Despite significant gains in understanding people's beliefs about various gender and ethnic groups in the U.S., several important research gaps remain. First, most stereotype research has focused on the beliefs associated with a single social identity such as ethnicity or gender. Studies have ignored the ways in which ethnic stereotypes may differ for men versus women. For example, how similar are stereotypes of Black men and Black women? This singular focus has led to an incomplete description of contemporary cultural stereotypes. Second, previous research has overwhelmingly emphasized personality traits while neglecting other stereotype components. Recent research suggests, however, that behaviors and physical characteristics may be more central to stereotypes than are personality traits (e.g., Miller et al., 2009). This personality-focused analysis of stereotypes misses the complexity of people's social perceptions and mischaracterizes the ways in which stereotypes may affect bias against social groups. Third, although past research has identified many stereotypic attributes, no studies have examined which attributes can reliably distinguish among ethnic groups, among ethnic men and among ethnic women.

To complete this project, I conducted two separate data collections. In the first data collection, a large sample of college student participants provided open-ended descriptions of the cultural stereotypes associated with men and women who are Asian American, Black, Latino, Middle Eastern American and White. In the second data collection, a different sample of college student participants was asked to assess stereotypes associated with three sets of target groups: (1) each of five ethnic groups including Asian American, Black, Latino, Middle Eastern American and White, (2) men from each of the five ethnic groups (e.g., Black men, Middle Eastern men) and (3) women from the five ethnic groups (e.g., Black women, Middle Eastern women). Details about these two data collections are provided in the two articles that follow.

Article 1, "An Intersectional Approach to the Study of Gender and Ethnic Stereotype Content," uses results from the first data collection to systematically compare the content of stereotypes associated with men and women who are Asian American, Black, Latino, Middle Eastern American and White. Grounded in Intersectionality theory, Social Dominance theory and the Black Exceptionalism Hypothesis, I tested and found consistent support for three main predictions. First, analyses confirmed the Ethnicity Hypothesis, demonstrating that stereotypes of ethnic groups (gender unspecified) were generally more similar to stereotypes of the men than the women of that ethnic group. Second, results supported the Gender Hypothesis . Gender stereotypes were more similar to those of White men and White women than to those of ethnic minority men and women. Third, consistent with the Intersectionality Hypothesis, I documented that intersecting gender and ethnic stereotypes contained unique elements that were not simply the result of adding gender stereotypes to ethnic stereotypes.

Article 2 is titled, "Multidimensionality and Diagnosticity of Ethnic and Gender Stereotypes." It reports findings from both of the data collections, which are presented as two studies. The first study uses the qualitative data to systematically chart the various stereotype domains (e.g., personality, physical appearance, and social status) that might comprise stereotypes of ethnic and gender groups. Drawing on multiple perspectives including evolutionary theories (Social Dominance theory; Sidanius & Pratto, 1999), socio-political theories (Black Exceptionalism Hypothesis; Sears, 1988) and social cognitive theories (e.g., Martin & Macrae, 2007), I predicted and found that the stereotypes of ethnic groups, ethnic men and ethnic women reflect many domains other than personality traits. Indeed, stereotypes fell into 17 distinct domains including not only personality but also physical descriptors, social status, occupations and ideology. Additionally, and consistent with social categorization research (e.g., Martin & Macrae, 2007), analysis revealed that certain domains—personality and physical descriptors—figured more prominently in stereotypes than others and appeared differentially accessible to participants. The second study draws on the quantitative data to identify attributes that could reliably distinguish among ethnic groups, among ethnic men and among ethnic women. In more technical terms, analyses determined which attributes were "diagnostic" for particular groups. To analyze the data, I used multidimensional scaling (MDS; Kruskal & Wish, 1978) to depict spatially the clustering of stereotypic attributes associated with each set of target groups. To assess diagnosticity, I used the Simpson's index of diversity (Simpson, 1949) to measure "diversity" in the pattern of responses. As predicted, results showed that visually prominent attributes such as physical descriptors were diagnostic of groups significantly more often than non-visual attributes such as personality traits. Results also demonstrated that both the gender of the target and the gender of the respondent played a role in attribute diagnosticity. These findings are presented and discussed in more detail in Article 2. (Abstract shortened by UMI.)