seattle Supersonics nearly won the NBA championship, and the year after that, they won. Previously, they’d been a mediocre team—then, suddenly, they changed coaches and became a sensation. Before 1978, journalists had generally written negatively about the Supersonics. Afterwards, the same journalists began to praise the players, celebrating what they’d previously described as the players’ weaknesses. In short, journalists began with a conclusion about the team—that it was good or that it was horrible—and then worked backwards to explain that conclusion.
Sports writers worked backwards from the conclusion that the Sonics were a bad team—as evidenced by their poor play—and then went on to rationalize their conclusion by blaming the individual players themselves.
In many ways, Steele says, social science is like sports writing: researchers start with a conclusion and then work backwards to justify it. Too often, researchers have explained the underperformance of black students by noting the students’ “psychic deficiencies.” The conventional wisdom said that black students were so used to hearing that they were inferior that they eventually accepted that they really were inferior.
Too often, social psychologists have explained minorities’ underperformance by claiming that minorities have been permanently damaged by prejudice. This defeatist way of thinking would suggest that there is nothing, or very little, that society can do to reduce the achievement gap—what’s done is done.
In 1978, Seattle sportswriters began to see the Sonics “for what they really were.” The Sonics’ winning record made it obvious that player deficiencies couldn’t be the reason for their failures in the past. Like sports journalists in 1978, Steele had found evidence that women performed as well as men on difficult math tests, provided that certain forms of stigmatization were minimized. The next step was to test if this applied to other minorities—for example, if black students would perform as well as white students on a test if researchers reduced stigmatization.
Steele himself works backward from the evidence that women can, under the right circumstances, perform at the same high level as men on math tests. Having studied this phenomenon, Steele goes on to devise further experiments to measure women’s performance on math tests in more detail, working from the hypothesis that the feeling of being stereotyped can be provoked and minimized.
In 1991, just as he was beginning to study stigmatization, Steele moved from Michigan to Stanford. His new research partner was a Princeton Ph.D. named Joshua Aronson. Together, Steele and Aronson tried to answer the question of whether all minorities (including "strong," motivated black students) were vulnerable to stigma pressure. They organized an experiment for black and white Stanford students, such that the students would be given a difficult verbal reasoning test. White students did significantly better on the test than black students, even controlling for grades and SAT scores.
Steele and Aronson turned from studying the effect of stereotyping on women to studying it in black students at one of the nation's most prestigious universities. As with his previous studies, Steele began by measuring the extent of the problem—i.e., the extent of black students’ underperformance.
Aronson and Steele’s findings posed new questions about why black students often underperformed. Steele proposed a novel way to lessen the stigma pressure of the exam: by telling the students that their exam was a "task" that specifically didn't measure intelligence. This, Steele and Aronson reasoned, would reduce the overall stigma pressure. In the end, the results of the experiment showed that black students performed at the same high level as white students when stigma pressure was lessened.
Steele and Aronson were still learning how to reduce the pressure of the stereotype threat, but they found a way to do so, with the encouraging result that black students matched white students’ performance, strengthening the idea that stereotypes could be controlled by altering students’ environment.
Steele and Aronson could now conclude three things: 1) stigma pressure was general, in the sense that it applied to at least two groups, women and black people; 2) stigma pressure was strong enough to affect the test performances of even excellent students; 3) stigma pressure can be isolated and studied in a laboratory setting—in other words, experimenters can provoke it and minimize it with simple verbal cues.
The results of Steele and Aronson’s research painted a different picture of stereotyping than the ones common in social psychology at the time. Minorities’ underperformance wasn’t the result of genetic, biological factors, nor was it caused by unalterable feelings of inferiority. Instead, stereotyping could be “turned on and off” in the course of the same experiment.
Moving forward, Aronson wanted to investigate the third conclusion of the experiment in more depth; he wanted to study what kinds of feelings stereotypes caused. He proposed asking black and white participants to rate the kinds of music they enjoyed, ranking their favorite and least favorite genres, immediately before taking a test. Aronson and Steele found that, when the test was presented to the subjects as measuring intellectual ability, black students were less likely to say they preferred traditionally black genres of music, such as jazz and hip-hop, but when the test wasn't presented as measuring intellectual ability at all, black students were more likely to say that they preferred these genres. Aronson and Steele interpreted their findings to suggest that black students were trying to avoid a stereotypical view of themselves, or “the spotlight of the negative group stereotype.” Steele began to realize that black students were taking the test “under the weight of history.”
The black students involved with the study give different answers to the survey questions based on the state of mind they’re in after taking the test. This illustrates an important concept in social psychology—the “operational definition.” Steele doesn’t have any way of understanding exactly what his subjects are thinking, but he can approximate their thoughts through the survey, advancing a makeshift, “operational’ definition of their feelings of being stereotyped. Based on these results, Steele sees that black students’ performances are impaired by stereotypes—but not in the uniform, unalterable way that previous social psychologists have described.
Steele and Aronson’s research suggested that black students of all kinds—not just poor students—were influenced by stigma pressure. Shortly after publishing the findings, Steele met with two students, Joseph Brown and Mikel Jollet. Joseph and Mikel proposed a question: Steele had studied elite Stanford students, but could he duplicate his findings in an inner-city Los Angeles school?
After every experiment Steele conducts, he considers ways of replicating his findings for different identity groups. For the time being, however, he’s restricted himself to studying test performance. This makes sense, because it’s easier to control subjects’ behavior and environment by proctoring a test—and therefore easier to see the relationship between environment and performance.
Steele, Mikel, and Joseph proceeded to study stigmatization in the Los Angeles high school Mikel had attended. Mikel proctored a half-hour exam for black and white students. To provoke stigmatization, Mikel told half of the students that the test measured verbal ability; he told the other half that the test was “an instrument to study problem solving in general,” which wasn’t intended to measure individual students’ abilities. Mikel also made a point of asking his subjects to complete a survey, in which they were asked how much they cared about school. In the end, Mikel found that black students who cared about school performed below white students when told that the test measured verbal ability. However, black students who said they didn’t care about school were less likely to underperform—and black and white students who said they didn’t care about school performed at the same level. The problem was, this was a consistently lower level.
With each experiment Steele organizes, he gets a better sense for his subjects’ thoughts and feelings during the study. Here, for example, Mikel improves on the results of Steele’s previous study by measuring the relationship between academic motivation, academic performance, and stereotyping.
Mikel’s findings suggested that ambitious, academically gifted black students were more likely to suffer from stigma pressure than less academically gifted or ambitious students. Talented black students at Mikel’s high school were like the Sonics in 1977—they had talent, but suffered from stereotype pressure. After Mikel’s experiment, Steele began using the term “stereotype threat” to describe the impact of stigmatization on people.
The experiment suggests that underperformance isn’t necessarily a sign of laziness or lack of motivation—underperformance is in fact more apparent in the most ambitious, driven students, because these are the kinds of students who think about the threat of stereotypes the most.
Steele had used social psychology to reach the conclusion that, contrary to what many people claim, universities and classrooms are different places for different people—different groups bring different mental and emotional baggage with them. But Steele’s findings also needed further research: he wanted to understand the range of behaviors with which stereotyping could interfere. He also wanted to understand how the effects of stereotypes could be canceled out, or at least limited.
As with his previous experiments, Steele’s inner-city Los Angeles high school study points the way toward further research. In this way, Steele advances his knowledge of stereotype threats and comes to understand the phenomenon in much more detail.