Sunteți pe pagina 1din 14
Child Development, January/February 2014, Volume 85, Number 1, Pages 250 – 263 The Roots of
Child Development, January/February 2014, Volume 85, Number 1, Pages 250 – 263 The Roots of

Child Development, January/February 2014, Volume 85, Number 1, Pages 250263

The Roots of Stereotype Threat: When Automatic Associations Disrupt Girls Math Performance

Silvia Galdi and Mara Cadinu

University of Padova

Carlo Tomasetto

University of Bologna

Although stereotype awareness is a prerequisite for stereotype threat effects (Steele & Aronson, 1995), research showed girls de cit under stereotype threat before the emergence of math gender stereotype awareness, and in the absence of stereotype endorsement. In a study including 240 six-year-old children, this paradox was addressed by testing whether automatic associations trigger stereotype threat in young girls. Whereas no indi- cators were found that children endorsed the math gender stereotype, girls, but not boys, showed automatic associations consistent with the stereotype. Moreover, results showed that girls automatic associations varied as a function of a manipulation regarding the stereotype content. Importantly, girls math performance decreased in a stereotype-consistent, relative to a stereotype-inconsistent, condition and automatic associations mediated the relation between stereotype threat and performance.

Stereotype threat research (Steele & Aronson, 1995) has shown that both adults and children underper- form in difcult tests when a negative domain- relevant in-group stereotype is made salient (see Inzlicht & Schmader, 2012, for a review). Despite this overwhelming evidence, the speci c require- ments for children s underperformance under ste- reotype threat are still unclear. With the aim of providing new insights into research on stereotype threat, this study investigates automatic associations as a potential requirement for the emergence of ste- reotype threat underperformance in children. Automatic associations are those associations between concepts (e.g., me woman), or between concepts and attributes (e.g., ower positive) that come to mind unintentionally when an individual encounters a relevant object, that are dif cult to con- trol once they have been activated, and that are not necessarily explicitly endorsed (e.g., Gawronski & Bodenhausen, 2006). Such automatic associations are often contrasted with conscious beliefs, which are those mental contents that an individual explicitly endorses as accurate (e.g., Gawronski & Bodenhau- sen, 2006). Most measures of automatic associations are based on participants performance on com- puter-based, speeded categorization tasks (see Gaw- ronski & Payne, 2010, for a review). These measures, commonly referred to as implicit measures, differ

Correspondence concerning this article should be addressed to Silvia Galdi, Dipartimento di Psicologia dello Sviluppo e della Socializzazione, Universit a di Padova, Via Venezia 15, Padova 35131, Italy. Electronic mail may be sent to silvia.galdi@unipd.it.

from self-reports, described as explicit measures , which assess conscious beliefs. This study employed implicit and explicit measures jointly to show that automatic associa- tions consistent with a negative in-group stereotype may lead to stereotype-induced underperformance at early stages of development, even though children do not possess the cognitive competencies for stereotype awareness yet, and in the absence of evidence that they endorse the stereotype content.

Prerequisites for Stereotype Threat Susceptibility

According to the stereotype threat model, for ste- reotypes to affect performance, children need to (a) have developed a concept of social categories ( cate- gory awareness ), (b) be able to identify themselves as members of a social category ( self-categorization ), and (c) know that the in-group category is negatively related to speci c domains or attributes ( stereotype awareness ; Aronson & Good, 2003). When children enter elementary school, they possess the cognitive competencies of category awareness and self-categoriza- tion (e.g., Martin & Ruble, 2010) but not stereotype awareness (e.g., McKown & Strambler, 2009). Research has shown that, between the ages of 3 and 4 years, children become aware of social categories, such as gender and race (e.g., Aboud,

© 2013 The Authors Child Development © 2013 Society for Research in Child Development, Inc. All rights reserved. 0009-3920/2014/8501-0017 DOI: 10.1111/cdev.12128

1988; Katz & Kofkin, 1997), and are able to identify themselves as members of these categories (e.g., Martin & Ruble, 2010; Quintana, 1998). Moreover, by this age children develop personal stereotypic beliefs about abilities and characteristics of ethnic (Aboud, 1988) and gender groups. For example, they hold personal stereotypic beliefs about gender differences in toy preferences, dressing, and aggres- sive versus prosocial behaviors (Martin & Ruble, 2010). However, personal stereotypic beliefs, referred to as stereotype endorsement, differ from the knowledge of stereotypes held by others (not neces- sarily personally endorsed), which is de ned as stereotype awareness (e.g., Martinot & D esert, 2007; McKown & Strambler, 2009). Thus, although in many cases personal beliefs overlap with socially shared stereotypes, this does not imply necessarily that young children are aware of stereotypes. Indeed, the literature on children s social perspec- tive taking (e.g., Selman, 1980), theory of mind (e.g., Perner & Wimmer, 1985), and person percep- tion (e.g., Rholes & Ruble, 1984) suggests that chil- dren develop the cognitive competencies for stereotype awareness only by middle childhood, and that up to then they do not distinguish their personal stereotypic beliefs from others beliefs. Consistently, children have been shown to be aware of common ethnic and gender stereotypes about academic abilities by the ages of 8 and 9 years (e.g., McKown & Weinstein, 2003; Quintana, 1998). The evidence that children do not possess stereo- type awareness when they enter elementary school has important consequences. As noted earlier, stereotype awareness is a speci c requirement for stereotype threat. Speci cally, the stereotype threat model posits that people s negative stereotypes hamper performance only in situations that induce targets to become concerned about being judged by others on the basis of relevant stereotypes (Steele, 1997). Consistently, research on performance decre- ments induced by ethnic stereotypes con rms that only children who are aware of broadly held stereo- types are vulnerable to stereotype threat effects (McKown & Strambler, 2009; McKown & Weinstein, 2003). Therefore, in principle we should not expect declines in performance under stereotype threat prior to 8 9 years of age, or even later (Aronson & Good, 2003). Nevertheless, Ambady, Shih, Kim, and Pittinsky (2001) showed that 5- to 7-year-old Asian American girls underperformed on a math task when their gender identity was made salient, as compared to children in a control condition. Similarly, Tomasetto, Alparone, and Cadinu (2011) found that gender identity activation hampered

Stereotype Threat Before Stereotype Awareness 251

math performance among 5- to 7-year-old Italian girls, and Neuville and Croizet (2007) obtained similar ndings among 7-year-old French girls. The last three studies raise a theoretical paradox:

How can stereotype-induced performance decre- ments be found in girls who do not possess stereo- type awareness yet? One could assume that for a negative stereotype to affect performance in chil- dren who have not developed stereotype awareness yet, it is suf cient that children hold the personal stereotypic belief that their in-group is less skillful in the relevant domain. Indeed, given that young children do not distinguish their own stereotypic beliefs from others , and rather assume that their beliefs are shared by others as well (e.g., Augousti- nos & Rosewarne, 2001), in principle children s personal negative beliefs about abilities of their in-group could be suf cient to trigger stereotype threat. If this were the case, stereotype endorsement would be the key to identify the sources of stereo- type threat in young children. To date, the potential role of children s endorse- ment of the math gender stereotype as a prerequi- site for stereotype threat effects does not t with the available evidence. For example, Ambady et al. (2001) found that 5- to 7-year-old American girls believe that boys and girls are equal at math (for similar results on Italian children, see Tomasetto et al., 2011); on the contrary, 5- to 7-year-old Amer- ican boys state that boys are better at math, thus exhibiting in-group favoritism (e.g., Powlishta, 1995; Yee & Brown, 1994). Similarly, other research showed no endorsement of the math gender stereo- type until 8 9 years of age among Italian and French children (Martinot & D esert, 2007; Muzzatti & Agnoli, 2007). To our knowledge, only one study found stereo- type endorsement as early as 6 7 years of age in a Western country (i.e., United States; Cvencek, Meltzoff, & Greenwald, 2011). However, this dis- crepancy from the other ndings might be due to the type of measures used. Whereas all the studies discussed elsewhere in the article assessed the endorsement of the gender stereotype regarding math ability, Cvencek et al. (2011) assessed the endorsement of the stereotype that boys like math and girls like language. Therefore, it is possible that 6- to 7-year-olds already hold the personal stereo- type that boys like math , but still do not endorse the stereotype about girls and boys math ability . Consistent with this reasoning, Steele (2003) found that 6- to 10-year-old children do not endorse the stereotype regarding boys and girlsabilities in math (see also Martinot, Bag es, & D esert, 2012), even

252 Galdi, Cadinu, and Tomasetto

though they endorse the same stereotype about men and women (stereotype strati cation). Thus, to date, there is no evidence that young children endorse the math gender ability stereotype, at least in Western countries. Nevertheless, research shows that it is speci cally the ability component of the math gender stereotype that affects women s per- formance, whereas other stereotypical beliefs (e.g., women are less interested in math) do not trigger stereotype threat effects (Thoman, White, Yam- awaki, & Koishi, 2008). To summarize, young girls show math underper- formance before being aware of the math gender stereotype, and in the absence of evidence that they endorse such belief as a personal stereotype. One possibility to explain this paradox is that previous ndings on children s math gender stereotype have been based only on self-reports, which may not capture all relevant aspects of the stereotype. Thus, in this research, we used implicit and explicit mea- sures jointly to investigate automatic associations as an alternative route for the emergence of stereo- type-induced underperformance in young children.

Insights From Social-Cognitive Research

Implicit measures, such as the well-known Impli- cit Association Test (IAT; Greenwald, McGhee, & Schwartz, 1998), have been developed to assess the strength of automatic associations between concepts or concepts and attributes. These procedures are based on the rationale that individuals should be faster at pairing concepts, or concepts and attributes that are strongly associated in their cognitive map, as compared to those that are weakly or not at all associated. Research has shown that men and women differ in the strength of their automatic associations between gender and academic domains (Nosek, Banaji, & Greenwald, 2002) and that auto- matic associations consistent with the math gender stereotype predict attitudes toward math, domain identi cation, and math performance (e.g., Kiefer & Sekaquaptewa, 2007; Nosek & Smyth, 2011). Together, these results show that automatic associa- tions play an important role in people s academic achievement, interest, and performance, and that the combined use of implicit and explicit measures may allow researchers to provide useful insights that neither implicit measures nor explicit measures alone would offer. Theorizing on the meaning of implicit and explicit measures posits that explicit measures re ect more recent as well as newly formed representations that are not strong enough yet to

be automatically activated (e.g., Rudman, 2004; Wilson, Lindsey, & Schooler, 2000), whereas impli- cit measures detect only those highly overlearned associations between concepts whose activation has become automatic over the course of long-term experiences. According to this widespread assump- tion, in terms of emergence, conscious beliefs re ect more recently formed representations, whereas automatic associations re ect only the older repre- sentations. However, an emerging body of research suggests that the opposite may be true as well. Recent ndings showing experimentally induced changes in implicit but not explicit measures (e.g., Gawronski & LeBel, 2008; Olson & Fazio, 2006) suggest that automatic associations may re ect more recently formed representations as compared to the corresponding conscious beliefs. These results are also consistent with studies using implicit and explicit measures in predicting future choices of decided and undecided individuals (for a review, see Gawronski & Galdi, 2011). These studies dem- onstrate that well-structured automatic associations about a speci c object are present also in the cogni- tive maps of those individuals who, on explicit measures, report being undecided and do not endorse well-de ned conscious beliefs yet. These ndings come from research on attitudes (i.e., dealing with the affective evaluation of an object; Eagly & Chaiken, 1993), whereas stereotypes are cognitive beliefs about the link between a cate- gory and an attribute (Hamilton & Trolier, 1986); nonetheless they suggest that in terms of emer- gence, automatic associations may precede conscious beliefs. Thus, there is reason to believe that implicit measures can also detect recently formed automatic associations between concepts and stereotypical attributes (e.g., boys math) that are not re ected on explicit measures yet.

Insights From Research on Malleability of Automatic Associations

Other research has demonstrated that external cues (e.g., situational stimuli) may increase or decrease the activation of automatic associations in adults. For example, Dasgupta and Asgari (2004) found that participants exposed to counterstereo- typical women showed lower activation of auto- matic associations consistent with gender role stereotypes, as compared to control participants. Similarly, women exposed to gender-stereotypical women in TV commercials showed increased acti- vation of automatic associations consistent with the traditional female stereotype, and this higher

activation of automatic associations led to worse math performance (Davies, Spencer, Quinn, & Ger- hardstein, 2002). The last results are consistent with research demonstrating that stereotypical automatic associations affect working memory resources needed to perform complex cognitive tasks. Speci - cally, Forbes and Schmader (2010) found that women trained via an IAT to associate their gender with being good at math showed higher working memory and increased math performance. Taken together, these ndings suggest that it may be spe- ci cally in automatic associations between group membership and stereotypical attributes, and in the malleability of such automatic associations, where we should look not only to investigate stereotype- induced underperformance but also to reduce stereotype threat effects. Unfortunately, to date no study has tested whether automatic associations are malleable in children as well, or whether changes in activation of stereotypical automatic associations may re ect changes in performance. Therefore, we hypothesized that situational cues making a nega- tive stereotype salient may automatically activate corresponding automatic associations. The activa- tion of such stereotypical automatic associations should in turn affect the performance of children who are stereotype targets, even in the absence of evidence for stereotype endorsement. Conversely, cues challenging the negative stereotype should reduce the activation of corresponding automatic associations, and such a reduction should lead to improved performance.

Aims of This Study

To test the hypotheses mentioned above, we employed explicit and implicit measures jointly to assess two aspects of the math gender stereotype, stereotype endorsement and automatic associations, and their differential relation with young children s stereotype threat vulnerability. Research with children has already used implicit measures to assess automatic associations (e.g., Baron & Banaji, 2006). Regarding math gender ste- reotypes, Steffens, Jelenec, and Noack (2010) found stereotypical automatic associations in 9- to 14-year- old girls and no automatic associations for boys at any age. Conversely, Cvencek et al. (2011) found stereotype-consistent automatic associations in both girls and boys as young as 6 7 years of age. How- ever, neither Steffens et al. nor Cvencek et al. assessed either performance or automatic associa- tions under stereotype threat. Therefore, these stud- ies are not directly relevant to our main hypothesis.

Stereotype Threat Before Stereotype Awareness 253

To our knowledge, only Ambady and collabora- tors used an implicit stereotype awareness task (Ambady et al., 2001, p. 387) to assess the math gender stereotype in a study on stereotype threat in young children. The measure was a memory recall task, in which the experimenter told a story about a student especially good at math, and recorded whether the participant, when repeating the story, used the pronoun he or she to identify the protago- nist. However, rather than a measure of stereotypi- cal automatic associations, this task should be considered as an indirect measure of children s stereotypes. Therefore, it is still unknown whether automatic associations may be responsible for ste- reotype threat performance de cits in young chil- dren. In this study, we focused on rst-grade children for two reasons. So far, the only study showing the presence of automatic associations consistent with gender stereotypes about math and language in young children (Cvencek et al., 2011) considered children from 6 to 7 years altogether (Grades 1 and 2). Second, studies showing stereotype-induced math performance decrements in young girls have addressed 5- to 7-year-old children (Ambady et al., 2001; Tomasetto et al., 2011) or 7-year-olds (Neuville & Croizet, 2007). Thus, the focus on 6-year-olds (Grade 1) allowed us to investigate stereotypical automatic associations as well as stereotype threat effects specically when children enter elementary school, thus becoming acquainted for the rst time with formal teaching of math (at least in the Italian school system). The study had four goals. First, using a child- friendly version of the IAT (Child IAT; Baron & Banaji, 2006) we tested whether gender stereotypes about math and language are present as automatic associations in 6-year-olds. Consistent with Cvencek et al. (2011), we expected stereotype-consistent auto- matic associations for both girls and boys. Second, we investigated the malleability of chil- dren s automatic associations. It was hypothesized that children s stereotypical automatic associations would increase or decrease in activation depending on the stereotype content of a prior manipulation task. We predicted an increased activation of stereo- typical automatic associations in a stereotype-consistent condition than in a stereotype-inconsistent condition, respectively, con rming or contradicting the tradi- tional gender stereotype about math, with results in the middle in a control condition, in which the math gender stereotype was not salient. Third, using an explicit measure, we investigated whether 6-year-olds endorse the belief that boys are

254 Galdi, Cadinu, and Tomasetto

better than girls at math. Three reasons led us to assess stereotype endorsement: (a) children rst develop stereotype endorsement and then stereo- type awareness (e.g., Aboud, 1988; Martin & Ruble, 2010), (b) 6-year-olds do not possess stereotype awareness yet (e.g., Rholes & Ruble, 1984; Selman, 1980), and (c) children s personal stereotypic beliefs could be suf cient to trigger stereotype threat, spe- ci cally because at this age children assume that their beliefs are shared by others as well (e.g., Augoustinos & Rosewarne, 2001). Consistent with previous research (e.g., Ambady et al., 2001; Muzz- atti & Agnoli, 2007), we expected no evidence of endorsement of the math gender stereotype. Finally, like in past research on stereotype threat including 6-year-olds (e.g., Tomasetto et al., 2011), we expected girls math performance to decrease under stereotype threat, that is, in the stereotype- consistent as compared to the control condition and the stereotype-inconsistent condition, which should lead to the highest performance. Importantly, we predicted the increased activation of stereotype- consistent automatic associations to mediate the decrease in girls performance under stereotype threat.

Method

Participants

Two hundred and seventy-six rst-grade chil- dren (143 girls) participated in the study. All of them were born in Italy and were of typical age for their grade ( M = 77.60 months, SD = 0.28). Chil- dren attended one of six elementary schools in one of three small towns in the northeast of Italy. Per- missions to conduct the study were granted by school principals and parents. Six female experi- menters were involved in the data collection, con- ducted during school hours. All children were tested in the same school period to avoid potential effects of variability in mathematics curricula across classes at the moment of data collection.

Procedure

Children were tested individually in a quiet room of their school, while sitting at a desk facing the experimenter. Each experimental session started by asking participants to color one of three pictures, depending on the experimental condition (stereo- type consistent, control, and stereotype inconsistent) to which they were randomly assigned. When chil- dren nished coloring, the experimenter asked them

to describe the picture, and then left the picture on one side of the desk. After the manipulation task, children were invited to play a computer game (i.e., the Child IAT) using a laptop computer. Partici- pants were told that they would see pictures during the game, and that they would press the red ( A key) or the green (L key) button of the computer board to indicate which picture they saw. After the Child IAT, children performed eight math calculations (i.e., math test). The experimenter asked one calcula- tion at a time (e.g., How much is 5 plus 5? ) and children had to respond within 5 s. For each ques- tion, the experimenter noted whether the response was correct or not (i.e., incorrect or no response by 5 s), without providing any feedback. Finally, chil- dren performed the explicit stereotype-endorsement task. They were shown a picture of a boy and a girl side by side and were told: These are a boy and a girl. They are 6-year-olds and they are good at school. Is the boy better at math, the girl better at math, or are they the same at math? Next, the experimenter recorded the response, and partici- pants were thanked, given a candy bar, and dis- missed. Thus, the administration of the implicit measure always followed the experimental manipulation and always preceded the math test followed by the expli- cit measure. This order of the tasks was chosen because we aimed at testing the effects of the experi- mental manipulation on automatic associations ruling out the possibility that counterbalancing implicit and explicit measures might weaken the effects of the experimental manipulation on auto- matic associations. Moreover, there is no evidence that performing the IAT before a self-report induces reactance or assimilation tendencies in the subse- quent self-report (Nosek, Greenwald, & Banaji, 2005).

Materials

Experimental manipulation. According to the stereotype threat model, it is the salience of a nega- tive domain-relevant in-group stereotype that impairs targets performance. A negative domain- relevant in-group stereotype can be made salient in three ways: (a) describing the task that targets will subsequently perform as diagnostic of the ability related to the negative in-group stereotype, (b) mak- ing targets in-group salient, or (c) making the con- tent of the negative stereotype salient (Inzlicht & Schmader, 2012). In this study, the manipulation choice was to make salient the stereotype content (math gender stereotype). Thus, participants were randomly assigned to one of three experimental

conditions by inviting them to color a picture, in which: (a) a boy correctly resolves a math calculation on a blackboard, whereas a girl fails to respond (stereotype-consistent condition); or (b) a girl correctly resolves the calculation, whereas a boy fails to respond (stereotype-inconsistent condition); or (c) a landscape was depicted (control condition). Implicit measure: The math(language)-gender Child IAT. A Child IAT (Baron & Banaji, 2006) was used to assess the relative strength of automatic associa- tions between the target category boy and the attribute category mathematics , and the target cate- gory girl and the attribute category language , as compared to the opposite pairings (i.e., boy lan- guage and girl mathematics). Categories and stimuli for the Child IAT. Four pic- tures of math-related objects (i.e., numbers) and four pictures of reading- and writing-related objects (i.e., letters), pretested with rst-grade children for famil- iarity and comprehension, were used as stimuli for the attribute categories mathematics and language . Four face-pictures of boys and four face-pictures of girls represented the target categories boy and girl. Procedure of the ChildIAT. Children were instruc- ted to respond fast and accurately to three simple- categorization (practice) blocks and two (third and fth) critical double-categorization blocks. Each practice block included 16 trials; each critical block included 32 trials. On each trial, a target or attribute picture appeared on the screen until participants gave their response. Children categorized each pic- ture by pressing a key on the computer board, the left-hand response A or the right-hand response L key. To simplify the task, the left-hand key ( A ) was colored in red, and the right-hand key ( L) was col- ored in green. The intertrial interval was 200 ms. A red cross, which remained on the screen for 200 ms, followed incorrect responses. In the rst block of the task, children were pre- sented with pictures for the target categories girl and boy , with each picture randomly presented twice. Participants had to indicate whether the pic- ture on the screen was a boy or a girl by pressing the red key for girl and the green key for boy. In the second block, children were presented with pic- tures for the attribute categories language and math ematics , with each picture randomly presented twice. Participants had to press the red key when the picture was a reading- or writing-related object and the green key when the picture was a math- related object. In the third double-categorization block, both pictures of boys and girls, and pictures of math-related and reading- or writing-related objects (with each picture representing each cate-

Stereotype Threat Before Stereotype Awareness 255

gory randomly presented twice) appeared on the screen. In this case, children pressed the red key when they saw either a girl or a reading- or writing- related object, and the green key when they saw either a boy or a math-related object. In the fourth block, participants categorized again the pictures representing the categories mathematics and lan- guage . However, different from the key assignment in the second block, participants had to press the red key when they saw a math-related object and the green key when they saw a reading- or writing- related object. Finally, in the fth double-categoriza- tion block, children categorized the same pictures of boys and girls and the same pictures of math-related and reading- or writing-related objects. However, participants now pressed the red key when they saw either a girl or a math-related object, and the green key when they saw either a boy or a reading- or writing-related object. Following the procedure employed in the third block, each picture represent- ing each of the four categories was randomly pre- sented twice. The order of the two critical blocks, third and fth, was counterbalanced across partici- pants to avoid order effects. Math test. Three days before the data collection, a math test was developed in collaboration with all teachers of the classes participating in the study, to create a set of dif cult math calculations taking into account the mathematics achievement of all classes. The math test was a retrieval of numerical facts including ve additions and three subtractions (i.e., 5 + 5, 8 4, 6 + 4, 10 5, 2 + 3, 2 + 2, 6 3, and 4 + 5). Children had to resolve each math calcula- tion in 5 s. Explicit stereotype endorsement. As in Ambady et al.s (2001) study, children were presented with a photograph of a boy and a girl described as good at school, and had to choose whether the boy or the girl is especially good at math or whether they are equal at math . The two photographs as well as the questions wording for the boy or the girl were presented in counterbalanced order.

Results

Preliminary Analyses

Twenty-three girls and thirteen boys were excluded only because of 30% or higher error rates in at least one critical (double-categorization) block of the Child IAT. Preliminary analyses con rmed that excluded participants were evenly distributed across conditions (i.e., 13 in the stereotype-consistent, 10 in the control, and 13 in the stereotype-inconsistent

256 Galdi, Cadinu, and Tomasetto

conditions), and that their responses did not differ from those of retained children on both the math test and the stereotype endorsement. The nal sample included 240 children (120 girls). Individual scores of automatic associations were calculated by means of the D-algorithm, designed for analyzing data with the IAT (Greenwald, Nosek, & Banaji, 2003). The D-algorithm compares the extent to which performance on the incompatible critical block (i.e., girl mathematics and boy language shar- ing the same response key) is impaired relative to performance on the compatible critical block (i.e., girl language and boy mathematics sharing the same response key), taking into account participants individual response latencies, standard deviations of latencies, and error rates in each of the two blocks. Scores were calculated so that positive scores re ect stronger boy mathematics and girl language automatic associations. To estimate reliability of the Child IAT, two Child IAT scores were calculated: one using the rst half and the other using the second half of the two critical blocks. Internal consistency was satisfac- tory (Cronbach s a = .84). A one-way analysis of variance (ANOVA) was then conducted on scores of automatic associations using conditions (stereotype consistent, control, ste- reotype inconsistent), gender (male, female), and the order of administration of the critical blocks (1 = third block: girl language and boy math, fth block:

girl math and boylanguage; 2 = third block: girl math and boy language, fth block: girl language and boy math ) as the independent variables. A main effect for order of administration was found, F (1, 234) = 6.64, p = .01, g p 2 = .03. However, because order of administration produced no signi cant interactions (all p s > .09), it is not further discussed. For each participant, correct responses to the math test (i.e., correct math calculation in 5 s) were coded + 1, and incorrect responses (i.e., no response or incorrect math calculation in 5 s) were coded 0. For each child, responses were then added up in a single math score. To correct for potential variabil- ity in mathematics pro ciency across classes, math scores were standardized within each class of the six schools. Internal consistency of the math test based on all eight calculations was satisfactory (Cronbach s a = .89). Separate one-way ANOVAs were conducted on scores of automatic associations and math scores, using condition (stereotype consistent, control, ste- reotype inconsistent), gender (male, female), and the schools as independent variables. Schools did not lead to any signi cant result (all p s > .70), and thus this variable is not further discussed.

Table 1 Zero-Order Correlations Between Automatic Associations, Math Per- formance, and Explicit Stereotype Endorsement for Girls and Boys

 

1

23

Girls (n = 120) Automatic associations Math performance Explicit stereotype endorsement Boys ( n = 120) Automatic associations Math performance Explicit stereotype endorsement

 

.267**

.115

.142

.045

 

.000

.034

Note. Pearsons r coefcients are reported for automatic associations and math performance. Spearmans r s coefcients are reported for the categorical variable explicit stereotype endorsement. **p < .01.

Zero-order correlations between all dependent variables (automatic associations, math performance, and explicit math gender stereotype endorsement), separately for girls and boys across experimental conditions, are presented in Table 1.

Automatic Associations and Their Malleability

We predicted stereotype-consistent automatic associations for both girls and boys in the control condition, in which the math gender stereotype was not made salient. In addition, we aimed at test- ing for the malleability of children s automatic asso- ciations. Higher scores of automatic associations were expected in the stereotype-consistent than in the stereotype-inconsistent condition, with results in the middle for the control condition. A two-way ANOVA was conducted on scores of automatic associations, with condition (stereotype consistent, control, stereotype inconsistent) and gender (male, female) as the between-participants variables. Results showed a signi cant Condition 9 Gender interaction, F (2, 234) = 6.35, p = .002, g p 2 = .05. To assess the effect of condition within gender, simple-effect analyses were conducted sepa- rately for girls and boys. Table 2 presents the aver- age scores of automatic associations in each experimental condition for both boys and girls, together with information about the planned con- trasts between the means. No effect of condition emerged for boys automatic associations ( p>. 30). Moreover, one-sam- ple t tests against 0 showed no effect in any condi- tion (all p s > .20). Conversely, simple-effect analysis on girlsautomatic associations revealed a signi - cant effect of condition, F (2, 117) = 7.75, p < .01,

Table 2 Average Scores of Automatic Associations and Math Scores as a Func- tion of Condition (Stereotype Consistent, Control, Stereotype Inconsis- tent) for Girls and Boys

 

Stereotype

 

Stereotype

consistent

Control

inconsistent

Girls (n = 120) Automatic associations Math test

**0.28 a ( SD = 0.45) 0.36 a (SD = 0.73)

*0.19 a

(SD = 0.78)

0.10 b (SD = 0.43)

 

0.16

b

Boys ( n = 120) Automatic associations Math test

0.01 ab (SD = 0.49)

(SD = 0.92)

0.03

a

0.10

a

(SD = 0.44) 0.28 a (SD = 1.03)

0.05 a (SD = 0.30)

(SD = 0.50)

0.05

a

 

(SD = 1.02)

0.03 a (SD = .75)

Note. Means within rows not sharing the same subscript are sig- ni cantly different from each other at the p < .05 level (Bonfer- roni test). *p < .02. **p < .001.

g p 2 = .06, linear trend, p < .001. Post hoc tests with Bonferroni correction of the alpha level for multiple comparisons showed that girls scores of automatic associations were lower in the stereotype-inconsis- tent as compared to the control ( p < .02) and the stereotype-consistent ( p < .01) conditions. As shown in Table 2, although girls automatic associations were the highest in the stereotype-consistent condition, scores in the control and the stereotype- consistent conditions did not differ from each other ( p > .90). Moreover, one-sample t tests against 0 showed that girls scores of automatic associations were different from 0 in the stereotype-consistent and in the control (both p s < .02) but not in the stereotype-inconsistent condition ( p = .20). Importantly, simple-effect analyses on the main effect of gender within conditions revealed that boys scores were lower than girls scores of automatic associations in the stereotype-consistent,

F (1, 234) = 5.90, p < .03, g p 2 = .02, and control con-

dition, F (1, 234) = 5.84, p < .03, g p 2 = .02. Although this difference fell short of signi cance, boys scores of automatic associations tended to be higher than those of girls in the stereotype-inconsistent condi- tion, F (1, 234) = 3.79, p = .053, g p = .01. In sum, although automatic associations consis- tent with math(language) gender stereotypes were not found for boys, girls revealed stereotype-consis- tent automatic associations in both the control and the stereotype-consistent conditions. Importantly, girls automatic associations showed the expected malleability, thus differing in activation as a function of the experimental condition: They were higher in

2

Stereotype Threat Before Stereotype Awareness 257

the stereotype-consistent and the control conditions than the stereotype-inconsistent conditions. No effect of the experimental manipulation was found for boys automatic associations.

Math Performance

For girls, we expected performance decrements in the stereotype-consistent condition as compared to the control and the stereotype-inconsistent condi- tions. Conversely, no effect of condition on boys math performance was hypothesized. An ANOVA on math scores, with condition (stereotype consistent, control, stereotype inconsis- tent) and gender (male, female) as the between- participants variables, was conducted. A Condi- tion 9 Gender interaction emerged, F (2, 234) = 4.69, p = .010, g p 2 = .04. Therefore, the effect of condition was tested separately for girls and boys. Simple- effect analyses on boys math scores (Table 2) showed that boys performed equally well across conditions ( p > .30). To the opposite, an effect of condition was found for girls, F (2, 117) = 3.66, p < .03, g p 2 = .03, linear trend, p < .01: Girls under- performed in the stereotype-consistent as compared to the stereotype-inconsistent condition (p < .03). On the contrary, math scores in the control and both in the stereotype-consistent and in the stereotype- inconsistent conditions were not different from each other (both p s > .20). Thus, whereas the salience of the negative in-group stereotype led girls to perform worse on the math test, the removal of the same stereotype led girls to perform better in the stereotype-inconsistent condition as compared to the stereotype-consistent condition.

Explicit Stereotype Endorsement

To assess stereotype endorsement, participants were shown a picture of a boy and a girl and were asked to say whether the boy or the girl is espe- cially good at math, or whether they are equal. For each child, stereotype-inconsistent response (i.e., the girl is better at math ) was coded 1, neutral choice (i.e., the boy and the girl are equal ) was coded 0, and

stereotype-consistent choice (i.e., the boy is better at math ) was coded + 1. We expected no evidence that children endorsed the math gender stereotype. A logistic regression for ordinal dependent mea- sures was performed on children s choices at the task

( 1 = stereotype-inconsistent , 0 = neutral , + 1 = stereo- type-consistent choice ), with gender (0 = male ,

1

= female ), condition ( 1 = stereotype inconsistent ,

0

= control, +1 = stereotype consistent), and the product

258 Galdi, Cadinu, and Tomasetto

term as predictors. A main effect of gender emerged, Wald v 2 = 11.49, p < .001, indicating that the proba- bility of a counterstereotypical versus neutral versus stereotypical response was not equal for boys and girls. Regardless of condition, 57% of boys and 57% of girls favored their own gender and indicated the boy or the girl, respectively, as the most talented for math. Only 12% of boys and 21% of girls responded that the boy and the girl were the same. Of the remaining children, 22% of girls identi ed the boy as being better at math, whereas 31% of boys believed that the outstanding math student was the girl. Thus, consistent with predictions, 6-year-olds did not endorse the math gender stereotype, but simply manifested gender in-group favoritism.

Mediation Analysis

Because of null ndings for boys on both auto- matic associations and math scores, a mediation analysis was conducted only for girls to test whether automatic associations mediated the rela- tion between condition and math performance. Given that the predictor (i.e., condition) was a cate- gorical variable with three levels, we created two dummy-coded variables to conduct the mediation analysis (Hayes & Preacher, 2012) with the stereo- type-consistent condition as the reference group. Speci cally, Contrast 1 tested the effect of the ste- reotype-consistent (coded 0) versus stereotype- inconsistent (coded 1) condition, with the control condition coded 0. Contrast 2 tested for the residual difference between the stereotype-consistent (coded 0) and the control (coded 1) conditions, with the stereotype-inconsistent condition coded 0. Consistent with the univariate analyses reported above, the effect of Contrast 1 on performance was

signi cant, b = .55, t (111) = 3.38, p < .01, whereas the effect of Contrast 2 fell short of signi cance,

b = .34, t (111) = 1.91, p = .06. Similarly, the effect of Contrast 1 on automatic associations was signi cant,

b = .37, t (111) = 3.56, p < .001, whereas the effect

of Contrast 2 was not, p > .40. Moreover, when automatic associations and the two contrasts were entered simultaneously in the model predicting performance, the effect of automatic associations was signi cant, b = .36, t (111) = 2.32, p < .03, indi- cating that automatic associations negatively affect performance. Importantly, the effect of Contrast 1 (i.e., stereotype-consistent vs. stereotype-inconsistent condition) was reduced, b = .38, t (110) = 2.06, p < .05, whereas the effect of Contrast 2 (i.e., stereo- type-consistent vs. control condition) remained not

signi cant, b = .31, t (111) = 1.77, p = .07. Figure 1

Automatic Associations -.36** -.37*** .38* (.55**) Experimental Math Performance Condition
Automatic Associations
-.36**
-.37***
.38* (.55**)
Experimental
Math Performance
Condition

Figure 1. Results of mediation analyses testing indirect effects of experimental condition (stereotype consistent = 0, stereotype incon- sistent = 1) on math performance via automatic associations in girls (n = 120). *p < .05. **p < .01. ***p < .001.

summarizes results for Contrast 1. To test for media- tion, we calculated a bias-corrected 95% con dence interval for the indirect effect (.13, SE = .06) using a bootstrapping technique (Preacher & Hayes, 2008). As the null hypothesis of no mediation states that the indirect effect is 0, the null hypothesis is rejected when the con dence interval does not include 0. In the present analysis, the con dence interval (with 5,000 resamples) for the estimate of the indirect effect of Contrast 1 on performance did not include 0, 95% CI [0.02, 0.28], thus con rming that auto- matic associations mediated the relation between condition (stereotype consistent vs. stereotype inconsistent) and girls performance.

Supplementary Analyses

Although main analyses showed that stereotype endorsement did not vary as a function of condition, supplementary analyses were conducted to test whether the relations among condition, girls automatic associations, and girls math per- formance were further moderated by stereotype endorsement. Two-way analyses of variance were conducted on both scores of automatic associations and math performance, with condition (stereotype consistent, control, stereotype inconsistent) and explicit stereotype endorsement (stereotype-consis- tent, neutral, stereotype-inconsistent response) as the between-participants variables. In both ANOVAs, explicit stereotype endorsement yielded no signi cant effects: The effect of condition was signi cant on both scores of automatic associa- tions, F(2, 111) = 7.76.12, p < .001, and math per- formance, F(2, 111) = 4.12, p < .02, and was not further qualied by explicit stereotype endorse- ment (all p s > .15). However, given that the rela- tively low number of girls who provided a neutral response or indicated the other gender group as

better at math resulted in unbalanced cell frequen- cies, this nding could be a consequence of low statistical power. To rule out this possibility, we repeated the same analyses by recoding explicit stereotype endorsement into a dichotomous vari- able, comparing girls indicating their gender group as better at math with those who either indicated boys as better or provided a neutral response. Results were very similar: The effect of condition remained signi cant on both automatic associa- tions, F (2, 114) = 6.83, p < .01, and math perfor- mance, F (2, 114) = 4.59, p < .01, and was not further qualied by explicit stereotype endorse- ment (all p s > .09). We also tested whether automatic associations may act as a moderator, rather than a mediator, of the effect of stereotype activation on girlsmath perfor- mance. An analysis of covariance (ANCOVA) was performed on math performance with condition (stereotype consistent, control, and stereotype incon- sistent), automatic associations, and the interaction between condition and automatic associations as predictors. The main effect of automatic associations was signicant, F(1, 114) = 5.57, p < .02, whereas the interaction between condition and automatic associa- tions was not (p > .20), thus conrming that auto- matic associations acted as a mediator of the relation between condition and performance. Finally, we tested whether automatic associations and stereotype endorsement may interact with each other in determining girl s vulnerability to stereo- type threat. We carried out a three-way ANCOVA on math performance with condition (stereotype consistent, control, stereotype inconsistent), auto- matic associations, and explicit stereotype endorse- ment response (stereotype consistent, neutral, stereotype inconsistent) as predictors. Again, only the main effect of automatic associations was signif- icant, F (1, 114) = 4.01, p < .05, whereas no other main effect or higher order interaction attained sig- ni cance (all p s > .10). Although boys showed null ndings on all dependent variables, the same supplementary anal- yses were conducted for them as well. In all analy- ses, neither main effects nor interactions reached signi cance (all p s > .08).

Discussion

Although the stereotype threat model identi es ste- reotype awareness as a requirement for stereotype threat effects, research has shown that gender iden- tity activation affects girls math performance before

Stereotype Threat Before Stereotype Awareness 259

the emergence of stereotype awareness, and in the absence of endorsement of the math gender stereo- type (Ambady et al., 2001; Neuville & Croizet, 2007; Tomasetto et al., 2011). The present research disentangles this paradox and extends the stereo- type threat model by demonstrating that automatic associations consistent with a negative in-group ste- reotype represent a key factor that may trigger stereotype threat in young children. Consistent with research showing that automatic associations can precede conscious beliefs (e.g., Gaw- ronski & Galdi, 2011), we demonstrated for the rst time that girls acquire the math gender stereotype as automatic associations before its emergence at the explicit level: Despite the absence of evidence that children endorsed the math gender stereotype, 6-year-old girls, but not boys, even in the control condition (i.e., in the absence of any experimental manipulation aimed at increasing or decreasing the salience of the stereotype) showed stereotype-consis- tent automatic associations between boy and mathe- matics, and girl and language . Moreover, consistent with research showing malle-ability of automatic associations in adults (e.g., Dasgupta & Asgari, 2004), for the rst time, malleability of girls auto- matic associations was found. After coloring a draw- ing of a boy correctly solving a math problem, girls stereotypical automatic associations were activated, as compared to the control condition. At the same time, after coloring a picture of a girl succeeding in math, stereotypical automatic associations were reduced. Importantly, the activation of automatic associa- tions in the stereotype-consistent condition ham- pered girlsmath performance as compared to the stereotype-inconsistent condition, in which, con- versely, the reduced activation of automatic associa- tions led girls to the highest performance. Furthermore, highlighting the role of counterstereo- typical information in counteracting stereotype threat effects, these ndings are parallel to research showing that providing information about equal gender abilities in math (i.e., stereotype-inconsistent information) is an effective strategy to prevent women s performance decits (e.g., Cadinu, Maass, Frigerio, Impagliazzo, & Latinotti, 2003). An open issue of this study concerns the poten- tial processes underlying girls underperformance in the stereotype-consistent versus -inconsistent condi- tion. Consistent with Forbes and Schmader (2010), one may speculate that activated stereotypical auto- matic associations in the stereotype consistent con- dition may have burdened working memory capacity, thus disrupting subsequent math performance.

260 Galdi, Cadinu, and Tomasetto

Conversely, the reduced activation in the stereo- type-inconsistent condition may have freed up working memory resources, thus enhancing subse- quent performance. However, no such conclusions can be drawn as this study did not test how girls automatic associations affect performance. There- fore, a direct test of the working memory hypothe- sis should be the goal of future studies. Another potential process underlying girls worse performance in the stereotype-consistent than the stereotype-inconsistent condition could be a mere priming effect , which would lead individuals to behave automatically consistent with a cognitively activated image (e.g., Wheeler & Petty, 2001). How- ever, if a priming effect were at work, one should expect underperformance even for boys in response to stereotype-related stimuli. To the contrary, nei- ther boys automatic associations nor performance were affected by the activation of the math gender (counter)stereotype. Thus, differently from a prim- ing effect, we argue that in this study it was the membership in a negatively stereotyped group (whose stereotype has been acquired via automatic associations) that was responsible for girls perfor- mance de cit under stereotype threat. Although girls automatic associations increased and decreased in activation depending on the ste- reotype content of the experimental condition, no changes emerged at the level of endorsement of ste- reotypical beliefs. Consistent with research showing gender in-group bias in young children, the major- ity of children across conditions indicated their gen- der as superior in math. We argue that this lack of endorsement of the math gender stereotype further strengthens the role of automatic associations in the process of stereotype acquisition in children. Relative to the latest point, the present pattern of ndings raises the tricky issue concerning the sources and mechanisms underlying the formation of automatic associations. Up to date, highly in uen- tial theorizing has posited that automatic associa- tions stem from attitudes and conscious beliefs that have become overlearned and automatized over time (e.g., Rudman, 2004; Wilson et al., 2000). However, recent research (e.g., Gawronski & Galdi, 2011), as well as the present results, suggests that also other underlying mechanisms may be responsible for the formation of automatic associations. Drawing on the distinction between associative and propositional learning (Gawronski & Bodenhausen, 2006), we argue that the automatic formation of stereotype-con- sistent associative links in children s cognitive map could stem from repeated co-occurrences of objects or stimuli in children s social environment. For

example, ample evidence shows that children are sensitive to adults nonverbal behavior and that such behaviors, often occurring in automatic and uncon- scious ways, may represent an important channel through which attitudes are transmitted (e.g., Rud- man, 2004; Walden & Ogan, 1988). Thus, even though adults may explicitly encourage nonstereo- typical interests in children, they may exhibit differ- ent patterns of nonverbal behaviors: Parents may purchase more games or manipulative materials related to math and science for their sons than for their daughters (Jacobs & Bleeker, 2004), or intrude more often in their daughters than in their sons math homework to offer unsolicited help (e.g., Bahnot & Jovanovic, 2005). As a result, such repeated co-occurrences of objects and events might promote the automatic formation of stereotype-consistent associative links in children s cognitive maps. In contrast to girls, boys did not reveal stereotypi- cal automatic associations in any conditions. Consistent with research demonstrating that elemen- tary school boys tend to manifest in-group favoritism regarding both math and language (Steele, 2003), these results could simply re ect the fact that boys showed in-group serving automatic associations about both math and language domains (Steffens et al., 2010), regardless of the stereotype content con- veyed by the experimental condition. Indeed, the presence of in-group-serving boy mathematics and boy language automatic associations would result in fast response latencies in both critical blocks of the Child IAT. Thus, given that the D -algorithm com- pares the extent to which performance on the incom- patible (i.e., girl math and boy language sharing the same response key) is impaired relative to the compatible critical block (i.e., girl language and boy math sharing the same response key), in-group serving automatic associations could have led to the small or null score of automatic associations that was found for boys across conditions. However, because both gender stereotypes regarding math and language contribute to the Child IAT score, and can- not be separated (Nosek et al., 2005), this possibility could not be tested in this study. To our knowledge, only one study used an implicit measure to assess math gender and language gender stereotypes separately (i.e., Go-No-Go association task [GNAT]; Nosek & Banaji, 2001) in a sample of 14-year-olds and university students. Consistent with the in- group serving explanation, Steffens and Jelenec s (2011) male participants revealed both boys math stereotypical automatic associations and boys lan- guage counterstereotypical automatic associations. However, as Steffens and Jelenec noted, the internal

consistency of the GNAT was low

clearly appears more sensitive (Steffens & Jelenec, 2011, p. 332). Thus, a goal of future studies should be to test whether other implicit measures that combine the measurement quality of the IAT with the separate measurement of math and language gender stereo- types, such as the GNAT, can be implemented and adapted for use with young children. Studies on the development of cognitive compe- tencies offer alternative explanations for boys lack of automatic associations. For example, previous research has shown earlier knowledge of gender categories (Zosuls et al., 2009) and earlier achieve- ment of gender constancy (Ruble et al., 2007) by girls than boys. At the same time, differences in gender categorization abilities between girls and boys could also be the result of socialization pro-

cesses. For example, women show stronger auto- matic associations between the self, their gender, and the in-group gender stereotype as compared to men (Cadinu & Galdi, 2012), and children from low-status groups are more likely to be aware of self-relevant stereotypes than children from high- status groups (McKown & Weinstein, 2003). Although consistent with Steffens et al. (2010), the result that boys did not manifest stereotypical automatic associations is inconsistent with Cvencek et al.s (2011) ndings, showing stereotypical auto- matic associations between gender and academic subjects in both 6- to 7-year-old boys and girls. However, this discrepancy might re ect cross- cultural variations in the strength of gender stereo- types at the societal level, which have been found to be associated with either smaller or bigger gen- der differences in the strength of stereotypical auto- matic associations (Nosek et al., 2009), as well as in the endorsement of gender stereotypes about academic abilities (e.g., Evans, Schweingruber, & Stevenson, 2002). Moreover, in a study involving Chinese, Japanese, and American children, Lummis and Stevenson (1990) found that whereas American and Chinese 6-year-olds do not believe that there are gender differences in math, 6-year-old Japanese children tend to believe that boys are better than girls at math (see also Del Rio & Strasser, 2013). These results show that cultural differences in the strength of gender stereotypes about academic abili- ties, in schooling, and in out-of-school experiences, may lead to different acquisition ages of the endorsement of these stereotypes, and suggest that the same could be true regarding the acquisition of stereotypical automatic associations. In conclusion, this study has the potential to con- tribute to the stereotype threat model by suggesting

and the IAT

Stereotype Threat Before Stereotype Awareness 261

automatic associations as a novel prerequisite for stereotype-induced underperformance in young children. Importantly, by showing also that chil- dren s automatic associations are malleable, these ndings are promising in terms of interventions to promote gender equality in math and science because they suggest that girls can be protected from the deleterious impact of math gender stereo- types. Relevant to this claim is a study (Dasgupta & Greenwald, 2001) demonstrating that Caucasian participants exposed to images of admired Black and disliked White exemplars showed lower pro- White automatic associations than participants exposed to images of admired Whites and disliked Blacks. Interestingly, such a decrease in pro-White automatic associations lasted for 24 hr, suggesting that relatively enduring changes in automatic associ- ations can be obtained. If so, repeatedly presenting girls with exemplars of successful women in math and science could promote the reduction in stereo- typical automatic associations, long before girls have acquired any awareness or endorsement of the ste- reotype favoring males in math. This study suggests that this strategy could protect girls performance in stereotype-threatening situations and potentially help them to expand their interests toward tradi- tionally male domains.

References

Aboud, F. E. (1988). Children and prejudice . New York:

Basil Blackwell. Ambady, N., Shih, M., Kim, A., & Pittinsky, T. (2001). Stereotype susceptibility in children: Effects of identity activation on quantitative performance. Psychological Science , 12, 385 390. doi:10.1111/1467-9280.00371 Aronson, J., & Good, C. (2003). The development and consequences of stereotype vulnerability in adolescents.

In F. Pajares & T. Urdan (Eds.), Adolescence and educa- tion: Vol. 2. Academic motivation of adolescents (pp. 299 330). Greenwich, CT: Information Age. Augoustinos, M., & Rosewarne, D. L. (2001). Stereotype knowledge and prejudice in children. British Journal of Developmental Psychology , 19, 143156. doi:10.1348/

026151001165912

Bahnot, R., & Jovanovic, J. (2005). Do parents academic gender stereotypes in uence whether they intrude on their children s homework? Sex Roles , 52 , 597607.

doi:10.1007/s11199-005-3728-4

Baron, A. S., & Banaji, M. (2006). The development of implicit attitudes. Evidence of race evaluations from ages 6 and 10 and adulthood. Psychological Science , 17, 5358. doi:10.1111/j.1467-9280.2005.01664.x Cadinu, M., & Galdi, S. (2012). The cognitive accessibility of womanhood: Implicit gender self-categorization

262 Galdi, Cadinu, and Tomasetto

leads to stronger self-stereotyping by women than men. European Journal of Social Psychology , 42, 546551. doi:

10.1002/ejsp.1881

Cadinu, M., Maass, A., Frigerio, S., Impagliazzo, L., & Latinotti, S. (2003). Stereotype threat: The effect of expectancy on performance. European Journal of Social Psychology , 33 , 267285. doi:10.1002/ejsp.145 Cvencek, D., Meltzoff, A., & Greenwald, A. G. (2011). Math-gender stereotypes in elementary school children.

Child Development , 82, 766779. doi:10.1111/j.1467-8624.

2010.01529.x

Dasgupta, N., & Asgari, S. (2004). Seeing is believing:

Exposure to counter-stereotypic women leaders and its effect on the malleability of automatic gender stereo- typing. Journal of Experimental Social Psychology , 40, 642658. doi:10.1016/j.jesp.2004.02.003 Dasgupta, N., & Greenwald, A. G. (2001). On the malleability of automatic attitudes combating automatic prejudice with images of admired and disliked individuals. Journal of Personality and Social Psychology , 81, 800 814. doi:10.1037/0022-3514.81.5.800 Davies, P. G., Spencer, S. J., Quinn, D. M., & Gerhard- stein, R. (2002). Consuming images: How television commercials that elicit stereotype threat can restrain women academically and professionally. Personality and Social Psychology Bulletin, 28 , 16151628. doi:10.1177/

014616702237644

Del Rio, M. F., & Strasser, K. (2013). Preschool children s

beliefs about gender differences in academic skills. Sex Roles , 68, 231238. doi:10.1007/s11199-012-0195-6 Eagly, A. H., & Chaiken, S. (1993). The psychology of atti- tudes. Fort Worth, TX: Harcourt Brace Jovanovich. Evans, E. M., Schweingruber, H., & Stevenson, H. W. (2002). Gender differences in interest and knowledge acquisition: The United States, Taiwan, and Japan. Sex Roles , 47, 153167. doi:10.1023/A:1021047122532 Forbes, C., & Schmader, T. (2010). Retraining attitudes and stereotypes to affect motivation and cognitive capacity under stereotype threat. Journal of Personality and Social Psychology , 99, 740754. doi:10.1037/a0020971 Gawronski, B., & Bodenhausen, G. V. (2006). Associative and propositional processes in evaluation: An integra- tive review of implicit and explicit attitude change. Psy- chological Bulletin, 132, 692731. doi:10.1037/0033-2909.

132.5.692

Gawronski, B., & Galdi, S. (2011). Using implicit mea- sures to read the minds of undecided voters. In M. Cadinu, S. Galdi, & A. Maass (Eds.), Social perception, cognition, and language in honour of Arcuri (pp. 203216). Padova, Italy: CLEUP. Gawronski, B., & LeBel, E. P. (2008). Understanding pat- terns of attitude change: When implicit measures show change, but explicit measures do not. Journal of Experi- mental Social Psychology , 44 , 13551361. doi:10.1016/

j.jesp.2008.04.005

Gawronski, B., & Payne, B. K. (Eds.). (2010). Handbook of implicit social cognition: Measurement, theory, and applica- tions . New York: Guilford Press.

Greenwald, A. G., McGhee, D. E., & Schwartz, J. K. L. (1998). Measuring individual differences in implicit cog- nition: The implicit association test. Journal of Personality and Social Psychology , 74, 14641480. doi:10.1037/0022-

3514.74.6.1464

Greenwald, A. G., Nosek, B. A., & Banaji, M. R. (2003).

Understanding and using the implicit association test: I. An improved scoring algorithm. Journal of Personality and Social Psychology , 85, 187216. doi:10.1037/0022-

3514.85.2.197

Hamilton, D. L., & Trolier, T. K. (1986). Stereotypes and stereotyping: An overview of the cognitive approach. In J. Dovidio & S. Gaertner (Eds.), Prejudice, discrimination, and racism (pp. 127163). San Diego, CA: Academic Press. Hayes, A. F., & Preacher, K. J. (2012). Statistical mediation analysis with a multicategorical independent variable . Man- uscript under review. Retrieved November 10, 2012, from http://www.afhayes.com/public/hp2012.pdf Inzlicht, M., & Schmader, T. (Eds.). (2012). Stereotype threat: Theory, process, and application . New York: Oxford University Press. Jacobs, J. E., & Bleeker, M. M. (2004). Girl s and boys developing interests in math and science: Do parents matter? New Directions for Child and Adolescent Develop- ment , 106 , 521. doi:10.1037/0022-0663.96.1.97 Katz, P. A., & Kofkin, J. A. (1997). Race, gender, and young children. In S. S. Luthar, J. A. Burack, D. Cicchetti, & J. Weisz (Eds.), Developmental psychopathology: Perspectives on adjustment, risk, and disorder (pp. 5174). New York:

Cambridge University Press. Kiefer, A. K., & Sekaquaptewa, D. (2007). Implicit stereo- types, gender identi cation, and math-related out- comes: A perspective study of female college students. Psychological Science , 18, 1318. doi:10.1111/j.1467-9280.

2007.01841.x

Lummis, M., & Stevenson, H. W. (1990). Gender differ- ences in beliefs and achievement: A cross-cultural study. Developmental Psychology , 26, 254263. doi:

10.1037/0012-1649.26.2.254

Martin, C., & Ruble, D. (2010). Patterns of gender devel- opment. Annual Review of Psychology , 61 , 353381.

doi:10.1146/annurev.psych.093008.100511

Martinot, D., Bag es, C., & D esert, M. (2012). French chil- dren s awareness of gender stereotypes about mathe- matics and reading: When girls improve their reputation in math. Sex Roles , 66, 210219. doi:10.1007/

s11199-011-0032-3

Martinot, D., & D esert, M. (2007). Awareness of a gender stereotype, personal beliefs and self-perceptions regard- ing math ability: When boys do not surpass girls. Social Psychology of Education, 10, 455 471. doi:10.1007/

s11218-007-9028-9

McKown, C., & Strambler, M. J. (2009). Developmental antecedents and social and academic consequences of stereotype-consciousness in middle childhood. Child Development , 80, 16431659. doi:10.1111/j.1467-8624.

2009.01359.x

McKown, C., & Weinstein, R. (2003). The development and consequences of stereotype consciousness in mid- dle childhood. Child Development , 74, 498515. doi:10.

1111/1467-8624.7402012

Muzzatti, B., & Agnoli, F. (2007). Gender and mathemat- ics: Attitudes and stereotype threat susceptibility in

Italian children. Developmental Psychology , 43, 747759.

doi:10.1037/0012-1649.43.3.747

Neuville, E., & Croizet, J. (2007). Can salience of gender identity impair math performance among 78 years old girls? The moderating role of task dif culty. European Journal of Psychology of Education, 22, 307316. doi:

10.1007/BF03173428

Nosek, B. A., & Banaji, M. R. (2001). The go/no-go associ- ation task. Social Cognition , 19, 625 666. doi:10.1521/

soco.19.6.625.20886

Nosek, B. A., Banaji, M. R., & Greenwald, A. G. (2002). Harvesting implicit group attitudes and beliefs from a demonstration web site. Group Dynamics , 6, 101115.

doi:10.1037/1089-2699.6.1.101

Nosek, B. A., Greenwald, A. G., & Banaji, M. R. (2005). Understanding and using the Implicit Association Test:

II. Method, variables, and construct validity. Personality and Social Psychology Bulletin , 31, 166180. doi:10.1177/

0146167204271418

Nosek, B. A., & Smyth, F. L. (2011). Implicit social cogni- tions predict sex differences in math engagement and achievement. American Educational Research Journal , 48,

11251156. doi:10.3102/0002831211410683 Nosek, B. A., Smyth, F., Sriram, N., Lindner, N., Devos, T., Ayala, A., et al. (2009). National differences in gender-science stereotypes predict national sex differences in science and math achievement. Proceed- ings of the National Academy of Sciences of the United States of America , 106 , 1059310597. doi:10.1073/pnas.

0809921106

Olson, M. A., & Fazio, R. H. (2006). Reducing automatically activated racial prejudice through implicit evaluative conditioning. Personality and Social Psychology Bulletin, 32, 421433. doi:10.1177/0146167205284004 Perner, J., & Wimmer, H. (1985). John thinks that Mary thinks that : Attribution of second-order beliefs by 5- to 10-year old children. Journal of Experimental Child Psychol- ogy, 39 , 437471. doi:10.1016/0022-0965(85)90051-7 Powlishta, K. K. (1995). Gender bias in children s percep- tions of personality traits. Sex Roles , 32, 1728. doi:

10.1037/0012-1649.31.5.781

Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and re- sampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods , 40, 879891. doi:10.3758/BRM.40.3.879 Quintana, S. M. (1998). Children s developmental under- standing of ethnicity and race. Applied and Preventive Psychology , 7, 2745. doi:10.1016/S0962-1849(98)80020-6 Rholes, W. S., & Ruble, D. N. (1984). Children s under- standing of dispositional characteristics of others. Child Development , 55, 550560.

Stereotype Threat Before Stereotype Awareness 263

Ruble, D., Taylor, L. J., Cyphers, L., Greulich, F. K., Lurye, L. E., & Shrout, P. (2007). The role of gender constancy in early gender development. Child Development , 78, 11211136. doi:10.1111/j.1467-8624.2007.01056.x Rudman, L. A. (2004). Sources of implicit attitudes. Cur- rent Directions in Psychological Science , 13 , 7982. doi:

10.1111/j.0963-7214.2004.00279.x

Selman, R. (1980). The growth of interpersonal understand- ing: Developmental and clinical analysis . New York: Aca- demic Press. Steele, C. M. (1997). A threat in the air: How stereotypes shape intellectual identity and performance. American Psychologist , 52, 613629. doi:10.1037/0003-066X.52.6.613 Steele, C. M., & Aronson, J. (1995). Stereotype threat and intellectual test performance of African Americans. Journal of personality and Social Psychology , 69 , 797811.

doi:10.1037/0022-3514.69.5.797

Steele, J. (2003). Childrens gender stereotypes about math: The role of stereotype stratication. Journal of Applied Social Psychology , 33, 25872606. doi:10.1111/

j.1559-1816.2003.tb02782.x

Steffens, M. C., & Jelenec, P. (2011). Separating implicit gender stereotypes regarding math and language:

Implicit ability stereotypes are self-serving for boys and man, but not for girls and women. Sex Roles , 64, 324

335. doi:10.1007/s11199-010-9924-x

Steffens, M. C., Jelenec, P., & Noack, P. (2010). On the leaky math pipeline: Comparing implicit math-gender

stereotypes and math withdrawal in female and male children and adolescents. Journal of Educational Psychol- ogy, 102, 947963. doi:10.1037/a0019920 Thoman, D., White, P., Yamawaki, N., & Koishi, H. (2008). Variations of gender-math stereotype content affect women s vulnerability to stereotype threat. Sex Roles , 58, 702712. doi:10.1007/s11199-008-9390-x Tomasetto, C., Alparone, F. R., & Cadinu, M. (2011). Girls math performance under stereotype threat: The moderating role of mothersgender stereotypes. Devel- opmental Psychology , 47, 943949. doi:10.1037/a0024047 Walden, T., & Ogan, T. (1988). The development of social referencing. Child Development , 59, 1230 1240. Wheeler, S. C., & Petty, R. E. (2001). The effects of stereo- type activation on behavior: A review of possible mech- anisms. Psychological Bulletin , 127, 797826. doi:10.1037/

0033-2909.127.6.797

Wilson, T. D., Lindsey, S., & Schooler, T. Y. (2000). A model of dual attitudes. Psychological Review, 107 , 101

126. doi:10.1037/0033-295X.107.1.101

Yee, M., & Brown, R. (1994). The development of gender differentiation in young children. British Journal of Social Psychology , 33, 183196. doi:10.1111/j.2044-8309.1994.

tb01017.x

Zosuls, K. M., Ruble, D., Tamis-LeMonda, C. S., Shrout, P. E., Bornstein, M. H., & Greulich, F. K. (2009). The acquisition of gender labels in infancy: Implications for sex-typed play. Developmental Psychology , 45, 688701.

doi:10.1037/a0014053