Gender and racial differences in standardized test scores have received a lot of coverage in the popular press. An article in yesterday's New York Times discussed how simply combining populations with different economic status can result in increased test scores—apparently just being around kids with different educational expectations can have an impact on performance.
We've talked here about research which indicates that simply reminding test-takers of stereotypes about their gender or racial group can impact their performance. But these tests are typically conducted on adults. Do children have the same stereotypes? And if they do, then how does it affect their performance?
Nalini Ambady led a team that examined those questions by studying a group of Asian-American children ranging from kindergarten through 8th grade. All the kids were given grade-appropriate questions from the math section of the Iowa Test of Basic Skills. To determine the effect of stereotypes on the children, before the test, they were asked to perform one of three activities based on age and gender. K through 2 girls colored one of three pictures: a girl holding a doll (gender identity), two Asian kids eating from a bowl with chopsticks (Asian identity), or a landscape (control condition). Older kids answered a set of questions designed to activate gender identity ("Do you think boys and girls are treated differently at your school?"), Asian identity ("Do you speak a foreign language at home?"), or neutral responses ("What is your favorite season?"). The questions and pictures were changed appropriately when boys took the test.
You might expect that younger children, less aware of stereotypes that boys are better at math than girls and Asians are better at math than non-Asians, would not be affected by the identity activation tasks. However, the actual results were quite different:
The youngest girls performed substantially worse when gender identity was activated compared to the neutral condition, and did better when Asian identity was activated. By third grade, however, girls did best when their gender identity was activated, and the Asian identity condition wasn't significantly different from the neutral condition. By middle school age, girls' performance matches that of adults.
For the youngest boys, both gender and race identity led to higher scores. In late elementary school, the Asian identity led to lower scores, but again, by middle school, their performance tended to match adult performance.
When the kids were explicitly asked about math skills, they tended to say that boys and girls, and Asians and Caucasians, performed the same. However, if they were shown pictures of an Asian and Caucasian girl, and asked which one was probably better at math, most children pointed to the Asian girl. So even though they seem to implicitly "know" that Asians are better, this stereotype doesn't always impact their test results the same way.
The bad news, Ambady et al. argue, is that stereotype threat can affect even very young children. The good news is that these stereotypes and their impacts are readily changed. Positive role models and positive stereotypes may be the best way to counter the negative stereotype threat. Perhaps some the success the Raleigh schools have had is related to this phenomenon.
Ambady, N., Shih, M., Kim, A., & Pittinsky, T. L. (2001). Stereotype susceptibility in children: Effects of identity activation on quantitative performance. Psychological Science, 12(5), 385-390.
[...] The impact of stereotypes clearly is complex ¡ª we've reported on positive, negative, and neutral effects (as in the case of gender here). Perhaps this experiment's findings on Asian-American women won't be replicated with other groups. What's certain is that stereotypes do have an important impact on performance. It's possible that the most important reminder this study may offer is not to put too much faith in the results of a single test, whether it purports to measure math ability, IQ, or some other skill. Cheryan, S., & Bodenhausen, G.V. (2000). When positive stereotypes threaten intellectual performance: The psychological hazards of ¡°model minority¡± status. Psychological Science, 11(5), 399-402. [...]