Gendered Races

Abstract

Six studies explored the overlap between racial and gender stereotypes, and the consequences of this overlap for interracial dating, leadership selection, and athletic participation. Two initial studies captured the explicit and implicit gender content of racial stereotypes: Compared with the White stereotype, the Asian stereotype was more feminine, whereas the Black stereotype was more masculine. Study 3 found that heterosexual White men had a romantic preference for Asians over Blacks and that heterosexual White women had a romantic preference for Blacks over Asians; preferences for masculinity versus femininity mediated participants’ attraction to Blacks relative to Asians. The pattern of romantic preferences observed in Study 3 was replicated in Study 4, an analysis of the data on interracial marriages from the 2000 U.S. Census. Study 5 showed that Blacks were more likely and Asians less likely than Whites to be selected for a masculine leadership position. In Study 6, an analysis of college athletics showed that Blacks were more heavily represented in more masculine sports, relative to Asians. These studies demonstrate that the gender content of racial stereotypes has important real-world consequences.

Keywords

stereotypes race gender attraction leadership stereotyped attitudes racial and ethnic attitudes and relations sex role attitudes

Racial stereotypes affect virtually every important life outcome, from job interviews to job placement, from housing to education, from police stops to prison terms. For example, in the United States, Blacks are less likely than identical White applicants to land job interviews (Bertrand & Mullainathan, 2004) and tend to receive harsher sentences compared with White perpetrators of similar crimes (Klein, Petersilia, & Turner, 1990; Pettit & Western, 2004). Experimental evidence suggests that these disparities can be partly attributed to stereotypes of Blacks as more aggressive and less intelligent than Whites (Sommers & Ellsworth, 2000).

Gender stereotypes also profoundly affect life outcomes. Among full-time U.S. workers, women earn 77% of what men earn (U.S. Census Bureau, 2005, Table 1), and only 4.2% of Fortune 500 CEOs are currently women (Catalyst, 2012), despite there being few differences between men’s and women’s actual leadership skills (Eagly, Karau, & Makhijani, 1995). The descriptive stereotype of women as gentle and nurturing (Eagly & Karau, 2002), combined with prescriptive stereotypes that punish female assertiveness (Rudman & Glick, 1999), can explain why women are more likely than men to be denied promotions despite equal performance (Blau & Devaro, 2007).

Table 1.

Uniformity Indexes for Stereotype-Content Studies

	Study
Target group	Katz and Braly (1933)	Gilbert (1951)	Karlins, Coffman, and Walters (1969)	Devine and Elliot (1995)	Current Study 1
Asians	—	—	—	—	12.88
Chinese	12.0	14.5	10.8	—	—
Japanese	10.9	26.0	9.4	—	—
Whites	—	—	—	—	17.00
Americans	8.8	13.6	9.6	—	—
English	7.0	9.2	8.0	—	—
Blacks	4.6	12.0	12.3	4.5	11.42

Note: Stereotype uniformity is computed by assessing the fewest number of traits needed to account for 50% of all possible trait allocations. For example, in our Study 1, the summed frequency of all trait allocations for Asians was 553. Thus, stereotype uniformity was computed by determining how many traits it took to account for 276.5 trait allocations (553/2 = 276.5). The uniformity value of 12.88 means that 276.5 trait allocations were captured by the first 12 to 13 most frequently cited traits. Given instructions to choose 10 representative traits out of 99 total traits, a stereotype-uniformity score of 5 would indicate perfect agreement, and a score of 49.5 would indicate perfect disagreement. Because participants in the four prior studies were encouraged to select 5, rather than 10, traits, a stereotype-uniformity score of 2.5, rather than 5, indicated perfect agreement. Because 30% of the sample in the current study (16 participants) provided either more or less than 10 attributes per ethnic group, trait allocations were divided by the number of traits chosen to control for the number of traits chosen for each racial group.

Despite these well-documented effects of racial and gender stereotypes on consequential life outcomes, both in and outside of organizations (Cuddy, Glick, & Beninger, 2011), only a handful of studies have addressed the possibility that racial and gender stereotypes may interact. Eagly and Kite (1987) were some of the first scholars to suggest the possibility that social categories can be gendered. Two recent studies have explored how race affects the sex categorization of faces. Goff, Thomas, and Jackson (2008) found that participants made more errors categorizing Black female faces than White female faces because participants saw Blacks as more masculine than Whites. Johnson, Freeman, and Pauker (2011) found that participants were better able to categorize Asian female and Black male faces, compared with Black female and Asian male faces. Not many studies have explored the real-world implications of the overlap between racial and gender stereotypes. Although some studies have explored the intersection of race and gender in the context of hiring, most of these studies have been correlational and have offered contradictory results. For example, some studies found evidence for the double-jeopardy hypothesis that individuals with multiple-minority status (e.g., non-White female) suffer the most discrimination and negative workplace experiences (Berdahl & Moore, 2006; Kulik, Roberson, & Perry, 2007; Nelson & Probst, 2004), but other research has found that Black men experience the worst outcomes in the courtroom, in the labor market, and in education (Sidanius & Veniegas, 2000).

Here, we extend the research on the intersection of race and gender in two important ways. First, we demonstrate that the overlap between racial and gender stereotypes goes beyond facial features and is captured in the content of stereotypes. That is, we show that racial stereotypes are gendered. To do so, we adapted the methods of the Princeton Trilogy (Gilbert, 1951; Karlins, Coffman, & Walters, 1969; Katz & Braly, 1933), and also utilized an implicit reaction time method, to measure the gender content of racial stereotypes. Second, we explored the implications of this association for interracial marriages, leadership selection, and athletic participation. In Study 3, we surveyed a national sample about their interracial dating habits and attraction to Asians and Blacks. Study 4 involved archival analyses of interracial marriage patterns in data from the 2000 U.S. Census. In Study 5, participants evaluated an Asian, White, or Black applicant and selected him or her for either a masculine or a feminine leadership position. Study 6 used the National Collegiate Athletic Association (NCAA) Student-Athlete Ethnicity Report to examine whether the perceived masculinity of a sport predicted racial differences in athletic participation in it.

Study 1: The Gender Content of Racial Stereotypes

Study 1 used the well-established methodology of the Princeton Trilogy for analyzing stereotype content (Gilbert, 1951; Karlins et al., 1969; Katz & Braly, 1933). In this paradigm, one group of participants evaluates the valence of attributes. A second group then assigns attributes from the same list to different ethnic groups. The valence ratings are then used to score the attributes attributed to each ethnic group, and an overall negativity score is calculated for each group. We followed the same procedure, but had the first group of participants assess the femininity-masculinity, rather than valence, of each trait. This methodology allowed us to calculate the overall masculinity of each ethnic group’s stereotype content.

Participants and method

Eighty-five participants (54 female, 31 male) completed an online survey. Participants’ self-reports indicated that 54 were White, 20 were Asian, 7 were Black, and 4 belonged to other racial categories. Participants’ mean age was 37.46 years.

We randomly assigned participants to evaluate the femininity-masculinity of 99 traits or to attribute those traits to three ethnic groups (Asians, Whites, and Blacks). The participants who rated the traits used a scale from 1 (extremely masculine) to 10 (extremely feminine); we recoded this variable so that higher values indicated greater masculinity. The traits included the 84 originally used by Katz and Braly (1933), 9 that were added in a replication (see Devine & Elliot, 1995), and 6 additional traits. A masculinity score was computed for each of the 99 attributes.¹ The participants who assigned attributes to the ethnic groups were asked to select 10 of the 99 attributes for each group.

We used the masculinity scores for the attributes to score the trait attributions for each ethnic group. Past studies involving the Princeton Trilogy method have used a stereotype- uniformity measure to determine the level of agreement across stereotypical traits for a given race or ethnicity. The stereotype-uniformity scores we obtained are consistent with those of prior studies (see Table 1).

Results

A 3 (target race: Asian, White, or Black) × 2 (participant’s gender: male or female) × 2 (participant’s race: White or non-White) mixed-measures analysis of variance (ANOVA) revealed that the stereotype content for Blacks was considered to be the most masculine (M = 6.34, SD = 0.53); the stereotype content for Whites was intermediate (M = 6.06, SD = 0.49), and the stereotype content for Asians was the least masculine (M = 5.41, SD = 0.39); the effect of target race was significant, F(2, 98) = 47.62, p = .000 (see Fig. 1). All means were significantly different from one another, all ts(52) > 2.99, all ps < .004.² Thus, using Princeton Trilogy methods, we found a substantial overlap between the contents of racial and gender stereotypes. That is, racial stereotypes differed in their gender content.

Fig. 1.

The gender content of racial stereotypes. The graph shows the mean masculinity score for each racial group in Study 1. Error bars represent ±1 SEM.

One potential concern is that the traits used in this study may differ on valence and collectivism-individualism, and that these dimensions may covary with femininity-masculinity ratings and differ by ethnicity. Therefore, we asked a separate group of participants (N = 53; 35 female, 17 male, 1 unidentified; 18 Whites, 21 Asians, 13 participants of other racial categories; mean age = 20.06 years) to assess each trait for both valence (1 = extremely negative, 10 = extremely positive) and individualism-collectivism (1 = extremely individualistic, 10 = extremely collectivistic). Then, we conducted a Spearman rank-order correlation for all 99 traits’ collectivism, valence, and masculinity scores. There was a negative correlation between the masculinity and collectivism scores, r(99) = −.34, p = .001, and between the masculinity and valence scores, r(99) = −.54, p = .000, but a positive correlation between the collectivism and valence scores, r(99) = .20, p = .043. Next, following the Princeton Trilogy method, we scored the trait attributions for each ethnic group according to their valence and collectivism ratings. The stereotype content for Asians was more positive (M = 6.98, SD = 0.80) than the stereotype content for Whites (M = 5.06, SD = 1.27) and Blacks (M = 4.70, SD = 1.26), ts(52) > 8.00, ps < .001, but the White and Black stereotypes did not differ in valence, t(52) = 1.23, p = .224. The collectivism content of the Asian stereotype (M = 4.67, SD = 0.34) differed from that of the White (M = 4.30, SD = 0.41) and Black (M = 4.32, SD = 0.25) stereotypes, ts(52) > 4.90, ps < .001, but the White and Black stereotypes did not differ in collectivism, t(52) = 0.420, p = .68.

Overall, the pattern of data was most consistent with the interpretation that our main results were due to overlapping content between gender and racial categories because the stereotype content of Whites and Blacks differed on masculinity but not on valence and collectivism. Our next study tested whether gender was associated with race when the collectivism and valence of the traits was held constant. Further, this second study used an implicit measure to test for spontaneous gender-race associations.

Study 2: Implicit Associations Between Racial and Gender Stereotypes

Our second study tested whether the association between racial and gender stereotypes exists at an implicit level. We subliminally primed participants with a word related to race (Asian, White, or Black) and then measured their reaction times to masculine or feminine words. We predicted that participants primed with Black would respond most quickly to masculine words, but those primed with Asian would respond most quickly to feminine words.

Participants and method

Seventy-two individuals (41 females, 31 males). According to self-reports, 34 participants were White, 20 were East Asian, 6 were South Asian, 8 were Black, and 4 were Hispanic. Participants’ mean age was 20.29 years. The experiment had a 3 (subliminal prime: Asian, Black, or White) × 2 (traits: masculine or feminine) between-participants design.

Following the procedure of Wittenbrink, Judd, and Park (1997), we first asked participants to identify the ethnic-group membership (Black, White, or Asian) of individuals on the basis of their first names. This task was used to strengthen participants’ associations to the ethnic labels that would subsequently be used as subliminal primes.

Next, participants performed a task in which they had to decide whether strings of letters constituted English words as quickly and accurately as possible. They were told that light flashes would be presented to distract them, but that they should concentrate on the center of the screen where the letter strings would appear. In fact, these flashes were presentations of the prime word (Asian, Black, or White). Our priming procedure followed the recommendations of Bargh and Chartrand (2000). On each trial, the prime was flashed for 86 ms at one of the four corners of the screen and then immediately masked for 14 ms. Right after the masking string, the target stimulus was presented. It remained on the screen until the participant pressed either the “word” or the “nonword” button. There were a total of 40 trials (17 nonwords, 23 words). We varied the location of the prime to ensure that participants could not focus their attention on it. Indeed, when asked after the experiment, none of the participants were able to recognize the prime word to which they had been exposed.

For participants randomly assigned to the masculine condition, five of the target letter strings were words referring to masculine attributes: masculine, vigorous, strong, muscular, and burly. In the feminine condition, five of the target letter strings were words referring to feminine attributes: feminine, graceful, gentle, beautiful, and delicate. We preselected physically gendered, rather than behaviorally gendered, attributes to make sure that the traits had a positive valence and did not have collectivistic associations.³

Following the criteria of Balota and Lorch (1986), we excluded from analysis response latencies less than 300 ms or greater than 1,300 ms (2.3% of all latencies), to reduce the distorting effect of outliers.

Results and discussion

We submitted response latencies to a 3 (prime: White, Black, or Asian) × 2 (traits: masculine or feminine) × 2 (participant’s sex: masculine or feminine) × 2 (participant’s race: White or non-White) between-participants ANOVA. This analysis revealed a significant interaction between prime and traits, F(2, 60) = 3.58, p = .03, which was not qualified by participant’s sex, F(2, 48) = 0.004, p = .996, or participant’s race, F(2, 48) = 0.34, p = .71.

Participants who were subliminally primed with Black responded significantly more quickly (M = 580.38 ms, SD = 64.78) to masculine words than did participants subliminally primed with Asian (M = 663.85 ms, SD = 88.89), t(31) = 2.09, p = .04, or with White (M = 690.04 ms, SD = 122.28), t(31) = 2.52, p = .02; response latencies to masculine words did not differ between participants primed with Asian and those primed with White, t(31) < 1, p = .50. In contrast, participants who were subliminally primed with Asian responded significantly more quickly to feminine words (M = 517.05 ms, SD = 56.89) than did participants subliminally primed with Black (M = 583.49 ms, SD = 85.00), t(35) = 2.28, p = .03, or with White (M = 597.65 ms, SD = 85.00), t(35) = 2.92, p = .006; response latencies to feminine words did not differ between participants primed with Black and those primed with White, t(35) < 1, p = .65.

Even at the implicit level, there was a link between racial and gender stereotypes. When participants were subliminally primed with Black, masculine traits became more accessible. Conversely, for participants subliminally primed with Asian, feminine traits became more accessible.

Study 3: Implications for Interracial Dating Patterns

We propose that the association between racial and gender stereotypes has important implications for interracial dating patterns because men tend to prefer women who seem to personify the feminine ideal and women tend to prefer men who seem to embody masculinity. Indeed, femininity confers an advantage to women in the heterosexual dating market, and masculinity offers a comparative advantage to men in attracting the attention of the opposite sex (Buss, 2000).

We hypothesized that men’s preference for femininity and women’s preference for masculinity drive interracial dating preferences. First, we predicted that White men would be more attracted to Asian than to Black women. In contrast, we predicted that White women would be more attracted to Black than to Asian men; because we expected Whites to be most attracted to members of their own race, we based our hypotheses on participants’ propensity to date Asians relative to Blacks. Second, we predicted that preferences for masculinity-femininity would mediate participants’ attraction to Asians and Blacks.

Participants and procedure

Two hundred sixty-eight heterosexual White participants (182 females, 86 males; mean age = 42.71 years) completed an online survey. We first asked participants the extent to which they were attracted to femininity and to masculinity in a mate (1 = not at all, 7 = very much so). Participants then indicated the extent to which they were attracted to Asians, Blacks, and Whites (1 = not at all, 7 = very much so) and the number of Asians, Blacks, and Whites they had dated.

Results

Men were more attracted to femininity (M = 5.69, SD = 1.26) than to masculinity (M = 2.71, SD = 1.93), t(84) = 10.22, p < .001, and women were more attracted to masculinity (M = 5.46, SD = 1.55) than to femininity (M = 2.66, SD = 1.71), t(181) = 14.68, p < .001.

As predicted, men were more attracted to Asians (M = 5.13, SD = 1.71) than to Blacks (M = 3.71, SD = 1.84), t(85) = 7.45, p < .001, but women were more attracted to Blacks (M = 3.30, SD = 1.87) than to Asians (M = 2.86, SD = 1.82), t(180) = 3.17, p = .002.⁴ Whereas 62% of the men had dated an Asian woman, only 49% had dated a Black woman. Conversely, only 27% of the women had dated an Asian man, whereas 37% had dated a Black man.

To test our mediational hypothesis, we measured relative preference for masculinity by subtracting femininity preference from masculinity preference. We determined relative preference for Blacks versus Asians by subtracting attraction to Asians from attraction to Blacks. We regressed the attraction measure on preference for masculinity, b = 0.14, SE = 0.04, p = .000, and participant’s sex (1 = male, 2 = female), b = −0.98, SE = 0.34, p = .004. Preference for masculinity mediated the association between participant’s sex and attraction to Blacks versus Asians, Sobel z = −3.36, p < .000, and the total indirect effect was significant, b = 0.83, SE = 0.25, p = .001, 95% bias-corrected and accelerated bootstrap confidence interval = [0.4077, 1.2955] (Fig. 2).

Fig. 2.

Results from Study 3: preference for masculinity as a mediator of the effect of participant’s sex on relative attraction to Blacks versus Asians. Standardized regression coefficients are shown; numbers in parentheses are the coefficients obtained when the two predictors were entered simultaneously in the model. Asterisks indicate significant paths (*p < .01, **p < .001).

Study 3 found that men had a romantic preference for Asians over Blacks and that women had a romantic preference for Blacks over Asians. Further, these interracial dating preferences were mediated by participants’ relative preferences for masculinity versus femininity in their romantic partners.

Study 4: Archival Analyses of Data from the 2000 U.S. Census

To examine whether the interracial dating preferences found in Study 3 would hold outside the lab and in longer-term relationships, we conducted archival analyses of data from the 2000 U.S. Census (U.S. Census Bureau, 2003) and found a similar pattern among interracial marriages. Among Black-White marriages, 73% (208,798/287,576) had a Black husband and a White wife. Among Asian-White marriages, a strikingly different pattern emerged, as 75% of the marriages (380,475/504,119) had a White husband and an Asian wife. An even stronger pattern occurred among Asian-Black marriages; 86% of these marriages (27,520/31,992) had a Black husband and an Asian wife, and Asian-Black marriages were significantly different from Asian-White marriages, χ²(1, N = 536,111) = 1,840.48, p < .001.

One might propose that this pattern of interracial marriages could be explained by existing status disparities among racial groups. Indeed, females tend to prefer high-status males (Buss, 1989). Black Americans possess lower socioeconomic status (SES; e.g., income, education) than White Americans (Iceland & Wilkes, 2006), but the relative SES of Asian Americans is less clear; although the income and education levels of Asians exceed those of Whites, the poverty rate is higher among Asians (Iceland & Wilkes, 2006), and they experience more prejudice (Maddux, Galinsky, Cuddy, & Polifroni, 2008) and discrimination (Kim & Lewis, 1994) than Whites.

If one assumes that Whites have the highest status, followed by Asians and then Blacks, then the status account would be consistent with the Black-male/White-female combination being the least common category of interracial marriage. The observed pattern from the Census data is virtually the opposite, however. The association between racial and gender stereotypes documented here, combined with masculinity-femininity preferences in mate selection, provides a parsimonious explanation for the gender composition of interracial marriages in the Census data.

Study 5: Implications for Leadership Selection

The next study explored the implications of the association between racial and gender stereotypes for leadership selection. We created two types of leadership positions, one that required feminine traits for success and one that required masculine traits for success. We asked participants to look at a single job applicant and place that person into one of these leadership positions. We predicted that relative to White applicants, Asians would be more likely to be selected for the feminine position and Blacks would be more likely to be selected for the masculine position.

Participants and procedure

One hundred forty-eight participants responded to an online survey (100 females, 48 males). According to self-reports, 79 were White, 38 were Asian, 7 were Black, and 24 were of other racial categories. Participants’ mean age was 19.78 years.

Participants were instructed to review the application of a candidate for a leadership position within a company.⁵ Gender and race were manipulated in two ways. First, we used check boxes on the application for gender and ethnicity. Second, we manipulated names so that they were distinctively Asian, White, or Black (female names: Ming Lee, Emily, and Lakisha, respectively; male names: Ming Hoa, Greg, and Jamal, respectively; Bertrand & Mullainathan, 2004). Participants were given a choice of selecting the candidate for either the feminine or the masculine leadership role. The feminine role called for someone who would be collaborative and would require relationship building. Conversely, the masculine role required the candidate to be fierce, competitive, and contentious.

Results

We analyzed the percentage of participants who nominated their candidate to each of the leadership positions by conducting a 2 (race of candidate: Asian, White, or Black) × 2 (gender of candidate: male or female) × 2 (position selected: feminine or masculine) log-linear analysis. This analysis revealed the predicted effect of the candidate’s race on the leadership position selected, χ²(1, N = 148) = 10.28, p = .006.⁶ Sixteen percent of participants nominated the Asian person, 37% nominated the White person, and 43% nominated the Black person for the masculine position (Fig. 3).

Fig. 3.

Percentage of participants who chose the masculine (vs. feminine) leadership position for the Black, White, and Asian candidates in Study 5.

There was also a marginally significant three-way interaction, χ²(1, N = 148) = 5.34, p = .07. Among the participants who evaluated female candidates, the pattern followed our predictions perfectly: Fourteen percent selected the Asian applicant, 25% selected the White applicant, and 54% selected the Black candidate for the masculine position. Among the participants who evaluated male candidates, 17% nominated the Asian candidate and 48% nominated the White candidate for the masculine position; however, only 32% nominated the Black candidate for that position. This latter percentage was the only anomaly in the results. It could have resulted because the feminine role was described using more positive characteristics than the masculine role; participants may have wanted to demonstrate a lack of prejudice by placing the Black male in the feminine role because it seemed more positive. This may also explain why two thirds of participants placed their applicant in the feminine role. However, this difference in positivity of the role description cannot explain the full pattern of participants’ choices. Overall, the pattern of selection to a leadership role conformed to our predictions in five of the six cells.

We wanted to rule out the possibility that the influence of the candidate’s race occurred because of perceptions that Asians are more collectivistic, or that Whites and Blacks are more individualistic. We tested this alternative interpretation using a similar sample of online participants (N = 84; 61 females, 23 males; 48 Whites, 26 Asians, 7 Blacks, and 3 people of other racial categories; mean age = 20.65 years). Participants saw the same applications as in the main study and rated their candidate using Oyserman’s (1993) scale for collectivism-individualism; on a scale from 1 (strongly disagree) to 5 (strongly agree), they rated their agreement with nine statements about the candidate’s collectivism (e.g., “This candidate probably believes that, in the end, a person feels closer to members of his/her group than to others”) and nine statements about the candidate’s individualism (e.g., “This candidate probably believes that a man/woman of character helps his/her group before all else”). We averaged the ratings from the nine collectivism statements to create a collectivism scale (α = .84) and the ratings from the nine individualism statements to create an individualism scale (α = .71). Neither scale revealed any main effects or interactions involving the candidate’s race, all Fs < 1.30, all ps > .279. Thus, participants perceived no differences in individualism or collectivism between the Asian, White, and Black candidates.⁷

Study 6: Archival Analysis of the NCAA Student-Athlete Ethnicity Report

The previous three studies established that as masculinity becomes more valued in the romantic or leadership domain, Blacks get selected more than Asians. Our final study extended these findings to the athletic domain. We predicted that as a sport is seen as more masculine, Blacks will be more heavily represented in that sport, relative to Asians. We chose to analyze racial differences in collegiate athletic participation in the United States because becoming a student-athlete is competitive and carries significant advantages (e.g., preferential admissions and scholarships). Participation in college athletics also involves a rigorous selection process, as most college athletes are selected by coaches through a recruitment process.

We analyzed archival data from the 2010 NCAA Student-Athlete Ethnicity Report, which breaks down the racial composition of 30 different collegiate sports in Divisions I, II, and III over 11 years, from 2000 through 2010. We analyzed the data collapsing across all years and divisions. Difference scores were computed to assess the relative frequency of Blacks over Asians in each sport.

Sixty-five undergraduate sports fans (46 females, 18 males, 1 unidentified; 32 Whites, 18 Asians, 5 Blacks, 8 people of other racial categories, 2 unidentified; mean age = 21.5 years) assessed the masculinity of each sport on a scale from 1 (extremely feminine) to 10 (extremely masculine).⁸ As predicted, the perceived masculinity of the sports was significantly associated with the relative number of Blacks in those sports, β = 0.37, b = 0.00000148, SE = 0.00, p = .047. Specifically, the more masculine a sport was perceived to be, the more likely Blacks were to be college athletes in that sport relative to Asians.⁹ To test whether individualism or collectivism was confounded with masculinity, we conducted another analysis in which we controlled for the individual or team nature of each sport. This variable had no significant effect on the relative numbers of Blacks and Asians in the sports, β = 0.10, b = 0.306, SE = 0.61, p = .564. Further, when this control was included, the predicted effect remained marginally significant, β = 0.35, b = 0.00000142, SE = 0.00, p = .062.

In summary, the masculinity of a sport predicted the relative number of Black and Asian college athletes in that sport. It is important to note that these differential levels of participation are likely driven by both selection processes in recruitment and self-selection processes.

Discussion

The current studies establish that there is an important overlap between racial and gender stereotypes. This overlap provides a parsimonious explanation for the gender composition of interracial couples, the nomination of candidates from different races to feminine versus masculine positions of leadership, and racial differences in athletic participation at the collegiate level.

Although we posit that racial stereotypes are gendered, one might argue instead that gender stereotypes are racialized. Although our studies cannot rule out this alternative possibility, both adults and children categorize other people on the basis of gender before race, which suggests that gender is the more primary category (Ocampo, Bernal, & Knight, 1993; Shutts, Banaji, & Spelke, 2010). Thus, we believe that an influence from gender stereotypes to racial stereotypes is more likely than the opposite pathway. It should be noted that the gender content of racial stereotypes represents the contemporary context within the United States. The gender content of racial stereotypes is not invariant and likely fluctuates across time, contexts, and countries and cultures.

The present research demonstrates that the intersection of race and gender has important real-world consequences. Considering the overlap between racial and gender stereotypes opens up new frontiers for understanding how stereotypes affect the important decisions that drive people’s most significant outcomes at work and home.

Footnotes

Declaration of Conflicting Interests

The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.

Notes

References

Balota

D. A.

Lorch

R. F.

(1986). Depth of automatic spreading activation: Mediated priming effects in pronunciation but not in lexical decision. Journal of Experimental Psychology: Learning, Memory, and Cognition, 12, 336–345.

Bargh

J. A.

Chartrand

T. L.

(2000). The mind in the middle: A practical guide to priming and automaticity research. In Reis

H. T.

Judd

C. M.

(Eds.), Handbook of research methods in social and personality psychology (pp. 253–285). New York, NY: Cambridge University Press.

Berdahl

Moore

(2006). Workplace harassment: Double jeopardy for minority women. Journal of Applied Psychology, 91, 426–436.

Bertrand

Mullainathan

(2004). Are Emily and Greg more employable than Lakisha and Jamal? A field experiment on labor market discrimination. American Economic Review, 94, 991–1013.

Blau

F. D.

Devaro

J. E. D.

(2007). New evidence on gender differences in promotion rates: An empirical analysis of a sample of new hires. Industrial Relations: A Journal of Economy and Society, 46, 511–550.

Buss

D. M.

(1989). Sex differences in human mate preferences: Evolutionary hypotheses tested in 37 cultures. Behavioral and Brain Sciences, 12, 1–49.

Buss

D. M.

(2000). The dangerous passion. New York, NY: Free Press.

Catalyst. (2012). Women CEOs of the Fortune 1000. Retrieved from http://www.catalyst.org/knowledge/women-ceos-fortune-1000

Cuddy

A. J. C.

Glick

Beninger

(2011). The dynamics of warmth and competence judgments, and their outcomes in organizations. Research in Organizational Behavior, 31, 73– 98.

10.

Devine

P. G.

Elliot

A. J.

(1995). Are racial stereotypes really fading? The Princeton trilogy revisited. Personality and Social Psychology Bulletin, 21, 1139–1150.

11.

Eagly

A. H.

Karau

S. J.

(2002). Role congruity theory of prejudice toward female leaders. Psychological Review, 109, 573–598.

12.

Eagly

A. H.

Karau

S. J.

Makhijani

M. G.

(1995). Gender and the effectiveness of leaders: A meta-analysis. Psychological Bulletin, 117, 125–145.

13.

Eagly

A. H.

Kite

M. E.

(1987). Are stereotypes of nationalities applied to both women and men? Journal of Personality and Social Psychology, 53, 451–462.

14.

Gilbert

G. M.

(1951). Stereotype persistence and change among college students. Journal of Personality and Social Psychology, 46, 245–254.

15.

Goff

P. A.

Thomas

M. A.

Jackson

M. C.

(2008). “Ain’t I a woman?”: Towards an intersectional approach to person perception and group-based harms. Sex Roles, 59, 392–403.

16.

Heilman

M. E.

Eagly

A. H.

(2008). Gender stereotypes are alive, well, and busy producing workplace discrimination. Industrial and Organizational Psychology, 1, 393–398.

17.

Iceland

Wilkes

(2006). Does segregation matter? Race, class, and residential segregation. Social Problems, 53, 248–273.

18.

Johnson

K. L.

Freeman

J. B.

Pauker

(2011). Race is gendered: How covarying phenotypes and stereotypes bias sex categorization. Journal of Personality and Social Psychology, 102, 116–131.

19.

Karlins

Coffman

T. L.

Walters

(1969). On the fading of social stereotypes: Studies in three generations of college students. Journal of Personality and Social Psychology, 13, 1–16.

20.

Katz

Braly

(1933). Racial stereotypes in one hundred college students. Journal of Abnormal and Social Psychology, 28, 280–290.

21.

Kim

P. S.

Lewis

G. B.

(1994). Asian Americans in the public service: Success, diversity, and discrimination. Public Administration Review, 54, 285–290.

22.

Klein

S. P.

Petersilia

Turner

(1990). Race and imprisonment decisions in California. Science, 247, 812–816.

23.

Kulik

C. T.

Roberson

Perry

E. L.

(2007). The multiple- category problem: Category activation and inhibition in the hiring process. Academy of Management Review, 32, 529– 548.

24.

Landy

F. J.

(2008). Stereotypes, bias, and personnel decisions: Strange and stranger. Industrial and Organizational Psychology, 1, 379–392.

25.

Maddux

W. W.

Galinsky

A. D.

Cuddy

A. J. C.

Polifroni

(2008). When being a model minority is good . . . and bad: Realistic threat explains negativity toward Asian Americans. Personality and Social Psychology Bulletin, 34, 74–89.

26.

National Collegiate Athletic Association. (2010). Student- athlete ethnicity: 1999-2000 – 2009-10. Retrieved from http://www.ncaapublications.com/p-4214-student-athlete-ethnicity-2009-10-ncaa-student-athlete-ethnicity-report.aspx

27.

Nelson

N. L.

Probst

T. M.

(2004). Multiple minority individuals: Multiplying the risk of workplace harassment and discrimination. In Chin

(Ed.), The psychology of prejudice and discrimination: Ethnicity and multiracial identity (Vol. 2, pp. 193–227). Westport, CT: Praeger.

28.

Ocampo

K. A.

Bernal

M. E.

Knight

G. P.

(1993). Gender, race, and ethnicity: The sequencing of social categories. In Bernal

Knight

(Eds.), Ethnic identity (pp. 31–46). Albany, NY: SUNY Press.

29.

Oyserman

(1993). The lens of personhood: Viewing the self, others and conflict in a multicultural society. Journal of Personality and Social Psychology, 65, 993–1009.

30.

Pettit

Western

(2004). Mass imprisonment and the life course: Race and class inequality in U.S. incarceration. American Sociological Review, 69, 151–169.

31.

Rudman

L. A.

Glick

(1999). Feminized management and backlash toward agentic women: The hidden costs to women of a kinder, gentler image of middle managers. Journal of Personality and Social Psychology, 77, 1004– 1010.

32.

Shutts

Banaji

M. R.

Spelke

E. S.

(2010). Social categories guide young children’s preferences for novel objects. Developmental Science, 13, 599–610.

33.

Sidanius

Veniegas

R. C.

(2000). Gender and race discrimination: The interactive nature of disadvantage. In Oskamp

(Ed.), Reducing prejudice and discrimination: The Claremont Symposium on Applied Social Psychology (pp. 47–69). Mahwah, NJ: Erlbaum.

34.

Sommers

S. R.

Ellsworth

P. C.

(2000). Race in the courtroom: Perceptions of guilt and dispositional attributions. Personality and Social Psychology Bulletin, 26, 1367–1379.

35.

U.S. Census Bureau. (2003). Hispanic origin and race of coupled households. Retrieved from www.census.gov/population/www/cen2000/briefs/phc-t19/index.html

36.

U.S. Census Bureau. (2005). Income, poverty, and health insurance coverage in the United States: 2005. Retrieved from http://www.census.gov/prod/2006pubs/p60-231.pdf

37.

Wittenbrink

Judd

C. M.

Park

(1997). Evidence for racial prejudice at the implicit level and its relationship with questionnaire measures. Journal of Personality and Social Psychology, 72, 262–274.