Stereotype threat, social class, gender, and academic ...

Social Psychology of Education 4: 295–310, 2001. © 2001 Kluwer Academic Publishers. Printed in the Netherlands.

295

Stereotype threat, social class, gender, and academic under-achievement: when our reputation catches up to us and takes over JEAN-CLAUDE CROIZET1, , MICHEL DÉSERT2, MARION DUTRÉVIS3 and JACQUES-PHILIPPE LEYENS4

1 Laboratoire de Psychologie Sociale de la Cognition (ESA CNRS 6024), Université Blaise Pascal, 34 avenue Carnot, 63000 Clermont-Ferrand, France 2 Catholic University of Louvain-La-Neuve, Belgium 3 Université Blaise Pascal, Clermont-Ferrand, France 4 Catholic University of Louvain-La-Neuve, Belgium

Abstract. According to Steele (1997), negative stereotypes about intellectual abilities can act as a threat that disrupts the performance of students targeted by bad reputations. Previous research on stereotype threat has showed that on a stereotype-relevant test, stigmatized group members (e.g., African Americans) performed worse than others on an intellectual verbal task. However, when the instructions accompanying the test did not create stereotype threat, stigmatized group members’ performance was equal to that of other participants. In this paper, we present studies documenting the effect of stereotype threat and discuss ways to counter it. Two strategies derived from SelfCategorization Theory (Turner & Oakes, 1989) and Self-Affirmation Theory (Steele, 1988) are presented, tested, and discussed.

1. Introduction Education has traditionally been viewed as the principal route to upward mobility in industrialized societies. Indeed, if in the past the position of a given person in the social hierarchy was partly fixed and determined by her or his race, social class, or gender, nowadays this position is mainly determined by her or his academic achievement. The higher people achieve in school, the higher their social status in society. Thus, today achievement is not a question of group inheritance; rather, it depends on people’s potential as individuals, on their ability, and work. In other words, the social hierarchy produced by schooling is no longer thought to be arbitrary; social status is the structural reflection of psychological differences in ability. Although academic achievement is assumed to be dissociated from social origins, researchers have often reported that in many societies members of lower status groups have relatively poor academic performance (for a review see Sidanius & Pratto, 1999). Author for correspondence: E-mail: [email protected]

296

JEAN-CLAUDE CROIZET ET AL.

Let us note briefly some examples concerning the under-achievement of three such groups: African Americans, students of low socioeconomic status (SES), and women. In the United States, on the 1997 SAT tests for high school seniors, out of a possible combined score of 1600 (for the verbal and math proportions combined), the average score for African Americans was 857 and for Whites, 1052. With regard to academic achievement, data from national assessments showed that 43% of 12th-grade African American students did not master the basic reading level to work proficiently at their grade level, which was the case for only 17% of their White peers (National Center for Education Statistics, 1999). Concerning mathematics, it appeared that 62% of 12th-grade Black students scored below the basic achievement level as compared to 21% of White students (National Center for Education Statistics, 1997a). This difference also extends to college where the U.S. national dropout rate for African American students (those who begin college but do not finish within 6 years) is 62% as compared to the national dropout rate of 40% for White students (Nettles, 1988). The academic under-achievement of students from low socioeconomic status is also well known. In the United States, 43% of the 12th-grade students whose parents did not finish high school scored below basic reading achievement levels as compared to 15% of those whose parents graduated from college (National Center for Education Statistics, 1999; the percentages for mathematics achievement are 58 and 21, respectively, National Center for Education Statistics, 1997a). In France, 58% of low SES students (i.e., students whose parents’ occupations are manual labor workers and unqualified employees) never reached 12th grade, while it was the case of only 10% of their high SES peers (i.e., students whose parents’ occupations were managers, other professionals, researchers, and college professors). Although low SES French students represent 40% of the people from 20–24 age group, they represent only 7% of the students from the prestigious public preparatory classes for the Grandes Ecoles (the attendance being tuition free). High SES students on the other hand, if they represent only 13% of the youth from the 20-24 age group, represent 51% of the students of these prestigious classes (DEPMEN, 1997). Finally, with regard to gender in the United States, the National Assessment of Education Progress (National Center for Education Statistics, 1999) reported that girls do better than boys on reading achievement tests and also that male students outperformed their female classmates in science (National Center for Education Statistics, 1997b). And, an extensive analysis of data from 1960 to 1994 showed that among students who scored in the top distribution on mathematics and science achievement tests, boys were overrepresented (up to two boys for one girl, see Nowell & Hedges, 1998). Many explanations have been proposed to account for these dramatic differences in scholastic achievement. Among the most convincing is the suggestion that economic inequalities which exist among these groups create differences in academic preparation. This inequality exists at both the district and the family levels.

STEREOTYPE THREAT AND ACADEMIC UNDER-ACHIEVEMENT

297

Because public school funding is based on local property taxes in the United States, district spending per pupil varies considerably from a poor neighborhood to a wealthier one (e.g., in the New York city area, the spending per pupil can vary from $7,299 to $15,084 – see Kozol, 1991). Poorer families also cannot provide for preschool program costs, educational resources such as books, or even for physiological necessities such as enough food or a safe environment (Ricciuti, 1993; Sigman, 1995). Some other authors have suggested that socialization differences among groups may contribute to students’ levels of motivation and interest in academic work (Bourdieu & Passeron, 1964, 1970; Eccles, 1994; Ogbu, 1994). Finally, certain scholars claim that intelligence has a genetic component that creates different levels of intellectual ability among social groups (Herrnstein & Murray, 1994). Though this latter explanation remains speculative (Gould, 1996), it is very popular among lay people, which produces pernicious effects. Our research program investigates one of these negative consequences: stereotype threat. In the present article, our goal is to demonstrate that the wide dissemination of reputations of lower ability that usually target low status group members can contribute to their academic under-achievement. To make this point, we focus on different low status groups taken from different national and cultural contexts in order to suggest some degree of cultural generality of the problem – African Americans, French students from low socioeconomic backgrounds, and women. We present research that shows how reputations of lower ability (e.g., stereotypes) can act as psychological threats that undermine academic performance. Our goal is to explain how these stereotypes can produce under-achievement and how this dynamic ultimately contributes to the justification of the inequalities between social groups. In that sense, we hope that the research we offer below contributes to debates about the reasons why some groups do not succeed as well as others.

2. Reputation of lower ability Certain social groups are targeted by reputations of low ability. Countless studies have demonstrated the existence of such beliefs (Fiske, 1998). African Americans have the reputation of being hostile, lazy, and stupid (Devine & Elliott, 1995). People from low socioeconomic backgrounds have the reputation of being ignorant and lazy (Bullock, 1995). As for women, several studies have shown that they were thought to have less ability in mathematics and related domains (Eccles, Jacobs, & Harold, 1990). There is widespread agreement within our societies about the personality traits that characterize these social groups (Devine, 1989). This consensual knowledge may stem from communicative processes that play a central role in the acquisition of stereotype (Ashmore & Del Boca, 1981; Bullock, 1995). Actually, stereotypes are so widely held that they are not identified by lay people as beliefs but as facts (Rothbart & Taylor, 1992).

298


Social psychology has demonstrated that these stereotypical beliefs were extremely powerful and that they could lead targets ultimately to confirm them (Snyder & Stukas, 1999). For example, researchers have documented that stereotypes of class (Gilly, 1984; Rist, 1970, Williams, 1976), of gender (Skrypnec & Snyder, 1982), and of African Americans (Word, Zanna, & Cooper, 1974) could lead the perceiver to behave toward targets differently and that this differential treatment may cause these targets to confirm the perceiver’s negative expectations (for such expectancy effects on academic achievement see Jussim, Eccles, & Madon, 1996). Because these negative reputations are known by anyone, research has started to examine their influence on the internal state of the targets of these stereotypes (Crocker & Major, 1989; Steele, 1992, 1997; Crocker, Major, & Steele, 1998; Swim & Stangor, 1998). Results have suggested that in some situations awareness of these negative stereotypes can influence targets to behave negatively, even in the absence of any concurrent negative treatment. 3. Stereotype threat model According to Steele (1992, 1997; Steele & Aronson, 1995), when a widely known negative stereotype (e.g., poor intellectual ability) exists about a group, it creates for its members a burden of suspicion that acts as a threat. This threat arises whenever individuals’ behavior could be interpreted in terms of a stereotype, that is, whenever group members run the risk of substantiating the stereotype. For example, when African Americans or students of low socioeconomic status (SES) take an academic exam, they are subject to the threat that if they fail their behavior may confirm a reputation of lower intelligence. When the stakes are high for the individual, the very possibility of substantiating a negative allegation may in fact impair performance because it is self-threatening enough to have disruptive effects on its own. In other words, evaluative situations create an extra pressure for stereotyped targets that may interfere with their performance (see Figure 1). Depending on the situation of threat, several processes may contribute to this inefficiency of processing – through distraction from the task (e.g., Easterbrook, 1959),

Figure 1. Schematic representation of the stereotype threat model.


299

interfering self-consciousness (e.g., Baumeister, 1984), evaluation apprehension (e.g., Geen, 1985), test anxiety (e.g., Sarason, 1972), and/or loss of motivation (e.g., Pyszczynski & Greenberg, 1983). In dramatic support for the stereotype threat hypothesis, research has shown that minor changes in testing situations that have altered the amount of stereotype threat for a group produced significant differences in performance on an identical task. When instructions described a test as a ‘measure of verbal abilities and limitations,’ Black college students performed worse than White students (Steele & Aronson, 1995). Under the same conditions, French students from low socioeconomic backgrounds performed worse than their high SES peers (Croizet & Claire, 1998). These gaps in performance were congruent with previously documented differences in academic achievement between Whites and Blacks, and between rich and poor students, as well as the commonly held stereotypes of Black and low SES academic abilities. However, instructions that described the identical test as a cognitive psychology study rendered the test irrelevant to the stereotype and erased the differences in performance between the two groups (see Figures 2 and 3). Analogous results were found for differences between men and women on a difficult math ability test (Spencer, Steele, & Quinn, 1999, see Figure 4). This research on stereotype threat showed that an interaction between the social identities of stereotyped group members and the immediate social context (i.e., the characterization and difficulty of the task) could produce a situational predicament (i.e., a sense of stereotype threat) that may disrupt reasoning and, as a consequence, depress performance (see also Huguet, Brunot, & Monteil, this issue).

Figure 2. Number of Verbal GRE items correctly solved (adjusted by SAT) as a function of race and characterization of the task (adapted from Steele & Aronson, 1995, Exp. 2).

300


Figure 3. Number of Verbal GRE items correctly solved as a function of socioeconomic status and characterization of the task (adapted from Croizet & Claire, 1998).

Figure 4. Number of Math GRE items correctly solved (corrected for guessing) as a function of sex and characterization of the task (adapted from Spencer, Steele, & Quinn, 1999).

4. Stereotype threat as a situational predicament Until recently, however, research on stereotype threat has been limited to stigmatized groups. Therefore, it was difficult to assess the extent to which stereotype threat may also be situational because it was possible that it may have tapped into an internalization of inferiority that is rooted in the way the person happened to be socialized, and that this is sometimes the cause for drop in performance


301

(e.g., Allport, 1954; Clark, 1965; Eccles, 1994). To examine this issue directly, we investigated whether members of a usually dominant group (i.e., White men) could be threatened by a negative stereotype that is not considered to be generally a stigma (i.e., deficiency in processing affective information). Our reasoning was that if such a group could experience stereotype threat, then it is very unlikely that chronic lower status or a history of stigmatization are preconditions of stereotype threat. This research extended stereotype threat effects outside of the academic domain and to a nonstigmatized group (Leyens, Désert, Croizet, & Darcis, 2001). Dominant groups are less likely to have stigmas, but they may still be characterized in negative stereotypic terms. For example, men are stereotyped as lacking of emotion and being insensitive. Though problematic and negative, the lack of affective skills is not considered crucially debilitating. Because stereotype threat is not dependent on chronic lower status and history of stigmatization, we expected that male students may experience stereotype threat when they are at risk of confirming a reputation of lower affective skills. To test this prediction, female and male undergraduate students performed three decision tasks. They had to decide as fast as possible whether stimulus words were indeed words or nonwords (i.e., a lexical task), whether they were positive or negative (i.e., a valence task), or whether the words were affective or not (i.e., an affective task). Half of the participants were told that, in general, men are poorer performers than women in affective processing tasks (i.e., a threat condition). The results showed no differences between conditions for the lexical and valence tasks. By contrast, for the affective task, threatened males made significantly more errors than participants in the other three conditions. More precisely, threatened males tended to accept as affective words that were not affective. Thus, male students were sensitive to the fear of confirming a negative reputation, and this fear had disruptive effects on their performance. Recent research has provided strong evidence in favor of this situational stance of stereotype threat (see Aronson, et al., 1999; Brown & Joseph, 1999; Stone, Lynch, Sjomeling, & Darley, 1999). For example, Aronson et al. (1999) showed that European American males were able to experience a disruptive stereotype threat during a mathematics exam when they were reminded of the superiority of Asian Americans’ ability in mathematics. Under such a threatening condition, their performance dropped dramatically. To sum up, these studies bring a converging evidence in favor of a situational interpretation of stereotype threat because slight situational changes can lead the dominant group members to underperform in a very similar way as African Americans, low SES students, and women usually do.

5. Discarding or reducing stereotype threat Though these previous studies do not tell much about the mediation of stereotype threat, they are clear enough in demonstrating that exposure to or even discrimination based on stereotypes does not diminish the abilities of stigmatized people

302


and that, given the right situation, they are able to demonstrate performance that are equal to that of their peers. This basic result is important for two reasons. First, it suggests that certain explanations for the under-achievement of certain groups (e.g., genetic deficiency) have utterly ignored a major force which generates differences in achievement. Second, it suggests that there is some hope for restoring more equality in school outcomes through appropriate situational changes. However, these previous efforts carry their own limitations. Even if the studies showed that Black, poor students, or women were able to perform as well as the others, they have also implied that these college students cannot infer they have higher intellectual abilities because, for them, increased performance is not relevant to academic achievement. Indeed, when it is the case (i.e., the task is presented as diagnostic of intellectual ability), the threatening negative allegations of the stereotype are lurking, always ready to impair performance. This shows how insidious reputations can prevent social change. To deal with these specific limitations, we have been searching for means to alleviate stereotype threat while keeping students in an ability diagnostic situation. Two research programs have been developed. The first one examines the impact of levels of self-categorization as a way to reduce stereotype threat (see also Turner & Oakes, 1989). The second program aims at examining whether stereotype threat can be reduced through self-affirmation (see Steele, 1988).

5.1. REDUCING STEREOTYPE THREAT THROUGH SELF - CATEGORIZATION Stereotype threat has been conceptualized as a fear of being seen as confirming the stereotype or to be judged only according to it (Steele & Aronson, 1995). Stated otherwise, people subjected to negative stereotyping are under the risk of no longer being seen as individuals but only as prototypical members of a group. As a consequence, stereotype threat should make people particularly willing to reduce the interpretive self-relevance of the stereotype. Initial research by Steele and Aronson (1995) provided evidence congruent with such a prediction. First, these researchers showed that a threatening situation led to a greater activation in memory of thoughts related to the stereotype. Black students in the diagnostic condition completed more words related to the stereotype of their group in a wordfragmented completion task. Second, participants under stereotype threat reported less preference for activities (e.g., basketball, rap music) and traits (e.g., aggressive) stereotypic of Blacks. The reverse was observed for the African American participants who were not threatened. Finally, when left with the option of reporting their race, only 25% of the African Americans did so in the diagnostic condition, whereas all participants in the other conditions did. Though not conclusive, these data suggest that threatened participants try to avoid looking like stereotypical members of their group. This is congruent with the conception of stereotype threat as a fear of being perceived only through the stereotype. Because


303

repeated attempts at reducing stereotype threat requires energy, this extra work may very likely be responsible for the observed drop in performance. Conversely, this implies that a reduction of the self-relevance of the stereotype (see Figure 1) should decrease the stereotype threat and, therefore, reduce its negative consequence on performance. According to the self-concept literature, the self-concept is a complex representation that includes a vast variety of available self-knowledge (Rhodewalt & Agustdottir, 1986; Baumeister, 1995). At a given time, only some aspects of this self-knowledge are accessible; they constitute the working self-concept (Markus & Wurf, 1987). Within this framework, stereotype threat could be conceptualized as the joint accessibility in memory of both the negative stereotype content and the self-knowledge related to the stereotyped social identity. Since this self-identity negotiation can undermine performance, appropriate changes in the working selfconcept may reduce the threat and its costly consequences. As pointed out earlier, this strategy is spontaneously used by threatened participants (Steele & Aronson, 1995, Expt. 3). However, a more radical way to reduce the threat may consist of increasing the accessibility of other aspects of one’s identity. In line with Social Identity Theory (Tajfel & Turner, 1979) and Self-Categorization Theory (Turner & Oakes, 1989), these other aspects may be related to another social identity (i.e., other group membership) or to specific elements of the personal identity which contribute to people’s sense of uniqueness. Focusing on another social identity or on one’s personal identity rather than on the threatened social identity should then reduce stereotype threat. To test this hypothesis, Belgian male and female college students were asked to solve some arithmetical problems. The task was relatively straightforward. Participants had to decide as fast as possible whether basic equations (e.g., 5 × 8 = 40, 9 × 8 = 72) were correct or not. The equations were successively presented on a computer, and participants were only given 3 s to answer. For one-half of the participants, the task was presented as a study designed to explore why women are under-represented in certain scientific and mathematics domains compared to men and whether it was related to some kind of lower ability (i.e., a threatening condition for women). For the remaining participants, the task was introduced as part of an investigation on short-term memory (i.e., a non-threatening condition). To test the hypothesis that increasing the accessibility of other aspects of one’s identity may reduce the self-relevance of the stereotype, other participants in a similar threatening condition were asked to describe themselves on a questionnaire just before starting on the test. The results turned out to be congruent with our hypothesis (see Figure 5). When women experiencing stereotype threat were given a chance to reduce the self-relevance of the stereotype by describing themselves as a unique person, their performance was restored (Désert, Leyens, Croizet, & Klopfenstein, 2001). This first result suggests that focusing on other aspects of oneself that render the stereotype temporarily less self-relevant may be an efficient strategy for escaping the threat.

304


Figure 5. Number of arithmetical equations correctly verified as a function of gender and characterization of the task (adapted from Désert, Leyens, Croizet, & Klopfenstein, 2001).

5.2. REDUCING STEREOTYPE THREAT THROUGH SELF - AFFIRMATION According to Steele (1997), stereotype threat is a general threat to the individual’s sense of self-integrity. It jeopardizes a self-perception of ‘adaptive and moral adequacy’. Research on self-affirmation theory (Steele, 1988) has demonstrated that threat to self-integrity resulting from dissonance manipulation (Steele, Spencer, & Lynch, 1993), or from negative feedback (Fein & Spencer, 1997), can be alleviated when individuals are offered the possibility to access alternative positive selfconceptions that restore self-image. To be self-affirmative, the knowledge people recruit needs to be of high personal importance (i.e., a central value) ‘at least as important to the individual’s perception of self-adequacy as are the negative images inherent in the threat’ (Steele, 1988, p. 291). If stereotype threat can be thought of as a threat to self-integrity, then one should expect that giving threatened participants the opportunity to affirm important aspects of their self-concept just before taking a test should reduce this sense of self-threat and, as a result, could alleviate its negative consequences on performance. To test this prediction, female college participants only were part of an experiment that was a close replication of the one described in the previous section. Again, they had to verify arithmetic equations, this time within 1850 ms. For one half of them, the task was described so as to trigger stereotype threat (see above). For the other half, it was presented as being designed to test hypotheses about short-term memory processes. Participants in the experimental condition under focus were required to fill out a questionnaire that was ostensibly unrelated to the present experiment (see Croizet, Désert, Dutrévis, & Leyens, 2001). The questionnaire emanated supposedly from the psychology department and took the form of a survey designed to understand why some students volunteer to participate in experimental studies beyond credit. This questionnaire was framed in such a way so that it led participants to indicate that they were


305

Figure 6. Number of arithmetical equations correctly verified by female participants as function of characterization of the task and the condition (adapted from Croizet, Désert, Dutrévis, & Leyens, 2001).

participating because they were good students (i.e., they liked helping graduate students and were curious). Because stereotype threat was related to the fear of being perceived as having a lower ability than others, we thought that self-affirmation on an academic value would be the best way to restore self-regard and to limit the negative consequences of stereotype threat on the performance. This expectancy was indeed confirmed by the results (see Figure 6). Women who thought that the test was designed to understand why men do better than women in some math-related domains under–performed compared to other female participants in the non-threatening condition (i.e., standard stereotype threat effect). However, when threatened participants completed a questionnaire designed to affirm their self-image in the academic domain, their performance increased. In accordance with our hypothesis, this suggests that the negative consequence of stereotype threat can be reduced if people experiencing stereotype threat are given a chance to reaffirm a sense of self-integrity. In other words, affirming positive aspects of the self may reduce the negative consequence of stereotype threat by restoring a positive self-image while participants are still under evaluative scrutiny.

6. Discussion Research on stereotype threat has convincingly demonstrated that reputations of lower ability in academic domains can significantly contribute to the underachievement of students targeted by such reputations (e.g., African Americans, low SES students, and women in math related domains). Probably one of the best illustrations of the significance of this contribution has been a recent decision by the Educational Testing Services to undertake a field experiment to assess the ecological validity of stereotype threat research.

306


Compared to more classical explanations, the stereotype threat hypothesis poses that this asymmetry in achievement results more from a situational predicament than from a stable predisposition, be it genetic or constructed through socialization. Although previous research has clearly demonstrated that situational changes may alleviate the threat, it has also implied that there was no room for increased performance when the test was diagnostic of ability. Because it seems difficult to convince students that academic tests are not diagnostic of their ability, it is important to look for new situational arrangements that may decrease stereotype threat or reduce its consequence on performance while students are still under evaluative scrutiny. Our research suggests that this can be achieved through self-accessibility manipulations. A first way to reduce stereotype threat directly may consist of reducing the self-relevance of the stereotype. Following Self-Categorization Theory (Turner & Oakes, 1989), this can be accomplished by switching from the intergroup level of self-categorization (e.g., women vs. men) to the subordinate level (e.g., myself as a person). This can also be achieved by making salient another social identity (for empirical evidence, see Désert, Leyens, Croizet, & Klopfenstein, 2001). In both cases, the accessibility of self-knowledge related to the stereotype is reduced in favor of an alternative one resulting in lower stereotype threat and heightened performance. The rationale of the other remediation to stereotype threat stems from a basic assumption that stereotype threat is a general self-threat that triggers effortful attempts to restore a sense of self-integrity. Thus, having the opportunity to affirm oneself after a stereotype threat induction may efficiently restore a positive self-image. This should make available more resources for performing the task. By showing that such a self-affirmation procedure restored performance, our last study suggested again that countering pernicious effects of stereotype threat in educational settings may be possible. Although the two strategies outlined above sound promising, they may also be difficult to implement. On the one hand, being the target of discrimination within the usual context of negative stereotypic beliefs may limit the flexibility of one’s self-definition. Such stereotypes may be so widely shared that they follow targets wherever they go to place limits on how these persons think about themselves. Thus, self-affirmation resources in educational settings (i.e., affirmation of either another social identity or of alternative aspects of the self-concept) should be less available for people targeted by stereotypes. On the other, we should not expect slight situational changes to easily overcome the problem of minority academic under-achievement. It is not because our understanding of the dynamic of stereotype threat is insufficient. It may have more to do with the fact that school by itself participates in what has been called institutional discrimination (see Bourdieu & Passeron, 1970; Sidanius & Pratto, 1999). Bearing these limitations in mind, however, we are convinced that our research suggests some ideas on how to improve the academic achievement of lower status


307

groups. Sharing Gerard’s (1983) point of view, we think that if laboratory research has generated interesting theories and empirical facts, the time has come to explore this research for applications in the real world. This will require time and money, but the challenge is worth it.

References Allport, Gordon W. (1954). The nature of prejudice. New York: Addison-Wesley. Aronson, Joshua, Lustina, Mickael, Good, Catherine, Keough, Kelly, Steele, Claude M., & Brown, Joseph L. (1999). When white men can’t do math: Necessary and sufficient factors in stereotype threat. Journal of Experimental Social Psychology, 35, 29–46. Ashmore, Richard D. & Del Boca, Frances K. (1981). Conceptual approaches to stereotypes and stereotyping. In Hamilton, D.L. (Ed.), Cognitive processes in stereotyping and intergroup behavior. Hillsdale, NJ: Erlbaum; pp. 1–36. Baumeister, Roy F. (1984). Choking under pressure: Self-consciousness and paradoxical effects of incentives on skillful performance. Journal of Personality and Social Psychology, 46, 610–620. Baumeister, Roy F. (1995). Self and identity: An introduction. In Tesser, A. (Ed.), Advanced social psychology. New York: McGraw-Hill; pp. 51–98. Bourdieu, Pierre & Passeron, Jean-Claude (1964). Les héritiers [The heirs]. Paris: Editions de Minuit. Bourdieu, Pierre & Passeron, Jean-Claude (1970). La reproduction (the reproduction). Paris: Editions de Minuit. Brown, Ryan Y. & Josephs, Robert A. (1999). A burden of proof: Stereotype relevance and gender differences in math performance. Journal of Personality and Social Psychology, 76, 246–257. Bullock, Heather E. (1995). Class acts: Middle-class responses to the poor. In Lott, B. & Maluso, D. (Eds.), The social psychology of interpersonal discrimination. New York: Guilford; pp. 118–159. Clark, Kenneth B. (1965). Dark ghetto: Dilemmas of social power. New York: Harper & Row. Crocker, Jennifer & Major, Brenda (1989). Social stigma and self-esteem: The self-protective properties of stigma. Psychological Review, 96, 608–630. Crocker, Jennifer, Major, Brenda, & Steele, Claude M. (1998). Social stigma. In Gilbert, D., Fiske, S.T. & Lindzey, G. (Eds.), Handbook of social psychology (4th edn., Vol. 2). Boston: McGrawHill; pp. 504–553. Croizet, Jean-Claude & Claire, Theresa (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. Croizet, Jean-Claude, Désert, Michel, Dutrévis, Marion, & Leyens, Jacques-Philippe (2001). Reducing stereotype threat through self-affirmation. Unpublished manuscript. DEP-MEN (1997). Direction de l’Evaluation et de la Prospective-Ministère de l’Education Nationale, Repères et références statistiques. Désert, Michel, Leyens, Jacques-Philippe, Croizet, Jean-Claude, & Klopfenstein, Thierry (2001). Reducing stereotype threat through re-individuation. Manuscript submitted for publication. Devine, Patricia G. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5–18. Devine, Patricia G. & Elliot, Andrew J. (1995). Are racial stereotype really fading? The Princeton trilogy revisited. Personality and Social Psychology Bulletin, 21, 1139–1150. Easterbrook, John A. (1959). The effects of emotion on cue utilization and the organization of behavior. Psychological Review, 66, 183–201. Eccles, Jacquelynne S. (1994). Understanding women’s educational and occupational choice. Psychology of Women Quarterly, 18, 585–609.

308


Eccles, Jacquelynne S., Jacobs, Janis E. & Harold, Rena E. (1990). Gender role stereotypes, expectancy effects, and parents’ socialization of gender differences. Journal of Social Issues, 46, 183–201. Fein, Steven & Spencer, Steven J. (1997). Prejudice as self-image maintenance: Affirming the self through derogating others. Journal of Personality and Social Psychology, 73, 31–44. Fiske, Susan T. (1998). Stereotyping, prejudice and discrimination. In Gilbert, D., Fiske, S.T. & Lindzey, G. (Eds.), Handbook of social psychology (4th edn., Vol. I). Boston: McGraw-Hill. Geen, Russell G. (1985). Evaluation apprehension and response withholding in solution of anagrams. Personality and Individual Differences, 6, 293–298. Gerard, Harold B. (1983). School desegregation: The social science role. American Psychologist, 38, 869–878. Gilly, Michel (1984). Psychosociologie de l’éducation (Psychosociology of education). In Moscovici, S. (Ed.), Psychologie sociale. Paris: Presses Universitaires de France; pp. 473–494. Gould, Stephen J. (1996). The mismeasure of man. London: Penguin Books. Herrnstein, Richard J. & Murray, Charles (1994). The bell curve: Intelligence and class structure in American life. New York: Free Press. Jussim, Lee, Eccles, Jacquelynne, & Madon, Stephanie (1996). Social perception, social stereotypes, and teacher expectations: Accuracy and the quest for the powerful self-fulfilling prophecy. In Zanna, M.P. (Ed.), Advances in Experimental Social Psychology (Vol. 28); pp. 281–288. Kozol, Jonathan (1991). Savage inequalities. New York: HarperPerennial. Leyens, Jacques-Philippe, Désert, Michel, Croizet, Jean-Claude, & Darcis, Catherine (2000). Stereotype threat: Are lower status and history of stigmatization preconditions of stereotype threat? Personality and Social Psychology Bulletin, 26, 1189–1199. Markus, Hazel & Wurf, Elissa (1987). The dynamic self-concept: A social psychological perspective. Annual Review of Psychology, 38, 299–337. National Center for Education Statistics (1997a). National assessment of educational progress 1996 mathematics report card for the nation and the states. Washington, DC:. Author. National Center for Education Statistics (1997b). National assessment of educational progress 1996 science report card for the nation and the states.Washington, DC: Author. National Center for Education Statistics (1999). National assessment of educational progress 1998 reading report card for the nation and the states. Washington, DC: Author. Nettles, Mickael T. (1988). Toward undergraduate student equality in American higher education. New York: Greenwood. Nowell, Ami & Hedges, Larry V. (1998). Trends in gender differences in academic achievement from 1960 to 1994: An analysis of differences in mean, variance, and extreme score. Sex Roles, 39, 1998. Ogbu, John U. (1994). From cultural differences to differences in cultural frames of reference. In Greenfield, P.M. & Cocking, R.R. (Eds.), Cross-cultural roots of minority child development. Hillsdale, NJ: Erlbaum; pp. 365–391. Pyszsynski, Tom & Greenberg, Jeff (1983). Determinants of reduction in effort as a strategy for coping with anticipated failure. Journal of Research in Personality, 17, 412–422. Ricciuti, Henry N. (1993). Nutrition and mental development. Current Directions in Psychological Science, 2, 43–46. Rist, Ray (1970). Student social class and teacher expectations: The self-fulfilling prophecy in ghetto education. Harvard Educational Review, 40, 411–451. Rhodewalt, Frederick & Agustsdottir, Sjöfn (1986). Effects of self-presentation on the phenomenal self. Journal of Personality and Social Psychology, 50, 47–55. Rothbart, Myron & Taylor, Marjorie (1992). Category label and social reality: Do we view social categories as natural kinds? In Semin, G. & Fiedler, K. (Eds.), Language, interaction and social cognition. London: Sage; pp. 11–36.


309

Sarason, Irwin G. (1972). Experimental approaches to test anxiety: Attention and the uses of information. In Spielberger, C.D. (Ed.), Anxiety: Current trends in theory and research (Vol. 2). New York: Academic Press. Sidanius, James & Pratto, Felicia (1999). Social dominance. Cambridge: Cambridge University Press. Sigman, Marian (1995). Nutrition and child development: More food for thought. Current Directions in Psychological Science, 4, 52–55. Skrypnek, Berna J. & Snyder, Mark (1982). On the self-perpetuating nature of stereotypes about women and men. Journal of Experimental Social Psychology, 18, 277–291. Snyder, Mark & Stukas, Arthur A. (1999). Interpersonal processes: The interplay of cognitive, motivational and behavioral activities in social interaction. Annual Review of Psychology, 50, 273–303. Spencer, Steven J., Steele, Claude M., & Quinn, Diane M. (1999). Stereotype threat and women’s math performance. Journal of Experimental Social Psychology, 35, 4–28. Steele, Claude M. (1988). The psychology of self-affirmation: Sustaining the integrity of the self. In Berkowitz, L. (Ed.), Advances in experimental social psychology (Vol. 21). New York: Academic Press; pp. 261–302. Steele, Claude M. (1992, April). Race and the schooling of black Americans. The Atlantic Monthly, 68–78. Steele, Claude M. (1997). A threat in the air: How stereotypes shape the intellectual identities and performance of women and African Americans. American Psychologist, 52, 613– 629. Steele, Claude M. & Aronson, Joshua (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology, 69, 797–811. Steele, Claude M., Spencer, Steven J., & Lynch, Mickael (1993). Self-image resilience and dissonance: The role of affirmational resources. Journal of Personality and Social Psychology, 64, 885–896. Stone, Jeff, Lynch, Christian I., Sjomeling, Mike, & Darley, John M. (1999). Stereotype threat effects on black and white athletic performance. Journal of Personality and Social Psychology, 77, 1213–1227. Swim, Janet K. & Stangor, Charles (Eds.) (1998). Prejudice: The target’s perspective. New York: Academic Press. Tajfel, Henri & Turner, John C. (1979), An integrative theory of intergroup relations. In Austin, W.G. & Worchel, S. (Eds.), The psychology of intergroup relations. Monterey, CA: Brooks-Cole; pp. 33–47. Turner, John C. & Oakes, Penelope J. (1989). Self-categorization theory and social influence. In Paulus, P.B. (Ed.), Psychology of group influence. Hillsdale, NJ: Lawrence Erlbaum Associates; pp. 233–275. Williams, Tannis (1976).Teacher prophecies and the inheritance of inequality. Sociology of Education, 49, 223–236. Word, Carl O., Zanna, Mark P. & Cooper, Joel (1974). The nonverbal mediation of self-fulfilling prophecies in interracial interaction. Journal of Experimental Social Psychology, 10, 109–120.

Biographical Notes Jean-Claude Croizet is assistant professor at the Université Blaise Pascal of Clermont-Ferrand, France. His major research interest is the social regulation of cognitive functioning. In this perspective, he is especially interested in the effect of stigmatization on academic achievement.

310


Michel Désert is researcher from the Fond National de la Recherche Scientifique at the Université Catholique de Louvain-La-Neuve, Belgium. His major research interest is stereotype threat. Marion Dutrévis is a Ph.D. student at the Université Blaise Pascal of ClermontFerrand, France. Her primary research interest focuses on stereotype threat and means to alleviate it. Jacques-Philippe Leyens is professor at the Université Catholique de Louvain-LaNeuve, Belgium. His main research interests are social perception and intergroup relations.