Stanford dean shares research on 'stereotype threat'

Why do black students, with equally good SAT scores, end up with worse college GPAs than their white counterparts?

Why do female math majors, going in with performance equal to that of male students, fall behind in advanced classes?

Social psychologist Claude Steele, the new dean of the Stanford University School of Education, shared his findings on those questions in a campus talk Monday (Oct. 31).

Steele has spent much of his career exploring performance anxiety based on perceived negative consequences of stereotypes.

His results shed light on the racial achievement gap. Data from the Palo Alto Unified School District, discussed by the Board of Education last week, showed that only 3 of the 20 African-American students who graduated in 2011 had completed the prerequisites for California's four-year state universities, compared to 80 percent of all district graduates.

When people are in situations where they feel they must refute a stereotype, they get distracted and anxious, wasting valuable cognitive resources in trying to perform, Steele said.

Thus, in one of Steele's early experiments, high-ability women performed worse than their male counterparts on a frustrating math test. But when a subsequent group of subjects was given the same difficult test -- but told in advance that men and women had always performed equally on the test in the past -- the women's performance rose to match that of the men, he said.

Steele has done similar work with what he calls "stereotype threat" as it pertains to race.

Told that a nonverbal test involving squares and designs was related to IQ, blacks performed a standard deviation worse than whites on the task.

"But if you give the exact same test and say it has nothing to do with your cognitive abilities -- 'just have fun' -- blacks score just the same as whites," he said.

"I think the best account of it is when an African-American is taking a difficult IQ test, there's that stereotype out there that's ancient about intelligence, and you allocate some of your cognitive resources to defending against that stereotype," Steele told more than 100 people attending his lunchtime talk.

In another context, engineering graduate students at Stanford underperform on a test when told that Asians generally do better than whites, he said.

Steele's talk was part of a series of brown bag seminars sponsored by the Stanford Center for Opportunity Policy in Education (SCOPE).

"We need a greater appreciation for the role of context in the way we function," said Steele, noting the long-term impacts of "stereotype threat" on a woman who chooses to go into math as a long-term career.

"We tend to think of the individual in a decontextualized way. When we're thinking about level of performance on a test or level of functioning in a class, we don't tend to think of the role that context can play," he said.

Steele recalled, as a child of 7 or 8, being confused when he learned that black kids couldn't swim at the neighborhood pool except on Wednesday afternoons, and couldn't use the roller rink except on Thursday nights.

"In hindsight, I was for the first time recognizing a condition of my life tied to race," he said.

Similarly, female students in the sciences might have been discouraged to hear then-Harvard University President Larry Summers suggest there could be a biological difference between men and women regarding math ability, Steele said.

"Stereotype threat comes when a person knows at some level they could be judged or treated in terms of a negative stereotype," he said.

"If you care about what you're doing, the prospect of being reduced automatically to a negative stereotype is upsetting and distracting."

As for remedies, Steele said, "We should do as much as we can to reduce the cues and contingencies tied to important situations, like a school classroom.

"There should be a self-conscious effort to examine a classroom, a workplace, for those cues (that could suggest people will be judged based on stereotypes)."

A good topic for future research would be how to help individuals develop "internal narratives" to help them withstand stereotype threats they may encounter in the future, he said.