• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

The influence of Afrocentric phenotype on promotion decisions for African American males

ProQuest Dissertations and Theses, 2011
Dissertation
Author: Jeanne Johnson Holmes
Abstract:
Although there are conflicting empirical findings regarding whether race has a significant impact on individual outcomes in the 21st century workplace, African Americans are still less visible in top leadership positions than would be expected on the basis of population base rates (The Alliance for Board Diversity, 2005; Corporate Board Initiative, 2006; Fortune, 2006; McCoy, 2007; Thomas & Gabarro, 1999). One potential explanation for these equivocal results is that much of this research assumes that race has an equal impact within a given racial group. Little attention has been given within the management literature to intra-racial differences, which may influence how and to what extent racial stereotypes are activated. This dissertation examines the role of feature-based bias in promotion evaluations. Specifically, Afrocentric phenotype bias is explored to gain a better understand of how within-category, feature-based differences influence promotability evaluations. This dissertation utilizes self-categorization theory (SCT: Turner, 1985, 1987) and Heilman's (1983, 1995, 2001) lack of fit model to examine the consequences of Afrocentric phenotype bias within the employment context. Finally, the availability of cognitive resources is investigated as a potential moderator of the influence of Afrocentric phenotype bias on promotion-related rating.

TABLE OF CONTENTS DEDICATION

..................................................................................................................................

III

ACKNOWLEDGEMENTS

..........................................................................................................

IV

ABSTRACT

........................................................................................................................................ V

LIST OF TABLES

......................................................................................................................... VIII

LIST OF FIGURES

...................................................................................................................... VIII

CHAPTER 1 INTRODUCTION

...............................................................................................

IX

CHAPTER 2 THEORETICAL BACKGROUND

................................................................ 8

2.1

S TEREOTYPE AND P ROTOTYPES ................................................................................. 9

2.2

P HENOTYPE AND S TEREOTYPE A CTIVATION ............................................................ 11

2.3

P HENOTYPE AND P ROTOTYPES ................................................................................. 15

2.4

P HENOTYPE AND W ORKPLACE O UTCOMES .............................................................. 16

2.5

P ERCEIVED L EADERSHIP E FFECTIVENESS ................................................................ 18

2.6

W ORKPLACE E VALUATIONS AS A FUNCTION OF PERSON - JOB FIT .............................. 19

2.7

L ACK OF F IT AND L EADERSHIP PROTOTYPES ............................................................ 21

2.8

C OGNITIVE P ROCESSING OF I NFORMATION AND S TEREOTYPE A CTIVATION ............. 22

CHAPTER 3 HYPOTHESES

....................................................................................................

25

3.1

P ROMOTION -R ELATED E VALUATIONS ..................................................................... 27

3.2

T HE I NTERACTION OF A FROCENTRIC P HENOTYPE AND C OGNITIVE P ROCESSING ..... 29

CHAPTER 4 METHOD

..............................................................................................................

32

4.1

P RETEST ................................................................................................................... 33

4.2

M AIN D ISSERTATION S TUDY .................................................................................... 36

4.3

M AIN S TUDY S TIMULI .............................................................................................. 38

4.4

M EASURES ............................................................................................................... 39

4.5

M ANIPULATION CHECK ............................................................................................ 42

CHAPTER 5 RESULTS

..............................................................................................................

44

5.1

S AMPLE R ESPONSE R ATE ......................................................................................... 44

5.2

S AMPLE REDUCTION ................................................................................................. 45

5.3

P ERCEIVED T IME P RESSURE M ANIPULATION C HECK ............................................... 46

5.4

D ATA A NALYSIS P LAN ............................................................................................. 47

5.5

C ONFIRMATORY F ACTOR A NALYSIS ........................................................................ 48

5.6

C OVARIATE A NALYSIS ............................................................................................. 52

5.7

S TATISTICAL A NALYSIS ........................................................................................... 54

5.8

H YPOTHESES T ESTING ............................................................................................. 60

vii

5.9

S UPPLEMENTAL A NALYSES ...................................................................................... 69

CHAPTER 6 DISCUSSION AND CONCLUSIONS

.........................................................

78

6.1

S UMMARY OF R ESULTS AND O VERALL D ISCUSSION ................................................ 78

6.2

A LTERNATE E XPLANATIONS .................................................................................... 80

6.3

L ACK OF FINDINGS FOR P HENOTYPE B IAS ................................................................ 80

6.4

S TUDY L IMITATIONS AND F UTURE R ESEARCH ......................................................... 81

6.5

I MPLICATIONS FOR T HEORY ..................................................................................... 87

6.6

C ONCLUSION ............................................................................................................ 90

REFERENCES

................................................................................................................................

92

APPENDIX A ................................................................................................................................ . 112

APPENDIX B

................................................................................................................................ . 113

APPENDIX C ................................................................................................................................ . 114

viii

LIST OF TABLES TABLE 1 DEMOGRAPHIC DATA OF SURVEYED INDIVIDUAL PARTICIPANTS VERSUS INDIVIDUAL PARTICIPANT POPULATION

............

49

TABLE 2 PERCEIVED TIME PRESSURE MANIPULATION CHECK MEANS, STANDARD DEVIATIONS, AND F STATISTICS

...........................................................

50

TABLE 3 CONFIRMATORY FACTOR ANALYSIS FOR THE DEPENDENT VARIABLES

....................................................................................................................................

53

TABLE 4 ROTATED COMPONENT MATRIX

................................................................ . 56

TABLE 5 SKEWNESS AND KURTOSIS OF KEY CONSTRUCTS

...........................

59

TABLE 6 LEVENE’S TES T OF EQUALITY OF ERROR VARIANCES

.................

61

TABLE 7 DESCRIPTIVE STATISTICS BY CONDITION

............................................

62

TABLE 8 CORRELATIONS AMONG INDEPENDENT AND DEPENDENT VARIABLES AND COVARIATES

.........................................................................................

63

TABLE 9 MANCOVA TESTS FOR HYPOTHESES 1 THROUGH 3

.........................

64

TABLE 10 MANCOVA TESTS FOR HYPOTHESES 4 THROUGH 6 INVOLVING MAIN EFFECTS, COVARIATES, AND INTERACTIONS

...............

66

TABLE 11 MANCOVA TESTS OF BETWEEN-SUBJECTS EFFECTS FOR HYPOTHESES 4 THROUGH 6

................................................................................................

67

TABLE 12 POST HOC EXPLORATORY ANALYSIS MANCOVA BETWEEN- SUBJECTS EFFECTS

..................................................................................................................

70

ix

LIST OF FIGURES FIGURE 1. PROPOSED MODEL OF AFROCENTRIC PHENOTYPE BIAS ACTIVATION

................................................................................................................................ . 26

FIGURE 2 THREE FACTOR MODEL

..................................................................................

55

FIGURE 3 SIMPLE SLOPES FOR THE INTERACTION OF TIME PRESSURE AND TARGET PHENOTYPE ON ALL PARTICIPANTS

.............................................

74

FIGURE 4 SIMPLE SLOPES FOR THE INTERACTION OF TIME PRESSURE AND TARGET PHENOTYPE ON CAUCASIAN PARTICIPANTS

...........................

75

FIGURE 5 SIMPLE SLOPES FOR THE INTERACTION OF TIME PRESSURE AND TARGET PHENOTYPE ON AFRICAN AMERICAN PARTICIPANTS

.......

76

1 CHAPTER 1 INTRODUCTION During the 2008 Presidential campaign, there were accusations that Hillary Clinton’s campaign attempted to highlight Barack Obama’s ―blackness‖ through several tactics. In addition to aligning him with Jeremiah Wright’s controversial comments regarding race, her campaign was accused of using the more clandestine and subconscious tactic of airing an artificially darkened video of Barack Obama in one of her television advertisements. This accusation mirrored the controversy involving Time

and Newsweek magazines in June of 1994. Both magazines used digitally altered mug shots of OJ Simpson that significantly darkened his skin to serve as the backdrop to their respective cover stories, which featured the report of Simpson’s arrest for his ex - wife’s murder. These events beg the question of why anyone would alter these images and what might they believe these alterations do to influence social perception. Did Hillary Clinton’s campaign really think that darkening Barak Obama’s skin in the video would make him seem more ―black‖ to voters ? If so, how does be ing ―blacker‖ influence voters’ perceptions of his ability to lead? The focus of this dissertation is the influence of Afrocentric phenotype bias (APB), which is the prejudice based on variation in physical features among African Americans, on promotion-related evaluations in the workplace and the role of cognitive processing in mitigating the negative effects of this phenomenon. This study seeks to

2

increase our knowledge of how the presence of certain physical features influences the extent to which category-based (i.e., race) stereotypes are activated when evaluating a target. In light of the multitude of research studies on race-based discrimination, a greater understanding of feature-based bias may help us understand how to close the racial gaps in career progression. As a society, the U.S. has strived to be more of a meritocracy over the past 50 years. Specifically, the racial demography of American business and industry has experienced considerable change since the legislation of equal employment opportunity in the U.S. workplace during the 1960s. There is reason to believe that U.S. business and industry has been successful to some degree in avoiding pure categorical discrimination (e.g., gender, racial); however, the empirical results are equivocal. For example, there is ample evidence of racial differences in promotion rates that reflect disadvantages for African Americans (Baldi & McBrier, 1997; Greenhaus, Parasuraman, & Wormley, 1990; Hersch & Viscusi, 1996; James, 2000; Kalleberg & Reskin, 1995; Powell & Butterfield, 1994). Although the number of entry-level and middle-management positions held by African Americans is a larger percentage than in the past (Tomaskovic-Devey, Zimmer, Stainback, Robinson, Taylor, & McTague, 2006; U.S. Bureau of Labor Statistics, 2004), African Americans are still less visible in top leadership positions than would be expected on the basis of population base rates (The Alliance for Board Diversity, 2005; Corporate Board Initiative, 2006; Fortune, 2006; McCoy, 2007; Thomas & Gabarro, 1999). Confirming this view, Ugorji (1997) found that African Americans were significantly more likely than Caucasian Americans to be denied training requests, be steered away from management track titles, be discouraged

3

from seeking professional promotion opportunities, and to have the evaluation criteria moved regarding promotion. Conversely, some scholars argue that overt racial bias has declined as the United States has adopted egalitarian social norms (McConahay, 1986; Sears, 1988). In fact, some assert that issues of racial difference in the workplace are no longer significant in modern U.S. organizations (Hurley, Fagenson-Eland, & Sonnenfeld, 1997) and contend that economic class is a more significant predictor of career mobility than race (Sakamoto & Tzeng, 1999; Wilson, 1980; Wilson, 1989). Indeed, not all studies agree that race has a significant influence on individual outcomes in the workplace. Some studies failed to find a direct effect of race on variables like promotions and access to mentoring (Nkomo & Cox, 1990; Powell & Butterfield, 1997; Sheridan, Slocum, & Buda, 1997). Similarly, Miller and colleagues (2008) did not find race to moderate the relationship between perceptions of organizational politics and job satisfaction or even job stress. One potential explanation for these equivocal results is that much of this research assumes that race and ethnicity have an equal impact within a given racial group. Little attention has been given within the management literature to int ra -racial differences, beyond behaviors, that may influence how and to what extent racial stereotypes are activated. Nevertheless, within recent years, psychology and social psychology research has taken a step beyond social category cues (i.e., race) to examine the extent to which within-category, feature-based (i.e., racial phenotype) differences influence interpersonal evaluations of African-Americans (e.g., Blair, Chapleau, & Judd, 2005; Blair, Judd, & Chapleau, 2004; Blair, Judd, & Fallman, 2004; Blair, Judd, Sadler, & Jenkins, 2002;

4

Deaux & Lewis, 1984; Dixon & Maddox, 2005; Livingston & Brewer, 2002; Maddox & Gray, 2002; Sczesny & Kuhnen, 2004). Racial phenotype is the observable physical characteristics (e.g., hair texture, skin tone, facial features, cheek bone structure, height) resulting from genetics and environment that are often used to differentiate individuals when identifying a racial group (OED, 2009). In other words, racial phenotype refers to physical features generally assumed to be common among a racial/ethnic group. Maddox (2004) coined the term Afrocentric phenotype bias to explain partiality based on differentiation in physical characteristics among African Americans. Specifically, he argued that negative attitudes and perceptions are associated with more Afrocentric features (e.g., darker skin, broader nose, fuller lips). This type of bias strays from the prevailing perspective on stereotype activation. The dominant model of stereotyping assumes that stereotypes are activated when perceivers sort targets into salient social categories (e.g., race, gender) based on perceptual cues such as facial features, hair style, or vocal characteristics (Hogg & Terry, 2000); however, researchers have found that feature-based stereotyping occurs independently of category-based stereotyping (Blair, Judd, & Chapleau, 2004). A strong example of this is found in the criminal justice system. In their analysis of criminal sentencing, Blair and colleagues (2004) found that the race of the offender did not account for a significant amount of variance in sentence length over and above the seriousness and number of offenses. However, race significantly predicted sentence length once Afrocentric features were taken into account. In fact, the presence of Afrocentric features, whether within African American or Caucasian inmates, was

5

associated with longer sentence lengths within the same racial group given equivalent criminal records. Given the wealth of evidence of phenotype bias in the social science literatures, an important and practical question for organizational scholars remains: Can differences in Afrocentric phenotype explain the ambiguous relationship between race and workplace outcomes? The goal of this dissertation is two-fold. First, it seeks to determine whether intra- racial differences (i.e., Afrocentric phenotype: observable physical characteristics such as skin color, hair texture, and nose structure that are often used in identifying ethnic membership) influence perceptions of leadership effectiveness, job performance, and promotability in a promotion evaluation context. And second, it explores whether more controlled processing within the evaluation process mitigates phenotype bias in promotability assessments for targets with high levels of Afrocentric phenotype. Examining the influence of within-category differences on promotion decisions and on race-based stereotype activation will help clarify our understanding of earlier equivocal findings on the influence of race on occupational advancement. Afrocentric phenotype is argued to have influence on the social judgments of both sexes and across racial/ethnic categories; however, a thorough review of the phenotype bias literature is beyond the scope of this dissertation. My goal is to address the influence of Afrocentric phenotype on the career progression of African American males in the context of promotion evaluations. In doing so, this dissertation attempts to advance the existing body of literature on human resource decision-making, both theoretically and empirically. First, drawing upon the social identity literature, it attempts to expand our understanding of how sub-category identity, specifically feature-based categorization,

6

influences the promotion decision-making process. While race-based stereotyping and bias has been analyzed extensively in the employment context, much less is known about whether variations in phenotype impact attitude formation about individual racial out- group members within the workplace context.

Additionally, I seek to clarify our understanding of the role cognitive processing plays in feature-based bias. The lack of attentional resources has shown to increase category-based bias; however, recent empirical evidence has failed to find similar significant results for feature-based bias (Blair, Judd & Fallman, 2004). Contrary to the findings of Blair and her colleagues (2004), I argue that cognitive processing will have a significant influence on feature-based bias. This dissertation suggests that the availability of attentional resources, which aid in cognitive processing, will only make a significant difference in the activation of Afrocentric phenotype bias directed toward targets with high levels of Afrocentric phenotype. I argue that this moderator’s influence will on ly be significant among targets with high levels of Afrocentric phenotype, because this sub- population is more susceptible to stereotype activation than their counterparts with lower levels. In the chapters that follow, I use social theories as a foundation for understanding the research on promotion decisions and to examine the relationship between phenotype and perceptions of promotability. In Chapter 2, I use self-categorization theory and person-job fit theory to lay a theoretical foundation for this study. In Chapter 3, I present hypotheses regarding the influence of Afrocentric phenotype and perceived time pressure on three work-related outcomes: perceived job performance, promotability, and perceived leadership effectiveness. I outline the methods used in the pretest and final dissertation

7

study in Chapter 4. In Chapter 5, I describe the results of the study. In the final chapter, I discuss the findings and provide some final conclusions.

8

CHAPTER 2 THEORETICAL BACKGROU ND

SELF-CATEGORIZATION THEORY Th e process and consequences of racial categorization have stimulated a wealth of theory and research over the past decades. Within the management literature, the interaction between race and career mobility has been well-documented, and research has provided evidence of the challenges experienced by minorities during their pursuit of senior-level positions (e.g., Jackson & O’Callaghan, 2009; Morrison and Von Glinow 1990) highlighting how these barriers influence a variety of factors that manifest in both conscious and sub-conscious discriminatory practices (Lee 2002 ; Martin 1991 , 1992 ).

Social identity theory and self-categorization theory argue that individuals prefer to interact with others like themselves and show bias towards those in the in-group and negative evaluations of the out-group (Tajfel, 1974; Tajfel & Turner, 1986). Supporting this view, Ferris and Judge (1991) found that promotion decisions appear to be influenced by in-group status within the employment context. In fact, similarity to the principle decision-maker has been found to positively influence performance ratings once an individual is already hired (Tsui & O ’ Reilly, 1989; Turban & Jones. 1988). Having shared or similar social or educational background with decision-makers has successfully predicted advancement in both qualitative and

9

quantitative studies (Markham, Harlan, & Hackett, 1987; Farh, Tsui, Xin, & Cheng, 1998). There is also some evidence of a same-sex bias where male raters are more apt than female raters to give high appraisals of men ’ s job performance (Bowen, Swim, & Jacobs, 2000). In the same way, research has found that being different has a significant negative influence on individual outcomes in organizations such performance ratings, organizational commitment, and social integration within the work unit (see Williams & O’Reilly, 1998). Given the influence of race on promotion outcomes, understanding the role of social identity in the evaluative process is central. 2.1 Stereotype and Prototypes The themes of categorization and its relation to stereotype formation have been studied at length (Corneille & Yzerbyt 2002; Dovidio & Gartner 2003; McGarty, Yzerbyt, & Spears 2002; Spears 2002; Stephan and Stephan, 1996). Self-categorization theory (SCT) squarely connects stereotyping to the process of identity development and maintenance (Mastro & Kopacz, 2006). The theory proposes that self-definition is formed and maintained through the assignment of oneself and others to multiple group categories that are both context- and motive-specific (Turner, 1985, 1987). Consequently, an individual’s identity is defined in reference to both the features of group membership and the individual’s own unique characteristics. When group membership is salient, an individual’s idiosyncrasie s become less relevant. Therefore, individual self-perception converts to a collective self-concept that is congruent with in- group members and divergent with out-group members (Abrams & Hogg, 1990; Turner, 1985, 1987).

10

The group categorization process emphasizes between-group variance and within- group homogeneity which often leads to over-generalized stereotypes (Tajfel & Wilkes, 1963) and the development of group membership prototypes . Prototypes are a set of characteristics considered representative of a group (Hogg & Terry, 2000). According to Hogg and Terry (2000), the prototypes become benchmarks in evaluating one’s self.

In addition, prototypes serve the dual purpose of being both descriptive and prescriptive in that they provide direction for attitudes, emotions, perceptions, and behaviors (Hogg, Cooper-Shaw, & Holzworth, 1993; Hogg & Hains, 1996). By defining others by their social role or in terms of the group to which they belong, others’ behavior becomes easier to predict and provides perceivers with the opportunity to reinforce self- esteem as well as adjust their own behaviors and attitudes (Hogg & Hains, 1996; Hogg & McGarty, 1990; Hogg & Terry, 2000). These predictions about others are often referred to as stereotypes . Stereotypes, ―generalized beliefs a bout social groups‖ (Madon , Guyll, Hilbert, & Kyriakatos, 2006), refer to the interpersonal judgment of how typical certain personality and/or behavior traits are of a group. Stereotype assumptions are assumed to activate as perceivers sort targets into salient social categories (e.g., race, gender) based on perceptual cues such as facial features, skin color, or vocal characteristics. When stereotypes influence interpersonal evaluations, the process is referred to as stereotyping. Activated stereotypes are known to bias judgment and behavior in a number of stereotype-consistent ways (see Wheeler & Petty, 2001). Empirical results have shown that strong negative evaluations are associated with racial out-groups. For example, African Americans are often stereotyped as generally less

11

intelligent (Devine, 1989) and as being more hostile than Caucasians (Fiske, 1998). In addition, activation of stereotypes for African Americans has been shown to increase ratings of hostility of African Americans’ behavi or (e.g., Devine, 1989), aggression (Devine & Elliot, 1995; Wittenbrink, Judd, & Park, 1997), and to provoke more hostile responses from interaction partners (Chen & Bargh, 1997). In summary, self- categorization leads to de-individuation of social perceptions and processing another by their similarity to the group prototype (Hogg & Hains, 1996). This view of stereotyping has been prevalent within the literature. Specifically, categorization has been viewed as the mechanism through which stereotypes operate. It asserts that traits associated with a specific group are uniformly applied to individuals who have been categorized as members of that group. Research has often contrasted this category-based process with the process of evaluating individuals based on unique, individual idiosyncrasies and attributes, which is often referred to as individuation (Brewer, 1988; Fiske & Neuberg, 1990). Stereotyping is assumed to be more automatic and heuristic in nature whereas individuation is assumed to be more conscious, labor-intensive, and integrative (see Bodenhausen, Macrae, & Sherman, 1999). Traits, such as phenotype, also assume differing roles based on the model of stereotype activation one adopts. 2.2 Phenotype and Stereotype Activation The role of phenotype in stereotyping has gained popularity and more thorough examination in the recent stereotyping literature. Two models of stereotype activation have arisen: the traditional and ecological perspectives. The ecological and traditional

12

perspectives differ in two important aspects (Friedman & Zebrowitz, 1992; Zebrowitz, 1996). First, there is the assumption within the traditional perspective that stereotyping is automatic and heuristic in nature whereas individuation (e.g., feature-based bias, phenotype bias) is assumed to be more conscious, labor-intensive, and integrative (see Bodenhausen, Macrae, & Sherman, 1999). However, a recent empirical examination of the automaticity of category (race-based) and feature-based (i.e., phenotype-based) evaluations found that individuals are aware of and are able to control race-based (i.e., category-based) stereotypes (Blair et al., 2004). Conversely, the ecological perspective argues that within category variation, such as phenotype, can lead directly to stereotype activation and is not more conscious and labor-intensive than category-based stereotyping. In fact, recent findings show that feature-based bias against those with more Afrocentric phenotype appeared to be unconscious and unavoidable even after the perceiver is given explicit information about feature-based processing (Blair et al., 2004). Second, the traditional perspective makes no assumption that a group’s physical appearance serves a necessary role in the stereotype or that it serves any evolutionary purpose; h owever, the ecological perspective ―assumes that features convey information through associations that have been developed in response to evolutionary demands‖ (Blair, Judd, Sadler, & Jenkins, 2002, p. 6). This perspective argues that physical features provide information that is categorical in nature. Therefore, the relationship between phenotype and the activation of a stereotype is indirect, because phenotype characteristics are assumed to operate as visual cues that lead to categorization. An individual ’s physical features indicate the category in which they belong and, subsequently, what traits they are

13

likely to possess. For example, the same stereotype will be activated regardless of whether an individual has withered or partially-amputated legs. Their disabled legs categorize them as being lame. However, several scholars argue that there are still substantial automatic responses that pull us toward stereotyping within group (Blair, Judd, Sadler, & Jenkins, 2002; Livingston & Brewer, 2002; Maddox & Gray, 2002; Sczeny & Kuhnen, 2003; Uhlmann, Dasgupta, Elgueta, Greenwald, & Swanson, 2002). In contrast, the ecological perspective predicts that features function in a continuous manner such that the more a feature is present, the more that person is assumed to have the related trait. For example, the more gray hair a person has, the older the person is assumed to be and the more the person is assumed to be reflective of age- related stereotypes such as moving slowly and lacking technology savvy. In addition, recent evidence suggests that categorization cues, such as phenotype, lead directly to stereotype activation (Blair, Judd, Sadler, & Jenkins, 2002; Livingston & Brewer, 2002; Maddox & Gray, 2002; Sczesny & Kuhnen, 2003; Uhlmann, Dasgupta, Elgueta, Greenwald, & Swanson, 2002), which aligns with the ecological perspective. Foundational evidence for this can be seen in the research on colorism , the prejudice and discrimination against African Americans with darker skin tone and, conversely, the benefits that are available to those with lighter skin tone. This phenomenon, which has primarily been documented by historians and sociologists, has generated a wealth of empirical evidence confirming the negative impact that darker skin tone has on economic and social outcomes as well as social judgment (ie., Edwards, 1959; Fraizer, 1957; Gullickson, 2005; Hill, 2000; Hughes & Hertel, 1990; Hunter, 1998;

14

Keith & Herring, 1991; Neal & Wilson, 1989; Russell, Wilson, & Hall, 1992; Seltzer & Smith, 1991). Although there are several cues for racial categorization, skin color still remains one of the most salient race-related phenotypic features used in social perception (Maddox, 2004). Amid the numerous historical and empirical examples of race-based stereotyping behavior, there are several historical and anecdotal reports of African American skin-tone bias (e.g., Drake & Cayton, 1945; Frazier, 1957; Russell, et al., 1992). Central to these skin-tone based historical accounts is that Caucasian or Eurocentric facial features in African Americans were perceived as evidence of Caucasian ancestry, which led to inferences of racial superiority (Russell et al., 1992). Lighter skin tone afforded better social, educational, and economic opportunities for African Americans in the time period preceding the abolition of slavery (Neal & Wilson, 1989; Russell et al, 1992). During that time, there were several accounts of systemic bias aimed at excluding darker African Americans from positions of higher socio-economic status (Hall, 1992; Maddox & Gray, 2002; Russell, et al., 1992). From slavery to modern time historians, anthropologists, psychologists, and sociologist have continued to documented that lighter-skinned African American have been afforded better educational and career opportunities and, ultimately, enjoy higher societal status than those with darker skin (Edwards, 1959; Fraizer, 1957a, 1957b ; Gullickson, 2005; Hill, 2000; Seltzer & Smith, 1991). In fact, the gap in education, occupational status, and income level between light- and dark-skinned African- Americans has remained even after controlling for factors such as parents’ socioeconomic status, gender, region of residency, proximity to urban locations, age,

15

marital status, and inheritance of wealth, and was documented well into the later half of the twentieth century (Hughes & Hertel, 1990; Hunter, 1998; Keith & Herring, 1991). Not surprisingly, similar stratification patterns have been found among Mexican Americans (Arce, Murguia, & Frisbie, 1987; Relethford, Stern, Gaskill, & Hazuda, 1983; Telles & Murguia, 1990). Colorism is seen in societies throughout the world (e.g., Arce, Murguia, & Frisbie, 1987; Harris, Consorte, Lang, & Byrne, 1993 ; Murguia & Telles, 1996; Uhlmann, Dasgupta, Elgueta, Greenwald, & Swanson, 2002). Research suggests that lighter skin is generally valued over darker skin and that skin tone has a broad influence on an individual’s social status. These findings, both within the U.S. and international societies, help echo the argument that racial phenotype may have a powerful influence on interpersonal perception (Maddox, 2004). 2.3 Phenotype and Prototypes A wide array of empirical evidence confirming the influence of a target’s physical appearance on person perception exists within the social psychology field (e.g., Berry & McArthur, 1985; Zebrowitz, 1996, 1997). The more prototypical an individual is to a salient positive category, the more positive an evaluation the target receives (Hogg et al., 1993; Hogg & Hains, 1996). This happens regardless of in-group or out-group status (Coover, 2001). In other words, the more an individual embodies a salient characteristic of the perceiver’s positive in -group, the more favorable the evaluation of the individual, regardless of whether they are an in-group or out-group member. In a recent study involving prototypicality and race, Mastro and Kopacz (2005) confirmed this by finding that prototypes play a central role in cognitive and affective

16

reactions to others. Their study investigated the influence of media-generated primes of prototypicality on Caucasian consumers’ liking of Caucasian and African American celebrities. Specifically, the more the celebrity was shown to be similar to the positive in- group (e.g., Caucasian) prototype along the primed characteristic, the greater the social attraction to the celebrity regardless of race. Social identity theory (SIT) and SCT assert that people prefer to interact with others like themselves and show bias towards those in the in-group. In her work on social identity and media exposure, Coover (2001) argues Caucasian consumers’ preferences for certain media-based depictions of race are based on the extent to which those representations support and are consistent with Caucasian racial norms. Racial prejudices are not the underlying force. Instead, she argues that racial identity is the fundamental mechanism that determines their responses to media characterizations of race. According to Coover’s (2001) findings, the greater the affirmation of Caucasian social identity, the more favorable evaluations of media content. In the context of phenotype-based bias, it could be argued that media depictions of African-Americans with dark skin or dread-locked hair as criminals would be favorable to Caucasian consumers because these depictions may be seen as representing the opposite end of the continuum from the Caucasian prototype, which affirms Caucasian status and privilege. These findings help corroborate assertions that racial phenotype bias mimics racial discrimination in terms of its influence on interpersonal outcomes (Maddox, 2004) and may provide more insight on how phenotype influences outcomes in the workplace. 2.4 Phenotype and Workplace Outcomes

Full document contains 124 pages
Abstract: Although there are conflicting empirical findings regarding whether race has a significant impact on individual outcomes in the 21st century workplace, African Americans are still less visible in top leadership positions than would be expected on the basis of population base rates (The Alliance for Board Diversity, 2005; Corporate Board Initiative, 2006; Fortune, 2006; McCoy, 2007; Thomas & Gabarro, 1999). One potential explanation for these equivocal results is that much of this research assumes that race has an equal impact within a given racial group. Little attention has been given within the management literature to intra-racial differences, which may influence how and to what extent racial stereotypes are activated. This dissertation examines the role of feature-based bias in promotion evaluations. Specifically, Afrocentric phenotype bias is explored to gain a better understand of how within-category, feature-based differences influence promotability evaluations. This dissertation utilizes self-categorization theory (SCT: Turner, 1985, 1987) and Heilman's (1983, 1995, 2001) lack of fit model to examine the consequences of Afrocentric phenotype bias within the employment context. Finally, the availability of cognitive resources is investigated as a potential moderator of the influence of Afrocentric phenotype bias on promotion-related rating.