Development of Stereotypes 1 The Development of Race, Gender, and Social Class Stereotypes in Black and White Adolescents

We examined race, gender, and social class stereotypes in fourth, sixth, and eighth grade European American and African American children. Participants reported their perceptions of the competence of rich, poor, Black, White, female and male children in academic domains, sports, and music. Sixth and eighth graders were more likely than fourth graders to report traditional stereotypes. Low status groups (girls and African American children) did not endorse stereotypes that reflected negatively on their own group, but were likely to report stereotypes that favored their social group. Correlations among stereotype scores indicated that children consistently favored one social group over another, and eighth graders, but not younger children, showed moderate consistency in their tendency to be traditional versus egalitarian. Our results show the importance of social status in children’s endorsement of stereotypes. Development of Stereotypes 3 The Development of Race, Gender, and Social Class Stereotypes in Black and White Adolescents Although holding stereotypes about social groups is often viewed negatively, stereotypes are grounded in our knowledge about the world and our attempts to differentiate among concepts (e.g., “boyness” versus “girlness”). Thus, children’s stereotypes—or their understanding of the attributes of various social groups—reflect their concept development (e.g., what is a “girl”) and are also important in the child’s emerging sense of self (e.g., I am a girl; therefore I wear dresses) and of social relationships (I am a girl; I play with girls). An early rudimentary understanding of social categories such as race, gender, and even social class eventually develops into a rich knowledge base and belief system regarding the behaviors and attributes ascribed to groups as well as the awareness that all group members do not share these characteristics. The present study focuses on the development of race (African American and European American), sex, and social class (rich, poor) stereotypes regarding academic skills. We had three goals. The first goal was to investigate whether children’s stereotypes/beliefs about race, sex, and social class differences in academic skills and the strength of those stereotypes change with age. The second aim was to examine whether there were race and sex differences in stereotypes. The third goal was to determine whether individuals who held traditional stereotypes in one domain (e.g., sex) were likely to hold traditional stereotypes in other domains (e.g., race). In the following discussion, we first briefly summarize the development of race, gender, and social class stereotypes. Next we discuss race, sex, and grade differences in stereotypes. We conclude with a description of our research questions. Development of Stereotypes 4 The Development of Social Stereotypes Stereotypes are cognitive structures that represent social groups (Hamilton, Stroessner, & Driscoll, 1994). As such, they develop through both cognitive processes and social experience (Bigler & Liben, 1992). From a cognitive developmental perspective, children’s understanding of stereotypes increases with certain cognitive competencies, such as multiple classification (Bigler & Liben, 1992), conservation (Serbin, Powlishta, & Gulko, 1993), and social cognition (Spencer, 1982). A strong cognitive theory would suggest that children’s stereotypes regarding members of different social groups should become more traditional and flexible over time and should share similar developmental trajectories. Few studies of stereotype development allow examination of this possibility, however, as most focus on a single social category such as gender or race, rather than examining the simultaneous development of several categories. A second set of theories of stereotype development suggests that environmental factors and social interaction strongly influence children’s adoption of stereotypes (Chafel, 1995; Huston, 1983; Katz, 1983, 1987; Leahy, 1983). According to these theories, stereotypes of different groups develop as individuals interact with members of the stereotyped group or obtain information about the group from other sources. For example, Quintana and de Baessa (1996, as cited in Quintana, 1998) found that children in Guatemala were aware of social class differences earlier than U.S. children, probably because the degree of economic stratification is greater in Guatemala than in the U.S. Such a social experiential theory allows for both intraindividual variability (i.e., variability in the strength of stereotypes across domains, but with individuals) and inter-individual variability (i.e., variability in stereotypes with domains among children of similar age). It is still possible, however, that the traditionality of different types of Development of Stereotypes 5 stereotypes converges at some point as a result of both cognitive maturation and social experience. Indeed, research in stereotypes across a variety of social categories and content domains has shown that children’s stereotypes increase in sophistication and flexibility throughout childhood and into adolescence (Signorella, Bigler, & Liben, 1993). In the following section we briefly examine the development of stereotypes in each of the three focal areas of this project: race, sex, and social class stereotypes regarding academic skills. The predominant academic racial stereotype in the United States is that African Americans are not as smart and do not do as well in school as their European American (and Asian American) counterparts. In addition, it is widely believed that African Americans’ natural talents lie in their musical, and especially, sports abilities. Children as young as 3 years old can reliably categorize individuals according to race (Aboud, 1988; Katz, 1983). Moreover, children report traditional race stereotypes (i.e., according greater status to Whites than Blacks) by age 5 (Clark & Mamie K. Clark, 1939). Although few studies have examined race-related academic stereotypes, research about race differences in general characteristics has shown that a pro-white bias develops in Black and White children at a young age. However, as children move into middle childhood, both Black and White children become more positive towards Blacks and view Whites less favorably (Doyle & Aboud, 1995; (Doyle, Beaudet, & Aboud, 1988); Freeman, 1997). Unfortunately, most of these studies include only participants who are less than 10 years old. Thus, it is unclear whether attitudes towards Blacks continue to grow more positive beyond this age or whether they remain constant. Gender is universally one of the most salient social categories for human beings and thus is a powerful influence on children’s developing understanding of the world and of themselves. Already by 30 months of age, children label themselves and others as male and Development of Stereotypes 6 female and show some limited knowledge of gender stereotypes (Fagot & Leinbach, 1989; Huston, 1983; Katz, 1983; Weinraub et al., 1984). Children’s awareness and reports of genderbased academic stereotypes have been researched much less than other stereotype categories such as occupations (e.g., nurses are female), personal qualities (e.g., boys are strong), and activities (e.g., girls like to jump rope). Although research on children’s reports of ability stereotypes is limited, a substantial body of research has shown that gender differences appear in self-perceptions of academic skill by late childhood or early adolescence. This research has shown that males tend to overestimate their performance across the board, and both males and females report self-perceptions in line with traditional stereotypes (i.e., girls report greater self competence in verbal domains whereas boys report greater self competence in math and science) (Bornholt, Goodnow, & Cooney, 1994; Lundeberg, Fox, & Puncochar, 1994; Stevenson & Newman, 1986; (Wigfield et al., 1997), Eccles, Yoon, Harold, Arbreton, Doan, & Blumenfeld, 1997). Elementary-aged girls’ and boys’ self-perceptions of sports and music abilities also show similarities to societal stereotypes: Girls believe that they are better in music and boys view themselves as better at sports (Wigfield et al., 1997). Researchers have not substantiated whether differences in individuals’ perceptions of self competence reflect differences in underlying beliefs about sex differences in ability, or whether these beliefs change as children enter adolescence. Unlike gender and race, social class is not biologically based, but socially derived, making its markers less stable and distinct, and perhaps less recognizable to children. Consequently, we might expect that stereotypes about social class would emerge later than sex and race stereotypes. Yet a growing body of research suggests that by age 5 children have at least a rudimentary schema for economic status (Ramsey, 1991), and by middle childhood they Development of Stereotypes 7 are able to distinguish among a variety of possessions and occupations on the basis of economic status cues (Brusdal, 1990; Emler & Dickinson, 1985; Estvan, 1952; Jahoda, 1959; Leahy, 1981, 1983; Mistry & Crosby, 2001; Stendler, 1949; Tudor, 1971). Much less is known about the development of children’s beliefs about the attributes of rich and poor people (see review by Chafel, 1995). It has been suggested, however, that children begin to form such beliefs at an early age (Chafel, 1997; Mistry & Crosby, 2001; Skafte, 1988; Stendler, 1949). In a classic study, Stendler (1949) reported age-related differences in first, fourth, sixth, and eighth graders’ conceptions of rich and poor individuals. Younger children tended to ascribe positive attributes to the rich and negative attributes to the poor, whereas adolescents attributed more negative characteristics to the rich and more positive attributes to the poor. More recent studies report findings that are both consistent with and in contradiction to Stendler's (1949) results. For example, in a study of second, fourth, and sixth graders, Mistry and Crosby (2001) reported that favorable evaluations of the rich (including ratings of academic skills) decreased with age, whereas positive evaluations of the poor increased with age. In contrast, Skafte (1988) observed that adolescents’ ratings of strangers were significantly lower on qualities such as intelligence, happiness, and ability to make friends if the stranger was depicted as being poor than if the stranger was depicted as being rich. In the present study, we examine whether children report social class differences in academic skills, whether these stereotypes change with the transition to adolescence, and whether there are sex or racial group differences in stereotypes. We turn next to a discussion of group differences in stereotypes. Group Differences in Stereotypes Social psychologists have long reported that both adults and children quickly and easily develop in-group biases, even when the groups are created arbitrarily (Nesdale & Flesser, 2001; Development of Stereotypes 8 Tajfel & Billig, 1974). One criticism of lab-based experimental studies of intergroup processes and stereotyping, though, is that artificially formed groups do not reflect real status differences between certain groups such as men and women and Blacks and Whites (Bigler, Brown, & Markell, 2001). The pervasive nature of race, gender, and social class stereotypes in America may make it difficult for members of negatively stereotyped groups to report an in-group bias regarding certain group attributes. Research by Tajfel (Tajfel, Nemeth, Jahoda, & Campbell, 1970; Tajfel et al., 1972) on ethnic group preferences demonstrates that although most individuals are biased toward their own group, members of low status groups (“underprivileged” or minority groups) do not show this in-group bias. In an innovative, nonlaboratory study by Bigler and her colleagues (2001), elementary school children were randomly assigned to different groups. A colored t-shirt worn during school hours denoted group membership. Children were subtly primed over time to be aware of the greater status of one group over the other. As in the work by Tajfel, children from high status, but not low status groups reported positive stereotypes in favor of their own group (Bigler et al., 2001). Students from low status groups were more egalitarian; that is they reported no differences between groups (cf. Signorella, Bigler, & Liben, 1993). Thus, it appears that although members of negatively stereotyped groups do not report in-group bias, they also fail to endorse the widely-held negative societal stereotypes. This egalitarian stance may reflect efforts to protect the self-esteem through group enhancement. In addition to failing to endorse negative stereotypes, low status groups may emphasize positive stereotypes of their group. For instance, rather than reporting egalitarian views of their abilities in sports, African American children may endorse strong positive stereotypes in favor of their group. In addition, girls may strongly emphasize the belief that girls are better than Development of Stereotypes 9 boys in reading and writing as a way for compensating for the negative views related to their relative math and science performance. Most studies of low status groups only evaluate groups on negative attributes. The second aim of our study—an examination of sex and racial differences in stereotypes--was undertaken to determine whether members of disparaged or low status groups would endorse negative and positive stereotypes of their own groups or would simply report no differences. In previous studies, low status groups reported no differences, but these studies did not tap into strongly entrenched societal stereotypes such as those related to race, gender, and social class in the United States, and few earlier studies with children have examined positive stereotypes of low status groups. An additional unexplored question relates to the interactions among race, gender, and status. Although Whites and males typically have higher status in the classroom than Blacks and girls, academic stereotypes of Black males tend to be less positive than those of Black girls (Graham, Taylor, & Hudley, 1998; Sidanius & Pratto, 1999; Sidanius & Veniegas, 2000). According to status theory of stereotyping, Black males should be most egalitarian in their academic stereotypes. On the other hand, some suggest that Black females suffer from double jeopardy (cf. Beale, 1970) from being at once female and Black. They may therefore be even less likely to endorse stereotypes negative stereotypes and more likely to report positive stereotypes of the groups to which they belong than others. Goals of this Study The present study examines the development of race, gender, and social class academic stereotypes in fourth, sixth, and eight grade European American and African American children. We examined these age groups to capture children’s stereotype beliefs on both sides of the transition to middle school, a time of rapid cognitive, social, and physical change. Young Development of Stereotypes 10 adolescents’ abilities to think abstractly and flexibly as well as their ability to integrate contradictory information lend themselves to more frequent use of broad, generalized stereotypes than younger children (Huston, 1983; Leahy & Shirk, 1985). In addition, social knowledge increases greatly and social relationships take on greater meaning during this period. These factors seem to culminate in an overall decline in self-perceptions of ability as children move into middle school (Anderman & Midgely, 1997; Eccles & Midgely, 1989; Wigfeld et al., 1997). This decline may result, in part, because students perceive a stronger focus on ability in middle school classrooms than in elementary school classrooms (Anderman & Midgely, 1997). The focus on ability in middle school, coupled with the increasing range of grades may serve to amplify societal stereotypes about race, class, and gender differences in academic ability. Little research has examined changes in stereotypes that occur in the midst of these cognitive, social, and physical changes. We had three main objectives in this study. First, we assessed if race, gender, and social class stereotypes of children entering adolescence vary systematically by age. We reasoned that if developmental factors are primarily driving stereotype development, eighth graders would hold stronger stereotypes than fourth graders. The second research question we addressed was whether members of each social group would be more likely to view their group more positively than the out-group. As outlined above, status theory of stereotyping would predict that children of lower status would be less likely to report negative stereotypes about their own social groups. For instance, girls would not report that boys are better than girls in math and science and Blacks would not report that Whites perform better academically than Blacks. Furthermore, low status children may try to Development of Stereotypes 11 buffer their self-esteem by strongly endorsing commonly held positive stereotypes of their group. The third purpose of the study was to compare the strength of stereotypes across domains to determine if children who have stronger stereotypes in one domain (e.g., SES) also have strong stereotypes in the other domains (race, gender). In other words, if stereotyping is primarily an individual difference construct whereby some individuals are more likely to hold stereotypes than others, we would expect positive relationships across the three social categories. We also assessed whether the relationships among stereotypes varied across age. We hypothesized that older children would be more consistently stereotyped or egalitarian across the three social categories than younger children.