Journal of Experimental Social Psychology 37, 158 –165 (2001) doi:10.1006/jesp.2000.1444, available online at http://www.idealibrary.com on

Black Experimenters and the Intellectual Test Performance of White Participants: The Tables Are Turned Henry A. Danso and Victoria M. Esses University of Western Ontario, London, Ontario, Canada Received November 23, 1999; revised June 7, 2000; accepted June 18, 2000 This research examined the effect of race of test administrator, White vs Black, on the intellectual test performance of White participants. Social dominance orientation was also assessed as a potential moderator of effects obtained. Results indicated that participants who were tested by a Black experimenter performed better on the ability test than did those who were tested by a White experimenter. This effect was moderated by social dominance orientation such that the tendency to perform better when tested by a Black experimenter was especially evident in participants who were higher in social dominance orientation. Results are interpreted in terms of perceived threat of competition from minority group members and the motivation of majority, or dominant, group members to maintain their group dominance. © 2001 Academic Press

nature of the test), test performance is markedly improved (Aronson et al., 1998; Croizet & Claire, 1998; Spencer et al., 1999; Steele, 1997; Steele & Aronson, 1995). In addition, it has recently been found that members of majority, or dominant, groups can also experience stereotype threat. Aronson et al. (1999) demonstrated that when White participants took a test of mathematical ability after reading that Asians (a group that is stereotyped to have higher intellectual ability than Whites) outperform Whites on tests of mathematical ability, they performed worse than when this information was not provided. This is consistent with the proposition that when the stereotype about a group’s lower intellectual ability is activated, the intellectual test performance of group members can be negatively affected (see also Brown & Josephs, 1999). The current research also examined group-relevant influences on the intellectual test performance of White participants that are elicited by the testing situation. In this research, however, we focused on the situational factor of the race of the test administrator: Black versus White. In a study designed for a different purpose from that discussed here, we came upon the effect of race of test administrator serendipitously. A main study was then designed to determine whether the basic effect evident in the first study would replicate and to begin to explore possible explanations for our findings. Thus, this research focuses on the effect of race of the test administrator on the intellectual test performance of White test takers.

It has recently been demonstrated that group-relevant factors present in a testing situation can systematically affect the test performance of certain minority groups (e.g., Blacks, women, students from low socioeconomic backgrounds) (Steele, 1997; see also Croizet & Claire, 1998; Shih, Pittinsky, & Ambady, 1999; Spencer, Steele, & Quinn, 1999). According to Steele (1997; see also Aronson, Quinn, & Spencer, 1998; Steele & Aronson, 1995), stereotype threat elicited by a testing situation—the threat of being negatively stereotyped and of perhaps confirming the stereotype— can impair the test performance of members of groups about whom the relevant negative stereotype exists. For example, stereotype threat may impair the test performance of Blacks when taking an apparent test of intellectual ability and may impair the test performance of women when taking a purported test of mathematical ability. When stereotype threat is removed (e.g., by changing the reported This research was supported by an Ontario Graduate Scholarship to the first author and a Social Sciences and Humanities Research Council of Canada grant to the second author. We thank Mike Ashton, Gordon Hodson, Jim Olson, Judith Harackiewicz, and three anonymous reviewers for their helpful comments on a previous version of this article. We also thank Mike Ashton, Brad Goodyear, Stephen Gyimah, David Kipka, and Obed Mfum-Mensah for acting as experimenters and Vanessa Lees, Erin Ross, and Sara Epstein for acting as research assistants. Address correspondence and reprint requests to Henry A. Danso or Victoria M. Esses, Department of Psychology, University of Western Ontario, London, Ontario, Canada N6A 5C2. E-mail: [email protected] or [email protected]. 158 0022-1031/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

BLACK EXPERIMENTERS AND WHITE PARTICIPANTS

There are several possible reasons to expect that the race of the test administrator might affect the intellectual test performance of White test takers. First, it is possible that Whites might perceive successful-appearing Black test administrators as a threat to their dominance in the intellectual domain and, therefore, be especially motivated to maintain their dominance through superior test performance (see also Esses, Jackson, & Armstrong, 1998). That is, threat of competition from minority group members and a desire to maintain group status may cause White students to perform better on an ability test when tested by a Black experimenter. Social dominance theory (e.g., Pratto, 1999; Sidanius & Pratto, 1999) would support such a prediction. This theory suggests that in social contexts in which groups have unequal power, ideologies form to justify and maintain group hierarchy. In the present context, White participants might believe that they are superior to Blacks in the intellectual domain and are entitled to such a position of superiority. In the presence of a Black experimenter in a position of authority, they might thus work especially hard to maintain the perception of White intellectual superiority. A related possibility is that White test takers might feel especially sure of themselves when they compare themselves to Blacks, in which case one might also expect better performance when the test is administered by a Black than by a White test administrator (though overconfidence might reduce performance). That is, participants who are tested by a Black experimenter may feel more confident in their ability and less anxious about the testing situation and thus perform especially well. In contrast, based on the intergroup anxiety literature (e.g., Stephan & Stephan, 1985; Wilder, 1993), one might expect that contact with a Black test administrator would lead to increased anxiety and arousal in White test takers. In addition, concerns about doing poorly on the test and resultant fear of embarrassment might be especially likely to occur in the presence of a Black test administrator, also leading to increased anxiety and arousal, perhaps especially when feedback is anticipated. In either case, anxiety and arousal elicited by the Black test administrator could lead to poorer performance on the relatively demanding test due to distraction and cognitive interference (see Wilder, 1993). Alternatively, the enhanced fear of embarrassment might cause White test takers to put more effort into the task and possibly perform better in the presence of a Black test administrator. Thus, one might expect White test takers to perform either better or worse on a standardized test of intellectual ability in the presence of a Black test administrator. Such effects would be of interest because they would provide further evidence that members of dominant groups are also systematically susceptible to subtle group-relevant factors present in a testing situation.

159

PILOT STUDY

The pilot study was part of a larger study designed for a different purpose from that described here. Participants were 95 White undergraduates at the University of Western Ontario, who were run individually by one of two Black or one of two White male graduate students. After participants were greeted by the experimenter, a female research assistant administered a measure of attitudes toward 10 groups (see Esses, Haddock, & Zanna, 1993), including Blacks, ostensibly as part of an independent study that she was conducting. Overall, participants’ attitudes toward Blacks were relatively favorable. The mean attitude score was 79.02 (SD ⫽ 18.78) with scores ranging from 30 to 100. A t test indicated that mean attitudes toward Blacks did not differ as a function of race of the experimenter, t(93) ⫽ 1.32, ns. Participants were then asked to complete a time-limited 20-item intellectual ability test (Necessary Arithmetic Operations test from the Educational Testing Service Kit; French, Ekstrom, & Price, 1963). We also manipulated whether participants expected to receive feedback on the test. In the feedback condition, participants were told that their answers to the test would be scored while they were completing some other measures and then feedback would be provided on their performance on the test. In the nofeedback condition, participants were not informed that they would receive feedback on the test. A multiple-regression analysis was run, predicting participants’ performance on the ability test from their attitudes toward Blacks (entered as a continuous variable, centered), race of the experimenter (coded ⫺1 ⫽ Black, 1 ⫽ White), feedback condition (coded ⫺1 ⫽ Feedback, 1 ⫽ No feedback), and the various interactions among these variables. Results indicated that the race of the experimenter who administered the test significantly predicted participants’ test scores, ␤ ⫽ ⫺.25, t(87) ⫽ ⫺2.49, p ⬍ .02. Participants who were tested by a Black experimenter performed significantly better on the ability test (M ⫽ 12.70, SD ⫽ 3.52) than did those who were tested by a White experimenter (M ⫽ 11.00, SD ⫽ 3.20). In addition, participants’ attitudes toward Blacks also tended to predict their ability test scores such that less positive attitudes toward Blacks were associated with higher performance on the ability test, ␤ ⫽ ⫺.19, t(87) ⫽ ⫺1.90, p ⫽ .061. Although suggestive of an effect of race of test administer, there were also procedural differences between the Black versus White experimenter conditions that could have partially accounted for the findings. Participants who were tested by a Black experimenter and expected to receive feedback were told that their test would be scored by the experimenter, whereas those who were tested by a White experimenter and expected to receive feedback were told that their test would be scored by either the experimenter or

160

DANSO AND ESSES

by a Black experimenter down the hall. Although there were no effects of expecting to receive feedback, this difference in procedure between the Black versus White experimenter conditions left some ambiguity as to whether the race of experimenter finding could be completely attributed to race per se. MAIN STUDY

We conducted the main study to determine whether the race of experimenter findings of the pilot study would replicate when other conditions were held constant, while also beginning to examine possible explanations for the effects. To examine the group competition and dominance explanation, we included the individual difference variable of social dominance orientation (Pratto, Sidanius, Stallworth, & Malle, 1994). Pratto et al. indicate that social dominance orientation is the tendency for individuals to desire that their in-group dominate and remain superior to out-groups, and, indeed, there is considerable evidence for this claim (e.g., Altemeyer, 1998; Pratto, 1999; Sidanius, Levin, & Pratto, 1996). In addition, Esses et al. (1998) have recently shown that social dominance orientation strongly predicts the perception of zero-sum competition among groups. In the present study, it was anticipated that if group competition and desire for group dominance were operating, White participants who were high in social dominance orientation would be especially motivated to perform well when tested by a Black experimenter because high-socialdominance-oriented individuals are especially likely to feel threat from Blacks. Thus, support for the group dominance hypothesis would be obtained if we found a significant interaction between social dominance orientation and race of experimenter. In particular, we expected that participants who scored higher on the social dominance orientation scale would perform better on the ability test than would those who scored lower on the social dominance orientation scale when tested by a Black experimenter. In contrast, when tested by a White experimenter, no effects of social dominance orientation were expected. We also took steps to provide consistent and clearer instructions regarding the expectation of receiving feedback. Participants in both the Black and White experimenter conditions were given the same information regarding provision of feedback or no feedback. Participants were told either that they would receive feedback on their individual performance or that their scores would be pooled with those of other participants. We assumed that if the results of the pilot study were due to concerns about receiving a poor grade from a Black experimenter (i.e., the embarrassment explanation), then we would obtain differences in performance between those who were expecting to receive feedback and those who were not. Specifically, if the embar-

rassment explanation held, those who were told that they would receive immediate feedback on their test performance would do better than those who were told that they would not receive feedback on their performance, especially when tested by a Black experimenter. In contrast, if group competition were operating, we predicted that no effect of expecting feedback would be evident. In this case, participants would be most concerned with the performance of White students as a group. Given that the vast majority of students at the university at which participants were recruited are White, participants would expect that their group performance would be evident irrespective of whether individual feedback was provided or their scores were pooled with those of other participants in the study. Method

Participants One hundred White undergraduates at the University of Western Ontario (21 male and 79 female) volunteered to participate in a study on “Personality correlates” in exchange for course credit. The mean age was 20.13 years. Measures Ability test. Participants took the time-limited, 20-question multiple-choice “Necessary Arithmetic Operations” test from the Educational Testing Service Kit (French et al., 1963), as in the pilot study. The questions concerned arithmetic operations that could be used to solve particular arithmetic problems rather than the actual solutions to the problems. The Necessary Arithmetic Operations test was selected by French et al. as a measure of reasoning ability, and subsequent data have shown it to be a good measure of general intelligence. For example, in an analysis of over 30 cognitive ability tests, Marshlek, Lohman, and Snow (1983) found the Necessary Arithmetic Operations test to have one of the highest loadings on the g, or general intelligence, factor. Social dominance orientation. Social dominance orientation was assessed using Pratto et al.’s (1994) group version of the Social Dominance Orientation scale. The scale consists of 16 items, 8 of which are worded in the pro-trait direction and 8 in the con-trait direction. Possible mean scores on the scale range from 1 to 7, with higher scores indicating greater social dominance orientation (with the con-trait items reverse-scored). The mean score for the present sample was 2.17 (SD ⫽ .95), with scores ranging from 1.00 to 5.44 (Cronbach’s ␣ ⫽ .89). Researchers have reported an average ␣ of .92 and an average cross-time correlation of .86 for the Social Dominance Orientation scale, indicating that this measure is reliable and stable over time (Pratto, 1999).

161

BLACK EXPERIMENTERS AND WHITE PARTICIPANTS

Results

Design and Procedure The study was a 2 (Race of experimenter, Black versus White) by 2 (Feedback condition, expected feedback versus no feedback) between-subjects design. Participants were randomly assigned to these four manipulated conditions. Social dominance orientation was also included in the design, as a continuous variable. To rule out the possibility that the results of the pilot study were due to any particular experimenter’s behavior, three Black and three White experimenters (all male graduate students, the original four experimenters plus two additional ones) administered the tests. Each experimenter ran at least 15 participants individually. When a participant arrived at the laboratory, he or she was greeted by one of the six experimenters, who informed the participant that the study consisted of taking an ability test and responding to several items assessing personality. The participant then read and signed an informed consent form. Manipulation of expected feedback and no-feedback conditions. All participants were informed that they would take an intellectual ability test. In the feedback condition, the experimenter informed the participants that “your answers to the ability test will be scored while you are responding to the personality items and you will then receive feedback on your individual performance.” In the no-feedback condition, participants were told that “your answers to the ability test will be scored at a later time and averaged with the responses of other participants so you will not receive any feedback on your performance. We are not interested in your individual performance.” Ability test. After a brief introduction, the experimenter gave a booklet containing the Necessary Arithmetic Operations test to participants and explained how to go about answering the questions. The experimenter explained that the questions involved indicating which arithmetic operations could be used to solve the 20 arithmetic problems, rather than actually solving the problems. The experimenter then gave participants some time to read a detailed explanation of the test, as well as two examples of the arithmetic questions, on the cover page of the booklet. When participants indicated that they understood the task, the experimenter timed them for 7 min. Additional measures and debriefing. Participants then completed some filler measures, followed by the social dominance orientation scale and questions soliciting demographic information. In the expected feedback condition, the experimenter left the room momentarily to do the “scoring” while participants responded to these measures. In neither condition were participants actually given any feedback on their test performance. Participants were then informed that the study was over, fully debriefed, and thanked for their participation in the study.

Ability Test Scores A multiple-regression analysis was run, predicting participants’ test scores from their social dominance orientation scores (entered as a continuous variable, centered), race of the experimenter (coded ⫺1 ⫽ Black, 1 ⫽ White), feedback condition (coded ⫺1 ⫽ Feedback, 1 ⫽ No feedback), and the various interactions among these variables. Consistent with the findings of the pilot study, the race of the experimenter significantly predicted participants’ ability test scores, ␤ ⫽ ⫺.29, t(92) ⫽ ⫺2.84, p ⬍ .01. Participants who were tested by a Black experimenter performed significantly better on the ability test (M ⫽ 13.16, SD ⫽ 3.50) than did those who were tested by a White experimenter (M ⫽ 11.94, SD ⫽ 3.05). In addition, social dominance orientation also predicted ability test scores, ␤ ⫽ .38, t(92) ⫽ 3.30, p ⬍ .005. The higher participants scored on social dominance orientation, the better they performed on the ability test. Of importance, as predicted, the social dominance effect was qualified by a significant interaction between social dominance orientation and the race of the experimenter, ␤ ⫽ ⫺.29, t(92) ⫽ ⫺2.61, p ⬍ .02. This interaction indicated that higher social dominance orientation was related to better test performance in the Black experimenter condition but not in the White experimenter condition (see Fig. 1). Number of Questions Completed and Percentage of Errors Made To gain further insight into the nature of the performance results, we examined separately whether the number of questions completed and the percentage of errors made by participants differed as a function of their social dominance orientation scores (treated as a continuous variable, centered) and the manipulated conditions. A multiple-regression analysis was first performed on the number of questions completed. Results indicated that the race of the experimenter significantly predicted the number of questions completed, ␤ ⫽ ⫺.38, t(92) ⫽ ⫺3.69, p ⬍.001. Those who were tested by a Black experimenter completed more questions (M ⫽ 18.24, SD ⫽ 2.16) than did those who were tested by a White experimenter (M ⫽ 16.48, SD ⫽ 2.54). No other effects were significant. A multiple-regression analysis was also performed on the percentage of errors made by participants on the ability test. There was a significant interaction between social dominance orientation and race of experimenter, ␤ ⫽ .30, t(92) ⫽ 2.59, p ⬍ .02. Higher social dominance orientation was related to a lower percentage of errors in the Black experimenter condition but not in the White experimenter condition (see Fig. 2).

162

DANSO AND ESSES

FIG. 1. Predicted mean ability test scores as a function of social dominance orientation (1 standard deviation below and above the mean) and race of experimenter. To plot this interaction, we conducted a regression analysis predicting ability test scores from social dominance orientation, race of experimenter, and the interaction between the two variables. The social dominance orientation ⫻ race of experimenter interaction was significant, ␤ ⫽ ⫺.27, t(96) ⫽ ⫺2.66, p ⬍ .01.

Social Dominance Orientation Social dominance orientation was assessed after participants had taken the ability test (the dependent measure) to ensure that participants would not have social dominance orientation primed before taking the test. To examine the effects on social dominance orientation, we therefore conducted a regression analysis predicting social dominance orientation scores from the race of experimenter, ability test scores (treated as a continuous variable, centered), and the interaction between the two variables. Results demonstrated a significant effect of race of experimenter, ␤ ⫽ .31, t(96) ⫽ 3.15, p ⬍ .01. Participants who were tested by a Black experimenter scored lower on social dominance orientation (M ⫽ 1.92,

SD ⫽ .72) than did those who were tested by a White experimenter (M ⫽ 2.42, SD ⫽ 1.09), perhaps because of social desirability. The fact that social dominance orientation scores were affected by the race of the experimenter may weaken the causal conclusions that can be drawn from this research. However, the effects were working in opposite directions. Black experimenters led to higher ability test performance but lower social dominance orientation scores (that were associated with lower ability test performance). In addition, of importance, there was not a significant interaction between race of experimenter and ability test scores, ␤ ⫽ ⫺.11, t(96) ⫽ ⫺1.16, ns. Thus, it was not the case that participants who were tested by a Black experimenter and performed well

FIG. 2. Predicted mean percentage of errors on the ability test as a function of social dominance orientation (1 standard deviation below and above the mean) and race of experimenter. To plot this interaction, we conducted a regression analysis, predicting percentage of errors on the ability test from social dominance orientation, race of experimenter, and the interaction between the two variables. The social dominance orientation ⫻ race of experimenter interaction was significant, ␤ ⫽ .26, t(96) ⫽ 2.46, p ⬍ .02.

BLACK EXPERIMENTERS AND WHITE PARTICIPANTS

then demonstrated higher social dominance orientation scores. GENERAL DISCUSSION

Our research demonstrates that White students’ performance on a standardized test of intellectual ability can be systematically affected by the race of the individual administering the test, particularly when they are high in social dominance orientation. Results from both the pilot and main studies revealed that White participants performed better on the intellectual ability test when they were tested by a Black experimenter than when they were tested by a White experimenter. In the pilot study, it was also found that those who had less positive attitudes toward Blacks performed better on the ability test than did those who had more positive attitudes toward Blacks. In the main study, the race of experimenter effect was moderated by social dominance orientation, indicating that participants did better on the ability test when they were tested by a Black experimenter and were higher in social dominance orientation. Participants who were tested by a Black experimenter completed more questions than did those who were tested by a White experimenter. In addition, when tested by a Black experimenter, higher social dominance oriented participants made a lower percentage of errors on the ability test than did lower social dominance oriented participants. Though we did not find an interaction between attitudes toward Blacks and race of experimenter in the pilot study, we obtained the predicted significant interaction between social dominance orientation and race of experimenter in the main study. This suggests that it is not negative attitudes toward Blacks per se that occasioned higher performance when tested by Blacks, but perhaps the need to dominate. Though higher social dominance orientation is associated with less favorable attitudes toward Blacks, the social dominance orientation measure also specifically taps the need to have one’s group remain superior to and dominate other groups (Pratto et al., 1994). In addition, because the Black experimenter condition was the only intergroup condition, it might have made the high-social-dominance-oriented participants’ orientation toward intergroup relations accessible and thus elicited their group competitiveness. Of importance, the feedback manipulation did not have an impact on participants’ test scores. This suggests that heightened concerns about possible embarrassment from a poor performance in the presence of a Black experimenter cannot account for the better performance of participants who were tested by a Black experimenter. Instead, it is likely that the feedback manipulation did not have an impact on participants’ test scores because participants, particularly those high in social dominance orientation, were more concerned with the performance of their White group as a whole than with their own individual performance. Because

163

the student population at the University of Western Ontario (where the participants were recruited) is predominantly White, telling participants in the no-feedback condition that their results would be averaged with those of others still implied that the pooled results would reflect the performance of Whites. As a result, the feedback manipulation did not affect participants’ motivation to perform well on the ability test. As suggested by research on group competition, at times individuals may be motivated by group interests and a desire to protect and/or enhance group resources and status (e.g., Brewer, 1986; LeVine & Campbell, 1972; Sherif & Sherif, 1979). This motivation is likely to be especially strong in high-social-dominance-oriented individuals, who believe in group inequality and desire a hierarchical structure to society (Pratto et al., 1994; Sidanius et al., 1996). As a result, when the hierarchy is potentially threatened, highsocial-dominance-oriented individuals may strive to prove themselves in the relevant domain (see Esses et al., 1998). In line with Steele’s (1997) proposition that group-relevant factors present in a testing situation can systematically affect social group members’ performance on a standardized test, we suggest that perceived threat of competition from Blacks affects the test performance of Whites, particularly those who are high in social dominance orientation. We propose that when these White test takers are put in a situation that makes salient the perception that Blacks are making progress in the academic domain (a domain that has typically been dominated by Whites), they are especially motivated to perform well in order to prove their superiority and maintain their group dominance. In the current research, this perception was probably made salient when a Black graduate student was in the role of experimenter and administered the ability test. Further research is needed to validate this group competition and need for dominance hypothesis. In addition, the causal connection between social dominance orientation and intellectual ability test performance should be further examined. It is important to note that we are not implying that Whites will necessarily do better on an intellectual ability test simply because a Black person is present in the room. Rather, we are suggesting that when members of a particular social group feel entitled to a position, and they perceive that their position is threatened, they are likely to attempt to assert their dominance in the relevant domain by working to demonstrate superior performance. This tendency is more likely to be demonstrated by individuals who endorse hierarchical stratification of society, such as high-social-dominance-oriented individuals (see also Van Laar, Sidanius, Rabinowitz, & Sinclair, 1999). Implications Our findings may have practical implications for relations between groups for whom there has previously been an

164

DANSO AND ESSES

unequal distribution of power and resources (e.g., between men and women in managerial position or between Blacks and Whites in the United States). In such situations, perceived shifts in power balance may represent a threat to the dominance of one group and, as a result, motivate the dominant group members to work to maintain their group dominance, especially if they desire an unequal distribution of resources (see Pratto, 1999, for a discussion of the maintenance of group inequality). For instance, in research conducted in South Africa, Heaven, Stones, and Rajab (1984) predicted that White South Africans would show high achievement motivation because of the threat of competition from the surrounding Black majority. However, Heaven et al. did not obtain support for this prediction, potentially because of the timing of their study: South Africa was still under an apartheid regime in 1984. Perhaps as Black South Africans become more capable of competing for resources, White South Africans will indeed be especially motivated to achieve in order to maintain group status. It is important to note that, whereas Steele and his colleagues have found that stereotype threat impairs the performance of minority group members for whom the stereotype is relevant, the current research suggests that threat of competition and the desire to maintain group dominance enhance the performance of dominant group members, particularly those who tend to generally see the world in terms of group competition. This creates a situation in which the academic performance of Blacks may be impaired when they are primed with race and group competition, whereas the academic performance of Whites may be enhanced. This type of behavioral asymmetry is likely to widen the difference in test scores between Whites and Blacks—a situation that helps maintain hierarchical relations between the groups (see Sidanius & Pratto, 1999). Conclusions In sum, our findings suggest that White students’ performance on a standardized test of intellectual ability can be systematically affected by the race of the test administrator, especially when they are high in social dominance orientation. In particular, White students high in social dominance orientation perform better when tested by a Black experimenter than when tested by a White experimenter. We propose that group competition and the desire to maintain group dominance are currently the most viable explanations for this finding. We suspect that future research in this area will uncover further evidence of the role of group competition in determining group members’ intellectual test performance, as well as additional group-relevant influences on test performance.

REFERENCES Altemeyer, B. (1998). The ‘other authoritarian personality.’ In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 30, pp. 47–92). San Diego: Academic Press. Aronson, J., Lustina, M. J., Good, C., Keough, K., Steele, C. M., & Brown, J. (1999). When White men can’t do math: Necessary and sufficient factors in stereotype threat. Journal of Experimental Social Psychology, 35, 29 – 46. Aronson, J., Quinn, D. M., & Spencer, S. J. (1998). Stereotype threat and the academic underperformance of minorities and women. In J. K. Swim & C. Stangor (Eds.), Prejudice: The target’s perspective (pp. 83–103). San Diego, CA: Academic Press. Brewer, M. B. (1986). The role of ethnocentrism in intergroup conflict. In S. Worchel & W. Austin (Eds.), The psychology of intergroup relations (pp. 88 –102). Chicago: Nelson. Brown, R. P., & Josephs, R. A. (1999). A burden of proof: Stereotype relevance and gender differences in math performance. Journal of Personality and Social Psychology, 76, 246 –257. Croizet, J. C., & Claire, T. (1998). Extending the concept of stereotype threat to social class: The intellectual underperformance of students from low socioeconomic backgrounds. Personality and Social Psychology Bulletin, 24, 588 –594. Esses, V. M., Haddock, G., & Zanna, M. P. (1993). Values, stereotypes, and emotions as determinants of intergroup attitudes. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition. and stereotyping: Interactive processes in group perception (pp. 137–166). New York: Academic Press. Esses, V. M., Jackson, L. M., & Armstrong, T. L. (1998). Intergroup competition and attitudes toward immigrants and immigration: An instrumental model of group conflict. Journal of Social Issues, 54, 699 – 724. French, J. W., Ekstrom, K. B., & Price, L. A. (1963). Kit of reference tests for cognitive factors. Princeton, NJ: Educational Testing Service. Heaven, P., Stones, C., & Rajab, D. (1984). Levels of achievement motivation in South Africa. The Journal of Social Psychology, 122, 277–278. LeVine, R. A., & Campbell, D. T. (1972). Ethnocentrism: Theories of conflict, ethnic attitudes, and group behaviour. New York: Wiley. Marshlek, B., Lohman, D. F., & Snow, R. F. (1983). The complexity continuum in the radex and hierarchical models of intelligence. Intelligence, 7, 107–127. Pratto, F. (1999). The puzzle of continuing inequality: Piecing together psychological, social, and cultural forces in social dominance theory. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 31, pp. 191–263). San Diego: Academic Press. Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67, 741–763. Sherif, M., & Sherif, C. W. (1979). Research on intergroup relations. In W. G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 7–32). Monterey, CA: Brooks/Cole. Shih, M., Pittinsky, T. L., & Ambady, N. (1999). Stereotype susceptibility: Identity salience and shifts in qualitative performance. Psychological Science, 10, 80 – 83. Sidanius, J., Levin, S., & Pratto, F. (1996). Consensual social dominance orientation and its correlates within the hierarchical structure of American society. International Journal of Intercultural Relations, 20, 385– 408.

BLACK EXPERIMENTERS AND WHITE PARTICIPANTS Sidanius, J., & Pratto, F. (1999). Social dominance: An intergroup theory of social hierarchy and oppression. New York: Cambridge Univ. Press. Spencer, S. J., Steele, C. M., & Quinn, D. M. (1999). Stereotype threat and women’s math performance. Journal of Experimental Social Psychology, 35, 4 –28. Steele, C. M. (1997). A threat in the air: How stereotypes shape intellectual identity and performance. American Psychologist, 52, 613– 629. Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology, 69, 797– 811.

165

Stephan, W. E., & Stephan, C. W. (1985). Intergroup anxiety. Journal of Social Issues, 41, 157–175. Van Laar, C., Sidanius, J., Rabinowitz, J. L., & Sinclair, S. (1999). The three Rs of academic achievement: Reading, ‘riting, and racism. Personality and Social Psychology Bulletin, 25, 139 –151. Wilder, D. A. (1993). The role of anxiety in facilitating stereotypic judgment of outgroup behavior. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp. 87–109). New York: Academic Press.