The University of Iowa
The University of Iowa News Services Home News Releases UI in the News Subscribe to UI News Contact Us


100 Old Public Library
Iowa City IA 52242
(319) 384-0011; fax (319) 384-0024

Release: Immediate

UI researchers find fear of success lowers IQ; may explain lower minority test scores

IOWA CITY, Iowa -- Students from disadvantaged backgrounds may earn lower scores on standardized tests because they fear that academic success will alienate them from friends or family members and arouse suspicion among teachers, a University of Iowa study has found.

Michael Lovaglia, a UI professor of sociology and the principle investigator of the study, found that a person's performance on a standardized test is strongly affected by what that person expects from a high score. For example, students from disadvantaged groups may expect that their teachers will suspect them of cheating or that their friends will consider them "nerds" if they achieve a high score on a standardized test.

The study provides a new window for research on the reasons many students from minority groups achieve lower scores than white students, even when their educational backgrounds are similar. Lovaglia suggests that many minority group members worry about facing a hostile reaction ­ both from other minority students as well as from white students -- if they demonstrate high mental ability and that concern causes them to score lower on standardized tests.

Lovaglia said that on average, black students score about 10 IQ points lower on standard tests of intelligence than do white students with similar social backgrounds. "People have been studying the differences in test scores for years, trying to determine what caused the disparity but could not find a social process that was responsible," he said. "In addition, those studies could not rule out genetic differences as a possible explanation. Our study was conducted under controlled laboratory conditions so that we could rule out genetics and other external factors. Our study is the first to demonstrate that an individual's perceptions of others' reactions to his or her success can have a negative effect on IQ scores."

Published in the current issue of "The American Journal of Sociology," this study is important for research on the biases of standardized tests because it is the first one that rules out factors unique to particular groups, such as genetic differences, Lovaglia said. Previous research conducted by professor Claude Steele of Stanford University had shown that reducing fears about the consequences of a high score on a standardized test could increase the test scores of

African-American students. However, it was unknown whether the effect was limited to African Americans.

This new study tackled this question to show that white students who expected negative consequences from a high standardized test score would also earn lower scores than students who expected positive consequences, Lovaglia said. This finding allows researchers to generalize that the expected consequences for the outcome of a test can affect the scores of a wide variety of test-takers who worry about a high test score, even those usually considered privileged.

An example of students who faced negative consequences from a high standardized test score that may be familiar to many Americans comes from the 1988 film, "Stand and Deliver," a true story starring Edward James Olmos as Jaime Escalante, a calculus teacher at Garfield High School in inner-city Los Angeles. The movie depicts students in his Advanced Placement Calculus class dealing with unsupportive parents and other family members and being taunted by friends for spending so much time studying. In addition, when they achieve high scores on the A.P. Calculus exam they are accused of cheating because officials at the company that scored the test do not believe that students from an impoverished background are capable of success on the rigorous exam.

"This example illustrates two important aspects of standardized test performance," Lovaglia said. "First, minority students often have good reason to fear the consequences of a high test score. Second, and more importantly, changing people's expectations about the consequences of a high test score, as Jaime Escalante did for his students, can change their score on a standardized mental ability test."

To reach this new conclusion, Lovaglia and his UI colleagues, Jeffrey Lucas, Jeffrey Houser, Shane Thye, and Barry Markovsky, conducted a series of experiments in which people were brought into the laboratory individually and then were randomly assigned to one of two social groups. Members of one group were treated as a privileged majority and members of the other as a disadvantaged minority. The effect was achieved by bringing people into the laboratory and individually testing their aptitude for work on a team project that would include both advantaged and disadvantaged group members.

The instructions explained that it was rare for a member of the disadvantaged group to score high enough to achieve the highest level of pay and responsibility. The instructions further explained that in the past, members of the disadvantaged group who had achieved the necessary score for high pay and responsibility had performed poorly in the team exercise and were therefore not trusted by team members. The participants were also told that team members were resentful of disadvantaged group members who had achieved the highest level of pay and responsibility.

After about half an hour of such instruction, all participants were given a standard test of mental ability, the Raven Progressive Matrices, under identical test conditions. Those people assigned by chance to the disadvantaged group scored seven to eight IQ points lower than did those people assigned to the advantaged group. Because participants were college students who had been randomly assigned to either the advantaged or disadvantaged group, members of the two groups could be expected to have similar IQ levels as the experiment began.

Lovaglia said that while the 7-point IQ difference produced by the study was not as large as the 10-point difference between blacks and whites in the United States, it was nonetheless a substantial difference. "Remember that the 7-point difference was created by half an hour of instructions whereas differences in expectations between African-Americans and European-Americans build up over years," he said.

The researchers concluded that differences in social status must be considered in any attempt to measure mental ability.

"You can't compare the standardized test scores of members of different social groups because something other than ability is affecting those scores," Lovaglia said. "We need a better way to decide who gets the 'good stuff' in society -- who gets to go to college, who gets the good jobs."

(Editors note: Lovaglia is available for interviews and is willing to demonstrate for reporters the test used to give his research subjects the feeling of being part of an "advantaged" or "disadvantaged" group. He can be reached at (319) 335-2494.)