Does Racial Phenotypicality Bias Apply to Black Women?

Does Racial Phenotypicality Bias Apply to Black Women? Exploring the Intersection of Racial Phenotypes and Gender in Stereotyping of Black Women A dissertation submitted by Kristin Nicole Dukes in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Psychology TUFTS UNIVERSITY May 2012 Committee Keith B. Maddox, PhD, Committee Chair Samuel R. Sommers, PhD Heather L. Urry, PhD Tufts University Phillip Atiba Goff, PhD University of California, Los Angeles © 2012, Kristin Nicole Dukes Abstract A limited amount of social psychological research on racial stereotyping and prejudice of Black Americans have considered the role of gender in these processes. Such is also the case with research on racial phenotypicality bias (Maddox, 2004)—the notion that within racial group variation in features indicative of race can result in increased stereotyping of group members with more of these features relative to group members with fewer of these features. This dissertation addresses a void in racial phenotypicality bias research by assessing stereotypic perceptions of Black women as a function of racial phenotypes (Afrocenrtric features). In line with previous evidence of racial phenotypicality bias toward Black men, I hypothesized that high Afrocentric Black women would be stereotyped to a greater degree than low Afrocentric Black women. In Experiment 1 participants evaluated the likelihood that several traits and behaviors stereotypic of Blacks as a racial group were characteristic of Black female targets varying in Afrocentricity. Analyses revealed relatively similar levels of stereotyping of high and low Afrocentric Black women, providing inconclusive evidence of racial phenotypicality bias toward Black women. Experiments 2 and 3 address the potential insensitivity of the measure used in Experiment 1 to assess stereotypic evaluations of Black women by identifying stereotypes associated with Black women specifically (Experiment 2) and reexaming racially phenotypicality bias toward Black women in light of these stereotypes (Experiment 3). Akin to Experiment 1, results of Experiment 3 ii provided inconsistent evidence of racial phenotypicality bias toward Black women, suggesting that racial phenotypes may not influence perceptions of Black women and Black men in the same manner. Theoretical considerations for the intersectional influence of racial phenotypes and gender on perceptions of Black women are addressed in the general discussion. iii Acknowledgments This dissertation is dedicated to Vina Green Dukes (1816 - ?) Emeley Harper (? - ?) Laura Tatum Walker (1863-1925) Della Hawkins (1890 - 1962) Nora Dukes (1926 - 2008) Florence Crenshaw (? - ?) Quida Mae Cole (1918 - 1967) Amandy Grimes (1848 - ?) Gertrude Houston (1892 - ?) Maud N. Jackson (1891 - ?) Mariah Johnson (1824 - 1919) Ann (Eliza) Johnson (1849 - ?) Sally Washington (1868 - ?) Emma Pinson (1903 - 2001) Vernice V. Butler (1920- ) Patsy E. Grimes (1940- ) Rosalin D. Rogers (1960- ) and all of the other "invisible women" that came before me Because of your sacrifices, I will be "seen" and I will be "heard" iv My Inspiration A continued source of inspiration…I strive to investigate empirically, what Nina Simone expresses with such precision lyrically. Black women are not a monolithic group destined for the shadows of society's interactions Four Women Written by singer, composer, pianist and arranger Nina Simone, 1966 My skin is black, my arms are long My hair is woolly, my back is strong Strong enough to take the pain, inflicted again and again What do they call me? My name is aunt Sarah My name is aunt Sarah, aunt Sarah My skin is yellow, my hair is long Between two worlds I do belong But my father was rich and white He forced my mother late one night And what do they call me? My name is Saffronia, my name is Saffronia My skin is tan, my hair fine My hips invite you, my mouth like wine Whose little girl am I? Anyone who has money to buy What do they call me? My name is Sweet Thing My name is Sweet Thing My skin is brown, my manner is tough I'll kill the first mother I see, my life has been rough I'm awfully bitter these days, because my parents were slaves What do they call me? My name is Peaches v Thank Yous! All thanks to my Heavenly Father for the strength, courage, and guidance to continue when my hope began to flicker and fade. Special Thanks To  Rosalin D. Rogers, my mother  Phillip M. Dukes, my father  Andrea D. Croskey, my sister  Shanigia D. Williams, my ―sister‖ and best friend  Dr. Keith B. Maddox, my adviser, committee chair, mentor, and friend  Drs. Heather Urry, Sam Sommers, and Phil Goff for being supportive and understanding committee members  Tufts University Social Cognition (TUSC) Lab, especially Jennifer Schultz  Ingar Moir, for the day-to-day encouragement during the last steps of this process  Drs. Michelle (Mikki) Hebl, David (Dave) Schneider, and Roland B. Smith, Jr. for starting me on my journey into academia  Karen Newcomb, Toni Royall, Judith Scott Davenport, and Pud Kearns for challenging me to dream big  My graduate school cohort: Elsie, Emily, Lara, Lisa, Martin, Matt, Nicole, Negin, and Nick-- thank you for being competitive while supportive at all of the right times  Dr. Lucille P. Fultz, for the day you told me I was brilliant in front of Fondren Library at Rice University—thank you for saying it because I didn’t know it vi  Cathi Clack, Sharon Bush, Christel Miller, Coryell LaRue, Christina Beatty, Anjeanette Gunter, and Akilah Mance for making sure I earned my B.A.  Dr. Sinaia Nathanson for giving me my first opportunity to teach  Clara Wilkins, Courtney Cogburn, Jessica Williamson, Nenah KowaiBell, Nicole Overstreet, Nicole Walden, Phia Salter, and Valerie Jones Taylor--my sisters in the field  Dr. Lisa Coleman, Yvette Dalton-McCoy, Katrina Moore, and Denise Phillips for becoming my extended family and support system at Tufts  Cynthia Goddard, Lidia Bonaventura, Madeline Amico, and Charlene Carle for all of your administrative support, patience, and kindness  Barbara Gentile, Ellen Birchander, Erin O‘Connor, Geoff Turner, Greg Feldman, Janie Ward, John Reeder, Lowry Pei, Rachael Galli, Renee White, Sarah Martin, and the rest of my Simmons College family for being so understanding during my transition  Drs. J. Nicole Shelton, Jacqueline Mattis, Jennifer Eberhardt, Jennifer Richeson , Stacey Sinclair , Tabbye Chavous, and Valerie Purdie-Vaughns for being excellent role models  And all of my other family members and friends that kept me in their thoughts and prayers vii Table of Contents Abstract ii Acknowledgements iv Introduction 3 Experiment 1: Racial Phenotypicality Bias in Stereotyping of Black Women Experiment 2: Examining Stereotypes of Black Women 9 16 Experiment 3: Revisiting Racial Phenotypicality Bias in Stereotyping of Black Women 24 General Discussion 35 Figures 47 Tables 51 Appendices 55 References 63 Does Racial Phenotypicality Bias Apply to Black Women? Exploring the Intersection of Racial Phenotypes and Gender in Stereotyping of Black Women Kristin Nicole Dukes Tufts University 2 Does Racial Phenotypicality Bias Apply to Black Women? Exploring the Intersection of Racial Phenotypes and Gender in Stereotyping of Black Women "I am a man of substance, of flesh and bone...I am invisible, understand, simply because people refuse to see me...They see only my surroundings, themselves, or figments of their imagination--indeed, everything and anything except me." ~ Ralph Ellison, Invisible Man, 1952 The quote above reflects feelings of social invisibility experienced by the purposely unnamed protagonist of Ralph Ellison‘s Invisible Man. Set during the Jim Crow era, the novel follows the life of a Black American man navigating a racially-divided society that considers him less than human. The novel illustrates the complex social position of Black Americans during the time period through the eyes of a Black man. One could generalize the experience of the ―invisible man‖ to that of all Blacks during this time period; however, with such a generalization, one would risk losing sight of the diverse experiences of Black Americans. For instance, the experiences of Black women during this time period differed markedly from that of Black men due to Black women‘s unique social position as both Black and female in an era of substantial oppression for both groups (Giddings, 1996; Harris, 2009; Hull, Bell-Scott, & Smith, 1993) One could also argue that the above quote describes the state of social psychological research examining perceptions of Black women. Social 3 psychological research on perceptions of Black Americans as a racial group has both implicitly and explicitly focused on Black men, rendering Black women ―invisible‖ (Purdie-Vaughns & Eibach, 2008; Sesko & Biernat, 2010). Although a considerable amount of research has examined stereotyping and prejudice toward Black Americans (e.g., Devine, 1989; Devine & Baker, 1991; Devine & Elliot, 1995; Dovidio, Evans, & Tyler, 1986; Dovidio, Kawakami, Johnson, Johnson, & Howard, 1997; Gaertner & McLaughlin, 1983; Gilbert, 1951; Karlins, Coffman, & Walters, 1969; Katz & Braly, 1933; Kawakami, Dion, & Dovidio, 1998), the majority of these investigations have not explored whether perceptions of Black women and Black men differ. As it may be problematic to generalize the experience of the ―invisible man‖ to that of Black women, assuming that Black women and Black men are perceived identically is an inaccurate approach to fully understanding how Black women are perceived. I argue that several previous social psychological investigations examining stereotyping and prejudice of Blacks as a racial group share this flaw. This research has disregarded or muted the role of gender in stereotyping and prejudice resulting in an impoverished understanding of how Black women are viewed. In the limited number of studies that have considered both race and gender in the perception of Blacks, divergent findings emerge for Black women and Black men. But for a few notable exceptions, many of these studies lack a theoretical framework. As a result, researchers are left with an incomplete picture of how race and gender may interact to affect judgments of Black Americans. 4 Overview The current research addresses a void of in racial stereotyping and prejudice research that considers the role of gender in these processes by explicitly assessing perceptions of Black women as a function of racial phenotypes (Afrocentric features) and the nature of racial phenotypicality bias (Maddox, 2004) toward Black women, an emerging area of social psychological research discussed in further detail in the next section. The majority of existing empirical research on racial phenotypically bias toward Black Americans has focused on impressions of Black men. I argue that similar to other stereotyping and prejudice research failing to account for the importance of gender, neglecting the influence of gender on racial phenotypicality bias results in a limited understanding of how racial phenotypes are incorporated into impressions of Black women. Racial Phenotypicality Bias Stereotype application has traditionally been theorized as a categorization driven process: first one sorts an individual into a specific category and then infers what characteristics the individual may have based on the stereotypes associated with the particular group in which the individual in question is placed (Bodenhausen & Macrae, 1998; Brewer, 1988; Fiske & Neuberg, 1990). With regard to race, it has been assumed that once individuals are categorized into a given category, they all receive the stereotypic evaluation of the group equally; that is, stereotypes are applied to the same degree to all members of a social group once categorization has occurred. For Black Americans, this viewpoint implies 5 that all Black Americans are stereotyped to a similar degree regardless of individual variation in skin tone or other features typically perceived as indicative of racial category membership. Supporting this categorization model, Secord, Bevan, and Katz (1956) found that Blacks with more European facial features and of lighter skin tone were stereotyped to the same degree as Blacks with fewer European facial features and of darker skin tone. However, recent research suggests within group variations in physical features indicative of race can have implications for stereotypic inferences, with individuals with fewer physical features typical of a racial group stereotyped to a lesser degree than individuals with more of these features (Blair, Judd, Sadler, & Jenkins, 2002; Maddox & Gray, 2002). The term racial phenotypicality bias, coined by Maddox (2004), has been used to describe this phenomenon of differential stereotyping, prejudice, and discrimination based on race-related physical feature variation within a racial category. Racial phenotypicality bias is apparent in evaluations of Black Americans. In an investigation of possible differences in the stereotyping of Blacks as a function of skin tone, Maddox and Gray (2002) found that negative cultural stereotypes of Blacks were more closely associated with dark-skinned Blacks than light-skinned Blacks suggesting that Blacks may be subcategorized as a function of skin tone. Racial phenotypicality bias is not limited to skin tone for Black Americans however. Within racial category variation in the entire collection of physical features indicative of Black racial category membership has implications for stereotyping and prejudice as well. These features, perceived to be typical of 6 people of Black African decent (e.g., dark skin, wide nose, full lips), are sometimes referred to as Afrocentric features and are often contrasted with Eurocentric features, defined as features that are perceived to be typical of people of European decent (e.g., pale or light skin, narrow nose, thin lips). Studies have shown that individuals with more Afrocentric features are stereotyped to a greater degree than individuals with fewer Afrocentric features (Blair, et al., 2002; Blair, Judd, & Fallman, 2004; Blair, Chapleau, & Judd, 2005; Blair, 2006). Further, Blacks with more Afrocentric features facilitate more automatic negative evaluations than Blacks with fewer Afrocentric features (Livingston & Brewer, 2002) and individuals have limited control of their use of these feature when forming impressions (Blair, Judd, & Fallman, 2004; Blair, 2006). The implications of racial phenotypicality bias for Black Americans are far reaching, with consequences in socioecomomic, health, and even criminal justice domains. In terms of socioeconomic status, some investigations have shown lower earned income and lower educational attainment among darkerskinned Black in comparison to lighter-skinned Blacks (Goldsmith, Hamilton, & Darity, 2007). In studies examining the influence of within group variance in physical markers of racial category membership on the health of Black Americans, higher blood pressure has been observed among darker complected Blacks in comparison to lighter-complected Blacks (Gleiberman, Harburg, & Cooper, 1995; Klag, et al., 1991; Sweet, McDade, Kiefe, & Lui, 2007). And within the realm of criminal justice, research utilizing actual criminal cases has revealed that Black defendants with more Afrocentric features received harsher 7 sentences than those with fewer Afrocentric features (Blair, Judd, & Chapleau, 2004), and that, in cases involving a White victim, Black defendants with more Afrocentric features were more likely to sentenced to death than Black defendants with fewer Afrocentric features (Eberhardt, Davies, Purdie-Vaughns, & Johnson, 2006). In summary, research on racial phenotypicality bias suggests that within racial category variation in physical features indicative of race, once considered inconsequential in racial category activation and stereotype application, does in fact influence these processes. Although it provides a more nuanced understanding of stereotyping and prejudice, existing research on racial phenotypicality bias has its limitations. Similar to other social psychological research on perceptions of Blacks as a racial group, to my knowledge, research in this area has almost exclusively examined Black males as targets of racial phenotypicality bias leaving us to speculate how Afrocentric features influence perceivers‘ impressions of Black women. Consequently, Experiment 1 expands the scope of racial phenotypicality bias research to examine the potential role of Afrocentric features in stereotypic evaluations of Black women. 8 Experiment 1: Racial Phenotypicality Bias in Stereotyping of Black Women Experiment 1 explored the influence of Afrocentric physical appearance on perceptions of Black women. Based on existing literature on racial phenotypicality bias, I hypothesized that Black women with more Afrocentric features would be stereotyped to a greater degree than their less Afrocentric appearing counterparts. Method Participants Forty-eight Tufts University undergraduate students (20 female, 12 male, 16 unknown; 19 White, 4 East Asian, 3 Multiracial, 3 South Asian, 1 African American, 1 Middle Eastern, 1 declined to answer and 16 unknown)1 participated in an experiment described as a study of individuals‘ ability to predict character and personality traits based on minimal information. Participants received partial course credit in exchange for their participation. Design Participants were shown photographs of either high Afrocentric or low Afrocentric Black women and asked to rate the likelihood that several traits or behaviors stereotypic of Blacks as a racial group were associated with each woman. Materials Target photographs. Pretest ratings were obtained for several photographs of women from various racial and ethnic backgrounds. A separate sample of 1 Demographic information for participants was not obtained during the experimental session due to experimenter error. The demographic information presented was ascertained from the Tufts University Psychology Department participant pool pretesting system. 9 participants assessed hair texture, nose width, lip fullness, skin tone, global assessments of Afrocentric and Eurocentric appearance, age, and facial affect using 1-7 scales (Appendix A). Afrocentricity was defined as the degree to which the individual pictured had features uniquely characteristic of people of African descent. Eurocentricity was defined as the degree to which the individual pictured had features uniquely characteristic of people of European descent. Targets photographs were selected from the highest and lowest thirds of mean Afrocentricity ratings. Photographs were then matched for age and facial affect across conditions and facial features within condition. Six Black female target photographs, three low Afrocentric and three high Afrocentric, were selected using these criteria (Appendix B). The three target photographs in the low Afrocentric condition had an average Afrocentricity rating of 3.67 (range: 3.0 – 4.0). The three target photographs in the high Afrocentric condition had an average Afrocentricity rating of 5.67 (range: 5.5 - 6). Eight filler photographs were also used: five White women, two Hispanic women, and one Asian woman. Stereotype Assessment. Participants were given a questionnaire composed of 34 actions, activities, and interests related to the stereotype of Black Americans as a racial group (Appendix C). The items were based on traits associated with the Black racial group stereotype as identified in previous research: aggressive, athletic, criminal, irresponsible/incompetent, lazy/unmotivated, ostentatious, poor, product of a broken home, religious, rhythmic/musical, sexually promiscuous, uneducated, and unintelligent (Blair, Judd, Sadler, & Jenkins, 2004; Devine, 1989; Maddox & Gray, 2002). Participants were asked to judge the likelihood 10 that each action, activity, or interest listed was a personality trait or character trait of the person pictured using a 5-point scale This assessment was the primary dependent measure. Individual differences scales. For exploratory purposes, participants were asked to complete several scales designed to measure individuals‘ differences in person perception, racial prejudice, tendency to notice and use physical features, social desirability, and conceptions of race as a social or biological construct. These scales included the following: Implicit Person Theory Scale (Levy & Dweck, 1998), Modern Racism Scale (McConahay, 1986), Perceptual Reliance Index (Livingston, 2001), Social Desirability Scale (Crowne & Marlowe, 1960), and Race Conceptions Scale (Williams & Eberhardt, 2002). Procedure An experimenter greeted participants in a laboratory room in the Tufts University Psychology building. The experiment was described as involving the prediction of character and personality traits based on minimal information. The experimenter read instructions from a script in order to maintain consistency across experimental sessions. After providing informed consent, participants received a study packet containing general instructions for the study and copies of the experimental materials, which included ten photographs (two targets and eight fillers) and stereotype assessments (one for each photograph presented) as well as the scales assessing various individuals‘ differences in person perception, prejudice, and cognitive accessibility of race. The experimenter then went over the general instructions for study with the participants stressing that he/she rate 11 the photographs in the order they were presented. Participants were instructed to rate the photographs first using the stereotype assessment and then to complete the remaining questionnaires. Two target photographs were presented in random order with the constraint that a target photograph did not appear first, last, or in succession. The eight filler photographs were presented in the same order in all conditions. Photographs were presented on 3x5 index cards separate from the stereotype assessment. The experimenter reassured participants that their responses would remain anonymous. After receiving these instructions, participants were directed to individual cubicles to complete the packet. Once finished, participants were debriefed and thanked for their participation. Results Raw scores from the stereotype assessment were transformed into an index score ranging from 1 to 5 with higher scores indicating greater stereotyping. Analyses focused on potential differences in stereotyping of Black women as a function of Afrocentricity. Employing an independent samples t-test, degree of Afrocentricity had a marginally significant influence on stereotyping of Black women. High Afrocentric (Mhigh = 2.86; SDhigh = .34) were stereotyped to a marginally greater degree than low Afrocentric Black women (Mlow = 2.68, SDlow = .33), t (46) = -1.86, p = .07, d = .54, r =.26 (Figure 1). None of the individual difference measures qualified this effect. 12 Discussion In light of previous research on racial phenotypicality bias, the findings of Experiment 1 suggest that Afrocentric features may not influence stereotypic evaluations of Black women in that same manner as they do stereotypic evaluations of Black men. Yet, what could account for this apparent gender asymmetry in the use of Afrocentric features in evaluations of Black women in comparison to Black men? Initial considerations began on a methodological level, examining the experimental materials used in the study. Target photographs used in the study were pretested, removing the validity of the Afrocentricity manipulation as a primary concern. Focus then turned to the primary dependent measure of the study. The stereotype assessment was created using Black racial group stereotypes identified in previous social psychological research—research which is limited in focus on stereotypes of Black women specifically. Taken with evidence suggesting that racial/ethnic stereotypes are more representative of men than women of a racial/ethnic group (Eagly & Kite 1987), the stereotypes comprising the measure in Experiment 1 may be more applicable to Black men than Black women. This could leave the instrument used a relatively insensitive measure of stereotypes associated with Black women, and consequently, racial phenotypicality bias toward Black women. The next section provides further support for discrepancies in stereotyping of Black women in comparison to Black men, in particular, in content of stereotypes held of each group. I then discuss two additional experiments 13 designed to address the potential methodological shortcomings identified in Experiment 1. Finally, the general discussion section expands upon theoretical considerations for the influence and use of racial phenotypes in evaluations of Black women. Why Gender Matters: Divergent Perceptions of Black Women and Black Men Although few in number, racial stereotyping and prejudice investigations that account for the influence of gender in these processes indicate that failing to consider the role of gender in racial group representations and perceptions may be problematic. For instance, an Eagly and Kite (1987) study examining if national group stereotypes apply to both men and women of a national group found differences in overlap between stereotypes of national groups and stereotypes of men and women of these national groups. Participants indicated if stereotypes of 28 national groups applied to men of each national group, women of each national group, and the national group as a whole. Eagly and Kite hypothesized that the social position of men and women within a society would impact what national group stereotypes were ascribed to them. Specifically, they hypothesized that men‘s social prominence and high social status renders them more visible than women, and subsequently, more representative of their national group than women. They argued that this greater perceived representativeness would result in greater overlap between stereotypes of men and their national group stereotypes than that between stereotypes of women and the national group. Consistent with predictions, greater similarity existed between stereotypes of men and stereotypes of their national group than between stereotypes of women and the national 14 group. Further, stereotypes unique to women of these groups relative to men were uncovered. Similarly, stereotypes of Black women vary from those of Black men to a certain degree. In a study by Neimann and colleagues (1994) on the content of racial/ethnic stereotypes across gender, participants reported cultural stereotypes associated with eight ethnic/gender groups: African American males, African American females, Anglo-American males, Anglo-American females, Asian American males, Asian American females, Mexican American males, and Mexican American females. Although similar terms were reported for both Black women and Black men, some distinctions emerged. For example the terms speak loudly, antagonistic, athletic, and dark skin were reported for both Black men and women. However, the terms muscular appearance and criminal activities were primarily associated with Black men, while sociable/socially active and unmannerly were associated with Black women. Gender differences in racial stereotype content for Black women and Black men also emerged in Maddox and Gray‘s (2002) investigation of the role of skin tone in perceptions of Black Americans. Although differences in perceptions of Blacks as a function of skin tone (i.e., dark- versus light-skinned Blacks) was the primary focus of the study, their results suggest that stereotypes of Blacks as a racial group are ascribed to Black women and Black men to different degrees. For instance, a greater percentage of participants reported athletic traits for Black men than for Black women. The same pattern emerged for the percentage of participants reporting criminal traits. By contrast, a greater percentage of 15 participants attributed bad attitude and self-assured traits to Black women than Black men. In summary, these studies support the possibility that unique stereotypes exist for Black women in comparison to those stereotypes about Black men. Experiment 2: Examining Stereotypes of Black Women Given evidence suggesting that stereotypes of Blacks as racial group identified in past social psychological research may better reflect societal perceptions of Black men and some degree of divergence in stereotypes of Black women and Black men, Experiment 2 sought to identify stereotypes explicitly associated with Black women and Black men. I predicted that traits reported for Black men would share greater overlap with Black racial group stereotypes identified in previous research than those reported for Black women. Further, I predicted that a number of stereotypes unique to Black women would emerge, distinct from those reported for Black men. Method Participants One hundred three Tufts University undergraduate students (66 female, 37 male; 66 White, 21 Asian, 8 Hispanic, 3 Black, 2 multiracial, 3 other/not listed) participated in an experiment described as a study of their knowledge about social groups. Participants received partial course credit in exchange for their participation. 16 Design Using a within-subjects design, participants were asked to report ―cultural stereotypes‖ associated with eight groups: Asian-American men and women, Black men and women, Hispanic men and women, and White men and women. Materials Stereotype Assessment. The Knowledge of Social Group Questionnaire was designed to ascertain stereotypes associated with racial/ethnic groups across gender (Appendix D). Each page was labeled with a specific social group: African American (Black) females, African American (Black) males, Asian American females, Asian American males, Caucasian (White) females, Caucasian (White) males, Latina/Hispanic females, Latino/Hispanic males. To encourage open, honest, and in depth responses, questionnaires described and instructed participants to list both cultural stereotypes and indicate whether each stereotype was consistent with their personal beliefs (see Devine, 1989). Cultural stereotypes were defined as general impressions of how a particular group of people is portrayed on a societal level while personal beliefs were defined as impressions of a group that are personally endorsed. An example differentiating cultural stereotypes from personal beliefs was provided. For each response listed, participants were asked to indicate if the response generated was consistent or inconsistent with their personal beliefs or if they were unsure of their position concerning the response. This option allowed participants to distance themselves from potentially volatile responses. While the question of personal endorsement is 17 interesting, here the focus was to encourage participant to accurately report their knowledge of the cultural stereotypes associated with these groups. Individual difference measures. For exploratory purposes, participants were asked to complete the Modern Racism Scale and /or Symbolic Racism Scale (Henry & Sears, 2002; Sears & Henry, 2003) and to provide demographic information (age, gender, race/ethnicity, year in school, and major). Procedure An experimenter greeted participants in a laboratory room in the Tufts University Psychology building. The experiment was described as an exploration of participants‘ knowledge of a variety of social groups. In order to maintain consistency across experimental sessions, the experimenter read instructions from a script. After providing informed consent, participants received a study packet containing general instructions for the study and copies of the experimental materials, including the Knowledge of Social Groups questionnaire reflecting the eight racial/ethnic and gender combinations (randomized) described above, followed by the individual differences measures. After completing the study packet, participants provided demographic information, were debriefed, and dismissed with thanks for their participation. Results Stereotype Assessment Coding Analyses focused on responses for Black women and Black men. These were coded by two panels, each composed of three raters. The coding scheme consisted of eleven stereotypes associated with the Black racial stereotype in 18 previous social psychological research (see Devine, 1989; Maddox and Gray, 2002) and used to create the Stereotype assessment used in Experiment 1. The stereotype categories included: athletic, criminal, dirty/smelly, inferior, lazy, ostentatious, poor, rhythmic, sexually aggressive, tough/aggressive, and undereducated/unintelligent. Responses that did not fit into these stereotype categories were coded as ―other‖. Raters coded responses individually and then conferred with the other two raters on their particular panel to come to a consensus on the coding of each response. Examples of responses coded into each category are presented in Table 1. Analyses Analyses primarily examined potential differences in the types of traits reported to describe Black women in comparison to those reported to describe Black men using stereotype categories identified in previous social psychological research as a frame of reference. I hypothesized that participants would be less likely to report traits falling into stereotype categories identified in previous research for Black women than for Black men and that participants would report a number of unique stereotypes for Black women, distinct from those reported for Black men. On average, participants reported more traits for Black men (Mmen = 6.33, SDmen = 2.35) than Black women (Mwomen = 5.61, SDwomen = 2.97), t (102) = 3.21, p = .002, d = .27, r = .13. The proportion of reported traits falling into to each stereotype category was examined. Potential frequency differences in the types of terms used to describe Black women in comparison to Black men were evaluated 19 using McNemar‘s test with Yates‘ continuity correction, a form of the chi-squared test for within-subjects designs (Lawal, 2003; Simonoff, 2003). Table 2 presents this information in terms of the percentage of traits reported falling into each stereotype category for all traits reported for Black women and all traits reported for Black men as well as the percentage of traits reported for Black women and Black men in each of these categories for traits reported overall. Frequency differences in the types of terms reported for targets were observed for all of the stereotype categories: athletic (χ2athletic (1, N = 1,249) = 326.57, p < .001, φ = .19), criminal (χ2criminal (1, N = 1,249) = 306.01, p < .001, φ = .49), dirty/smelly (χ2dirty (1, N = 1,249) = 569.04, p < .001, φ = .67), inferior(χ2inferior (1, N = 1,249) = 365.08, p < .001, φ = .54), lazy (χ2lazy (1, N = 1,249) = 486.03, p < .001, φ = .62), ostentatious (χ2ostentatious(1, N = 1249) = 414.21, p < .001, φ = .58), poor (χ2poor (1, N = 1,249) = 391.02, p < .001, φ = .56), rhythmic (χ2rhythmic (1, N = 1,249) = 411.10, p < .001, φ = .57), sexually aggressive (χ2sexuallyaggressive (1, N = 1,249) = 506.77, p < .001, φ = .64), tough/aggressive (χ2tough (1, N = 1,249) = 318.05, p < .001, φ = .50), and uneducated (χ2uneducated (1, N = 1249) = 361.27, p < .001, φ = .29). Differences were also observed in the number of traits reported that did not fall into any of these categories, (χ2other (1, N = 1,249) = 67.25, p < .001, φ = .23). Although statistically significant, the direction and magnitude of the effects described above could not be determined with a chi-squared test alone. Consequently, odd ratios were calculated to determine directionality and better understand the rates at which participants described Black women and Black men using terms falling into each stereotype category relative to the total number of 20 traits reported for Black women or Black men respectively (Table 3). Participants were more likely to describe Black men than Black women using athletic traits (ORathletic = 5.18:1, p <.001) and criminal traits (ORcriminal = 14.23:1, p<.001) as well as marginally more likely to do so for lazy traits (ORlazy = 1.68:1, p <.08). Conversely, participants were more likely to describe Black women than Black men using ostentatious traits (ORostentatious = 2.72:1, p <.001) and sexually aggressive traits (ORsexuallyaggressive = 4.93:1, p <.001) as well as marginally more likely to do so for rhythmic traits (ORrthymic = 1.36:1, p <.10). Participants were also more likely to describe Black women than Black men using traits that did not fall into any of the stereotype categories (ORother = 1.73:1, p <.001). The frequency of participants using each stereotype category listed to describe Black women and Black men was analyzed using McNemar‘s test with Yates‘ continuity correction. Table 4 presents this information in terms of the percentage of participants reporting at least one trait falling into each stereotype category for Black women and Black men. More participants reported at least one trait falling into the following categories for Black men than Black women: athletic (χ2athletic (1, N = 206) = 19.76, p < .001, φ = .30; ORathletic = 3.05:1, p =.001), criminal (χ2criminal (1, N = 206) = 5.37, p < .05, φ = .16; ORcriminal = 24.46:1, p <.001), and lazy (χ2lazy(1, N = 206) = 39.43, p < .001, and φ = .44; ORlazy = 3.35:1, p = .002). And although McNemar‘s test statistic did not reach statistical significance for an overall difference, more participants appeared to report at least one tough/aggressive traits (ORtough = 2.52:1, p<.001) and 21 uneducated traits (ORuneducated = 1.54:1, p = .08) for Black men relative to Black women. The converse was relationship was true for the stereotype categories ostentatious (χ2ostentatious (1, N = 206) = 8.00, p < .01, φ = .19; ORostentatious = 3.31:1, p <.001) and sexually aggressive (χ2sexuallyaggressive (1, N = 206) = 50.83, p <.001, φ = .50; ORsexuallyaggresive = 3.14:1, p = .003), with more participants reporting at least one trait falling into these categories for Black women than Black men. More participants also reported traits that did not fit into any of the stereotype categories used as frame of reference for Black women than Black men (χ2other (1, N = 206) = 36.69, p < .001, φ = .42; ORother = 2.53:1, p = .006). Discussion The goal of Experiment 2 was to identify stereotypes associated with Black women specifically. In line with evidence suggesting divergence in resemblance of overall racial stereotypes with men and women of that racial group, I hypothesized that stereotypes reported for Black men would have greater overlap with stereotypes identified for Blacks as a racial group in previous research in comparison to those stereotypes reported for Black women. I also hypothesized that participants would report traits uniquely associated with Black women. Consistent with predictions, the findings of Experiment 2 suggest that stereotypes reported for Black men tend to resemble overall Black racial group stereotypes identified in previous research to a greater degree than those reported for Black women. This pattern emerged for five of the eleven stereotype categories examined: athletic, criminal, lazy, poor, tough/aggressive, and 22 undereducated/unintelligent. Further supporting hypotheses, a greater proportion of participants reported stereotypes for Black women falling outside any of the previously identified stereotype categories than Black men, again suggesting that these stereotype categories poorly correspond with representations and perceptions of Black women. Conversely, the opposite trend emerged for the stereotype categories ostentatious and sexually aggressive with participants‘ responses for Black women overlapping with overall Black racial group stereotypes than responses for Black men. However, these patterns may be linked the high frequency of a specific of responses falling into these categories. For instance, the stereotype loud, categorized as ostentatious by coders, represented 54% of the traits coded as ostentatious for Black women and 37% of all reported traits coded as ostentatious. Similarly, the stereotypes have children at a young age, single mother¸ and have sex at a young age, categorized as sexually aggressive by coders, represented 43% of traits coded at sexually aggressive for Black women and 34% of all reported traits coded as sexually aggressive. Further, consistent with predictions, some traits emerged as unique to Black women. One of the most notable of these were the stereotype [has an] attitude and related responses (e.g., have lots of attitude; are highly defensive about their beliefs, sassy/lots of attitude, feisty, highly opinionate/forceful). 23 Experiment 3: Revisiting Racial Phenotypicality Bias in Stereotyping of Black Women Experiment 3 revisits the primary research question of this dissertation: does racial phenotypicality bias apply to Black women? In light of evidence garnered in Experiment 2 that some stereotypes of Blacks of a racial group are applied to Black women and Black men at different rates and sometimes in a diverging manner, the question was approached in two ways: 1) evaluations made with the Stereotypic Evaluations of Black Assessment used in Experiment 1 were reanalyzed on a trait and item level to focus on stereotypes most likely to be representative of participants‘ views of Black women, and 2) an additional sample of participants made evaluations of high and low Afrocentric Black women using an updated version of the Stereotypic Evaluations of Black Assessment composed of new items specific to the stereotypes of Black women revealed in Experiment 2. Experiment 3a: Trait and Item-level Analysis of the Stereotypes of Black Assessment Using the evaluations made with the stereotype assessment used in Experiment 1, analyses examined potential instances of racial phenotypicality bias on a trait and singular item level. Consistent with previous research investigating racial phenotypicality bias, I hypothesized that high Afrocentric Black women would be stereotyped to a greater degree than their low Afrocentric counterparts along stereotypic traits identified to be more associated with Black women than 24 Black men. Specifically, high Afrocentric Black women would be rated more ―sexually aggressive‖ and ―ostentatious‖ than low Afrocentric Black women. Results Raw scores from the Stereotypic Evaluations of Black Assessment were transformed into an index score ranging from 1 to 5 with higher scores indicating greater stereotyping. Analyses focused on potential differences in stereotyping of Black women as a function of Afrocentricity of a stereotypic trait and singular item level. Contrary to predictions, participants viewed high (Mhigh = 2.36, SDhigh = .62) and low Afrocentric targets (Mlow =2.25, SDlow = .44) as similarly sexually aggressive, t (46) = -.68, ns (Figure 2). The stereotype ―sexually aggressive‖ was qualified by three statements: two capturing beliefs about childbearing (―Believes it important to wait until marriage to have children‖ and ―Has had children with more than one man‖) and one capturing beliefs about dating and sexual activity (―Described as a ‗player‘ by her friends‖). No statistically significant differences emerged for rating of high and low Afrocentric women for any of these items individually. Similarly, contrary to predictions, difference did not emerge in evaluations of high (Mhigh = 2.80, SDhigh = .60) and low Afrocentric (Mlow = 2.87, SDlow = .34) for the stereotype ostentatious, t (46) = .49, ns (Figure 2). This stereotype was reflected by two items, each designed to capture beliefs about flashy appearance and consumerism: ―Is not interested in material things‖ and ―Drives a car with expensive tires, rims, and sound system‖. No statistically significant differences 25 emerged for rating of high and low Afrocentric women for either item individually. However, analyses did reveal significant differences for the traits ―poor‖, discussed here in terms of socioeconomic status. High Afrocentric Black women (Mhigh = 3.57, SDhigh = .53) were perceived as being of a lower socioeconomic status than their low Afrocentric counterparts (Mlow = 3.05, SDlow = .57), t (46) = 3.31, p =.002, d = .67, r = .44 (Figure 2). An item level analysis of the stereotype ―poor‖ revealed significant differences in perceptions of high and low Afrocentric Black women as well. The stereotype ―poor‖ or lower socioeconomic status was qualified by statements regarding employment status and aspirations (―Has been unemployed for the past six months and struggling to find employment‖ and ―Aspires to be an investment banker like her father‖) racial composition of target‘s neighborhood (―Lives in a neighborhood comprised mostly of minorities‖ and ―Grew up and continues to live in an upscale, suburban neighborhood‖) and cultural/musical preferences (―Has a season subscription to the Boston Symphony‖). While there were no differences in perceived unemployment for high (Mhigh = 2.62, SDhigh = .53) and low Afrocentric women (Mlow = 2.54, SDlow = .78), t (46) = -.40, ns, high Afrocentric Black women (Mhigh = 3.86, SDhigh = .97) were rated as less likely to have aspirations of becoming an investment banker than low Afrocentric Black women (Mlow = 3.41, SDlow = .67) t (46) = -1.84, p<.07, d =.54, r = .26. High Afrocentric women (Mhigh = 3.60, SDhigh = .78) were rated as more likely to live in a neighborhood comprised mostly of minorities than low Afrocentric Black women (Mlow = 2.98, SDlow 26 =.80), t (46) = -2.72, p = .009, d = .77, r = .36, and less (Mhigh = 3.88, SDhigh = .83) likely to live in an upscale, suburban neighborhood than low Afrocentric Black women (Mlow = 3.00, SDlow = .71), t (46) = -3.93, p <.001, d = 1.14 r = .50. Finally, high Afrocentric Black women (Mhigh = 3.90, SDhigh = .76) were rated likely to have a season subscription to the Boston symphony than their low Afrocentric counterparts (Mlow = 3.30, SDlow = .81), t (46) = -2.63, p <.01, d = .76, r = .36. Analyses also revealed a pattern consistent with racial phenotypicality bias toward Black women for the stereotype ―product of a broken home‖, discussed here in terms of the marital status of the target‘s parents and being raised by extended family. High Afrocentric Black women (Mhigh = 3.07, SDhigh = .41) were also rated as more likely to be a ―product of a home‖ than low Afrocentric Black women (Mlow = 2.70, SDlow = .45), t (46) = -3.03, p = .004, d = .84, r = .39 (Figure 3). An item level analysis for the stereotype ―product of a broken home‖ also revealed differences in perceptions of high and low Afrocentric Black women. This stereotype was qualified by statements regarding the marital status of the target‘s parents (―Plans to get married soon and hopes to have a lasting marriage like her parents‖) and being raised by extended family members (―Was raised by grandparents and other extended family members‖). While there were no differences in perceptions of high (Mhigh = 3.14, SDhigh = .68) and low Afrocentric Black women‘s (Mlow = 2.80, SDlow = .75) parent‘s marital status, t (46) = -1.62, ns, high Afrocentric Black women (Mhigh = 3.00, SDhigh = .76) were evaluated as more likely to have been raised by their grandparents or other 27 extended family members than low Afrocentric Black women (Mlow = 2.59, SDlow = .65), t (46) = -2.01, p = .05, d = .58, r = .28. Experiment 3b: Evaluations of High and Low Afrocentric Black Women Using a Stereotyping Assessment Specific to Black Women Employing the same paradigm and stimuli as Experiment 1, participants evaluated high and low Afrocentric Black women using a revised stereotyping assessment measure composed of stereotypes uniquely associated with Black women as identified in Experiment 2. Using what should be a more sensitive measure of stereotypic evaluations of Black women, participants were asked to rate the likelihood that several stereotypic traits or behaviors were associated with high or low Afrocentric Black women. With this more sensitive measure, I hypothesized that Black women with more Afrocentric features would be stereotyped to a greater degree than their less Afrocentric appearing counterparts. Method Participants Twenty-eight Tufts University undergraduate students (15 male, 13 female; 19 White, 5 Asian, 3 Hispanic, and 1 other) participated in an experiment described as a study of individuals‘ ability to predict of character and personality traits based on minimal information. Participants received partial course credit in exchange for their participation. 28 Design, Stimuli, and Procedure Similar to Experiment 1, participants were shown photographs of either high Afrocentric or low Afrocentric Black women and asked to rate the likelihood that several traits or behaviors were associated with each woman. Materials Stereotypes of Black Women Assessment. An updated version of the Stereotypes of Blacks Assessment used in Experiment 1 was composed using of several statements describing actions, activities, and interests related to the cultural stereotype of Black women identified in Experiment 2. For instance, an item addressing the stereotype ―loud‖ was included: ―Has had people comment or complain that she speaks loudly in public‖. Also, an item addressing the stereotype of being a ―single mother‖ was included: ―Is a single mother of two or more children with different fathers‖. Additionally, two items reflecting the stereotype ―[has a] bad attitude‖ were introduced: ―Comes across as pleasant and friendly when initially meeting others‖ and ―Is described by other as having a bad/snappy attitude or as having an attitude problem.‖ This stereotype was among traits used by participants to describe Black women exclusively. Other changes were made to the wording of items in an effort to better reflect perceived differences in behavior for Black women and Black men on the basis of gender. For example, for the stereotype ―ostentatious‖, the item ―Drives a car with expensive tires, rims, and sound system‖ was replaced with the item ―Is a sharp dresser and makes sure to wear the latest fashions.‖ Or in some cases, items reflecting stereotypes shown to be less relevant to Black women were removed 29 and replaced with items better reflecting traits used to describe Black women in Experiment 2. For example, the item ―has been charged with drug possession‖ reflecting the stereotype ―criminal‖ was replaced with the item ―has received food stamps or other government assistance‖ reflecting the stereotype ―poor‖. Identical to the instructions given for the Stereotypic Evaluations of Black Assessment used in Experiment 1, participants were instructed to judge the likelihood that each action, activity, or interest listed was characteristic of the person pictured using a 5-point scale (1-not at all likely to 5-very likely) (Appendix E). Results Raw scores from the Black stereotype assessment scale were transformed into an index score ranging from 1 to 5 with higher scores indicating greater stereotyping. Analyses focused on potential differences in stereotyping of Black women as a function of Afrocentricity of a stereotypic trait level and singular item level. Contrary to predictions, high Afrocentric (Mhigh = 3.01, SDhigh = .27) and low Afrocentric targets (Mlow =2.88, SDlow = .33) as were stereotyped to a similar degree, t (26) = -1.19, ns (Figure 3). Additional, analyses focused on potential differences in stereotyping of Black women as a function of Afrocentricity on a stereotypic trait level and singular item level with particular attention to items included based on stereotypes identified in Experiment 2. Counter to predictions, high (Mhigh = 2.69, SDhigh = .49) and low Afrocentric targets (Mlow = 2.60, SDlow = .52) were evaluated similarly for the stereotype ―loud‖, t (26) = -.44, ns (Figure 7). Additionally, high 30 Afrocentric (Mhigh = 2.85, SDhigh = .60) and low Afrocentric targets (Mlow = 2.59, SDlow = .54) were rate similarly in sexual aggressiveness, t (26) = -1.21, ns (Figure 4). Significant differences in evaluations of ostentatiousness emerged but in the direction opposite from predicted. Low Afrocentric Black women (Mlow = 3.53, SDlow = .34) were rated more ―ostentatious‖ than their high Afrocentric counterparts (Mhigh = 2.96, SDhigh = .56), t (26) = 3.04, p = .005, d = 1.23, r = .52 (Figure 4). An item level analysis of the stereotype ―poor‖ revealed significant differences in perceptions of high and low Afrocentric Black women as well. The stereotype ―ostentatious‖ was qualified by the items ―Is not interested in material things‖ and ―Is a sharp dresser and makes sure to wear the latest fashions.‖ Low Afrocentric Black women (Mlow = 3.88, SDlow = .65) were rated as ―more interested in material things‖ than high Afrocentric Black women (Mhigh =3.21, SDhigh = .66), t (26) = 2.67, p = .01, d = 1.02, r = .46. Likewise, low Afrocentric Black women (Mlow = 3.19, SDlow = .54) were rated as more likely to be ―sharp dressers‖ than high Afrocentric Black women (Mhigh = 2.71, SDhigh = .75), t (26) = 1.96, p <.06, d = .73, r = .34. Analyses also revealed significant differences for the traits ―poor‖, discussed here in terms of socioeconomic status. Consistent with previous research on the nature of racial phenotypicality bias, high Afrocentric Black women (Mhigh = 3.37, SDhigh = .53) were perceived as being of a lower socioeconomic status than their low Afrocentric counterparts (Mlow = 2.98, SDlow = .35), t (26) = -2.31, p =.03, d = .87 r = .40 (Figure 4). 31 An item level analysis of the stereotype ―poor‖ revealed differences in perceptions of high and low Afrocentric Black women as well. The stereotype ―poor‖ or lower socioeconomic status was qualified by statements regarding employment status and aspirations (―Has been unemployed for the past six months and struggling to find employment‖ and ―Aspires to be an English professor like her mother‖), reliance on government assistance (―Has received food stamps or other government assistance‖) racial composition of target‘s neighborhood ( ―Lives in a neighborhood comprised mostly of minorities‖ and ―Grew up and continues to live in an upscale, suburban neighborhood‖) and cultural/musical preferences (―Has a season subscription to the Boston Symphony‖). There were no differences in perceived unemployment for high (Mhigh = 2.67, SDhigh = .62) and low Afrocentric women (Mlow = 2.31, SDlow = .54), t (26) = -1.61, ns, or perceived aspirations of be an English professor for high Afrocentric Black women (Mhigh = 3.29, SDhigh = .69) and low Afrocentric Black women (Mlow = 3.25, SDlow = .58) t (26) = -.17, ns. There were also no differences in perceived reliance on government assistance for high Afrocentric (Mhigh = 2.61, SDhigh = .81) and low Afrocentric targets (Mlow = 2.41, SDlow = .58), t (26) = -.99, ns. Likewise, high Afrocentric Black women (Mhigh = 3.90, SDhigh = .76) and their low Afrocentric counterparts (Mlow = 3.30, SDlow = .81) were rated equally likely to have a season subscription to the Boston symphony, t (26) = -1.01, ns. However, consistent with Experiment 3a, high Afrocentric women (Mhigh = 3.83, SDhigh = .72) were rated as more likely to live in a neighborhood comprised mostly of minorities than low Afrocentric Black women (Mlow = 3.13, SDlow =.56), t (26) = -2.93, p = .007, d = 32 1.09, r = .48, and less (Mhigh= 3.67, SDhigh = .79) likely to live in an upscale, suburban neighborhood than low Afrocentric Black women (Mlow = 3.00, SDlow = .61), t (26) = -2.55, p <.02, d = .95, r = .43. Finally, high Afrocentric Black women (Mhigh = 4.06, SDhigh = .48) were rated marginally more likely to be ―religious‖ than low Afrocentric Black women (Mlow = 3.72, SDlow = .48), t (26) = -1.87, p = .07, d = .71, r = .33 (Figure 4). The stereotype ―religious‖ consisted of two items: ―Attends a local Baptist church regularly and is very involved in church activities‖ and ―Disagrees with most organized religions and recently became agnostic‖. High Afrocentric targets (Mhigh = 3.96, SDhigh = .58) were rated as marginally more likely to ―attend…church regularly…‖ than low Afrocentric targets (Mlow = 3.56, SDlow = .63), t (26) = -1.70, p =.10, d = .66, r = .31. However, high Afrocentric Black women (Mhigh = 4.17, SDhigh = .75) and low Afrocentric Black women (Mlow =3.88, SDlow = .62) were rated equally likely to have ―…recently become agnostic‖, t (26) = -1.13, ns. Discussion The goal of Experiment 3 was to reexamine the nature of racial phenotypicality bias toward Black women in light of stereotypes identified to be specific to Black women in Experiment 2. Although racial phenotypes did not influence stereotyping of Black women for the predicted stereotypes of ―loud‖ or ―sexually aggressive‖, racial phenotypes appeared to influence of evaluations of Black women in some instances, specifically for the stereotypes ―poor‖, ―product of a broken home‖ and ―ostentatious‖. Further examination of the items used to 33 represent these particular stereotypes suggests that the specific wording of these items may underlie these significant effects. For instance, the items used to reflect the stereotypes ―poor‖ and ―product of a broken home‖ described the target‘s upbringing and background, factors largely outside of the target‘s control, rather than the target‘s current actions possibly perceived to be within the target‘s control. It is possible that these items are seen as better reflecting current stereotypes of Black women in comparison to other items or that participants were more willing to endorse items perceived as being outside of the target‘s control than those items reflecting personal attributes perceived as being within the target‘s control. Additionally, the wording of the items representing for the stereotype ―ostentatious‖ may have indirectly tapped into participants‘ stereotypes about attractiveness for Black women. Previous research has shown that Black Americans with fewer Afrocentric features are perceived as more physically attractive than Black American with more Afrocentric features, a pattern particularly pronounced for Black women (see Maddox, 2004 for a review). The items representing ―ostentatious‖ queried beliefs about materialism and fashionableness—both potential proxies for perceived attractiveness. Consequently, participants may have viewed low Afrocentric targets as more attractive than high Afrocentric targets, thus explaining the greater stereotyping of low Afrocentric Black women relative to high Afrocentric Black women for the stereotype ―ostentatious‖. 34 However, in spite of these significant findings and given the methodological considerations addressed in response to Experiment 1 in the current study, I argue that the inconclusive evidence of racial phenotypicality bias toward Black in Experiments 1 and 3 may be due to differences in the processing of racial phenotypes by perceivers when forming impressions of Black women in comparison to the processing of the information when forming impressions of Black men. This argument is addressed in greater detail in the general discussion. General Discussion Racial Phenotypicality Bias toward Black Women As of late social psychological research has come to accept that Black Americans are not perceived as a monolithic group. In particular, work on racial phenotypicality bias has examined the role of within racial category variation in physical features indicative of race in stereotyping and prejudice. Research exploring racial phenotypicality bias suggests that of Blacks with more Afrocentric features are stereotyped to a greater degree than Blacks with fewer Afrocentric features. Although this area of research provides a more nuanced understanding of how Black Americans are perceived, racial phenotypicality bias research to date is limited by its focus on Black men. The current research aimed to address this limitation by exploring the role of Afrocentric features in perceptions of Black women, and ultimately, how racial phenotypes and gender may jointly influence impressions of Black women. Experiment 1 was an initial step toward isolating the potential use of Afrocentric features in stereotyping of Black women. Results of Experiments 1 35 provided somewhat inconclusive evidence for racial phenotypicality bias applying to perceptions of Black women, with high Afrocentric Black female targets and low Afrocentric Black female targets stereotyped to a marginally different degree. Experiments 2 and 3 tackled potential methodological explanations for the lack of evidence of racial phenotypicality bias in stereotypic evaluations of Black women. I argued that the stereotypes comprising the stereotyping assessment used in Experiments 1 may not accurately represent current perceptions of Black women. Experiment 2 attempted to address the validity and sensitivity of the stereotypes used in the stereotyping assessment used in Experiment 1 by identifying stereotypes explicitly associated with Black women. Consistent with predictions for Experiment 2, stereotypes reported for Black men shared greater overlap with racial group stereotypes for Black Americans identified in previous social psychological research in comparison to Black women. Further, some stereotypes were identified for Black women exclusively. Experiment 3 used stereotypes specific Black women identified in Experiment 2 to reevaluate the influence of Afrocentric features in evaluations of Black women. Though evidence emerged for racial phenotypicality bias for some cases, the inconsistency in greater stereotyping of high Afrocentric Black women in comparison to low Afrocentric Black women for the majority of stereotypes examined, in particular those identified as specific to Black women, leaves inconclusive evidence racial phenotypicality bias toward Black women. 36 Overall, the findings of the current research suggest that racial phenotypes may not be used in the same manner when forming impressions of Black women as they are used when forming impressions of Black men--a conclusion I argue is consistent with other investigations suggesting divergent views Black women and Black men. Literature showing this difference is reviewed in the next section. Divergent Cognitive Representation: Processing Speed and Accuracy Social psychological research on person perception has highlighted the importance of cognitive representations of social groups in impression formation processes (Fiske & Neuberg, 1990; Brewer, 1998; Bodenhausen & Macrae, 1998). Cognitive representations inform the ways in which we view others, servings as templates by which we categorize others (e.g., by race, gender, or age) and consequently, making stereotypes associated with these groups readily assessable for use. Researchers have investigated cognitive representations of individuals falling into multiple social categories and the implications for stereotyping; for instance, how being categorized simultaneously by gender and race impacts stereotyping. Although limited in number, studies investigating cognitive representations of Black women versus those of Black men suggests some differences in the ways in which these groups are possessed in comparison to one another. Three notable investigations are discussed below. In an examination of how race and gender impact the social categorization of Black and White men and women, Zárate and Smith (1990) presented participants with a series of social category labels one at a time (e.g. Black, White, man, and woman) and then presented images of Black and White male and 37 female targets. Participants indicated whether the pictured individuals fit into the previously shown social category. Analysis of categorization speed and accuracy revealed that female targets were categorized by gender faster than male targets while male targets were categorized by race faster than female targets. Zárate and Smith interpreted these results as evidence of a male cultural norm bias, later expanded to a ―White Male Norm‖ bias to include racial normalcy as a factor in social judgments (Smith & Zárate, 1992). According to the White Male Norm bias, deviating from White male normalcy (being White and male) along any dimension attracts attention, facilitating processing along the deviating dimension but interfering with processing of other social category information. For instance, because women deviate from the male gender norm, perceivers attend to their gender which in turn interferes with the processing of women‘s racial category information. Applying the White Male Norm bias to cognitive representations and perceptions of Black women specifically, Black women‘s perceived gender and racial non-normalcy, has the potential to both interfere with and facilitate their racial and gender categorization. By contrast, Black men deviate from ―White male normalcy‖ by race only, facilitating racial categorization but interfering with gender categorization. The differing degrees of perceived gender and racial nonnormalcy for Black women relative to Black men may translate to divergent cognitive representations as well. An investigation of race and gender categorization by Stroessner (1996) also revealed evidence for divergent cognitive representations for Black women and Black men as a consequence of their relative gender and racial non-normalcy. 38 In Experiment 1, participants were shown photographs of Black and White men and women and asked to categorize the individuals by race and gender separately (e.g., Black or White, male or female). Incorporating Zárate and Smith‘s (1990; Smith & Zárate, 1992) notion that White males serve as a cultural norm, Stroessner predicted that targets would be categorized along dimensions deviating from White male normalcy. For example, it was predicted that Black males would be categorized by race more quickly than White males due to their racial deviance from the White male cultural norm. This rationale also suggests divergent categorization of Black women in comparison to Black men. However, since Black women deviate from White male normalcy in both race and gender (while Black men deviate racially only), it was not clear how this might impact processing. Consistent with predictions, race and gender categorization reaction times for Black men and Black women differed in comparison to White men. However, the nature of these differences in race and gender categorization relative to White males varied for Black women in comparison to Black men. Black women were categorized more slowly by race as well as by gender in comparison to White men but Black men were categorized by race more quickly than White men suggesting divergent cognitive representations for Black women and Black men. A second experiment provided additional evidence for differing cognitive representations for Black women and Black men. In Experiment 2, participants were shown photographs of Black and White men and women and asked to categorize the individuals by race and gender simultaneously (e.g., Black male, 39 Black female, White male, White female). Again, categorization reaction times for Black women and Black men differed in comparison to White men. The nature of these differences relative to White males varied for Black women and Black men as well. Black female targets were categorized as ―Black women‖ more quickly than White males targets were categorized as ―White men‖ but Black male targets categorized as ―Black men‖ more slowly than White male targets were categorized as ―White men‖. Work by Ito and Urland (2003) examining the automaticity of attention to and encoding of race and gender information also suggests that cognitive representations of Black women differ from those of Black men. Ito and Urland use event-related potentials (ERPs) to evaluate cognitive processing of Black and White men and women during race and gender categorization tasks. ERP data indicated preferential attention to Black targets very early in processing, about 100 milliseconds after the stimulus was introduced. Additionally, Black targets elicited ERP component activation indicative of early selective attention. Consistent with research suggesting that early components such as the N100, P200, and N200 are associated with early selective attention, Black targets elicited larger N100s and P200s than White targets. However, gender differences emerged in these effects with P200s largest for Black males, suggesting dissimilar processing of Black women and Black women. In summary, these studies provide evidence that cognitive representations differ for Black women and Black men, and consequently, that support the notion that Black women are most likely perceived differently than Black men. I argue 40 that these divergent cognitive representations for Black women and Black men have meaningful implications for racial phenotypicality bias research and give credence to potential differences in the ways in which Afrocentric features are use in forming impressions of Black women versus they ways in which these features are use in forming impressions of Black men. Further, I argue that it is essential to develop a model of racial phenotypicality bias that carefully considers the dual influence of both racial phenotypes and gender. Such an approach has been referred to as an intersectional approach, outlined in the next section. Bridging the Theoretical Divide through an Intersectional Approach While the current research does not explicitly utilize an intersectional approach, the results do bring importance of using an intersectional approach to the forefront. ―An intersectional approach to race and gender is one in which consideration is given to the unique positions that exist for people on the basis of the combination of their race/ethnicity and gender; it recognizes that gender and race/ethnicity can only be experienced simultaneously with an individual‖ (Settles, 2006, p. 589). Issues of identity, stereotyping, prejudice, and discrimination are often studied by social psychologists; yet, an intersectional approach has been used by a limited number of psychologists investigating Black women (Cole, 2009). A theoretical piece by Purdie-Vaughs and Eibach (2008) discusses the importance and implications of an intersectional approach. They propose a model of intersectional invisibility. The model proposes that ―possessing multiple subordinate-group identities renders a person invisible relative to those with a 41 single subordinate-group identity (p.377).‖ Purdie-Vaughns and Eibach argue that androcentric, ethnocentric, and heterocentric ideologies cause individuals with multiple-subordinate social identities to be considered non-prototypical members of their social groups. In contrast, individuals with single-subordinate social identities are viewed as prototypical of their social group. For example, a Black lesbian, subordinate in her racial, gender, and sexual identities, is considered non-prototypically ―homosexual‖, ―female‖, and ―Black‖ while a gay White man, subordinate in only his sexual identity, would be considered the prototypical ―homosexual‖ group member, a heterosexual White woman, subordinate in gender identity only, is viewed as the prototypical ―female‖, and a heterosexual Black man, subordinate in racial identity only, is viewed as the prototypical ―Black‖ group member. Recent empirical work on perceptions of Black women has also noted the utility of an intersectional approach in research on individuals with multiple social identities. For instance, Goff and colleagues (2008) examined the effect of intersecting race and gender identities by asking participants to make racial and gender categorizations of Black and White male and female targets. Complimentary to work of Purdie-Vaughns and Eibach, Goff and colleagues contend that ―intersectional categories serve as basic units of person perception (p.394).‖ They hypothesized that in the case of Black women, that race actually ―erases‖ perceived femininity from the perceptual equation. As a result, they hypothesized that, when asked to make gender categorizations of Black and White men and women, participants would make more gender categorization errors 42 when evaluating the gender of Black women in comparison to making gender evaluations of Black men and White men and women. In addition to gender categorization, participants were asked to rate the femininity/masculinity and attractiveness of Black and White men and women. Congruent with predictions, participants made more gender categorization errors when evaluating Black women in comparison to Black men and White men and women; participants were more likely to erroneously categorize Black women as men than misattributing gender for any other group. Further, in line with the intersectional invisibility model‘s assessment of Black women as non-prototypically female in comparison to White women, participants rated Black women as less feminine and less attractive than White women. Additionally, recent empirical work by Sesko and Biernat (2010) provides further support for an intersectional approach to understanding perceptions of Black women. Sesko and Biernat suggest that the relative non-prototypicality of Black women in comparison to Black men (as the prototypical ―Blacks‖) and White women (as the prototypical ―women‖) has implications for the perceiving Black women. They argue that Black women‘s unique intersectional social position renders them invisible. In two studies addressing this claim, the faces of Black women went ―unnoticed‖ and their voices ―unheard‖ relative to Black men, White men, and White women. In a memory task for the faces and speech contributions of Black and White men and women, participants were least likely to remember Black women‘s faces and were most likely to misattribute statements made by Black women to other targets. 43 Usefulness of an Intersectional Approach in Racial Phenotypicality Bias and other Social Psychological Research The literature reviewed above highlights the importance of considering both race and gender in stereotyping and prejudice research. Though a substantial amount of social psychological research has focused on stereotyping and prejudice of Black Americans as a racial group, limited emphasis has been placed on gender. Gender, however, may factor into perceptions of Black Americans, in particular impressions of Black women. Previous research on stereotyping and prejudice of Black Americans has been generalized and applied to our understanding of how to both Black women and Black men are viewed. I argue that such an approach is problematic because it neglects the role of gender in racial stereotyping of and prejudice resulting in an incomplete depiction of how Black women are viewed. This argument can also be made for the incorporation of an intersectional approach in future racial phenotypicality bias research. Failing to consider the potential gender differences in the reliance on Afrocentric features to form impressions of Black Americans and results in an incomplete understanding of racial phenotypicality bias toward Black women. The current research also brings to the forefront the importance of taking an intersectional approach to researching individuals with multiple subordinate group identities in general. A recent American Psychologist essay by Elizabeth Cole discusses the use and importance of an intersectional approach in psychological research. Cole (2009) illustrates psychologist‘s limited use of an intersectional approach by highlighting a study conducted by Silverstein (2006) 44 showing that only a minority of publications dealing with either race or gender indexed in PsycINFO between 2002 and 2004 investigated both race and gender. Cole suggests that an intersectional approach is necessary to understanding how social identities such as race, gender, class, or sexuality impact perceptions of others. She argues that ―to understand any one of these dimensions, psychologists must address them in combination (p. 179).‖ Although an intersectional approach was not investigated empirically in this dissertation, the current research did acknowledged the potential importance of intersecting social identities in how Black women are perceived by questioning if race and gender dully impact the use of racial phenotypes in the stereotyping of Black women. By doing so, this dissertation highlighted how perceptions of Black women may differ from those of Black men, and suggests why acknowledging and accounting for the influence of both race and gender in research on stereotyping, prejudice, and discrimination of Black Americans is essential. The questions posed in this dissertation reach beyond the realm of research on Black women. These issues can be applied to the study of any individual and their experience in the world. How are people viewed when we consider the intersection of multiple category memberships such as race, gender, age, socioeconomic status, and so on and so forth? We are all members of multiple social categories simultaneously and any of these identities have the potential to influence how we view ourselves and how we are viewed by others in our social world. 45 Future Directions Although the findings of the current research did not support many of the predicted hypotheses, some interesting trends did emerge. For instance, though significant differences in stereotyping of high and low Afrocentric Black women did not occur in for the predicted traits Experiment 3, differences did emerge for the stereotypes ―poor‖, ―product of a broken home‖, and ―ostentatious‖. These finding suggest that racial phenotypicality bias toward Black women may exist for specific stereotypes or in specific domains. Future research in this area should further examine why or how racial phenotypes influenced evaluations of Black women in these particular domains. The similarities and differences between these stereotypes and other stereotypes associated with Black women, such as those identified in Experiment 2, should be investigated in an effort to determine and understand the types of evaluations or domains in which racial phenotypes may be used to form impressions of Black women. 46 Figure 1. Stereotyping of Black women as a function of Afrocentricity Experiment 1: Stereotyping of Black women as a Function of Afrocentricy 5.0 Stereotypes of Blacks Assessment Index 1-Not at all to 5-Very much so 4.5 4.0 3.5 3.0 2.86 2.68 2.5 2.0 1.5 1.0 High Afrocentric Low Afrocentric Afrocentricity t (46) = -1.86, p = .07, d = .54, r =.26 47 Figure 2. Trait level analysis of stereotyping of Black women as a function of Afrocentricity using the Stereotypes of Blacks Assessment Experiment 3a: Trait Level Stereotyping of Black women as a Function of Afrocentricy Stereotypes of Blacks Assessment Index 1-Not at all to 5-Very much so 5.0 High Afrocentric 4.5 ** 4.0 Low Afrocentric 3.57 ** 3.5 3.05 2.802.87 3.0 2.5 3.07 2.70 2.36 2.25 2.0 1.5 1.0 Sexually Aggressive * p<.05 Ostentatious **p<.01 Poor ***p<.001 48 Product of a Broken Home Figure 3. Stereotyping of Black women as a Function of Afrocentricity using the Stereotypes of Black Women Assessment Experiment 3b: Stereotyping of Black women as a Function of Afrocentricy 5.0 Stereotypes of Black Women Assessment Index 1-Not at all to 5-Very much so 4.5 4.0 3.5 3.01 2.88 3.0 2.5 2.0 1.5 1.0 Afrocentricity High Afrocentric Low Afrocentric t (26) = -1.19, ns 49 Figure 4. Trait level analysis of stereotyping of Black women as a function of Afrocentricity Experiment 3b: Trait Level Stereotyping of Black women as a Function of Afrocentricy 5.0 High Afrocentric Low Afrocentric * 4.06 ** Assessment Index 1-Not at all to 5-Very much so Stereotypes of Black Women 4.5 4.0 ** 3.53 3.72 3.37 3.5 2.85 3.0 2.69 2.60 2.98 2.96 2.59 2.5 2.0 1.5 1.0 Loud * p<.05 Sexually Ostentatious Aggressive **p<.01 Poor ***p<.001 50 Religious Table 1. Examples of responses coded into each category Stereotype Category Responses for Black Men Responses for Black Women Athletic Athletic Basketball/football players Can jump high/run fast Athletic African American females are all athletic Good at sports that require stamina Criminal Cause of high crime rates Criminal ―Thugs‖ Delinquent Start drinking at young ages Use drugs Dirty/Smelly Nappy hair They are dirty Unrefined/un-cultured Inferior ―Troubled‖-in need of assistance Less Successful Not as powerful Are less educated than white females Dependent In need of ―saving‖ Lazy Lack motivation Lazy Underachieving Are unmotivated Lack Ambition Lazy Ostentatious Listen to loud music (lots of bass in car) Love fancy cars, bling, etc. Loud Less wealthy Lower class ―Ghetto‖ Animated Fashionable Loud Rhythmic Good at rap Good dancers Talent in singing/music Good dancers Good singers Like hip-hop/R&B/rap Sexually Aggressive Leering/catcalling Hypersexualized Players ―Easy‖ Become sexually active at a young age Promiscuous Tough/Aggressive Aggressive Dangerous Violent Aggressive Confrontational Sassy Uneducated/ Unintelligent Dumb Uneducated Unintelligent Less educated Less intelligent Uneducated Poor 51 From lower-income neighborhoods Lower class On welfare Table 2. Percentage of traits reported coded into each stereotype category Stereotype Category Responses for Black Men Responses for Black Women % for Black Men % of Total % for Black Women % of Total Athletic 14.29 7.69 3.12 1.44 Criminal 16.67 8.97 1.39 .64 Dirty/Smelly .30 .16 .17 .08 Inferior 9.38 5.04 8.49 3.92 Lazy 4.02 2.17 2.43 1.12 Ostentatious 4.91 2.64 12.31 5.68 Poor 8.04 4.32 7.45 3.44 Rhythmic 6.40 3.44 8.49 3.92 Sexually Aggressive 1.49 .80 6.93 3.20 Tough/Aggressive 12.56 6.00 12.49 5.76 Uneducated/Unintelligent 9.38 5.04 9.19 4.24 Other 34.04 13.77 37.26 17.21 52 Table 3. Odd ratios for frequency of traits reported coded into each stereotype category Reported for Black Men to Black Women Reported for Black Women to Black Men Athletic*** 5.18 .19 Criminal*** 14.23 .07 Dirty/Smelly 1.72 .50 Inferior 1.11 .90 Lazy^ 1.68 .59 Ostentatious*** .37 2.72 Poor 1.09 .92 Rhythmic^ .74 1.36 Sexually Aggressive*** .20 4. 93 Tough/Aggressive .88 1.13 Uneducated/Unintelligent 1.02 .98 Other .58 1.73 Stereotype Category ^p < .1 * p<.05 **p<.01 53 ***p<.001 Table 4. Percentages and odd ratios for number of participants reporting at least one trait coded into each stereotype category for Black women and Black men Stereotype Category % Ps Reporting for Black Men %Ps Reporting for Black Women Odds Ratio Black Men to Black Women Odds Ratio Black Women to Black Men Athletic 35.92 15.53 3.05*** .33 Criminal 64.08 6.80 24.46*** .04 Dirty/Smelly 1.94 .97 2.02 .50 Inferior 27.18 33.01 .76 Lazy 24.27 8.74 3.35 ** .30 Ostentatious 24.27 51.46 .30 3.31 Poor 45.63 35.92 1.49 .67 Rhythmic 34.95 30.10 .80 1.24 Sexually Aggressive 9.71 25.24 .31 3.14** Tough/Aggressive 55.34 33.01 2.51 *** .40 Uneducated/ Unintelligent 50.49 39.81 1.54 ^ .65 Other 69.90 .40 2.53** ^p < .1 * p<.05 85.44 **p<.01 54 ***p<.001 1.32 *** Appendix A. Stimuli Pretest Section One Rate each photo on the following dimensions. For this section please focus on the face of the individual pictured, ignoring his or her attire and background of the photo. Rate features relative to the entire face. 1. Gender (circle one) Male 2. Race/Ethnicity (circle one) Asian or Female Black Hispanic White 3. Age (estimate in years) _______ 4. Texture of Hair? (If individual is bald, circle bold BALD here) Kinky 1 Loose Curls 2 3 Wavy 4 5 Straight 6 7 5. Width of nose? Very Narrow 1 Somewhat Narrow 2 3 Somewhat Broad 4 5 Very Broad 6 7 6. Fullness of lips? If upper lip and lower lip are different sizes, average the two when making your rating. Very Thin 1 Somewhat Thin 2 3 Somewhat Full 4 5 Very Full 6 7 7. Skin tone? Very Light 1 Somewhat Light 2 3 Somewhat Dark 4 5 Very Dark 6 7 8. Afrocentricity? Afrocentric features refer to features that a uniquely characteristic of people of African descent. Not at all Afrocentric 1 Moderately Afrocentric 2 3 4 55 Very Afrocentric 5 6 7 9. Eurocentricity? Eurocentric features refer to features that a uniquely characteristic of people of European descent. Not at all Eurocentric 1 Moderately Eurocentric 2 3 4 Very Eurocentric 5 6 7 10. Physical Attractiveness? Very Unattractive 1 Somewhat Unattractive 2 3 Somewhat Attractive 4 5 Very Attractive 6 7 11. Baby-faceness? Very Mature 1 Somewhat Mature 2 3 Somewhat Baby-faced 4 5 Very Baby-faced 6 7 12. Friendliness? Very Unfriendly 1 Somewhat Unfriendly 2 3 Somewhat Friendly 4 5 Very Friendly 6 7 13. Pleasantness? Very Unpleasant 1 Somewhat Unpleasant 2 3 Somewhat Pleasant 4 5 Very Pleasant 6 7 14. Facial Expression? Frowning 1 Somewhat Frowning 2 3 Somewhat Smiling 4 5 Very Smiling 6 7 15. Hostility? Not at all Hostile 1 Moderately Hostile 2 3 4 56 Very Hostile 5 6 7 Section Two Rate each photo on the following dimensions. For this section please focus on the background of the picture pictured, ignoring the individual pictured and his or her attire. 1. How distracting is the background? Not at all Distracting 1 Moderately Distracting 2 3 4 Very Distracting 5 6 2. Did the background affect your evaluation of the person pictured or his or her attire? No Moderate Effect Effect 1 2 3 4 57 5 6 7 Large Effect 7 Appendix B. Target Photographs High Afrocentric Targets Low Afrocentric Targets 58 Appendix C. Stereotypes of Blacks Assessment 59 Appendix D. Knowledge of Social Groups Questionnaire Knowledge of Social Groups Instructions: In this study we are interested in your knowledge of a variety of social groups. We are also interested in the difference between cultural beliefs and personal beliefs. Cultural Beliefs are your general impressions of how a particular group of people is portrayed on a societal level. Cultural beliefs come from television, newspapers, movies, books, jokes, etc. They can also come from people you know such as your friends, family, teachers, and others you have interacted with. These sources are a reflection of the way that society as a whole views a group. Personal Beliefs are those impressions of a group that you personally endorse. These beliefs may or may not be the way society thinks about a group. Here is a concrete example. Society may see lawyers as generally intelligent and also sleazy. These characteristics reflect Cultural Beliefs. However, while you think that lawyers are intelligent, you personally do not believe that lawyers are sleazy. These characteristics reflect your Personal Beliefs. On the following pages we would like you to report your knowledge about characteristics associated with the groups listed in terms of Cultural Beliefs and Personal Beliefs. Characteristics can be personality traits, physical traits, behaviors, occupations, likes and dislikes--anything that you believe that society associates with the social group. In the space provided you will write down the characteristics associated with Cultural Beliefs about the group written at the top of the page. Next, you will be asked to indicate if the characteristic is consistent or inconsistent with your personal beliefs. 1) Put a ―C‖ next to those characteristics that are CONSISTENT with your Personal Beliefs. 2) Put an ―I‖ next to those characteristics that are INCONSISTENT with your Personal Beliefs. 3) Put a ―U‖ next to those characteristics that you aren‘t sure whether to you agree or disagree with. 60 In the space below, write down Cultural Beliefs associated with the social group listed. 1) Put a ―C‖ next to those characteristics that are CONSISTENT with your Personal Beliefs. 2) Put an ―I‖ next to those characteristics that are INCONSISTENT with your Personal Beliefs. 3) Put a ―U‖ next to those characteristics that you aren‘t sure whether to you agree or disagree with. Social Group: African American (Black) Females __________________________________________________________________ _________________________________________________________________ __________________________________________________________________ __________________________________________________________________ __________________________________________________________________ __________________________________________________________________ __________________________________________________________________ __________________________________________________________________ __________________________________________________________________ __________________________________________________________________ 61 Appendix E. Stereotypes of Black Women Assessment 62 References Blair, I.V. (2006). The efficient use of race and Afrocentric features in inverted faces. Social Cognition, 24, 563 - 579. Blair, I. V., Chapleau, K. M., & Judd, C. M. (2005). The use of Afrocentric features as cues for Judgment in the presence of diagnostic information. European Journal of Social Psychology, 35, 59 – 68. Blair, I. V., Judd, C. M., & Chapleau, K. M. (2004). The influence of Afrocentric facial features in criminal sentencing. Psychological Science, 15(10), 674679. Blair, I. V., Judd, C. M., & Fallman, J. L. (2004). The automaticity of race and Afrocentric facial features in social judgments. Journal of Personality and Social Psychology, 87, 6 763 - 778. Blair, I. V., Judd, C. M., Sadler, M. S., & Jenkins, C. (2002). The role of Afrocentric features in person perception: Judging by features and categories. Journal of Personality & Social Psychology, 83(1), 5-25. Bodenhausen, G. V., & Macrae, C. N. (1998). Stereotype activation and inhibition. In R. S. Wyer Jr. (Ed.), Advances in social cognition (Vol. 11, pp. 1-52). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Brewer, M. B. (1988). A dual-process model of impression formation. In T. K. Srull & R. S. Wyer (Eds.), Advances in social cognition: a dual process model of impression formation (pp. 1-36). Hillsdale, NJ: Erlbaum. Cole, E. R. (2009) Intersectionality and research in psychology. American Psychologist, 64, 170-180. 63 Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent ofpsychopathology. Journal of Consulting Psychology, 24,349-354. Devine, P. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5-18. Devine, P.G., & Baker, S. M. (1991). Measurement of racial stereotype subtyping. Personality and Social Psychology Bulletin, 17, 44-50. Devine, P G., & Elliott, A. J. (1995). Are racial stereotypes really fading? The Princeton Trilogy revisited. Personality and Social Psychology Bulletin, 21, 1139-1150. Dovidio, J. F., Evans, N., & Tyler, R. B. (1986). Racial stereotypes: The contents of their cognitive representations. Journal of Experimental Social Psychology, 22, 22-37. Dovidio, J.F., Kawakami, K., Johnson, C., Johnson, B., & Howard, A. (1997). The nature of prejudice: Automatic and controlled processes. Journal of Experimental Social Psychology, 33, 510-540. Eagly, A. H. & Kite, M. E. (1987). Are stereotypes of nationalities applied to both women and men? Journal of Personality and Social Psychology, 53, 451462. 64 Eberhardt, J.L., Davies, P.G., Purdie-Vaughns, V. & Johnson, S.L. (2006). Looking deathworthy: Perceived stereotypicality of Black defendants predicts capital sentencing outcomes, Psychological Science, 17 (5), 383388. Ellison, R. (1952). Invisible man. New York: Random House, Inc. Fiske, S. T., and Neuberg, S. L. (1990). A continuum of impression formation, from category-based to individuation processes: Influences of information and motivation on attention and interpretation. Advances in Experimental Social Psychology, 23, 1- 74. Gaertner, S., & McLaughlin, J. (1983). Racial stereotypes: Associations and ascriptions of positive and negative characteristics. Social Psychology Quarterly, 46, 3-30. Giddings, P. J. (1996). When and where I enter: The impact of Black women on race and sex in America. New York: Amistad. Gilbert, G. M. (1951). Stereotype persistence and change among college students. Journal of Personality and Social Psychology, 46, 245-254. Gleiberman, L., Harburg, E., & Cooper, M. L. (1995). Skin color, measures of socioeconomic status, and blood pressure. Annals of Human Biology, 22(1), 69-73. 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. 65 Goldsmith, A., Hamilton, D., & Darity, Jr., W. (2007). From dark to light: Skin color and wages among African Americans. Journal of Human Resources, 42(4), 701-738. Harris, D. (2009). Black feminist politics from Kennedy to Clinton. New York: Palgrave Macmillan. Henry, P. J., & Sears, D. O. (2002). The symbolic racism 2000 scale. Political Psychology, 23 (2), 253-283. Hull, G., Bell-Scott, P., & Smith, B. (1982). All the women are White, all the Blacks are men, but some of us are brave: Black women’s studies. New York: Feminist Press. Ito, T. A., & Urland, G. R. (2003). Race and gender on the brain: Electrocortical measures of attention to the race and gender of multiply categorizable individuals. Journal of Personality and Social Psychology, 85, 616-626. Karlins, M., Coffman, T.L., & Walters, G. (1969). On the fading of social stereotypes: Studies in three generations of college students. Journal of Personality and Social Psychology, 13, 1-16. Katz, D., & Braly, K (1933). Racial stereotypes of one hundred college students. Journal of Abnormal and Social Psychology, 28, 280-290. Kawakami, K., Dion, K. L., & Dovidio, J. F. (1998) Racial prejudice and stereotype activation. Personality and Social Psychology Bulletin, 24, 407-416. 66 Klag, M. J., Whelton, P. K., Coresh, J., Grim, C. E., & Kuller, L. H. (1991). The association of skin color with blood pressure in U.S. Blacks with low socioeconomic status." Journal of the American Medical Association, 265 (5), 599-602. Lawal, B. (2003). Categorical data analyasis with SAS and SPSS applications. Mahwah: N.J.: Lawrence Erlbaum Associates, Publishers Levy, S. R, & Dweck, C. S. (1998). Trait- versus process-focused social judgment. Social Cognition, 16, 151-172. Livingston, R. W. (2001). What you see is what you get: Systematic variability in perceptual based social judgment. Personality and Social Psychology Bulletin, 27, 1086-1096. Livingston, R. W., & Brewer, M. B. (2002) What are we really priming?: Cuebased versus category-based processing of facial stimuli. Journal of Personality and Social Psychology, 82, 5-18. Maddox, K. B., & Gray, S. A. (2002). Cognitive representations of Black Americans: Reexploring the role of skin tone. Personality and Social Psychology Bulletin, 28, 250 –259. Maddox, K. B. (2004). Perspectives on racial phenotypicality bias. Personality and Social Psychology Review, 8, 383-401. McConahay, J. B. (1986). Modern racism, ambivalence, and the modern racism scale. In J. F. Dovidio and S. L. Gaertner (Eds.), Prejudice, discrimination and racism (pp. 91- 126), New York: Academic. 67 Niemann, Y. F., Jennings, L., Rozelle, R. M., Baxter, J., & Sullivan, E. (1994). Use of free responses and cluster analysis to determine stereotypes of eight groups. Personality and Social Psychology Bulletin, 20, 379-390. Purdie-Vaughns, V. J., & Eibach, R. P. (2008). Intersectional invisibility: The distinctive advantages and disadvantages of multiple subordinate-group identities. Sex Roles, 59, 377-391. Sears, D. O., & Henry, P. J. (2003). The origins of symbolic racism. Journal of Personality and Social Psychology, 85, 259-275. Secord, P., Bevan, W., & Katz, B. (1956). The Negro stereotype and perceptual accentuation. Journal of Abnormal and Social Psychology, 59, 309-314. Sesko, A. K., & Biernat, M. (2010). Prototypes of race and gender: The invisibility of Black women. Journal of Experimental Social Psychology, 46, 356-360. Settles, I. H. (2006). Use of an intersectional framework to understand Black women‘s racial and gender identities. Sex Roles, 54, 589–601. Silverstein, L. B. (2006). Integrating feminism and multiculturalism: Scientific face or science fiction? Professional Psychology: Research and Practice, 37, 21-28. Simonoff, J. S. (2003). Analyzing categorical data. New York: Springer-Verlag, Inc. Smith, E. R., & Zárate, M.A. (1992). Exemplar-based model of social judgment. Psychological Review¸99¸3-21. 68 Stroessner, S. J. (1996). Social categorization by race or sex: Effect of perceived non-normalcy on response times. Social Cognition, 14, 247-276. Sweet, E., McDade, T. W., Kiefe, C. I., & Liu, K. (2007). Relationships between skin color, income, and blood pressure among African Americans in the CARDIA Study. American Journal of Public Health, 97, 2253–2259. Williams, M. J., & Eberhardt, J. L. (2008). Biological conceptions of race and the motivation to cross racial boundaries. Journal of Personality and Social Psychology, 94, 1033-1047. Zárate, M. A., & Smith, E. R. (1990). Person categorization and stereotyping. Social Cognition, 8, 161-185. 69