Wednesday, October 24, 2007

Stereotype Threat, NCLB, and the Half-Way Solution

From Ed Week:

Stanford University psychologist Claude M. Steele made headlines in 1995 with a study that introduced the phrase “stereotype threat” into the national lexicon. Put simply, it’s the idea that people tend to underperform when confronted with situations that might confirm negative stereotypes about their social group.
Could there ever be devised a more consistent, hammering confirmation of stereotype for minority children and parents living economically disadvantaged lives: the more you struggle, the steeper the hill gets over time, and the more likely you are to fail the high stakes tests on which NCLB is built?

It is sadly interesting to note, too, that now as we finally start talking about interventions to address stereotypes and the achievement gap, there is a focus almost entirely on psychological interventions--as if the poverty that drives the achievement gap can be fixed by tinkering inside the heads of students and teachers. While surely the psychological space is critical in terms of shaping learning and education, fixation on the psychological can lead to a debilitating blindness to the equally-strong sociological realities that shape children's lives.

All the tinkering inside the head can only go so far in getting to a full solution to the educational achievement gap, as John Dewey knew over a hundred years ago. Focusing on they psychological, alone, can bring more negative consequences, in fact:
--this educational process has two sides--one psychological and the one sociological--and that neither can be subordianted to the other , or neglected, without evil results following (My Pedagogic Creed, Article I).

Again, from Ed Week, an interesting chunk of the article:

Mr. Steele’s original research involved black college students whose test performance faltered when they were told they were taking an exam that would measure intellectual ability. But the effect has since been documented in more than 200 studies involving all sorts of situations.

Scholars have found evidence of “stereotype threat” occurring, for example, among elementary school girls taking mathematics tests, elderly people given a memory test, and white men being assessed on athletic ability. Even something as subtle as asking students to indicate their race or gender on a test form can trigger the phenomenon, some of those studies have suggested.

Now comes a new line of research aimed at figuring out what to do about the problem. Focusing mostly on middle schools, the second generation of work is starting to point to tools and techniques that show promise in countering stereotype threat in the classroom and improving the academic achievement of students who are most likely to suffer from its effects, such as African-Americans, Latinos, and girls.

The hope is that such interventions might one day narrow persistent achievement gaps between many minority students and their higher-achieving white and Asian American peers, and expand the ranks of young women who pursue high-level studies in mathematics and science.

“You can scare people so much in the lab that you can make existing gaps wider,” said Joshua Aronson, an associate professor of applied psychology at New York University, who is conducting much of that research. “But the real message of this work is that you also can make the gaps narrower.”

. . . .

Mr. Aronson was the co-author, with Mr. Steele, of the study that propelled the “stereotype threat” idea into the national achievement-gap discussion more than a decade ago.

In that experiment, the researchers divided black and white Stanford undergraduates into two groups and gave each group the same test. One group got the message that the test had IQ-like diagnostic properties; the other group was told that the test did not measure intellectual ability.

The African-Americans in the group performed dramatically better in the latter situation, while white students performed equally well under both conditions. Mr. Steele and his colleagues contend the black students tested badly in the first instance because they feared their poor performance would confirm a stereotype that blacks were intellectually inferior.

Few groups, though, seem to be immune from stereotype threat. Mr. Aronson found in a 1999 study, for instance, that he could induce the same sort of effect in white male engineering and math majors with “astronomical” SAT scores simply by telling them that scores from their laboratory tests would be used to study Asian students’ apparent superiority in mathematics.

Faced with the risk of confirming a negative stereotype about themselves, students react in different ways, according to researchers. While some underperform, others redouble their efforts.

Most worrisome, though, is the large number of students from vulnerable groups who, over time, begin to avoid situations that seem potentially threatening. In one of Mr. Aronson’s experiments with middle schoolers, for example, Latino students—but not non-Hispanic white students—chose easier problems when they understood that the test they were taking measured mathematical ability.

“I think we need to worry about how students’ vulnerability can lead to real differences in ability,” said Mr. Aronson. “As students shy away from those things that could make them smarter, it becomes sort of a negative spiral.”

“We think middle school is when problems emerge,” he added, “and when you see minority kids start to disengage and girls start to get anxious about math.” . . . .

No comments: