a

 

1

 

1
L.

           

 

   

:

flaw”
- {I . ‘
.2 s 8

L50} «we ‘ v. _

..
. .
1.1-...
. ‘ ha
kart
it“ I

an A
a .wammwg

 

  

3.1,»;

2%.

  

‘ 'I l
.SaWfimM... ,
.. . “bun..."shu: ides .
..! i... a I p 4 F. n TILH .2”:
.¢..u5rw-anw«a , H.393 . ‘
. , .

 

 

          

 

    

 

 

     

      

 

€32.91 x ,. ,a,
:1... 2.} ANN

V fiafih’é‘sfflufif‘)‘ 9, )
nut: 8 iii 1"»

f :3- 1: . 1..
’43... 3 .11..

.31: 51.1

  

 

       

..n 5
, ‘ 5‘
“.5569? ..

 

      

  

as... .3
.. dink“ .

0‘

     

 

.
3: 4:1:-
. 3...: 3.! .

 

  

firs}
yak:

V. v7!

    

  

~II’\£-I.l§’i;
5....9.L...%z;..n. ‘
.. C...rt....2.v.c...inmmflvrt
.. .2 $1.». unalnwghfin
, t he}... ..
n. £58.).

)3. a...
1‘. Din.

 

 

          

A 3. .
x 1...: .luln...
{iii}. Attaébl
att||~.!, 1‘}: ,

          

 

1:152:62. i... 2'3...
L... salsgifisl‘fl
.- » ;. . .

\l!!r 4" I. \‘l‘ .

 

  

‘ x.
3‘...» It!!!
galsiyii I!§t¢l&sD-II!.
cu,us;r.w
t t .2 V «I... ‘ 9
«In... 95"....“ ...§.( xvusr I is? ’2‘
.i1iiufiallp Site}.
Ill.- r|

     

  

3"
ill}. It???

 

          

 

. 232 rt! z!
2v: 5?. . O
.t. with: Ll:\i\&..|.6l.ll§!
¥?l§sl¥i\s . ‘
If}!
15..-}.

(‘AQV‘stbleVOOI
20’ .l ‘a...(- ll;- 30"..1“
'1. a . Y n’;. .b-Qflvytllu

 

 

 

This is to certify that the
dissertation entitled

THE EFFECTS OF BLACKS’ PHYSICAL
CHARACTERISTICS ON WHITES’ EVALUATIONS OF
BLACKS AND BLACKS' EXPERIENCES WITH
DISCRIMINATION: SEPARATING THE EFFECTS OF
FACIAL FEATURES FROM SKIN TONE

presented by

Nao Hagiwara

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Psychology

 

 

124/414 4%,. ri/M/w

Major Professor’ 5 Signature
5/ 7/ /0
/ /

Date

MSU is an Affirmative Action/Equal Opportunity Employer

 

 

c-.—-—-a-o-u—--_-—-u-n-n-o—.-.-.-—-.-—.-c-o---o-—.-.-.-.-.—._.-._.-.-.—.-.-.-.-.-,-,-.-.-.-.-.-,-.-._.-._._._.-.-

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:lej/Achres/ClRC/DateDue.indd

 

 

THE EFFECTS OF BLACKS’ PHYSICAL CHARACTERISTICS ON WHITES’
EVALUATIONS OF BLACKS AND BLACKS’ EXPERIENCES WITH
DISCRIMINATION: SEPARATING THE EFFECTS OF FACIAL FEATURES FROM
SKIN TONE
By

Nao Hagiwara

A DISSERTATION
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILSOPHY
Psychology

2010

ABSTRACT
THE EFFECTS OF BLACKS’ PHYSICAL CHARACTERISTICS ON WHITES’

EVALUATIONS OF BLACKS AND BLACKS’ EXPERIENCES WITH

DISCRIMINATION: SEPARATING THE EFFECTS OF FACIAL FEATURES FROM
SKIN TONE
By
Nao Hagiwara

People in the same social groups share set of physical characteristics. However,
even within the same groups, there is real variation in physical characteristics. Two
literatures have been developed to determine how perceivers respond to differences in
physical characteristics among Black targets. Research on skin tone suggests that people
rely on skin tone to make inferences of Black individuals’ attributes. Research on
Afrocentric features suggests that facial features, such as eye color, shapes of nose and
lips, and texture of hair, in addition to skin tone, influence people’s perceptions of and
reactions to Black targets. Although these two lines of research argue that one type of
physical characteristic may be more important than the other when predicting people’s
perceptions of and reactions to Black targets, they have failed to separate these two
physical characteristics. Four studies were conducted to examine how variation in Black
target’s skin tone and facial features influences people’s affective (Studies 1, 3b, and 4)
and cognitive (Studies 2, 3b, and 4) reactions. This research also examined whether
variation in their own appearance relates to Blacks’ experiences with discrimination
(Study 4). Results showed that skin tone (lighter versus darker) and facial features (less
prototypical of Blacks versus more prototypical of Blacks) had additive effects on

Whites’ feelings toward Black men and that they affected how much Whites disliked

Black men at both implicit and explicit levels. The effects of skin tone and facial features

were different when Black targets were women. Skin tone affected how much Whites
liked Black women, whereas facial features had no effect on Whites’ feelings toward
Black women. The effects of Skin tone and facial features were also moderated by
perceiver race (White versus Black), and this was true only at implicit level. There was
little evidence that skin tone and facial features influenced people’s cognitive reactions
(i.e., stereotype activation) toward Blacks, but problems with measure validity prevent
clear conclusions with respect to this point. There was some evidence that Blacks’
experiences with discrimination, sensitivity to racism, and tendency to attribute
ambiguous negative events to racism were influenced by their own skin tone and facial
features. However, due to low statistical power, additional data is required to further
examine the nature of the skin tone and facial features effects on Blacks’ own
experiences with racism. The present studies contribute to a better understanding of the
effects of Black targets’ physical characteristics on people’s perceptions of and reactions
to them, by pointing to the benefits and importance of identifying factors that explain
who, when, and under what circumstances targets are more likely to experience prejudice

and discrimination.

I would like to dedicate this to my family.

iv

ACKNOWLEDGEMENTS

I would like to thank my committee members, Drs. Deborah Kashy, Joseph
Cesario, Nobert L. Karr, and Gwen M. Wittenbaum for their time and thoughtful
comments on my dissertation project. I would like to especially acknowledge the patience
and support that Dr. Deborah Kashy provided me throughout the graduate training at
Michigan State University. She has offered much guidance not only in academics but also
in my personal life when I was going through major life transitions as well as hardships.
Without her support and guidance, 1 would not be here.

I must thank my family for their support, especially my parents, Junichi and Mari
Hagiwara, who have always believed in me. I would also like to thank my son, Akiharu
Ethan Livingston, and my husband, Brian L. Livingston, who have given me a constant
companionship while I was working on my dissertation. My son’s smile at the end of the

day has always energized me and encouraged me to keep going.

I owe thanks to my undergraduate research assistants in the Cesario Laboratory
who helped me with participant recruitment and data collection. I also want to thank
graduate students in the Social/Personality interest group for their support. I would like
to especially acknowledge Michael Mott for helping me with the analyses of target

stimuli.

TABLE OF CONTENTS

LIST OF TABLES ................................................................................... x
LIST OF FIGURES ................................................................................ xiv
INTRODUCTION .................................................................................. l
The Effects of Skin Tone .................................................................. 4
The Effects of Afrocentric Features. .9
Unexamined Gender Effects In the Current AfrOcentrrc Feature Research ......... l8
Unexamined Target Effects In the Current Afrocentric Feature Research .......... 21
CURRENT RESEARCH .................................................................................................. 23
PILOT STUDY ................................................................................................................ 24
METHOD ........................................................................................................................ 25
Procedures ............................................................................................................ 25
RESULTS ........................................................................................................................ 28

Steps 1 & 2 (Examinations of Natural Skin Tone and Natural Facial Features)...29
Step 5 (Examinations of Skin Tone Selection and Facial Feature Manipulation).29

STUDY 1 .......................................................................................................................... 31
METHOD ......................................................................................................................... 31
Participants ............................................................................................................ 32
Procedures .............................................................................................................. 32
Measures ............................................................................................................... 32
COMPUTING FACILITATION SCORES ...................................................................... 37
RESULTS ....................................................................................................................... 38
Implicit Negative Attitudes toward Black Men .................................................. 38
Implicit Positive Attitudes toward Black Men .................................................... 39
Black vs. White Targets ...................................................................................... 4O
Explicit Affective Reactions to Black Men ......................................................... 41
DISCUSION ................................................................................................................... 4]
STUDY 2 ........................................................................................................................ 43
METHOD ....................................................................................................................... 43
Participants ......................................................................................................... 44

vi

Procedures ............................................................................................................ 44

Measures .............................................................................................................. 45
COMPUTING RESPONSE LATENCIES ...................................................................... 46
RESULTS ........................................................................................................................ 47

Stereotype Activation among Whites .................................................................. 47

Black vs. White Targets ....................................................................................... 48

Explicit Affective Reactions to Black Men ......................................................... 49
DISCUSION .................................................................................................................... 50
STUDY 3 ......................................................................................................................... 51
STUDY 3a ....................................................................................................................... 52
METHOD ........................................................................................................................ 52

Participants .......................................................................................................... 52

Procedures ............................................................................................................ 53
RESULTS ........................................................................................................................ 53

Implicit Negative Attitudes toward Black Women ............................................. 53

Implicit Positive Attitudes toward Black Women ............................................... 54

Black vs. White Targets ....................................................................................... 55

Explicit Affective Reactions to Black Women .................................................... 55
STUDY 3b ....................................................................................................................... 56
METHOD ........................................................................................................................ 56

Participants .......................................................................................................... 56

Procedures ............................................................................................................ 56
RESULTS ........................................................................................................................ 57

Stereotype Activation among Whites .................................................................. 57

Black vs. White Targets ...................................................................................... 58

Explicit Affective Reactions to Black Women ................................................... 59
DISCUSION .................................................................................................................... 6O
STUDY 4 ......................................................................................................................... 63
METHOD ........................................................................................................................ 63

Participants .......................................................................................................... 63

Procedures ............................................................................................................. 63

Measures .............................................................................................................. 65

Pre-Laboratory Questionnaires ................................................................ 65

vii

Measures Used in the Laboratory Session ............................................... 68

CODING OF PARTICIPANTS’ PHOTOS ..................................................................... 70
ANALYSIS PLAN .......................................................................................................... 71
RESULTS ........................................................................................................................ 73
Attributions of Ambiguous Negative Events ....................................................... 74
Attribution to Racism ............................................................................... 74
Attribution to Sexism ............................................................................... 75
Attribution to Skin Tone .......................................................................... 75
Attribution to Facial Features .................................................................. 77
Attribution to Internal Factors ................................................................. 78
Attribution to External Factors ................................................................ 78
Experiences with Racial Discrimination ............................................................. 79
Race-Based Rejection Sensitivity ........................................................................ 80
Other Criterion Variables ..................................................................................... 83
Implicit Affective Reactions to Black Men ......................................................... 83
Implicit Negative Attitudes ...................................................................... 84
Implicit Positive Attitudes ....................................................................... 85
Black vs. White Control .......................................................................... 85
Implicit Affective Reactions to Black Women ................................................... 86
Implicit Negative Attitudes ..................................................................... 86
Implicit Positive Attitudes ....................................................................... 86
Black vs. White Control .......................................................................... 87
Explicit Affective Reactions to Fellow Ingroup Members .................................. 87
Liking toward Black Men ........................................................................ 87
Liking toward Black Women ................................................................... 88
Stereotype Activation among Blacks following Black Male Targets .................. 88
Stereotype Activation among Blacks following Black Female Targets .............. 89
DISCUSION .................................................................................................................... 92
Attributions for Ambiguous Negative Events ..................................................... 92
Experience with and Sensitivity to Racism ......................................................... 95
The Effects of Black TargetS’ Characteristics on Black Perceivers’ Attitudes
toward Them ........................................................................................................ 96
The Effects of Black Targets’ Characteristics on Stereotype Activation for Black
Perceivers ............................................................................................................. 97
GENERAL DISCUSSION .............................................................................................. 98
Independent Effects of Skin Tone and Facial Features ....................................... 99
The Effects of Target Gender ............................................................................. 100
The Effects of Perceiver Race ............................................................................ 101
Implicit Attitudes-Explicit Attitudes Relation ................................................... 102
Black Person’s Own Physical Characteristics and Their Experiences with
Racism ................................................................................................................ 103

viii

Limitations and Future Directions ...................................................................... 106

CONCLUSIONS ........................................................................................................... 1 13
FOOTNOTES ................................................................................................................ 114
APPENDICES ............................................................................................................... 168
Appendix A: Objective Measures of Facial Features .......................................... 168
Appendix B: An Example of Outlined Pictures ................................................. 169
Appendix C: The Final 64 Black Male Target Pictures ..................................... 170
Appendix D: The Final 8 Black Female Target Pictures ................................... 172
Appendix E: A Design of Phase 4 in the Sequential Priming Task ................... 173
Appendix F: Selection of the Target Words for the Lexical Decision Task ...... 174
Appendix G: A List of Pre-Laboratory Questionnaires ..................................... 178
Appendix H: Selection of Attribution Items and Scenarios for the Attribution
Measure in Study 3 ............................................................................................ 185
Appendix I: Race-Based Rejection Sensitivity ................................................. 195
Appendix J: Example of Average Skin Colors .................................................. 199
REFERENCES ............................................................................................................. 200

ix

LIST OF TABLES

Table 1. Means and Standard Deviations of Subjective Ratings and Objective Measures
of Black Faces by Race .................................................................................................. l 17

Table 2. Correlations among Subjective Ratings and Objective Measures of Black Faces
for Black Men ................................................................................................................. 1 l 8

Table 3. Correlations among Subjective Ratings and Objective Measures of Black Faces
for Black Women ............................................................................................................ 1 19

Table 4. Means, Standard Deviations, and Mean Diflerences of Subjective and Objective
Measures of Black Men ’s Physical Features for Dark vs. Light Skin Tone
Condition ....................................................................................................................... 1 20

Table 5. Means, Standard Deviations, and Mean Diflerences of Subjective and Objective
Measures of Black Women ’s Physical Features for Dark vs. Light Skin Tone
Condition ....................................................................................................................... l 21

Table 6. Mean Differences in Prototypicality Ratings between Strong Facial Feature
Faces and Weak Facial Feature Faces ......................................................................... 122

Table 7. Means and Standard Deviations of Facilitation Scores for the Negative Words
as a Function of Black Men ’s Skin Tone and Facial Features ...................................... 123

Table 8. Means and Standard Deviations of Facilitation Scores for the Positive Words as
a Function of Black Men 's Skin Tone and Facial Features .......................................... 124

Table 9. Means and Standard Deviations of White Perceivers ’ Explicit Liking toward
Black Men as a Function of Skin Tone and Facial Features (Study 1) ......................... 125

Table 10. Means and Standard Deviations of Response Latencies for the Stereotype and
Non-Stereotypic Words as a Function of Black Men ’s Skin Tone and Facial F eature.. 126

Table 11. Means and Standard Deviations of White Perceivers ’ Explicit Liking toward
Black Men as a Function of Skin Tone and Facial Features (Study 2) ......................... 128

Table 12. Means and Standard Deviations of Facilitation Scores for the Negative Words
as a Function of Black Women ’s Skin Tone and Facial Features ................................. 129

Table 13. Means and Standard Deviations of Facilitation Scores for the Positive Words
as a Function of Black Women ’s Skin Tone and Facial Features ................................. 130

Table 14. Means and Standard Deviations of White Perceivers ’ Explicit Liking toward
Black Women as a Function of Skin Tone and Facial Features (Study 3a) .................. 131

Table 15. Means and Standard Deviations of Response Latencies for the Stereotype and
Non-Stereotypic Words as a Function of Black Women ’s Skin Tone and Facial
Features ......................................................................................................................... 1 32

Table 16. Means and Standard Deviations of White Perceivers ’ Explicit Liking toward
Black Women as a Function of Skin Tone and Facial Features (Study 3b) .................. 134

Table 17. Means, Standard Deviations, and Correlations among the Predictor Variables
for Black Male Participants ’ Pictures ........................................................................... 135

Table 18. Means, Standard Deviations, and Correlations among the Predictor Variables
for Black Female Participants ' Pictures ....................................................................... 136

Table 19. Means, Standard Deviations, and Correlations among All the Major Criterion
Variables ........................................................................................................................ 1 37

Table 20. The Eflects of Skin Tone, Facial Features, and Participants ' Gender on
Attribution to Racism ..................................................................................................... 138

Table 21. The Effects of Skin Tone, Facial Features, and Participants ' Gender on
Attribution T o Sexism ..................................................................................................... 139

Table 22. The Eflects of Skin T one, Facial Features, and Participants ' Gender on
Attribution T 0 Skin Tone ................................................................................................ 140

Table 23. The Eflects of Skin Tone, Facial Features, and Participants ’ Gender on
Attribution T o Facial Features ...................................................................................... 141

Table 24. The Eflects of Skin Tone, Facial Features, and Participants ’ Gender on
Internal Attribution ......................................................................................................... 142

Table 25. The Effects of Skin Tone, Facial Features, and Participants ' Gender on
External Attribution ....................................................................................................... 143

Table 26. The Effects of Skin Tone, Facial Features, and Participants ' Gender on
Experiences With Racial Discrimination In The Past One Year ................................... 144

Table 27. The Effects of Skin Tone, Facial Features, and Participants ' Gender on
Lifetime Experiences With Racial Discrimination ......................................................... 145

Table 28. The Effects of Skin T one, Facial Features, and Participants ’ Gender on Race-
Based Rejection Sensitivity ............................................................................................ 146

xi

Table 29. The Eflects of Skin Tone, Facial Features, and Participants ’ Gender on Racial
Identification .................................................................................................................. 147

Table 30. The Effects of Skin Tone, Facial Features, and Participants ' Gender on
Perception Of Discrimination ........................................................................................ 148

Table 31. The Effects of Skin Tone, Facial Features, and Participants ' Gender on Overall
Self-Esteem .................................................................................................................... 149

Table 32. The Eflects of Skin Tone, Facial Features, and Participants ' Gender on Social
Self-Esteem .................................................................................................................... l 50

Table 33. The Effects of Skin Tone, Facial Features, and Participants ’ Gender on
Appearance Self-Esteem ................................................................................................ 151

Table 34. Means and Standard Deviations of Facilitation Scores for the Negative Words
as a Function of Black Male Targets ' Skin Tone and Facial Features ......................... 152

Table 35. Means and Standard Deviations of Facilitation Scores for the Positive Words
as a Function of Black Male Targets ’ Skin Tone and Facial Features ......................... 153

Table 36. Means and Standard Deviations of Facilitation Scores for the Negative Words
as a Function of Black Female Targets ' Skin Tone and Facial Features ..................... 154

Table 37. Means and Standard Deviations of Facilitation Scores for the Positive Words
as a Function of Black Female Targets ' Skin Tone and Facial Features ..................... 155

Table 38. Means and Standard Deviations of Black Perceivers ' Explicit Liking toward
Black Male Targets as a Function of Skin Tone and Facial Features .......................... 156

Table 39. Means and Standard Deviations of Black Perceivers ' Explicit Liking toward
Black Female Targets as a Function of Skin Tone and Facial Features ...................... 157

Table 40. Means and Standard Deviations of Response Latencies for the Stereotype and
Non-Stereotypic Words as a Function of Black Male Targets ’ Skin Tone and Facial
Feature .......................................................................................................................... 158

Table 41. Means and Standard Deviations of Response Latencies for the Stereotype and

Non-Stereotypic Words as a Function of Black Female Targets ’ Skin Tone and Facial
Feature .......................................................................................................................... 160

Table I. Means Differences in Stereotypicality and Negativity Ratings between
Stereotypic Words and Non-Stereotypic Words within Each Gender ........................... 177

Table 11. Means and Standard Deviations of the Nine Attribution Items for Each of the 10
Scenarios ....................................................................................................................... 1 91

xii

Table 111. Internal Consistency across the Nine Attribution Items within Each of the 10
Scenarios ......................................................................................................................... 192

Table IV. Correlation between Internal and External Racism and Internal and External
Sexism Attribution Items across the 10 Scenarios .......................................................... 193

Table V. Internal Consistency across the 10 Scenarios within Each of the 10 Attribution
Items ................................................................................................................................ 194

xiii

LIST OF FIGURES

Figure 1. Subjectively Measured Black Participants’ Skin Tone Moderates the Effect of
Subjectively Measured Participant Facial Features on their Tendency to Attribute
Ambiguous Negative Events to Skin Tone ..................................................................... 162

Figure 2. Objectively Measured Black Participants’ Skin Tone Moderates the Effect of
Objectively Measured Participant Facial Features on their Tendency to Attribute
Ambiguous Negative Events to Facial Features ............................................................. 163

Figure 3. Subjectively Measured Black Participants’ Skin Tone Moderates the Effect of
Subjectively Measured Participant Facial Features on their Tendency to Attribute
Ambiguous Negative Events to Internal Factors ........................................................... 164

Figure 4. Subjectively Measured Black Participants’ Skin Tone Moderates the Effect of
Subjectively Measured Participant Facial Features on their Experiences With Racial
Discrimination in the Past 1 Year ................................................................................... 165

Figure 5. Objectively Measured Black Participants’ Skin Tone Moderates the Effect of
Objective] y Measured Participant Facial Features on their Sensitivity to Racism ........ 166

Figure 6. Subjectively Measured Black Participants’ Skin Tone Moderates the Effect of

Subjectively Measured Participant Facial Features on their Sensitivity to Racism ........ 167
Figure 1. Objective Measurements of Face Width, Face Length, Nose Width, and Lip

Thickness ........................................................................................................................ 168
Figure 11. An Example of Outlined Pictures ................................................................... 169
Figure 111. Black Male Target Pictures Used in the Main Studies ................................. 170
Figure IV. Black Female Target Pictures Used in the Main Studies .............................. 172
Figure V. Examples of Average Skin Colors ............................................................. - ..... 199

Images in this dissertation are presented in color.

xiv

The Effects of Blacks’ Physical Characteristics on Whites’ Evaluations of Blacks
and Blacks’ Experiences with Discrimination: Separating the Effects of Facial

Features from Skin Tone

People have mental representations (i.e., concepts) of a variety of social groups
(Kunda, 1999), and they use these concepts when categorizing target persons into social
groups and forming impression of target persons (Brewer, 1988; F iske, Lin, & Neuberg,
1999). For instance, people have concepts of how typical men and women should look in
terms of eyebrow Shape, cheekbones, stubble, and hair style, and use them when
identifying an individual’s sex (Brown & Perrett, 1993; Burton, Bruce, & Dench, 1993;
Goshen—Gottestein & Ganel, 2000; Macrae & Matrin, 2007; Martin & Macrae, 2007). In
fact, the major models of impression formation (e.g., see Brewer, 1988 for the “Dual
Process Model” and Fiske et al., 1999 for the “Continuum Model”) posit that
categorization of people into different social groups based on salient physical cues is the
first step in the process of impression formation. These models further argue that people
tend to form their impressions of a target person based on their attitudes toward the social
category to which the target person was categorized, even before getting know the person
(Brewer, 1998; Fiske et al., 1999).

Skin color plays an important role when people attempt to identify an individual’s
race (Brown, Dane, & Durham, 1998). In fact, the importance of skin color for making
racial distinctions is reflected in the long history of the use of color terminology to label
racial groups (i.e., Black, Brown, Red, White, and Yellow). Racial classification based on

these five colors has been used since Johann Friedrich Blumenbach came up with the five

color typology for humans in the 18th century (Gossett, 1997).

Interestingly, researchers have shown that people use skin color not only to
categorize people into different racial groups but also to make inferences about others’
attributes within the same racial groups (Dixon & Maddox, 2005; Maddox & Chase,
2004; Maddox & Gray, 2002; Montalvo, 2004; Uhlmann, Dasgupta, Greenwald, &
Swanson, 2002; Wade & Bielits, 2005). For instance, it is well documented that Black
individuals with darker Skin tone are perceived to possess more negative attributes, such
as aggressiveness and loudness, as compared to Black individuals with lighter Skin tone
(Dixon & Maddox, 2005; Maddox & Chase, 2004; Maddox & Gray, 2002; Wade &
Bielits, 2005). This phenomenon is often referred to as colorism (Hunter, 2002). Blacks
are not the only social group who face colorism. Incidents of colorism have been
observed among Hispanic Americans in the United States (Banerji, 2006; Hill, 2002;
Hochschild, 2007; Morales, 2008; Russell, Wilson, & Hall, 1992; Uhlmann et al., 2002)
as well as in societies outside the United States, such as Brazil (Harnandez, 2006), South
Asia (Rondilla & Spickard, 2007), and Arabic Speaking countries (Mukhtar, 2007).

Are there other physical cues that influence people’s impressions of others within
the same racial groups? Facial features (e.g., nose, lips, cheekbones, eyes) may be such
cues. Indeed, several researchers have recently found evidence that people also make
inferences about Black individuals’ characteristics based on facial features, such as nose,
lips, eyes, hair texture, in addition to skin tone (Blair, 2006; Blair, Chapleau, & Judd,
2005; Blair, Judd, & Fallma, 2004b; Blair, Judd, Sadler, & Jenkins, 2002; Eberhardt,
Goff, Purdie, & Davies, 2004; Maddox, 2004; Oliver, Jackson, Moses, & Dangerfield,
2004). Some researchers refer to these additional characteristics together with skin tone

as “Afrocentric features.”

One problem with the Afrocentric feature research is that the definition of
Afrocentric features includes skin tone. In fact, Afrocentric features are often confounded
with skin tone. That is, Black individuals who have stronger Afrocentric features (i.e.,
fuller lips, a wide nose, and coarse hair) are likely to have darker skin tone, whereas
Black individuals who have weaker Afrocentric features (i.e., thin lips, a narrow nose,
and soft hair) are likely to have lighter Skin tone. Therefore, it is unclear in the current
Afrocentric feature literature whether the effects of Afrocentric features on Whites’
evaluations of and reactions to Black individuals are driven by facial features in addition
to skin tone, or by only skin tone.

It is important to separate facial features from skin tone for two reasons. First, if
the effects of Afrocentric features reported previously are due to skin tone, then
researchers are Simply reinventing the wheel and calling Skin tone “Afrocentric features.”
Second, if physical characteristics that influence people’s perceptions of and reactions to
Black individuals could be pinpointed, this information can be used for informing and
educating people about the factors that automatically influence attitudes toward and
discrimination against Black individuals.

The present studies aimed to examine whether facial features influence how
people perceive and react to Black individuals above and beyond the well-documented
skin tone effects. In the current research, facial features unconfounded with skin tone (i.e.,
lips and nose) are referred to as “facial features.” In contrast, when the term “Afrocentric
features” is used, it indicates that facial features are confounded with skin tone.

Beyond separating facial features from Skin tone, the present research addresses

three important research questions that will further our understanding of the effects of

multiple physical characteristics on impression formation within a certain racial group
(i.e., Blacks). First, it extends the previous Afrocentric features research by looking at
female Black targets. In fact, no one has ever examined how Black women’s Afrocentric
features influence people’s impressions of them. Previous studies have exclusively used
Black men as targets and measured how Whites respond to Black men who possess
different degrees of Afrocentric features. The current study investigates how people
respond to Black women with different degrees of skin tone and facial features. Second,
it also extends the previous Afrocentric features research by looking at how Blacks' own
physical characteristics influence their experiences with intergroup relations. If Whites do
differentially perceive and treat Blacks who vary in Skin tone and Afrocentric features,
how Blacks construe and approach intergroup relations may differ depending on these
physical cues. Previous researchers have focused exclusively on how Black targets are
perceived and treated by Whites and have failed to look at how Black targets respond to
differential perceptions and treatments by Whites. The present research will investigate
the effects of skin tone and facial features from the target’s perspective. Finally, the
present research further examines the effects of Skin tone and facial features by looking at
Black perceivers. Researchers have previously focused on how Whites react to Blacks
with different degrees of Afrocentric features, and no researchers to date have examined
how Blacks perceive and react to their fellow ingroup members who vary in Afi'ocentric
features. Thus, the current study investigates the effects of skin tone and facial features
within the same social group.

The Effects of Skin Tone

There is a large literature that has shown that skin tone has a strong influence on
how Black individuals are perceived and treated by both Whites and fellow Blacks
(Dixon & Maddox, 2005; Maddox & Chase, 2004; Maddox & Gray, 2002; Wade &
Bielits, 2005). The term “Skin tone” is used to describe the level of lightness or darkness
of Skin color (Maddox & Chase, 2004). Skin tone is often measured with Blacks' self-
ratings (Averhart & Bigler, 1997; Coard, Breland, & Raskin, 2001; Harvey, LaBeach,
Pridgen & Gocial, 2005; Klonoff & Landrine, 2000; Wade, 1996) or with perceivers’
subjective ratings (Hill, 2000; 2002; Keith & Herring, 1991; Thompson & Keith, 2001),
both on Likert-type scales. Very few researchers use objective measures, such as the
percentage of light reflected from the surface of the Skin measured by a
spectrophotometer, to assess skin tone (Sweet, McDade, Kiefe, & Liu, 2007).

Previous studies have shown that both White and Black people perceive Black
individuals with darker skin tone more negatively, as compared to those with lighter skin
tone (Anderson & Cromwell, 1977; Averhadt & Bigler, 1997; Hall, 1992; Maddox &
Gray, 2002; Dixon & Maddox, 2005; Porter, 1991). For instance, Maddox and Gray
(Study 2, 2002) asked both White and Black participants to provide cultural beliefs about
“dark-skinned Black women,” “dark-skinned Black men,” “light-skinned Black women,”
and “light-skinned Black men.” They found that participants listed a greater number of
negative, compared to positive, attributes (e.g., criminal, lazy, aggressive, unintelligent)
for Black targets with darker Skin tone, whereas they listed a greater number of positive,
compared to negative, attributes (e.g., kind, educated, wealthy, motivated) for Black

targets with lighter skin tone, regardless of target gender.

In another study, Dixon and Maddox (2005) have Shown that Black men with
darker skin tone are associated with negative traits by demonstrating that participants
reported that Black men with darker skin tone were more memorable as a murderer. In
this study, participants (73% Whites) were asked to watch a lS-min news clip that
contained several stories. Among these stories was a 22-sec story about either a Black or
a White criminal who was being sought for the murder of a race-unidentified police
Officer. A picture of the perpetrator of this murder was displayed for 3 sec and was
described as “a White,” “a light-skinned Black,” “a medium-skinned Black,” or “a dark-
skinned Black” man. Participants were then asked to respond to a series of dependent
measures that included a measure of memory (i.e., how memorable each character Shown
in the news clip was). The results showed that the dark-skinned Black perpetrator was
more memorable than the White perpetrator as a murderer.

Averhadt and Bi gler (1 997) have further shown that even Black children have
biased memory based on skin tone. Specifically, Black children could remember
stereotypic information about skin tone better than counterstereotypic information.
Averhadt and Bigler (1997) had Black children listen to Six stories in which Black
individuals with different levels of Skin tone were described in terms of either positive or
negative traits (e. g., honest/dishonest, nice/mean, rich/poor, clean/dirty) and high or low
status occupations (e.g., a company president/janitor, a lawyer/parking lot attendant, a
factory owner/factory worker, a ballet dancer/maid). The results demonstrated that Black
children Showed better memory for stories that depicted Black individuals with dark skin
tone with negative traits and in occupations of lower status than the stories that depicted

them with positive traits and in occupations of higher status. Likewise, Black children

showed better memory for stories that depicted Black individuals with light skin tone
with positive traits and in occupations of higher status than stories that depicted them
with negative traits and in occupations of lower status. Taken together, these results
indicate that people, whether White or Black, associate positive traits and higher status
among Black individuals with lighter skin tone and negative traits and lower status with
Black individuals with darker skin tone.

Previous research has also shown that people treat Black individuals with darker
Skin tone more negatively, as compared to Black individuals with lighter skin tone (Hall,
2003; 2005; Wade, Romano, & Blue, 2004). For instance, Wade et a1. (2004) had White
participants review a set of application materials, ostensibly completed by one of four
Black applicants (fair-Skinned or dark-Skinned, man or woman), that included the
applicant’s résumé and a written description of the applicant. Participants were told that
their task was to take the role of a manager and evaluate the applicant. Wade et a1. (2004)
found that participants were more willing to accept fair-skinned Black individuals than
dark-skinned Black individuals, regardless of the gender of the applicant.

A difference in Skin tone among Black individuals is associated with economic,
vocational, and educational status as well (Blackwell, 1985; Frazier, 1966; Hill, 2000;
Hughes & Hertel, 1990; Keith & Herring, 199]). For instance, Keith and Herring (1991)
looked at the relation between skin tone and educational attainment, occupation, and
income, using the National Survey of Black Americans. In this data set, skin tone was
assessed by the interviewer’s subjective rating, using a scale ranging from 1 (a very dark
brown skin color) to 5 (a very Ii ght brown or very light Skin). The study Showed that

Black individuals with lighter skin tone had higher education, occupational status, and

income, as compared to Black individuals with darker skin tone, when controlling for
gender, age, marital status, parental socioeconomic status, region of residency, and
urbanicity. Similarly, in his archival study, Hill (2000) has Shown that skin tone predicted
the occupational status of Black men. In this study, participants’ skin tone was
determined by the ratings provided by census enumerators on a dichotomous scale (in the
1920 census): Blacks of “full blood” vs. Blacks “having some proportion of white
blood.” Then, participants’ occupational status was determined by the information
provided on the US. Standard death certificate (collected between 1980 and 1985). The
results showed that Black men who were identified as Blacks having some proportion of
White blood (lighter skin tone) were more likely to have attained a white collar
occupation, which includes professional, managerial, sales, and clerical workers, than
Black men who were identified as Blacks of full blood (darker skin tone).

Differential treatments of Black individuals based on their skin tone are further
reflected in their reports of experiences with racial discrimination. Klonoff and Landrine
(2000) asked Black participants to complete a questionnaire that included the Schedule of
Racist Events (SRE: Landrine & Klonoff, 1996; Klonoff & Landrine, 1999) and a
measure of skin tone. The SRE assesses both the frequency and types of racial
discrimination as well as people’s appraisal of the discriminatory events. The results
showed that Black individuals with darker skin tone reported experiencing racial
discrimination 11 times more often and appraising their experiences with racial
discrimination as more stressful, as compared to Black individuals with lighter skin tone.

In addition to the self-reported experiences with discrimination, several studies

have used more objective measures, such as average blood pressure, to show that

experiences of Black individuals with darker skin tone are different fiom those of Black
individuals with lighter skin tone (Boyle, 1970; Dressler, I991; Harburg, Gleibermann,
Roeper, Schork, & Schull, 1978; Sweet et al., 2007). For instance, Sweet et a1. (2007)
demonstrated that income was negatively associated with average blood pressure among,
Black individuals with lighter Skin tone (paralleling the well-documented pattern among
Whites), whereas income was not associated with average blood pressure among Black
individuals with darker Skin tone. These results indicate that the experience of Black
individuals with lighter skin tone are more similar to Whites (at least with respect to
blood pressure) and that the protective effects of economic success may not be available
to Black individuals with darker skin tone.

Taken together, a large body of research on skin tone has demonstrated that
people perceive and react differently to Black individuals with different levels of Skin
tone: people perceive and react more negatively to Black individuals with darker skin
tone than to Black individuals with lighter Skin tone. However, is skin tone the only
physical characteristic people use when making their judgments Of Black individuals? Do
people use more subtle physical characteristics, such as facial features, to form
impressions of and react to Blacks?

The Effects of Afrocentric Features

An increasing number of researchers have argued that people use facial features in
addition to Skin tone when they form impressions of Black individuals (Blair, 2006; Blair
et al., 2005; Blair et al., 2004b; Blair et al., 2002; Eberhardt et al., 2004; Oliver et al.,
2004). Research has shown that Afrocentric features, defined as visible physical

characteristics that differentiate Blacks from other racial groups, such as filler lips, a

wider nose, more coarse hair, and darker eye and skin color (Blair, 2006; Blair et al.,
2005; Blair, Judd, & Chapleau, 2004a; Blair et al., 2004b; Blair et al., 2002; Eberhardt et
al., 2004; Eberhardt, Davies, Purdie-Vaughns, & Johnson, 2006), influence how Black
men are perceived by Whites (Blair, 2006; Blair et al., 2005; Blair et al., 2004b; Blair et
al., 2002; Eberhardt et al., 2004; 2006; Oliver et al., 2004).

More specifically, researchers have shown that Black men who have stronger
Afrocentric features are perceived more negatively, as compared to Black men with
weaker Afrocentric features. Livingston and Brewer (2002) showed this effect on an
implicit (i.e., unobtrusive) measure of attitudes by having participants complete a
sequential priming task designed to assess affective reactions to Black individuals. This
task assesses how quickly participants can identify positively-valenced and negatively-
valenced adjectives, when these adjectives are preceded by Black and White faces.
Responding more quickly to negatively-valenced adjectives following Black faces is
indicative of negative affective reactions to Blacks. The researchers found that White
participants responded to negative words more quickly after seeing Black male faces with
stronger Afrocentric features, as compared to after seeing Black male faces with weaker
Afrocentric features. Furthermore, Whites’ affective responses to Black male faces with
weaker Afrocentric features did not differ from their responses to White male faces.
Livingston and Brewer (2002) argue that these results indicate that Whites’ affective
responses to Black men were influenced by within-group differences in Afrocentric
features, rather than by between- group differences in race.

Blair and her colleagues (Blair, 2006; Blair et al., 2005; Blair et al., 2004b; Blair

et al., 2002) have firrther shown that Afrocentric features influence not only Whites’

10

affective responses to Black men, but also their cognitive responses to Black men (i.e.,
stereotype activation). In Blair et al.’s (2002) study, participants read scenarios that
differed along two crossed dimensions: valence of stereotypes and degree of
stereotypicality for Black people (i.e., positive stereotypic, positive non-stereotypic,
negative stereotypic, negative non-stereotypic). Participants were told to read each
scenario very carefully and try to form a mental image of what the person described in
the scenario might look like. After they read a given scenario, participants were presented
with Black male faces (one at a time on the computer screen) that differed in the degree
of Afrocentric features. They were instructed to estimate the probability that the person
presented on the computer screen was in fact the person who had been described in the
scenario. After completing the probability ratings for all the target faces for this scenario,
participants went through the same procedures for each of the remaining three scenarios.
Blair et al. (2002) reported that Black men with stronger Afrocentric features were judged
as significantly more likely to be the actor in the stereotypic scenarios than in the non-
stereotypic scenarios, and in negative scenarios than in positive scenarios. These results
suggest that White people associate Afi'ocentric features with behaviors that are
stereotypic of Black men in our society.

Blair and her colleagues further showed that the relation between Afrocentric
features and stereotype activation is robust and efficient (Blair, 2006; Blair et al., 2005;
Blair et al., 2004b). In a series of studies, Blair et al. (2004b) had participants complete
the same probability rating task described above while they gave participants clear
instructions to avoid using stereotypes associated with Black people in general (Study 2)

or stereotypes associated with Afi'ocentric features specifically (Study 3). They found that

11

White participants, regardless of whether they received the instructions or not, judged
Black men with stronger Afrocentric features as significantly more stereotypic of Blacks
than those with weaker Afrocentric features. White participants’ inability to control
stereotype activation based on Afrocentric features even with the instructions suggests
that the effects of Afrocentric features on their perceptions of Black men may be
triggered automatically.

Furthermore, researchers have shown that the effects of Afrocentric features on
stereotype activation can be found when participants are under high cognitive load (i.e.,
when controlled processing is difficult; Blair et al., Study 1, 2004b). Under high-load,
participants were told to pay attention to letters presented on the computer screen while
they completed the probability rating task and were further instructed to press the
spacebar every time they saw the letter sequence G-X—Q. Blair et al. (2004b) found no
difference between the high-load and no-load conditions. That is, in both conditions,
Black men with stronger Afrocentric features were perceived to possess more stereotypic
attributes than Black men with weaker Afrocentric features. Previous research has also
Shown that the effects of Afrocentric features on stereotype activation were found when
the Black male faces were presented up-side-down (Blair, 2006), Changing the
orientation of a face makes controlled processing difficult (Diamond & Carey, 1986;
Leder & Bruce, 2000; Valentine, 1988; 1991) In another study (Blair, 2006), participants
completed the same probability rating task either in the upright face condition or in the
inverted face condition. The results showed that Black men with stronger Afi'ocentric
features were stereotyped to a greater degree than Black men with weaker Afrocentric

features, regardless of face orientation. These findings indicate that stereotype activation

12

based on Afrocentric features is efficient, because this process operates even when the
task is made difficult by introducing cognitive load or changing the orientation of faces
(Bargh, 1994).

Afrocentric features have also been found to influence Whites’ behavioral
responses to Black men (Blair et al., 2004a; Eberhardt et al. 2006; Kahn, Davies,
Eberhardt, & Correll, 2008). Blair et al. (2004a) presented faces of Black and White male
prison inmates to participants and asked them to make a single, global assessment of the
degree to which “each face had features that are typical of African Americans.” They
found that inmates who were rated to possess stronger Afrocentric features had received
harsher sentences than those inmates who were rated to possess weaker Afrocentric
features, regardless of whether inmates were Black or White. This was true even when
controlling for criminal histories, attractiveness, and babyfacedness. Similarly, Eberhardt
et al. (2006) found that Black male defendants with stronger Afrocentric features were
more likely to be sentenced to death than their counterparts with weaker Afrocentric
features, when the victim was White. This result was found above and beyond a number
of factors that were found to be related to sentencing decisions (i.e., aggravating
circumstances, mitigating circrunstances, severity of the murder, the defendant’s
socioeconomic status, the victim’s socioeconomic status, and the defendant’s
attractiveness).

Behavioral responses that are less controlled, or more automatic, have also been
Shown to be affected by Afrocentric features. Kahn et al. (2008) had White participants
complete a shooter bias video game. The goal of this game is to shoot (by pressing

labeled keys) targets carrying a gun and not to shoot targets carrying neutral objects

13

(either a cell phone or a wallet). Kahn et al. (2008) used a Black man with stronger
Afrocentric features, a Black man with weaker Afrocentric features, and a White man as
targets. They found that participants were more likely to incorrectly shoot an unarmed
Black man with stronger, compared to weaker, Afrocentric features. Likewise,
participants were more likely to correctly Shoot an armed Black man with stronger,
compared to weaker, Afrocentric features. Taken together, these findings suggest that
Black men with stronger Afrocentric features elicit more negative behavioral responses at
both the explicit and implicit levels, as compared to Black men with weaker Afi'ocentric
features.

Previous studies appear to suggest that Afiocentric features have robust effects on
how White individuals feel about, think about, and treat Black men. However, it is
unclear whether the effects of Afrocentric features on Whites’ evaluations and treatments
of Black individuals are driven by facial features in addition to Skin tone or by skin tone
only, because the definition of Afrocentric features includes skin tone. Indeed, facial
features tend to covary with skin tone, such that Black men with stronger facial features
are likely to have darker skin tone, whereas Black men with weaker facial features are
likely to have lighter skin tone. If facial features and skin tone naturally covary, why iS it
important to separate the effects of facial features from skin tone?

Although facial features and Skin tone correlate with one another, the size of the
correlation is moderate, suggesting that there is still large variability among Black
individuals, such that some Blacks with stronger facial features have lighter skin tone,

whereas some Blacks with weaker facial features have darker Skin tone. The previous

14

research that does not separate the two effects cannot predict how people feel about and
treat these Black individuals with “mismatched” physical cues.

Additionally, there is a debate over which physical characteristics play more
important roles in determining people’s perceptions of Blacks. Some researchers have
found that facial features are more important than skin tone when making racial
distinctions (Deregowski, Ellis, & Shepherd, 1975; Sorce, 1979), whereas other
researchers have found that people explicitly use Skin tone over facial features to make
racial distinctions (Brown et al., 1998). For instance, Sorce (1979) had children freely
categorize a number of sketches of a male face displaying a variety of racial
characteristics (i.e., Skin tone, facial features) and non racial features (i.e., color of a shirt).
He found that White children used facial features (especially hair features) rather than
skin tone to classify the sketches. In another study, Deregowski et al. (1975) had
participants describe a number of color pictures of faces, which differed in gender and
race and counted the number of times that participants mentioned particular features of
each face. They found that White participants mentioned eye color, hair color, and hair
texture significantly more often than other features. These studies suggest that White
individuals tend to use facial features more than skin tone in interracial perceptions.

In contrast, Brown et al. (1998) found that both White and non-White participants
explicitly reported that skin color is the most important feature when they make decisions
about the race of other people, as compared to other features such as nose fullness and
nose placement on face. However, all these studies used self-report measures. As several
researchers have demonstrated (Blair, 2006; Blair et al., 2005; Blair et al., 2004b), the

process by which Afrocentric features influence people’s perceptions of and responses to

15

Black individuals is likely to be automatic. Self-report measures would not be sufficient
to address the question of whether people use facial features in addition to Skin tone when
they perceive Black individuals because they would not be aware of this process. In
addition, self-report measures are susceptible to social desirability. More specifically,
people may think that it is inappropriate to mention skin tone of other people, because
Skin tone is one of the most salient features that differentiate many racial groups.

More recently, some researchers have argued that Whites use Blacks’ facial
features and Skin tone when they make social judgments of Black people using implicit
measures (Blair, 2006; Blair et al., 2002; 2004b; 2005; Eberhardt et al., 2004; Maddox,
2008; Oliver et al., 2004). However, this argument has no empirical support because the
Operational definition of Afrocentric features used by social psychologists includes not
only full lips, a wide nose, coarse hair, and dark eye color, but also dark Skin tone (Blair,
2006; Blair et al., 2002; 2004a; 2004b; 2005; Eberhardt et al., 2004; 2006; Livingston &
brewer, 2002). The commonly used procedures to assess the degree of Afrocentric
features ask participants to rate the degree of Afrocentiricity of target pictures using skin
tone, nose, lips, eyes, and hair as indicators. For instance, Blair et al. (2002) told
participants that some individuals have features that are “more typical of Afiican
Americans than others, in terms of skin color, hair, eyes, nose, cheek, lips, etc.” and that
their task was to rate Afrocentric features for each Black male face they viewed on a
scale of 1 (not at all) to 9 (very much). Similarly, Livingston and Brewer (2002) told
participants that “judgments of African American prototypicality primarily entail skin
color but should also take into account facial features, eye color, hair, texture, and so on”

and that their task was to rate each Black male face on a scale of 1 (low prototypic) to 5

16

(high prototypic). Although participants in both studies were told to attend to facial
features in addition to skin tone, there is no way to test whether participants used all
available facial information or used only skin tone information when making Afrocentric
feature judgments because target pictures presented to participants included both types of
information.

In opposition to the researchers who argue that facial features in addition to skin
tone play an important role in Whites’ perceptions of Black individuals, several skin tone
researchers argue that people use skin tone as a primary cue in social perception and that
skin tone (without any other facial features) is sufficient to activate the “Black” racial
category and stereotypes associated with it (Dixon & Maddox, 2005; Maddox & Chase,
2004; Maddox & Gray, 2002). Maddox and his colleagues (Dixon & Maddox, 2005;
Maddox & Chase, 2004; Maddox & Gray, 2002) argue that because they manipulated
Skin tone while keeping other features constant, they can conclude that skin tone, not
additional facial features, is the primary cue in social perception. However, it is important
to note that in Dixon and Maddox’s study (2005) they used only one Black face and
manipulated its skin tone. In this case, there is only one level of facial features. Without
variation in facial features, whether additional features have additive or interactive effects
in social perception cannot be tested. In two other studies (Maddox & Chase, 2004;
Maddox & Gray, 2002), Maddox and his colleagues used six Black faces. Even though
there might have been variation in additional facial features among these six faces, they
did not measure the degree of variation in facial features. Therefore, it is still possible

that all Six faces were high or low in stereotypical facial features. In an absence of clear

17

differences in degree of facial features in these studies, whether facial features,
independent of skin tone, play a critical role in social perception cannot be examined.

Taken together, although both Afrocentric features researchers and skin tone
researchers have been trying to come up with an answer for the debate over “which
physical cues are more important for impression formation, Skin tone or facial features?,”
, neither of them successfully has answered the question, because they have failed to
separate empirically these two factors.
Unexamined Gender Effects in the Current Afrocentric Feature Research

Prior research on Afrocentric features has had a glaring omission, which is the
role of target gender. Existing studies have exclusively focused on Black men as targets,
and to my knowledge, no Afrocentric features research has ever used Black women as
targets. Therefore, it is unclear what the effects of Afrocentric features are for Black
female targets. It is important to examine the potential gender effect on Afrocentric
features for two reasons. First, gender differences have been reported in skin tone
research. A number of studies have shown that many effects of skin tone (e.g., on
perceptions of attractiveness, self-esteem, and self-efficacy) are different for Black men
and Black women (Coard et al., 2001; Hill, 2002; Maddox, 2004; Ross, 1997; Thompson
& Keith, 2001; Wade, 1996; Wade & Bielitz, 2005; Wade et al., 2004). Therefore, it is
reasonable to assume that some effects of Afrocentric features as well may be different
for Black men and Black women. Second, different perspectives, such as social
dominance theory (Sidanius & Pratto, 1999) and multiple-identity perspective (Buchanan

& Fitzgerald, 2008; Buchanan & Ormerod, 2002; Fleming, 1983; King, 2005; Landrine,

l8

Klonoff, Alcaraz, Scott, & Wikins, 1995; McLeod & Owens, 2004; Rederstorff,
Buchanan, & Settles, 2007; Settles, 2006), predict different gender effects.

Researchers have reported that the effects of skin tone can be different between
Black men and Black women in some situations (Bond & Cash, 1992; Coard et al., 2001;
Hill, 2002; Maddox, 2004; Ross, 1997; Thompson & Keith, 2001; Wade, 1996; Wade &
Bielitz, 2005; Wade et al., 2004). In particular, it is well documented that the effects of
Skin tone on perceptions of attractiveness and mating preferences differ between men and
women, such that Black men with darker Skin tone are perceived by fellow Black people
to be more attractive than Black men with lighter skin tone. In contrast, Black women
with lighter skin tone are perceived to be more attractive than Black women with darker
skin tone (Coard et al., 2001; Hill, 2002; Maddox, 2004; Ross, 1997; Wade, 1996; Wade
& Bielitz, 2005; Wade et al., 2004). Thompson and Keith (2001) have also shown that
skin tone predicted the level of self-efficacy for Black men, such that light skin tone was
associated with higher self-efficacy, but not for Black women. In contrast, skin tone
predicted the level of self-esteem for Black women, such that lighter skin tone was
associated with higher self-esteem, but not for Black men. Given that there are gender
differences in the effects of skin tone in some contexts, it is possible that there are gender
differences in the effects of Black individuals’ other physical characteristics in some
situations as well.

In addition, different theoretical perspectives predict different gender effects on
Afrocentric features. For instance, social dominance theory predicts gender differences,
such that Black men would experience the negative consequences of having stronger

Afrocentric features, whereas Black women would not. According to social dominance

l9

theory, arbitrary-set social hierarchy (e.g., hierarchy based on race, nation, or religion)
mainly concerns the control of subordinate men by coalitions of dominant men, which is
known as the “subordinate male target hypothesis” (Haley, Sidanius, Lowery, &
Malarnuth, 2004; Sidanius, Levin, Liu, & Pratto, 2000; also see Sidanius & Pratto, 1999;
and Pratto, Sidanius, & Levin, 2006 for extensive review). Indeed, it has been
demonstrated that men rather than women are the primary victims of racial discrimination
in a variety of contexts, such as the labor market, housing market, and criminal justice
system (Sidanius & Pratto, 1999). Based on this perspective, it can be predicted that
Black women would not experience negative treatment by White individuals based on
their Afrocentric features to the same extent as Black men do.

In contrast, a multiple-identity perspective predicts a different pattern of gender
effects. This perspective argues that individuals who possess more than one socially
disadvantaged goup membership (e. g., Black women, low socioeconomic status Latinos)
are likely to experience discrimination based on an aggregation of all disadvantaged
statuses (Buchanan & Fitzgerald, 2008; Buchanan & Ormerod, 2002; Fleming, 1983;
King, 2005; Landrine et al., 1995; McLeod & Owens, 2004; Rederstorff et al., 2007;
Settles, 2006). According to this perspective, Black women, who possess two socially
disadvantaged group memberships (i.e., race and gender), would experience both racism
and sexism (Buchanan & Fitzgerald, 2008). Thus, this perspective predicts that Black
women would be treated negatively by White individuals based on their Afrocentric
features to the same extent as Black men or to a greater degree than Black men due to the

combined effects of racism and sexism. Thus, it is unclear whether the effects of

20

Afrocentric features found in the existing research can be generalized to the entire
“Black” social category.
Unexamined Target Effects in the Current Afrocentric Feature Research

Existing studies on Afrocentric features have focused exclusively on the effects of
Afrocentric features on Whites as perceivers, ignoring two important processes involved
in intergroup relations. First, previous research has failed to investigate how one’s own
Afrocentric features influence Black individuals’ experiences with intergroup relations.
Second, no one has examined how Blacks as perceivers perceive and react to their fellow
ingroup members who vary in the degree of Afrocentric features.

The “target’s perspective” (Swim & Stangor, 1998) suggests that Black
individuals are active agents who contribute to the quality of interracial interactions and
that personal, situational, and structural factors influence how Black individuals perceive
and experience discrimination (Major & O’Brien, 2005; Major, Quinton, & McCoy, 2002;
Shelton & Richeson, 2006). Afrocentric features can be considered as personal factors
that may influence how Black individuals perceive and react to discrimination. More
specifically, because White individuals tend to treat Black men with stronger Afi'ocentric
features more negatively than Black men with weaker Afrocentric features (Blair et al.,
2004; Eberhardt et al. 2006; Kahn et a1. 2008), Black individuals with stronger
Afrocentric features may be able to report more personal incidents with racism, as
compared to their counterparts with weaker Afrocentric features. If they have
experienced more racism throughout their life, Black individuals with stronger
Afrocentric features might have become more sensitive to discrimination and may tend to

construe even ambiguous negative events in terms of their race. AS a result, Black

21

individuals with stronger Afrocentric features may have more negative perceptions of
interactions with Whites in general and may even be more likely to avoid interactions
with Whites, as compared to Black individuals with weaker Afrocentric features (i.e., a
self-firlfilling prophecy, Merton, 1968). So, it is important to examine how one’s own
physical characteristics influence Blacks’ feelings and thoughts in intergroup relations.

How people perceive and evaluate individuals from different social groups is
often different from how people perceive and evaluate individuals from their own group.
Specifically, people tend to perceive members from the same social group (i.e., ingroup
members) more positively than members from different social groups (i.e., outgroup
members; Ashbum-Nardo, Volis, & Monteith, 2001; Brewer, 1999; Perdue, Dovidio,
Gurtman, & Tyler, 1990). In fact, research using minimal groups demonstrates that
favorable perceptions and treatments of ingroup members over outgroup members occur
even when group membership is defined by meaningless characteristics, such as colors Of
shirts or pens (Brewer, I979; Otten & Wentura, 1999: Tajfel, Billing, Bundy, & Flament,
1971). Given these well-documented findings, how Blacks feel about fellow ingroup
members with different levels of skin tone and facial features may be different from how
Whites feel about Black targets. Thus, it is important to examine whether the effects of
physical features on people’s affective and cognitive reactions are the same for ingroup
and outgroup perceivers.

To conclude, there are several gaps in the current literature. First, it is unclear
whether there are facial features effects on Whites’ reactions to Blacks above and beyond
Skin tone effects. Second, no study to date has looked at whether the effects of physical

characteristics on White individuals’ perceptions are same for Black male targets and

22

Black female targets. Third, it is unknown how Black individuals’ physical
characteristics influence their own experiences with discrimination and perceptions of
interracial interactions with White individuals. Additionally, no one has ever examined
how Black individuals perceive and react to fellow ingroup members who vary in
physical characteristics. The current studies aim to fill these gaps.

Current Research

In the current project, four studies were conducted to address four research
questions: 1) Do facial features have effects on Whites’ affective and cognitive reactions
to Black individuals above and beyond Skin tone effects? 2) Does target gender moderate
the effects of physical features (i.e., skin tone and facial features) on Whites’ perceptions
of Blacks? 3) Are the effects of physical features on Black perceivers different from
White perceivers? and 4) What are the effects of a Black individual’s physical features on
his or her perceptions of and experiences with discrimination?

Study 1 was designed to address question 1. In this study, White individuals’
affective reactions, at both implicit and explicit levels, to Black men who vary in facial
features and skin tone were examined. Study 2 was also designed to address question 1.
In this study, Whites’ cognitive reactions (i.e., stereotype activation), instead of affective
reactions, to Black men who differ in their facial features and skin tone were examined. It
is important to note that for both studies 1 and 2, facial features of Black male targets
were manipulated independently from skin tone, and therefore these studies were able to
examine whether there are facial feature effects above and beyond skin tone and to probe
for interactions between these variables. Study 3 was designed to address question 2. In

this study, Whites’ affective and cognitive reactions to Black women, instead of Black

23

men, who vary in skin tone and facial features were examined. Finally, Study 4 was
designed to address questions 3 and 4. This study examined whether Black individuals
with more prototypical facial features, darker Skin tone, or a combination of the two, as
compared to those with less prototypical facial features, lighter Skin tone, or a
combination of the two, are more likely to attribute ambiguous negative events to racism,
report experiences with racial discrimination, and be sensitive to racism. In addition,
Study 4 also examined how Black individuals affectively and cognitively react to their
fellow ingroup members who vary in skin tone and facial features. Prior to these four
major studies, a pilot study was conducted to create the target stimuli, which are pictures
of Black men and Black women with different levels of skin tone and with manipulated
facial features.
Pilot Study

The purpose of this pilot study was to create four groups Of target faces of Black
men and Black women that would be used in the studies: dark-skinned faces with more
prototypical facial features, dark-Skinned faces with less prototypical facial features,
light-skinned faces with more prototypical facial features, and light-skinned faces with
less prototypical facial features. First, pictures were selected based on their actual skin
tone---a group of pictures with darker skin tone was naturally and objectively different
from the other group of pictures with lighter skin tone. Then, facial features of each of the
selected pictures were digitally manipulated, so that one group of pictures had more
prototypical facial features and the other group of pictures had less prototypical facial
features. Manipulation of facial features ensured that facial features are separate item

skin tone in these pictures.

24

Method
Procedures
The pilot study consisted of five steps. Step 1 involved creation of the pool of
potential target pictures. Pictures of 120 college-age Black men and 50 college-age Black

women were collected from several university athletic web Sites, the Productive Aging

. 1
Lab Face Database (Minear & Park, 2004), and the MSU campus community. All

pictures were in a standard head-and-shoulder format, and all faces had a neutral facial
expression, with no visible accessories (e.g., eye glasses, hats). Then, pictures were
digitally edited so that all background features were replaced by a solid gray background,
and all clothing was replaced with a plain black t-shirt. Once the pictures were edited,
four pieces of facial information (i.e., facial width and length, nose width, and lip
thickness) were obtained in millimeters (see Appendix A). Then, the ratio of nose width
(i.e., nose width divided by face width) and lip thickness (i.e., lip thickness divided by
face length) were computed for each face. The average luminosity was also computed for
each face by averaging the luminosity value obtained from each pixel across a given face.
Therefore, all 170 pictures had the following information: the ratio of nose width, the
ratio of lip thickness, and the average luminosity.

In Step 2, 170 pictures were presented to a group of White participants to assess
perceived racial group membership, attractiveness, babyfacedness, hostility, and degree
of overall Afrocentric features (i.e., the ratings used by previous Afrocentric features
researchers). Fourteen White participants were asked to categorize Black men and 35
White participants were asked to categorize Black women into one of the three possible

categories: Black, White, or Racially Ambiguous. This ensured that all the faces used in

25

the main studies would be perceived as being unambiguously Black. Then, 190 White

participants rated Black men, and 35 White participants2 rated Black women in terms of

their Afrocentric features, attractiveness, babyfacedness, and hostility. The order of the
four ratings was randomized across participants. In addition, the presentation of the faces
within each rating category was randomized across participants. The ratings of
attractiveness were used when selecting dark vs. light skin tone faces to ensure that these
two groups did not differ in attractiveness. Previous studies have found that attractiveness
covaries with Afrocentric features (Blair et al., 2002) and several researchers have
controlled for attractiveness when selecting their target stimuli (Kahn et al., 2008;
Livingston & Brewer, 2002). The ratings of hostility and babyfacedness were also
obtained because previous studies have either controlled for them in stimuli selection or
measured them for use as important variables that may covary with Afrocentric features
(Blair et al., 2002; Livingston & Brewer, 2002).

Step 3 involved the selection of target pictures that would be used in each of the
three main studies. First, the average luminosity was used to select 20 dark-skinned and
20 light-skinned faces for Black male targets in order to create dark vs. light skin tone
conditions. Because there were only 50 Black female pictures to begin with, all female
pictures were still used in the following steps. Next, the ratios of nose width and lip
thickness of each of the 40 selected male faces and all female faces were digitally altered,
using Adobe Photoshop CS3, to create more vs. less prototypical facial feature conditions
using the grand means and standard deviations (computed separately for Black men and
Black women) of the ratio of nose width and the ratio of lip thickness computed in Step 1.

For the more prototypical facial features condition, the nose and lips from each picture

26

were manipulated to reflect ratios of nose width and lip thickness that were one standard
deviation above the grand mean. Likewise, for the less prototypical facial features
condition, the nose and lips from each picture were manipulated to reflect ratios of nose
width and lip thickness that were one standard deviation below the grand mean. Again,
for this manipulation, gender-specific grand means and standard deviations (as opposed
to means and standard deviations across all targets) were used. After the manipulation, all
pictures were digitally transformed into outlined pictures, using Adobe Photoshop CS3,
in which there is no luminosity, saturation, or hue information-~only outlines of facial
features remained. All shadows on foreheads and checks in each picture were also erased
(see Appendix B for an example picture).

In Step 4, White participants, who did not participate in Step 2, were asked to rate
the outlined pictures in terms of a degree of Afrocentric features and naturalness of the
picture to ensure that the manipulation has successfully produced two sets of photos that

appear natural: more vs. less prototypical facial features. F ifty—ei ght participants rated

outlined Black male pictures, and 44 participants rated outlined Black female pictures.3

For the facial feature rating, participants were instructed to indicate how prototypical
each of the young men and women in the pictures was in terms of "Afiican American-
ness." They were further told that judgments of African American prototypicality entail
facial features (e.g., lip thickness, nose width), hair texture, and so on. These instructions
were Similar to the ones in Livingston and Brewer’s (2002) study, except that “skin tone”
information was excluded from the current instructions. High prototypicality indicated
more "African type" facial features, and low prototypicality indicated less "Afiican type"

facial features. The scale ranged from 1 (Low Prototypic) to 5 (High Prototypic). For the

27

rating of naturalness, participants were told that the pictures had been transformed into
outlined pictures from their original colored pictures, and that some parts of the face in a
small number of pictures had been digitally altered. Participants were asked to indicate,
for each of the target faces, whether they thought features on a face have been changed
(in addition to transformation from color format into outline format) by choosing “Yes, I
think the facial features on this face have been changed” or “No, I don't think the facial
features on this face have been changed.”

In Step 5, based on the information obtained in Steps 1 through 4, the final target
pictures of 32 Black men and four Black women that would be used in Studies 1 - 3 were
selected. Because each face was manipulated to reflect both strong and weak facial
features, the final selection includes 64 Black male and eight Black female pictures (see
Appendixes C and D for the actual pictures). The final set of female targets included only
four dark-skinned faces and four light-skinned faces due to a lack of available female
pictures that could be unambiguously categorized as Black. Importantly, some
participants categorized those four pictures into White or rated as racially ambiguous;

therefore, the present study serves as a preliminary study to examine target gender

differences in the effects of skin tone and facial features.4 Once 64 Black male and 8

Black female pictures were selected, it was examined 1) whether dark skin tone faces
were objectively darker than light Skin tone faces, and 2) whether faces manipulated to
reflect more prototypical facial features were perceived to be significantly more
prototypical of “Afi'ican type” than faces manipulated to reflect less prototypical facial
features even when the pictures were presented in outlines (i.e., no skin tone information).

Results

28

Steps 1 & 2 (Examinations of Natural Skin Tone and Natural Facial Features)

Means and standard deviations of subjective ratings (i.e., overall Afrocentric
features, attractiveness, babyfacedness, and hostility) and objective measures (i.e., nose
ratio, lip ratio, and the average luminosity) of all faces in the original pool are presented
separated by gender in Table 1. In order to get a broad sense of how each of the variables
are related to one another, correlations among all variables are provided in Tables 2 and 3
separately for Black male targets and Black female targets, respectively. These
correlations indicate that subjective ratings of Afrocentric features were significantly
related to Objective measures of facial features and Skin tone such that those who were
rated high on Afrocentric features naturally had significantly wider nose, thicker lips, and
darker skin, as compared to those who were rated low on Afrocentric features, regardless
of their gender.

In addition, Afrocentric features were negatively related to attractiveness for both
Black men and Black women, such that those with stronger Afrocentric features were
perceived to be less attractive than those with weaker Afrocentric features. Furthermore,
for Black men, Afrocentric features were significantly and negatively related to
babyfacedness and positively related to hostility. Thus, Black men with stronger
Afrocentric features were perceived to be less babyfaced and more hostile. For Black
women, Afrocentric features were not significantly related to babyfacedness and hostility.
This suggests that the meaning of having Afrocentric features may differ for Black men
and women.

Step 5 (Examinations of Skin Tone Selection and Facial Feature Manipulation)

29

In order to examine whether the faces that were selected to represent the “dark
Skin tone” condition are significantly and objectively darker than faces that were selected
to represent the “light skin tone” condition, independent groups t-tests were conducted on
the average luminosity between 16 dark-skinned Black men and 16 light-skinned Black
men, and between 2 dark-Skinned Black women and 2 light-skinned Black women. In
addition, the same analyses were done on the other objective (i.e., lip and nose ratios) and
subjective (i.e., Afrocentric features, attractiveness, babyfacedness, and hostility) ratings
to see whether there were systematic differences between dark- and light-skinned faces
on these characteristics. Means, standard deviations, and results of the t-tests are
presented in Tables 4 and 5 for Black men and Black women, respectively. Independent
group t-testS indicated that the selection of dark- vs. light-skinned pictures based on the
average luminosity was successful, such that both Black men and women in the dark Skin
tone condition had significantly lower luminosity scores than those men and women in
the light skin tone condition. In addition, Black men in the dark skin condition and Black
men in the light skin condition differed significantly in their Afrocentric features and
hostility, such that darker-skinned men were perceived to possess more Afrocentric
features and to be more hostile, as compared to their light-skinned counterparts. In
contrast, Black women in dark skin tone condition and those in light Skin tone condition
did not differ in these characteristics. However, it is important to note that df for Black
women was 2, so that the tests were seriously underpowered. In fact, mean differences
between dark- vs. light-skinned pictures in Afrocentric features and hostility were bigger

for Black women and for Black men.

30

Next, it was examined whether faces manipulated to reflect more prototypical
facial features were perceived to be Significantly more prototypical of “Afiican type”
than faces manipulated to reflect less prototypical facial features, even without Skin tone
information. Average scores for the more prototypical feature faces and less prototypical
feature faces were computed within each participant, thus each participant was associated
with two average prototypicality scores: one for more prototypical feature faces and one
for less prototypical feature faces. Paired t-tests indicated that participants reported that
the faces that were manipulated to reflect more prototypical facial features possessed
more prototypical Afiican American facial features than the faces that were manipulated
to reflect less prototypical facial features for both male and female targets (see Table 6),
even when there was no skin tone information. The results indicate that the manipulation
of facial features independent of Skin tone was successful. Taken together, 64 Black male
pictures (see Appendix C) and eight Black female pictures (see Appendix D) were
successfully selected to create four experimental conditions: dark-skinned more
prototypical facial features, light-Skinned more prototypical facial features, dark-Skinned
less prototypical facial features, and light-Skinned leSS prototypical facial features.

Study 1

Study 1 aims to examine the independent effects of skin tone and facial features
on Whites’ affective reactions to Black men at both implicit and explicit levels. The study
was a 2 (Skin Tone: Dark, Light) x 2 (Facial Features: Strong, Weak) x 2 (Prime Set: Set
1, Set 2) mixed design, with the first two being within-participants variables and the last
being a between-participants variable.

Method

31

Participants

Two hundred and thirty eight self-identified White undergraduate students, who

. . 5 .
also took part in Study 3b, participated in exchange for partial course credit. Fifty

participants (21.0%) were excluded from the analyses due to at least one of the following
reasons: 1) computer malfunction (39 participants), 2) non-White participants (eight
participants), 3) mentioned “Skin color” or “Skin tone” when asked about the purpose of
the study (nine participants), and 4) lack of attention during the task (one participant).
This resulted in 186 analyzable cases.
Procedures

Up to six participants reported to the laboratory session and were greeted by an
experimenter (one of the eight White and one Asian experimenters), who led participants
to a room equipped with Six computers. Participants were informed that the study
examined people’s memory and judgment skills. After obtaining consent, the
experimenter explained to the participants that their task was to complete a number of
short tasks on the computer; participants were then left in the room to complete the tasks.
Participants first completed a modified version of a sequential priming task to assess
automatic affective reactions to the targets and then completed self-report measures of
liking toward the targets. They also provided their demographic information at the end of
the computer task. Participants were randomly assigned to view one of the two target sets.
Participants read all of the instructions on the computer as they completed each task, and
the presentation of experimental stimuli and data collection were controlled by MediaLab
and DirectRT software.

Measures

32

Modified sequential priming task The sequential priming task used in the
Livingston and Brewer’s (2002) study, which resembles the task created by Fazio,
Jackson, Dunton, and Williams (1995), was modified for this study. The task is designed
to assess automatic affective reactions to social categories that the primes represent. More
specifically, the task assesses participants' reaction times to positive and negative words
when preceded by primes (i.e., dark-Skinned Blacks with more prototypical facial
features, light-Skinned Blacks with more prototypical facial features, dark-skinned Blacks
with less prototypical facial features, light-Skinned Blacks with less prototypical facial
features, or control Whites). Faster reaction times to negative words following a
particular prime indicate more negative affective reactions to that prime type, and faster
reaction times to positive words following a particular prime indicate more positive
affective reactions to that prime. The faster participants respond to negative words
following more prototypical facial feature Black faces or darker skin tone Black faces
relative to other primes, the more we can infer negative feelings toward Black individuals
with more prototypical facial features or with darker skin tone. Likewise, the faster they
respond to positive words following less prototypical facial feature Black faces or lighter
Skin tone Black faces relative to other primes, the more we can infer positive feelings
toward Black individuals with less prototypical facial features or with lighter skin tone.

The task consisted of five phases, with the first and forth phases being the
important phases. The task in the first phase was described as a word-meaning task, and
participants were instructed to make a judgment about whether a word presented on the
computer screen was a good word or a bad word as quickly and accurately as possible by

pressing appropriate keys labeled “good” or “bad.” Words were eight positive and eight

33

negative adjectives. The purpose of this phase was to obtain baseline reaction times for
these words that would then be used as target words in the fourth phase.

The details of the task in the first phase are as follows. First, a fixation point
“*********” appeared on the computer screen for 315 ms. Then, one of the 16 words
(Eight good words: beauty, joy, love, paradise, romance, smile, success, vacation, and
eight bad words: cockroach, despair, disgust, garbage, pest, poison, sewage, vomit) were
randomly presented on the screen until participants respond by pressing keys labeled as
either “good” or “bad.” Once participants responded to that trial, the word disappeared,
and the next trial started, with a 2.5 sec inter-trial interval (ITI). Response latency for
each word was recorded in milliseconds. Participants first completed a practice block and
then completed two blocks of trials with the key 16 words. The average latency of the
two trials for each word was computed, and those values serve as the baseline latencies
for each of the 16 words.

The second and third phases provided a cover story for why participants were
presented with pictures (i.e., primes) and why they needed to attend to those faces. In the
second phase, participants completed a task described as a face-learning exercise. In this
task, participants were asked to simply attend to eight faces (Six Black faces and two
White faces that were not included in the target stimuli used in the critical phase)
presented on the computer screen for 2.5 sec. Participants looked at the eight faces twice.
The order of eight faces was randomized within each of the two sets across participants.

The third phase of the study was described as a face-detection exercise, and
participants were told that their task was to indicate whether they recognize the faces

from the previous task by pressing a key labeled “yes” or a key labeled “no.” In this

34

phase, eight faces (four previously presented faces: three Black faces and one White face,
and four new faces: three Black faces and one White face) were randomly presented on
the computer screen until participants responded by pressing appropriate keys, with 2.5
sec ITI.

The fourth phase was the critical phase that involved the actual priming task. In
this task, participants were told that the previous tasks would be combined to examine the
extent to which the judgment of word meaning is an automatic process. Participants were
further told that if judging word meaning was automatic and easy, they should perform
just as well at it as they did in the first phase, even if they had to learn faces at the same
time. The instructions and procedures were the same as the first phase with two
exceptions. First, the fixation point “*********” was replaced by target faces (i.e.
primes). Second, participants were told that it was very important for them to attend to
the faces presented because they would be asked to indicate whether they recognize the
faces in a following face-detection task.

There were two sets of target pictures, and participants were randomly assigned to
complete one of the two sets. In each target set, the target pictures consisted of 40 young
male faces (eight faces for each of the following five categories: dark-skinned more
prototypical facial features (DM), light-skinned more prototypical facial features (LM),
dark-Skinned less prototypical facial features (DL), light-Skinned less prototypical facial
features (LL), and White control (W; these faces were unaltered and matched in
attractiveness with the 32 target Black pictures). Again, dark vs. light skin tone were

selected based on the actual variation in skin color, and more vs. less prototypical facial

35

features were experimentally manipulated, so that it was ensured that skin tone and facial
features were independent of one another.

First, a face was presented on the computer screen for 315 ms as a prime,
followed by a 135 ms interval before onset of the target word. The target word was
presented on the screen until participants responded to the trial by pressing the
appropriate key identifying whether the word was “good” or “bad.” Once participants
made a response, the word disappeared, and the next trial began. There was 2.5 sec ITI
with a blank screen between each trial. Participants first completed a practice block that
included five trials in which they were exposed to different faces and words from the
actual trials. Then they completed four blocks of trials. Each block consisted of 40 trials
in which each of the 40 faces appeared once followed by one of the 16 target words. Over
the course of the four blocks, each face was paired with two positive and two negative
words. Although trials within each block were presented in a random order, each of the
40 faces was paired with specific positive and negative words (see Appendix E for the
table summary of trials and blocks).

Finally, in the fifth phase, participants were asked to complete another face-
detection task that they were led to expect from the instructions for the fourth phase. In
this task, participants were asked to look at eight faces (four used in phase 4 and four new)
and indicate whether they recognized the faces from the task they had just completed, by
pressing appropriate keys labeled as either “yes” or “no.”

Liking toward targets. Participants’ liking toward each target presented in Phase
4 was assessed with an explicit self-report measure. More specifically, participants were

asked to indicate how much they thought they wOuld like the person in the picture. The

36

scale ranged from 0 (I would DISLIKE this person very much.) to 5 (I would LIKE this
person very much), with higher scores indicating more liking toward targets.

Demographic information. Participants were asked to provide their age, gender,
and race. Although only participants who self-identified as White were recruited for this
study using an online recruitment program, participants were also asked to provide their
racial information in order to ensure that they were White. Due to a computer
malfunction, participants’ age and gender were not recorded.

Computing Facilitation Scores

Incorrect trials as well as response latencies faster than 300 ms and Slower than
2.5 standard deviations above the individual’s mean were deleted. In order to compute
useable facilitation scores, first, the log transformed response latency for any given target
word following a given face (as measured in the fourth phase) was subtracted from the
log transformed baseline latency of that word (as measured in the first phase) to arrive at
a facilitation score. Therefore, positive values indicate that responses to the target word
are facilitated compared with baseline, and negative values indicate that responses to the
target word are inhibited compared with baseline. Next, mean facilitation scores for
positive target words and negative target words were computed for each face (each face
was followed by 2 positive and 2 negative words). Then, mean facilitation scores for
positive words were averaged across the eight faces for each participant to arrive at five
facilitation scores (i.e., DM, LM, DL, LL, or W faces followed by positive words). Mean
facilitation scores for negative words were also averaged across the eight faces for each
participant to arrive at the remaining five facilitation scores (i.e., DM, LM, DL, LL, or W

faces followed by negative words). Thus, each participant ended up with 10 facilitation

37

scores. Higher facilitation scores for negative words suggest more automatic negative
feelings toward preceding primes. Likewise, higher facilitation scores for positive words
suggest more automatic positive feelings toward preceding primes.

Facilitation scores were computed separately for positive and negative words,
because it is well-documented that positive and negative affect are not a unidimensional
construct, but independent of one another (Bradbum, 1969; Diener & Emmons, 1984;
Goldstein & Strube, 1994; Warr, Barter, & Brwonbriedge, 1983). In fact, the Positive
Affect Negative Affect Schedule was developed to capture the independent nature of
positive and negative affective reactions of people (Watson, Clark, & Tellegen, 1988),
and this measure has been used by many personality and positive psychologists.
Additionally, separately examining facilitation scores for positive and negative words
allows us to see the direction of the change in attitudes. In other words, if Black men with
darker skin tone are evaluated more negatively than Black men with lighter skin tone,
examination of facilitation scores of both positive and negative words allows us to see
whether the change in the mean reflects increased positivity or decreased negativity
toward Black men with lighter skin tone, or increased negativity or decreased positivity

toward Black men with darker skin tone. With a unidimensinal attitude measure, it is hard
to conclude the directions of the changes in attitudes.6
Results
Implicit Negative Attitudes toward Black Men
The means and standard deviations of facilitation scores for the negative words
for each prime category in both raw scores and log transformed scores are presented in

Table 7. Higher numbers mean that participants responded faster to negative words

38

following primes, as compared to baseline responses, indicating more negative affective
reactions to given primes. A three-factor mixed model ANOVA, treating skin tone (dark,
light) and facial features (more prototypical, less prototypical) as within-participants
factors and prime set (set I, set 2) as a between-participants factor, was computed on the
facilitation scores for negative words. The main effect of Skin tone was significant, F (1 ,
184) = 6.65, MSE = .01, p = .01, d = .09, such that participants responded Significantly

faster to negative words following dark-Skinned Black male primes (M = 37.76, SD =

70.11),7 as compared to light-Skinned Black male primes (M = 32.64, SD = 74.34). The

main effect of facial features was also Significant, F(1, 184) = 4.15, MSE = .002, p < .05,
d = .07, such that participants responded faster to negative words following Black male
primes with more prototypical facial features (M = 37.23, SD = 71.54), as opposed to
Black male primes with less prototypical facial features (M = 33.17, SD = 72.90). These
results suggest that participants evaluated dark-Skinned Black men more negatively than
light-skinned Black men, and they evaluated Black men with more prototypical facial
features more negatively than Black men with less prototypical facial features. The
interaction between skin tone and facial features was not significant, F (1, 184) = .04,
MSE = .002, p = .85. This analysis also revealed that there was no main effect of prime
set type, F(1, 184) = .01, MSE = .04,p = .94.
Implicit Positive Attitudes toward Black Men

Next, participants’ responses to positive words following male target pictures
were examined using the same three-factor mixed model ANOVA. See Table 8 for the
means and standard deviations for each category. Higher numbers mean that participants

responded faster to positive words following primes, as compared to baseline, indicating

39

more positive affective reactions toward given primes. The analysis revealed that none of
the main effects of Skin tone (F (I, 184) = 2.60, MSE = .002, p = .1 l), facial features (F (l ,
184) = .25, MSE = .001, p = .62), or set type (F(1, 184) = .40, MSE = .04, p = .53) was
significant. In other words, there iS some evidence that having lighter skin tone or having
less prototypical facial features does not increase Whites’ positive evaluations. Again, the
interaction between skin tone and facial features was not significant, F (l , 184) = 2.15,
MSE = .003,p = .14.
Black vs. White Targets

Next, in order to examine whether evaluations of Black men are different from the
White control, a two-factor mixed model ANOVA, treating target race (Black, White) as
within-participants factor and prime set (set 1, set 2) as between-participants factor on
facilitation scores for negative words, was conducted. In order to compute facilitation
scores for Black targets, regardless of skin tone and facial features, facilitation scores for
each of the four prime categories were averaged. The ANOVA revealed that the main
effect of target race was Significant, F (l, 184) = 11.78, MSE = .002, p < .001, d = .14,
such that participants responded faster to negative words following Black primes (M =
35.20, SD = 66.87), as compared to following White primes (M = 35.48, SD = 72.12).
The results indicate that White participants affectively react more negatively toward
Black men than toward White men. There was no main effect of prime set, F (I , 184)
= .10, MSE = .02,p = .75.

The same analysis was conducted on facilitations scores for positive words and
revealed no Significant main effect of target race (F (l , 184) = .02, MSE = .002, p = .89)

or main effect of set type (F (l, 184) = .48, MSE = .02, p = .49).

40

Explicit Affective Reactions to Black Men

Finally, the effects of skin tone and facial features on White participants’ explicit
liking toward Black male targets were examined, and the means and standard deviations
are presented in Table 9. A three-factor mixed model ANOVA, treating Skin tone (dark,
light) and facial features (more prototypical, less prototypical) as within-participants
factors and prime set (set I, set 2) as a between-participants factor, was conducted to
examine the effects of Skin tone and facial features on Whites’ explicit reports of liking
toward Black men. The ANOVA revealed that the main effect of skin tone was
significant, F(1, 184) = 36.68, MSE = .08, p < .001, d = .18, such that participants liked
lighter-skinned Black men significantly more (M = 3.12, SD = .73) than dark-skinned
Black men (M = 2.99, SD = .70). The main effect of facial features was also significant,
F (1, 184) = 13.56, MSE = .06, p < .001 , d = .09, such that participants liked Black men
with less prototypical facial features significantly more (M = 3.09, SD = .71) than those
with more prototypical facial features (M = 3.02, SD = .72). The interaction between skin
tone and facial features was not significant, F (1 , 184) = .97, MSE = .07, p = .33. Finally,
the main effect of set type was not Significant, F(1, 184) = .99, MSE = 1.80, p = .32.

Discussion

The goal of Study 1 was to examine whether facial features, above and beyond
the skin tone, influence how Whites affectively react to Black men at both implicit and
explicit levels. The results showed that, at the implicit level, White individuals responded
Significantly faster to negative words following Black men with darker Skin tone, as
compared to following Black men with lighter Skin tone. This indicates that Whites feel

more negatively toward Black men with darker Skin tone than toward Black men with

41

lighter skin tone. This finding is consistent with previous studies reporting that Whites
perceive Black individuals with darker skin tone more negatively than those with lighter
skin tone (Anderson & Cromwell, I977; Averhadt & Bigler, 1997; Hall, 1992; Maddox
& Gray, 2002; Dixson & Maddox, 2005; Porter, 1991).

Importantly, the results fiom the present study also Showed that White individuals
responded Significantly faster to negative words following Black men with more
prototypical facial features than following Black men with less prototypical facial
features. This indicates that Whites felt more negatively toward Black men with more
prototypical facial features than toward those with less prototypical facial features. Thus,
the results suggest that White individuals do detect such subtle differences in facial
features (in addition to the differences in Skin tone) in a very Short amount of time (i.e.,
315 ms), and the differences in facial features influence how they feel about Black male
targets. Taken together, the effects of skin tone and facial features on Whites’ affective
reactions toward Black men were additive, and having darker Skin tone or more
prototypical facial features increases White individuals’ negativity toward Black men.

Interestingly, there was no difference in how fast White individuals responded to
positive words following primes. In other words, there is evidence that having lighter
skin tone or less prototypical facial features does not increase Whites’ positive feelings
toward Black men. The findings suggest that the previously reported findings that Black
men with darker Skin tone and stronger Afrocentric features were perceived and treated
more negatively (Blair, 2006; Blair et al., 2005; Blair et al., 2002; Dixon & Maddox,

2005; Livingston & Brewer, 2002; Maddox & Chase, 2004; Maddox & Gray, 2002;

42

Wade & Bielits, 2005) are due to increased negative perceptions of such individuals,
rather than increased positive perceptions of Black men who are less prototypical.

The results for explicit liking were consistent with those of implicit attitudes, such
that both Skin tone and facial features independently influenced how much White
individuals liked Black male targets. More specifically, Whites liked Black men with
lighter Skin tone more than those with darker Skin tone, as well as Black men with less
prototypical facial features more than those with more prototypical facial features. Taken
together, the findings from Study 1 suggest that facial features are as informative as skin
tone when predicting White individuals’ feelings about Black men both at implicit and
explicit levels. Specifically, skin tone and facial features seem to affect how much White
individuals dislike Blacks.

Past studies have shown that skin tone and Afrocentric features influence not only
Whites’ affective reactions toward Blacks, but also their cognitive reactions toward
Blacks (i.e., stereotype activation; Averhadt & Bigler, 1997; Blair, 2006; Blair et al.,
2002). In the next study, the independent effects of Skin tone and facial features on
stereotype activation following Black male primes among Whites were examined. I

Study 2

The study was a 2 (skin tone: dark, light) x 2 (facial features: more prototypical,
less prototypical) x 2 (word type: stereotypic, non-stereotypic) x 4 (prime set: set la, set
lb, set 2a, set 2b) mixed design, with the first three independent variables being within-
participants factors, and the last independent variable being a between-participants
variable. Furthermore, the study examined explicit liking toward Black men.

Method

43

Participants

Two hundred and forty six self-identified White undergraduate students, who also
took part in Study 3a, participated in this study. Forty-four participants (17.9 °/o) were
excluded from data analyses based on the same exclusion criteria as Study 1 (i.e.,
computer malfunction, non—White participants, mentioned “skin color” or “skin tone”
when asked about the study purpose, and lack of attention). Thirty participants were
excluded due to computer malfunction, six participants were excluded because they were
not White, four were excluded because they mentioned “skin color” or “skin tone” when
asked about the purpose of the study, and seven were excluded due to lack of attention
during the computer task. This resulted in 202 analyzable cases.
Procedures

Participants reported to the laboratory session in a group (up to Six participants
per session) and were greeted by either a White American or an Asian American
experimenter. The experimenter led participants to a room equipped with six computers
and informed them that the purpose of the study was to examine people’s memory and
judgment skills. After Obtaining consent, participants were told that they would complete
a few tasks on the computer, and they were left in the room to complete the computer
tasks. Participants first completed a modified version of a lexical decision task, which
assesses the degree of activation of stereotypes associated with Black men. After
completing the lexical decision task, participants filled out self-report measures of liking
toward the targets and provided their demographic information (i.e., age, gender, and

race). Again, all the instructions were provided on the computer screen as participants

proceeded at their own pace, and the presentation of experimental stimuli and data
collection were controlled by MediaLab and DirectRT software.
Measures

Modified lexical decision task (LDT). The original LDT created by Wittenbrink,
Judd, and Park (1997) was modified for this study. The task is designed to assess
participants' reaction times to negative words that are either stereotypic or non-
stereotypic of Black men when preceded by primes (i.e., dark-skinned Blacks with more
prototypical facial features, light-Skinned Blacks with more prototypical facial features,
dark-skinned Blacks with less prototypical facial features, light—Skinned Blacks with less
prototypical facial features, or control Whites). The faster participants respond to
stereotypic words following more prototypical facial features or dark-skinned Black faces
as opposed to less prototypical facial features or light-skinned Black faces, the more we
can infer that stereotypes associated with the category “Black” are activated.

In the LDT, a fixation point “*********” first appeared on the computer screen
for 1.0 sec, and it was immediately followed by the prime (i.e., a target face). The target
faces presented during the task differed depending on the assigned prime set. However,
each set included the five target categories.

The prime stayed on the screen for 315 ms, and a target word (one of the 24
words or 24 non-words) followed (see Appendix F for the detailed explanation of how
the target words were selected). Participants’ task was to identify whether the target word
presented on the computer screen was an actual word or a non-word as quickly and
accurately as possible by pressing appropriate keys labeled “W” or “NW.” The target

word stayed on the screen until participants responded by pressing the appropriate key.

45

Once participants responded to that trial, the next trial started with a 2.5 sec ITI.
Response latency for each word/non-word was recorded in milliseconds.

Each of the five prime categories occurred on 48 different trials. On 12 trials,
target words were real words related to negative Black male stereotypes (i.e., creepy,
dangerous, ignorant, incompetent, lazy, mean, noisy, poor, reckless, threatening,
untrustworthy, violent). On another 12 trials, target words were real words that are
negative but unrelated to Black male stereotypes (i.e., annoying, arrogant, boring,
condescending, greedy, im'tating, judgmental, selfish, smelly, stuckup, ugly, weak). On
the remaining 24 trials, target words were non-words. Thus, participants completed a
total of 240 trials. Participants firSt completed a practice block including 10 trials, and
then they completed the actual trial.

In order to conceal the actual purpose of the LDT, participants were told that the
task was designed to examine their ability to perform two different tasks (i.e., a memory
task and a judgment task) Simultaneously. Furthermore, participants were told that they
should try to memorize the faces (i.e., primes) while completing the word/non-word
judgment task because they would perform a memory task at the end of the task. In
actuality, participants did not complete the memory task.

Self-report measures. The same measures used in Study 1 (i.e., liking toward
targets and demographic information) were used in Study 2.
Computing Response Latencies

First, incorrect trials as well aS response latencies faster than 300 ms and Slower

than 2.5 standard deviations above the individual mean were deleted. After performing

the log transformation, average response latency of 12 negative words that were

46

stereotypic of Black men and average response latency of 12 negative words that were
non-stereotypic of Black men were computed for each of the five prime categories. Thus,
10 response latency scores were computed for each participant.
Results

Stereotype Activation among Whites

Table 10 presents means and standard deviations of response latencies in both raw
and log transformed scores for each prime category for the stereotypic and non-
stereotypic words. In order to examine the effects of skin tone and facial features on
stereotype activation, a four-factor mixed model of ANOVA, treating skin tone (dark,
light), facial features (more prototypical, less prototypical), and word type (stereotypic,
non-stereotypic) as within-participants factors and prime set (set 1a, set lb, set 2s, set 2b)
as a between-participants factor, was conducted on the response latencies. The main
effect of word type was Significant, F(1, 198) = 25.73, MSE = .004,p < .001, d = .l I,
such that participants responded significantly faster to stereotypic words (M = 583.09, SD
= 98.63) than to non-stereotypic words (M = 593.12, SD = 98.55). Neither the main effect
of skin tone (F (l , 198) = .l l, MSE = .004, p = .74) nor the main effect of facial features
(F (l, 198) = .71, MSE = .003, p = .40) was significant. The ANOVA revealed that there
was no main effects of prime set, F(3, 198) = .82, MSE = .17, p = .48, suggesting that
picture sets and presentation of order did not have effects on participants’ responses.

None of the two-way interactions between skin tone and facial features (F (l , 198)
= 1.65, MSE = .004, p = .20), between Skin tone and word type (F(1, 198) = .11, MSE
= .004, p = .74), and between facial features and word type (F(1, 198) = .01, MSE = .004,

p = .94) was significant. However, the three-way interaction between skin tone, facial

47

features, and word type was significant, F (l, 198) = 7.20, MSE = .004, p < .01, indicating
that the relationship between skin tone and facial features on response latencies differed
depending on whether the words were stereotypic or non-stereotypic.

Examination of the simple interactions between skin tone and facial features at
different levels of word type revealed that the two-way interaction was significant in the
non-stereotypic word condition (F (l, 198) = 7.66, MSE = .004, p < .01), but not in the
stereotypic word condition (F (I, 198) = .90, MSE = .004, p = .34). In order to further
examine the nature of the two-way interaction between skin tone and facial features in the
non-stereotypic word condition, the simple main effect of facial features was examined in
the dark Skin tone condition and in the light skin tone condition separately. The analyses
revealed that the simple main effect of facial features was significant in the light skin
tone condition, t(201) = -2.25, p < .05, d = .10, such that participants responded faster to
non-stereotypic negative words following Black men with strong facial features than
following Black men with weak facial features. The simple main effect of facial features
in the dark skin tone condition was not Significant, t(201) = 1.58, p = .12.

Black vs. White Targets

Next, the degree of stereotype activation following Black male primes was
compared with stereotype activation following White male primes. A three-way mixed
model ANOVA, treating target race (Black, White) and word type (stereotypic, non-
stereotypic) as within-participants factors and prime set (set 1a, set 1b, set2a, set2b) as a
between-participants factor, was conducted on response latencies. The analysis revealed
that that the main effect of word type was significant, F (l, 198) = 11.94, MSE = .002, p

< .001, d = .07, such that White participants responded faster to stereotypic words (M =

48

583.07, SD = 94.83), as compared to non-stereotypic words (M = 590.03, SD = 92.48),
regardless of target race. The main effect of target race was marginally significant, F (1,
198) = 3.21, MSE = .002, p = .08, d = .03, such that participants responded faster to
negative words, regardless of word type, following White primes (M = 584.92, SD =
95.60) than following Black primes (M = 588.18, SD = 91.72). Neither the main effect of
prime set (F (3, 198) = .99, MSE = .08, p = .40) nor the interaction between target race
and word type (F(l , 198) = .82, MSE = .003, p = .48) was significant. The absence of
mean difference between Black primes and White control in the stereotypic word
condition (which has been reported in the previous studies), suggests that the lexical
decision task in this study might not have been effective.
Explicit Affective Reactions to Black Men

Finally, a three-factor mixed model ANOVA, treating skin tone and facial
features as within-participants factors and prime set as a between-participants factor, was
conducted on explicit reports of liking toward Black men. The means and standard
deviations of Whites’ explicit liking toward Black men as a function of skin tone and
facial features are presented in Table 11. The main effect of skin tone was Significant,
F(1, 198) = 58.07, MSE = .06, p < .001, d = .18, such that White participants liked light-
Skinned Black men significantly more (M = 3.10, SD = .74) than dark-skinned Black men
(M = 2.97, SD = .73). Although the pattern of the means was in the same direction as
Study 1a, the main effect of facial features was not significant, F(l , 198) = 1.86, MSE
= .O7,p = .17. Neither the main effect of prime set (F(3, 198) = 1.83, MSE = 1.95, p = .14)
nor the interaction between skin tone and facial features (F (1 , 198) = .07, MSE = .04, p

= .79) was Significant.

49

Discussion

Study 2 examined whether facial features and Skin tone independently affect
stereotype activation among Whites. More specifically, it investigated how Black male
targets, who vary in Skin tone and facial features, influence Whites’ stereotype activation.
The results Showed that White individuals responded significantly faster to stereotypic
negative words, as compared to non-stereotypic negative words. At first glance, the
findings suggest that negative stereotypes associated with Black men were activated
following Black primes, regardless of the degree of Skin tone and prototypicality of facial
features. However, it is important to note that White individuals’ response time to
stereotypic negative words following White controls were not different from those
following Black primes. The absence of mean differences in reaction time to stereotypic
words between Black primes and White controls suggests that the modified lexical
decision task might have failed to assess stereotype activation. Potential explanations for
the low validity of this task are presented in the general discussion.

Study 2 also looked at White individuals’ explicit affective reactions toward
Black men, replicating Study 1. The results showed that Skin tone influenced Whites’
liking toward Black men, such that light-skinned Black men were liked more than dark-
skinned Black men. Although the patterns of the mean liking scores between Black men
with more prototypical facial features and Black men with less prototypical facial
features were consistent with Study 1 (i.e., White individuals liked Black men with more
prototypical facial features less than those with less prototypical facial features), the

effects of facial features was not Significant in this study.

50

Studies 1 and 2 examined how Black men’s skin tone and facial features influence
Whites’ affective and cognitive reaction toward them, and they demonstrated that the
effects of skin tone and facial features are additive and that having either darker skin tone
or more prototypical facial features increase Whites’ negative feelings. However, it is still
unclear whether there are additive effects Of skin tone and facial features when the targets
were Black women, as previous research has exclusively focused on Black male targets.
In the next study, it was examined whether the effects Of skin tone and facial features
would be moderated by target gender.

Study 3

Study 3 examined the independent effects of skin tone and facial features on
Whites’ affective and cognitive reactions toward Black women. There are two sub-studies,
with Study 3a investigating Whites’ affective reactions to Black women at both implicit
and explicit levels, and Study 3b examining Whites' stereotype activation following
Black female primes. Study 3b also looks at Whites’ explicit liking toward Black women,
replicating Study 3a. Study 3a was a 2 (Skin Tone: Dark, Light) x 2 (Facial Features:
More Prototypical, Less Prototypical) x 2 (Prime Set: Set 1, Set 2) mixed design, with the
first two being within-participants factors and the last being a between-participants factor.
Study 3b was a 2 (Skin Tone: Dark, Light) x 2 (Facial Features: More Prototypical, Less
Prototypical) x 2 (Word Type: Stereotypic, Non-Stereotypic) x 2 (Prime Set: Set I, Set 2)
mixed design, with the first three being within-participants factors and the last being a
between-participants factor.

It is important to note that the participants in this study also completed either

Study 1 or Study 2 first. Participants who completed the sequential priming task with

51

Black male targets (i.e., Study 1) first were asked to complete the lexical decision task
with Black female targets (i.e., Study 3b) second, and those who completed the lexical
decision task with Black male targets (i.e., Study 2) first were asked to completed the
sequential priming task with Black female targets (i.e., Study 33) second. Although the
ideal design would have been to counterbalance the order of the target gender, all
participants completed the task with Black female targets after they completed tasks
involving the Black male targets. This approach was taken because there were only a few
Black female target photographs that could be used in the study, and therefore results
based on the female targets should be considered as preliminary. Note that the small
number of female targets resulted in considerably shorter sequential priming and lexical
decision tasks, as compared to the tasks created for Black male targets.

Study 3a

Methods
Participants

Two hundred and forty Six self-identified White undergraduate students

participated in this study. Again, the participants completed the lexical decision task (i.e.,
Study 2) before completing this study. The same exclusion criteria as Studies 1 and 2
were used (i.e., computer malfunction, non-White participants, mentioned “skin color” or
“skin tone” when asked about the study purpose, and lack of attention) with one added
requirement that participants perceived female targets as unambiguously Black. This
requirement was added because female target pictures were not perceived as
unambiguously Black by everyone in the pilot study. Participants who categorized the

target pictures as either racially ambiguous or White were excluded from the analyses.

52

Based on these six exclusion criteria, 95 participants (40.4%) were excluded from the
analyses (50 participants due to computer malfunction, six participants who were not
Whites, four participants who mentioned “Skin color” or “skin tone” when asked about
the purpose of the study, seven participants due to lack of attention, and 47 participants
who categorized the target women as non-Black). Thus, 140 participants were included in
the following analyses.
Procedures

The procedures were the same as Study 1 except for two modifications. First,
length of the sequential priming task in this study was considerably shorter than the task
in Study 1. The sequential priming task for female Black targets included only eight
target words (Four good words: paradise, romance, smile, success, and four bad words:
despair, pest, poison, sewage), instead of 16 words. This was due to the lack of useable
female target stimuli. Second, there were only five target pictures in this task (i.e., one
target picture for each prime category: DM, DL, LM, LL, W) in each target picture set,
unlike Study 1 in which there were 40 targets pictures in total in each target picture set.

After completing the sequential priming task with female target primes,
participants also completed the explicit liking measure and provided demographic
information. Again, due to a computer malfunction, participants’ age and gender were not
recorded.

Results

Implicit Negative Attitudes toward Black Women

Participants’ responses to negative words following female target pictures were

examined using a three-factor mixed model ANOVA, treating skin tone (dark, light) and

53

facial features (more prototypical, less prototypical) as within-participants factors and
prime set (set 1, set 2) as a between-participants factor. See Table 12 for the means and
standard deviations for each prime category. The analysis revealed that none of the main
effects of Skin tone (F(l , 138) = .17, MSE = .01, p = .68), facial features (F (1, 138)
= .0001, MSE = .01,p = .98 or set type (F(1, 138) = 1.84, MSE = .05,p = .18) was
Si gnificant. Thus, there is evidence that having darker Skin tone or more prototypical
facial features do not increase negative evaluations of Black women. The interaction
between skin tone and facial features was not Si gnificant either, F (l, 138) = .41 , MSE
= .01 , p = .33.
Implicit Positive Attitudes toward Black Women

Next, participants’ responses to positive words were examined using the same
three-factor mixed model of ANOVA. The means and standard deviations for each prime
category are presented in Table 13. The analysis revealed that the main effect of skin tone
was Significant, F(1, 138) = 10.43, MSE = .01,p < .01, d = .15, such that participants
responded faster to positive words following light—skinned Black female primes (M =
38.05, SD = 88.29) than following dark-skinned Black female primes (M = 23.72, SD =
98.17). The results suggest that Ii ght-Skinned Black women were evaluated more
positively than dark-Skinned Black women. In contrast, the main effect of facial features
was not Significant, F (l, 138) = .02, MSE = .01, p = .90, indicating that facial features did
not have an additive effect on Whites’ positive attitudes toward Black women. The
interaction between skin tone and facial features was not Significant, F (l, 138) = 1.13,
MSE = .01, p = .29. Again, there was no main effect of set type, F(1, 138) = .01, MSE

= .07,p = .94.

54

Black vs. White Targets

In order to examine whether evaluations of Black women are different from the
White control, a two-factor mixed model ANOVA, treating target race (Black, White) as
a within-participants factor and prime set (set I, set 2) as a between-participants factor,
on facilitation scores for negative words was conducted first. Again, in order to come up
with facilitation scores for Black women, regardless of skin tone and facial features,
facilitation scores for four prime types were averaged. The ANOVA revealed no
Significant main effect of target race, F (1, 138) = .48, MSE = .004, p = .49). The main
effect ofprime set was marginally Significant, F(1, 138) = 3.81, MSE = .03,p = .05,
suggesting that participants’ negativity toward Black, as compared to White, targets
differed depending on pictures. This might be due to the fact that there were only four
Black female pictures and one White female picture in each set.

The same analysis was conducted on facilitation scores for positive words and
revealed that neither the main effect of target race, F (l, 138) = .14, MSE = .004, p = .71,
nor the main effect of prime set, F(1, 138) = .001, MSE = .03, p = .98, was significant.
Explicit Affective Reactions to Black Women

The means and standard deviations of Whites’ explicit liking toward Black
women as a function of skin tone and facial features are presented in Table 14. A three-
factor mixed model ANOVA, treating skin tone and facial features as within-subjects
factors and prime set type as a between-subjects factor, was conducted to examine the
effects Of skin tone and facial features on explicit reports of liking toward Black women.
The ANOVA revealed that the main effect of skin tone was significant, F (1 , 138) = 17.29,

MSE = .50, p < .001, d = .27, such that participants liked light-Skinned Black women

55

more (M = 3.02, SD = .92) than dark-Skinned Black women (M = 2.78, SD = .91). The
main effect of facial features was not Significant, F(1, 138) = 1.36, MSE = .37, p = .25.
The interaction between skin tone and facial features also was not Si gnificant, either, F (1,
138) = .02, MSE = .45, p = .89. Finally, the main effect of prime set was not Significant,
F(1, 138) = .07, MSE = 1.88, p = .80. The results of explicit liking were consistent with
the results of implicit positive attitudes.
Study 3b

Next, Study 3b examined the independent effects of skin tone and facial features
on stereotype activation among Whites. The study also examined Whites’ explicit
attitudes toward Black women.

Methods

Participants

Two hundred and thirty eight self-identified White students participated in this
study. Again, the participants also completed the sequential priming task with Black male
targets (i.e., Study 1) before completing this study. The same exclusion criteria as Study
3a were used in this study. Based on the six exclusion criteria, 63 participants (30.3 %)
were excluded from the analyses (39 participants due to computer malfunction, eight
participants who were not Whites, nine participants who mentioned “skin color” or “Skin
tone” when asked about the purpose of the study, one participants due to lack of attention,
and 29 participants who categorized the target women as non-Black). Thus, 145
participants were included in the following analyses.

Procedures

56

The procedures were the same as Study 2 with two modifications. First, the
lexical decision task for female Black targets included only six words stereotypic of
Black women (i.e., annoying, initating, noisy, nosey, rude, stubborn) and Six words non-
stereotypic of Black women (i.e., exploitative, selfish, sheltered, Sloppy, slow, stuckup).
Second, only five target pictures (i.e., one target picture for each prime category: DM,
DL, LM, LL, W) were used in each target set due to lack of usable Black female pictures.
After completing LDT, participants also filled out the same self-report measures used in
the previous studies (i.e., liking toward targets and demographic information).

Results
Stereotype Activation among Whites

Table 15 presents means and standard deviations of response latencies in both raw
and log transformed scores for each prime category. A four-factor mixed model of
ANOVA, treating skin tone (dark, light), facial features (more prototypical, less
prototypical), and word type (stereotypic, non-stereotypic) as within-participants factors
and prime set (set 1, set 2) as a between-participants factor, was conducted on the
response latencies. The main effect of word type was Significant, F (1, 143) = 7.70, MSE
= .01, p < .01, d = .11, such that participants responded Significantly faster to stereotypic
words (M = 625.91, SD = 109.12) than to non-stereotypic words (M = 641.34, SD =
151.47), suggesting that stereotypic words were more accessible to White participants
than non-stereotypic words. Neither the main effect of skin tone (F (1, 143) = .01, MSE
= .01,p = .92) nor the main effect of facial features (F(1, 143) = .23, MSE = .01, p = .63)
was Significant. The ANOVA revealed that there was no main effects of prime set, F ( 1 ,

143) = .95, MSE = .23,p = .33.

57

The two-way interactions between skin tone and facial features (F (1, 143) = .28,
MSE = .01, p = .60) and between skin tone and word type (F (l, 143) = .02, MSE = .01, p
= .89) were not significant. However, the two-way interaction between facial features and
word type reached marginal Significance, F (1 , 143) = 3.60, MSE = .01 , p = .06. In order
to investigate the nature of the two-way interaction, the Simple main effect of facial
features was examined within each of the word condition. In the stereotypic word
condition, the main effect of facial features was not significant, t(144) = .88, p = .38. In
contrast, in the non-stereotypic word condition, the main effect of facial features was
marginally significant, t(144) = 1.75, p = .08, d = .08, such that participants responded
somewhat faster to non-stereotypic negative words following Black women with less
prototypical facial features (M = 635.85, SD = 134.83), as compared to following Black
women with more prototypical facial features (M = 646.83, SD = 147.61), regardless of
the female targets’ Skin tone. Finally, the three-way interaction between Skin tone, facial
features, and word type was not significant, F(1, 143) = .39, MSE = .01, p = .53.
Black vs. White Targets

Next, a three-way mixed model ANOVA, treating target race (Black, White) and
word type (stereotypic, non-stereotypic) as within-participants factors and prime set (set 1,
set 2) as a between-participants factor, was conducted on response latencies in order to
examine whether the degree of stereotype activation following Black primes was
different fi'om stereotype activation following White primes. In order to come up with
response latencies following Black primes, regardless of physical characteristics,
response latencies for four Black prime categories were averaged. The ANOVA revealed

significant main effect oftarget race, F(1, 143) = 8.79, MSE = .01,p < .01, d = .10, such

58

that White participants responded faster to negative words, following White primes (M =
620.24, SD = 143.99) than following Black primes (M = 633.63, SD = 127.32), regardless
of whether the words were stereotypic or non-stereotypic. The main effect of word type
was also Significant, F(1, 143) = 26.84, MSE = .01,p < .001, d = .23, such that White
participants responded faster to stereotypic words (M = 611.55, SD = 121.37) than non-
stereotypic words (M = 642.32, SD = 149.94).

However, these main effects were qualified by the Significant interaction between
target race and word type, F (1 , 143) = 8.10, MSE = .01 , p < .01. In order to examine the
nature of the two-way interaction, the effect of target race was examined within each
word type. In the stereotypic word condition, White participants responded faster
following White primes (M = 597.19, SD = 124.44), as compared to Black primes (M =
625.91, SD = 118.31), t(l44) = 3.93, p < .001, d = .24. This suggests there was greater
stereotype activation when participants looked at White female pictures, as compared to
when they looked at Black female pictures. In contrast, there was no difference in
response latencies between Black (M = 641.34, SD = 136.34) and White targets (M =

643.29, SD = 163.54) when White participants responded to non-stereotypic words, t(l44)

.08, p = .94. The main effect of prime set was not significant, F(1, 143) = .2584, MSE

.11, p = .62. The findings that participants responded faster to stereotypic words that
are supposed to be associated with Black women following White primes, as compared to
Black primes, suggest that the lexical decision task was not effective.

Explicit Affective Reactions to Black Women

The means and standard deviations of Whites’ explicit liking toward Black

women as a function of skin tone and facial features are presented in Table 16. The three-

59

SU

factor mixed model ANOVA on explicit reports of liking toward Black women revealed
that the main effect of skin tone was significant, F (1, 143) = 10.56, MSE = .40, p < .001,
d = .22, such that White participants liked light-skinned Black women significantly more
(M = 2.92, SD = .41) than dark-skinned Black women (M = 2.73, SD = .93). Neither the
main effect of facial features (F (l, 143) = .31, MSE = .40, p = .58) nor the main effect of
prime set (F(1, 143) = .22, MSE = 2.12,p = .64) was Significant.

However, the two-way interaction between skin tone and facial features was
marginally Significant, F (l, 143) = 3.1 l, MSE = .35, p = .08. Examination of the mean
suggests that White participants liked Black women with more prototypical facial
features to the same degree as Black women with less prototypical facial features when
Black women had lighter skin tone. In contrast, when Black women had darker skin tone,
White participants liked Black women with less prototypical facial features more than
Black women with more prototypical facial features. The pattern of the means suggest
that facial features have negative effects on Black women when they have darker skin
tone but not when they have lighter skin tone.

Discussion

The goal of Study 3 was to examine how Skin tone and facial features influence
Whites’ affective and cognitive reactions toward Black women and investigates the
potential moderating role of target gender. Interestingly, the results from Study 3a
suggest that the effects of Skin tone and facial features are different between Black female
targets and Black male targets. Study 3a showed that there was no difference in how fast
White individuals responded to negative words following the primes. Remember that

White participants responded faster to negative words following Black men with darker

60

skin tone or those with more prototypical facial features. The absence of the mean
differences in negative attitudes toward Black women measured at the implicit level
suggests that having darker skin tone or more prototypical facial features do not
necessarily increase Whites’ negative feelings toward Black women.

Instead, Whites responded Significantly faster to positive words following Black
women with lighter Skin tone, as compared to following Black women with darker skin
tone, indicating that Whites felt more positively toward Black women with lighter skin
tone than toward those with darker skin tone. Again, the findings suggest that there are
differences between Black female targets and Black male targets, as Study 1 showed that
having lighter Skin tone does not increase Whites’ positive feelings toward Black men. In
addition, unlike Black male targets, facial features did not have an additive effect on
Whites’ attitudes toward Black women.

The results for explicit liking toward Black women were consistent with those of
implicit attitudes. White participants liked Black women with lighter Skin tone more than
Black women with darker Skin tone. Although Study 3a showed that there were no effects
of facial features on Whites’ explicit liking toward Black women, the effects of facial
features emerged in Study 3b. Interestingly, there were facial features effects when Black
women had darker skin tone, but not when they had lighter skin tone. More specifically,
dark-Skinned Black women with less prototypical facial features were more liked by
Whites, as compared to dark-Skinned Black women with more prototypical facial features.
However, these effects were marginally significant. Thus, the results of explicit attitudes
were consistent between Studies 3a and 3b. The findings that Whites liked Black women

with lighter skin tone than those with darker skin tone are consistent with the previous

61

studies Showing that people perceive and treat Black women differently based on their
skin tone (Keith & Herring, 1991; Maddox & Gray, 2002; Wade et al., 2004). However,
unlike the previous studies arguing that Black women with darker skin tone are perceived
and treated more negatively, the present study (which was able to examine the direction
of attitude change) demonstrates that Black women with lighter skin tone are perceived
and treated more positively.

Just like Study 2, Study 3b showed no evidence that skin tone or facial features
influenced how Whites cognitively react to Black women, as White individuals
responded significantly faster to stereotypic negative words, as compared to non-
stereotypic negative words, following Black female primes, regardless of skin tone and
prototypicality of facial features. Again, it is important to point out that the lexical
decision task modified for this study might have failed to capture activation of
stereotypes associated with Black women. This is because White participants responded
significantly faster to stereotypic words associated with Black women following White
controls, as compared to following Black primes.

Taken together, Studies 1 and 3a demonstrated that the nature of the skin tone
effects is different between Black men and Black women, such that having darker skin
tone increases Whites’ negativity for Black men, but not for Black women, and that
having lighter skin tone increases Whites’ positivity for Black women, but not for Black
men. In addition, these studies showed that White individuals use facial features when
forming impressions of Blacks only when the targets were men.

Studies 1 through 3 examined the independent effects of skin tone and facial

features on how Whites affectively and cognitively react to Blacks. These studies

62

demonstrated that different physical characteristics affect White perceivers’ feelings and
thoughts about Black individuals in different ways and that these effects are different
depending on target gender. However, there are still missing pieces of information in the
current research—the effects of target physical characteristics on Black perceivers. In the
next study, the effects of Skin tone and facial features are investigated from the Blacks’
perspectives.
Study 4

Study 4 examined how one’S own Skin tone and facial features influence Black
individuals’ 1) tendency to attribute ambiguous negative events to racism, 2) experiences
with racism, and 3) sensitivity to racism. Furthermore, Study 4 also investigated how
Blacks respond, both affectively and cognitively, to fellow ingroup members who vary in
skin tone and facial features.

Method

Participants

Eighty-four self-identified Black undergraduate students (women = 85.7 %, M age
= 19.51, SD = 2.44 years old) participated in this study. One female participant was
excluded from the data, because she did not agree to release her photo.
Procedures

Prior to the laboratory session, participants were required to complete an online
questionnaire that included measures of racial group identification, self-esteem, past
experience with discrimination, perception of discrimination, and demographic
information (see Appendix G for a list of measures and items for each measure).

Participants later reported to the laboratory session in a group (up to five participants per

63

session) and were greeted by either a White American or an Asian American
experimenter. The experimenter led participants to a room equipped with five computers
and informed participants that the purpose of the study was to understand how
personality influences the way people interpret and respond to certain situations.

After providing consent, participants were told that they would complete a few
tasks on the computer. All instructions were provided on the computer screen as
participants proceeded at their own pace. In the first task, participants were told to read a
series of scenarios and imagine that they had experienced the negative events described
in the scenarios. Participants were then asked to attribute each negative event to several
potential causes. In the second task, participants were asked to complete a race-based
rejection sensitivity measure in which they were asked to read several scenarios that
described negative events and indicate how concerned they would be about the events.

After completing the attribution and rejection sensitivity measures, participants
completed the modified sequential priming and lexical decision tasks from Studies 1
through 3. Participants were told that these computer tasks examine people’s memory and

judgment Skills. Of 83 participants, 41 participants completed the sequential priming task
with male target pictures and the lexical decision task with female target pictures.9 The

other 42 participants completed the sequential priming task with female target pictures

and the lexical decision task with male target pictures. 10 Again, all participants
completed the tasks with Black female targets second because these tasks were
considerably shorter than tasks for Black male targets and they were preliminary. At the

end of the computer tasks, all participants were also asked to report their liking toward

both male and female targets.

64

Once all participants in a session completed the computer tasks, the experimenter
took three different photographs of them. For the first picture, participants were told to
make a neutral facial expression. For the second picture, participants were asked to think
of a recent event that made them really happy and facially express that emotion. For the
third picture, participants were asked to think of a recent event that made them really
angry and facially express that emotion. Three different pictures were taken to create a
pool of facial stimuli that will be used in future research. However, in the current study,
only the neutral faces of these participants were used to record their Skin tone and facial
features.

After obtaining the pictures, participants were fully debriefed. More specifically,
participants were told that the research examined the relation between participants’
physical characteristics (i.e., facial features and skin tone) and their sensitivity to racial
discrimination. Participants were further told that, in order to examine the relation, it was
important for the researchers to analyze participants’ facial pictures in terms of facial
features and skin tone. After the debriefing process, participants were asked whether their
pictures could be used for analyses of Skin tone and facial features. If they agreed to
release their pictures for physical feature analyses, the experimenter obtained
participants’ signature on the picture release form.

Measures

Pre-Iaboratory questionnaires.

Group identification. The importance of being Black to participants’ sense of self
was assessed with Luhtanen and Crocker’s (1992) four-item identity centrality subscale:

“Overall, my racial group membership has very little to do with how I feel about myself

65

(R),” “The racial group I belong to is an important reflection of who I am,” “The racial
group I belong to is unimportant to my sense of what kind of person I am (R),” and “In
general, belonging to my racial group iS an important part of my self-image.” The scale
ranged from 1 (Strongly Disagree) to 7 (Strongly Agree), with higher numbers indicating
that participants strongly identified with Black (a = .68).

Self-esteem. Participants’ self-esteem was assessed with Rosenberg’s (1965) 10-
item measure and with Heatherton and Polivy’s (1991) 20-item state self-esteem scale,
which consists of three subscales: social, performance, and appearance. Example items
are: “I feel that I am a person of worth, at least on an equal basis with others,” “ I am
worried about what other people think of me,” and “I feel displease with myself.” The
scale ranged from 1 (Strongly Disagree) to 7 (Strongly Agree), with higher scores
indicating higher self-esteem. Scale reliabilities for subscales were: a = .85 for social, a
= .84 for performance, and a = .80 for appearance. Scale reliability for overall self-
esteem was a = .90.

Past experiences with discrimination. Participants’ experiences with racial
discrimination in the past were assessed with Landrine and Klonoff’s (1996) l8-item
Schedule of Racist Events. Each of the first 17 items provided a Specific type of racial
discrimination and asked three questions. The first two questions asked participants how
often they experienced this type of racial discrimination in the past year (i.e., experiences
with discrimination in the past 1 year) and in their entire life (i.e., lifetime experiences
with discrimination), respectively. Example items include “How many times have you
been treated unfairly by teachers and professors because you are Black?,” and “How

many times have you been treated unfairly by your employers, bosses, and supervisors

66

because you are Black?.” The scale ranged from 1 (Never) to 6 (Almost All of the Time).
The third question asked participants how stressful each type of racial discrimination was
for them (i.e., appraisal of discrimination), and the scale ranged from 1 (Not at All) to 6
(Extremely). Finally, the last item in the measure asked participants how different their
life would be if they were not treated in an unfair way in the past year and in their entire
life. For this last item, the scale ranged from 1 (Same as Now) to 6 (Totally Different).
Higher scores indicate more frequent and stressful experiences with racial discrimination.
Scale reliability for experiences with discrimination in the past 1 year was a = .94, and
that for lifetime experiences with discrimination was a = .94. Finally, scale reliability for
appraisal of discrimination was a = .95.

Perception of discrimination. Participants’ perception of their experiences with
racial discrimination were assessed with five items: “I experience discrimination because
of my race,” “I feel that I am discriminated against because of my race,” “I personally
have been a victim of racial discrimination,” “I feel like I am personally a victim of
society because of my race,” and “I consider myself a person who is deprived of
Opportunities that are available to others because of my race.” The last two items were
used in Branscombe, Schmitt, and Harvey’s study (1999). The scale ranged from 1
(Strongly Disagree) to 7 (Strongly Agree), with higher scores indicating that participants
are more likely to perceive discrimination based on their race (a = .89).

Demographic information. Participants were asked to provide the following basic
demographic information: age, gender, race, year in school, number of same-race and
different-race close fiiends on and off the MSU campus, family’s socioeconomic status,

and current overall GPA.

67

Measures used in the laboratory session.

Attribution to discrimination. Participants read six different scenarios that
described negative events in different situations that could be attributed to racial
discrimination or to other causes. They were asked to imagine that they have experienced
these ambiguous negative events and think of the cause of the negative events. Example
scenarios are: “Imagine that you went to a ‘fancy’ restaurant. Your server seemed to be
taking care of all the other customers except you. You were the last person whose order
was taken,” and “Imagine that you were to look at an apartment for rent. The manager of
the building refused to show it to you, saying that it has already been rented.” After
reading each scenario, participants were asked to rate their perception of the cause of the
experience for each of the following six attribution categories: 1) racial discrimination, 2)
gender discrimination, 3) internal (e.g., their personality, mood, abilities, or intelligence),
4) external (e. g., interaction partners’ personality, social norm, lack of opportunity), 5)
skin tone, and 6) facial features. The scales ranged from 1 (Very Unlikely the Cause of
the Experience) to 7 (Very Likely the Cause of the Experience). Attribution scores for

each of the six different attribution categories were computed by averaging over the six

different scenarios.11 Please see Appendix H for the detailed explanations of the

procedure used in the selection of the scenarios and attribution items.

Race-based rejection sensitivity. Participants’ sensitivity to rejection by others
based on their racial group membership was measured with Mendoza-Demon, Downey,
Purdie, David, and Pietrazk’ (2002) 12-item Race-Based Rejection Sensitivity scale.
Each item provided participants a scenario in which participants could be rejected and

asked two questions: 1) how concemed/anxious participants would be, and 2) how likely

68

participants think that they would be rejected. Example items include “Imagine that you
are in a pharmacy, trying to pick out a few items. While you’re looking at the different
brands, you notice one of the store clerks glancing your way,” “Imagine you are riding
the bus one day. The bus is full except for two seats, one of which is next to you. As the
bus comes to the next stop, you notice a woman getting on the bus,” and “Imagine that
you are standing in line for the ATM machine, and you notice the woman at the machine
glances back while she’s getting her money” (see Appendix I for the complete list of the
items). The scale ranged from 1 (Very Unconcemed/Very Unlikely) to 6 (Very
Concerned/Very Likely). The rejection sensitivity score was computed first by
multiplying the two scores for each scenario. Then, product scores for each of the 12
scenarios were averaged to come up with a total score. Therefore, scores range between 1
and 49. Higher total scores indicate that participants are more sensitive to race-based
rejection (or = .86).

Modified sequential priming task. In order to assess Black participants’
automatic affective reactions toward their fellow ingroup members, the same sequential
priming task as Studies 1 and 3a were used in this study.

Modified lexical decision task. The same lexical decision task in Studies 2 and
3b were used to examine stereotype activation among Black individuals when exposed to
fellow ingroup members.

Explicit liking toward targets. Black participants’ liking toward Black male and
Black female targets presented during the sequential priming and lexical decision tasks
was assessed with an explicit, self-report measure. The scale was the same as the

previous studies and ranged from 0 (I would DISLIKE this person very much) to 5 (I

69

would LIKE this person very much), with higher scores indicating more liking toward
targets.
Coding of Participants’ Photos

Before analyzing skin tone and facial features, pictures of 83 participants were
edited so that all background features were replaced by a solid gray background, all
clothing was replaced with a plain black t-shirt, and all pictured were transformed into an
outlined format.
Objective Skin Tone

The average luminosity was computed for each face by averaging the luminosity
value obtained from each pixel across a given face.
Subjective Skin Tone

The average skin colors were created for each face first. In order to create the
average skin color for each face, color information values (i.e., values of red, blue, and
green) were computed for each pixel and averaged over a face, so that each face was
associated with only one value for each red, blue, and green. These three values were
entered into a color information window on Adobe Photoshop CS3 to create one solid
color panel that represented the average skin color for a face. The average skin colors
were used for this task, because skin color even within the same face differs depending
on the areas being looked at (e. g., skin color of the forehead is usually lighter than that of
below the eyes). Then, the solid color panels that represent average skin tone were
presented to 153 self-identified White participants (women = 77.1 %, M age = 19.48, SD

= 2.21 years old). Participants were asked to rate each of the average skin colors in terms

70

 

of their darkness/lightness, using a scale ranging from 1 (Extremely Dark) to 7
(Extremely Light). See Appendix J for examples of the average skin color panels.
Objective Facial Features

Two independent researchers measured the length and width of a nose, lips, and a
face in millimeter for each participant’s picture. Then, the ratio of nose width (i.e., nose
width divided by face width) and lip thickness (i.e., lip thickness divided by face length)
were computed for each face. In order to create a single score that represents the overall
facial feature rating, raw scores of both nose and lip ratios were transformed into 2 scores
and then they were averaged.
Subjective Facial Features

Fifty-five self-identified White participants (women = 63.6 %, M age = 19.47, SD
= 1.36 years old) looked at the outlined pictures on a computer screen. They were asked
to rate each of the 83 faces in terms of the degree of prototypicality of facial features. In
addition, participants also rated each face in terms of attractiveness, babyfacedness, and
hostility, and these measures serve as control variables in the following analyses. The
instructions for these ratings were the same as those used in Pilot Study 1, Step 4.

In sum, each of the 83 Black participants’ pictures was associated with four main
scores: luminosity (i.e., objective skin tone measure), rating of color panels (i.e.,
subjective skin tone measure), average a score of nose and lip ratios (i.e., objective facial
feature measure), and rating of Afrocentric features (i.e., subjective facial feature
measure), as well as scores for the three control variables.

Analysis Plan

71

In order to examine whether skin tone and facial features influence Black
individuals’ experiences in intergroup relations, two sets of moderated regression
analyses were conducted for each criterion variable. The first set of analyses looked at the
effects of objective ratings of participants’ physical characteristics (i.e., skin tone
computed with luminosity and facial features measured by experimenters), and the
second set looked at the effects of subjective ratings of participants’ physical
characteristics (i.e., skin tone and facial features rated by White participants).

In each set of analyses, four separate moderated regression analyses were run. The
first model included the main effects of participants’ skin tone, facial features, and gender,
and two-way interaction between skin tone and facial features. Because there were only
12 Black male participants, the interaction between skin tone and gender and between
facial features and gender could not be entered into the model. Participants’ gender was
coded as 0 (women) and 1 (men), and facial feature and skin tone were centered using
grand means before being entered into the regression model. The second through fourth
model included control factors that were reported to be related to skin tone and
Afrocentric features in the previous studies (i.e., attractiveness, babyfacedness, and
hostility). Because attractiveness, babyfacedness, and hostility were significantly related
(see Tables 17 and 18 for Black men and Black women separately), these control
variables were entered into a regression model separately, instead of being entered into
one model simultaneously. It is important to note that only the results from the first
model are reported in the following sections as well as in the tables. When the inclusion

of a control variable in the model changes the results from the first model, the nature of

72

the changes are discussed in the results section, but the results reported in the tables are
still from the first model.
Results

Means and standard deviations of objective and subjective measures of skin tone
and facial features as well as control variables (i.e., attractiveness, babyfacedness, and
hostility), and correlations among all variables are provided in Tables 17 and 18
separately for Black male and Black female participants, respectively. These correlations
showed that the subjective ratings of skin tone and prototypicality of facial features (i.e.,
skin tone and facial features rated by White participants) closely mirror the objective
measures of skin tone and facial features (i.e., skin tone computed with luminosity and
facial features measured by experimenters), respectively, suggesting that White
participants could accurately perceive these physical characteristics.

Skin tone was significantly and positively related to attractiveness and
babyfacedness for Black men such that Black men with darker skin tone were perceived
by White individuals to be less attractive and less babyfaced, as compared to Black men
with lighter skin tone. Unlike the the pilot study, facial features were not related to
attractiveness, babyfacedness, or hostility. However, it is important to note that there
were only 12 Black male pictures in Study 4 as oppose to 120 pictures in the pilot study.

In contrast, luminosity was negatively related to hostility among Black women,
such that Black women with darker skin tone were perceived to be more hostile than
those with lighter skin tone. In addition, ratings of facial features were significantly
related to attractiveness, babyfacedness, and hostility, such that Black women who were

rated as more prototypic of Afiican Americans were perceived to be less attractive, less

73

babyfaced, and more hostile, as compared to Black women who were rated as less
prototypic of African Americans.

In order to get a broad sense of how each of the criterion variables is related to
one another, correlations among all variables are provided in Table 19. Table 19 also
presents the means and standard deviations of the criterion variables. The correlations
indicate that racism attributions and race-based rejection sensitivity were significantly
correlated with all other criterion variables, such that the more participants attribute the
negative events to racism (or the more sensitive they are), the more they attribute the
negative events to other causes and the more they report experiences with discrimination.
The results suggest that people who are more sensitive to racism are more likely to be
extreme in their attribution judgments in general, as compared to those who are less
sensitive to racism. Overall, the examination of the correlations suggests that all criterion
variables are highly related to one another.

Attributions of Ambiguous Negative Events

It the following section, it was examined whether Black participants’ tendency to
attribute ambiguous negative events to certain causes are influenced by their skin tone
and facial features.

Attribution to racism. Table 20 presents the results of racism attribution. The
first model, including the objective measures of the Black participants’ skin tone and
facial features, revealed that the model does not account for variance in attribution to
racism more than by chance, and none of the main effects or the interaction between skin

tone and facial features was significant.

74

The same analysis was run replacing objective measures of the Black participants’
skin tone and facial features with subjectively measured participants’ skin tone and facial
features. The results showed that the main effect of facial features was significant, such
that as rated prototypicality of facial features increased by one unit (i.e., being rated as
more prototypical of Blacks), Blacks’ tendency to attribute ambiguously negative events
to racism decreased by .67.

Attribution to sexism. Neither the first set of models including objective
measures of skin tone and facial features nor the second set of models including
subjective measures of skin tone and facial features predicted Black individuals’
tendency to attribute ambiguously negative events to sexism (see Table 21). Although it
would be appropriate to examine whether the effects of Black participants’ skin tone and
facial features on their tendency to make sexism attributions are moderated by participant
gender, the model could not include the interaction between skin tone and gender and
between facial features and gender due to lack of Black male participants.

Attribution to skin tone. Table 22 presents the summary of the results of skin
tone attribution. The model including the objective measures of Black participants’ skin
tone and facial features accounted for about 11.8 % of variance in skin tone attribution
scores. In this model, both the main effect of facial features and the main effect of
participant gender were significant. When holding all other variables constant, as the
average of standardized nose and lip ratios increases by one unit, Blacks’ tendency to
attribute negative events to skin tone decreased by .51. In addition, when holding all
other variables constant, Black men were more likely to attribute negative events to skin

tone, as compared to Black women.

75

Next, the moderated regression analysis including the subjective measures of the
Black participants’ skin tone and facial features was conducted. The model accounted for
about 17.6 % of variance in the skin tone attribution scores. Similarly to the model
reported above, the main effect of facial features and the main effect of participants’
gender were significant. As rated prototypicality of their faces increased by one unit,
Black participants’ tendency to attribute negative events to skin tone decreased by 1.03,
when controlling for all other variables constant. In addition, Black male participants
were more likely to attribute negative events to skin tone, as compared to Black female
participants, when holding other variables constant. The results suggest that Black men
are more likely to be aware of the negative consequences of skin tone than Black women.
Because the majority of the sample (85.7 %) was women, the negative relationship
between facial features and skin tone attribution may reflect the experiences of women,
who do not experience the negative consequences of skin tone and facial features as the
results from Study 3 suggest.

When three addition models, in which control variables were entered in addition
to the main effects of subjective measures of the participants’ skin tone, facial features,
and participant gender, all three models resulted in the marginally significant two-way
interaction between skin tone and facial features. Because the pattern of the results were
the same for the all three additional models, simple slopes of facial features was
examined using the second model in which rated attractiveness was entered as a control
variable. Figure 1 shows the nature of the interaction between participants’ skin tone and
facial features. Simple slopes of facial features were examined at 1 standard deviation

above the mean (i.e., lighter skin tone) and at 1 standard deviation below the mean (i.e.,

76

darker skin tone). The analyses revealed that the simple slope of facial features was
significant when participants had darker skin tone, B = -1.54, SE = .49, t(81) = -3.13, p

< .01, such that the more prototypical facial features participants had, the less likely they
were to make skin tone attribution. In contrast, the simple slope of facial features was not
significant when participants had lighter skin tone, B = -.53, SE = .37, t(81) = -1.43, p

= .16 (see Figure 1).

Attribution to facial features. Table 23 presents the summary of the results of
facial feature attribution. In the first model in which attribution to facial features was
predicted by objective skin tone and facial features of Black participants, the two-way
interaction between participants’ objective skin tone and facial features was marginally
significant. Examination of the patterns suggest that the relationship between facial
features and tendency to attribute negative events to facial features was negative when
patricians had darker skin tone, but it was positive when participants had lighter skin tone
(see Figure 2). In other words, participants with darker skin tone and less prototypical
facial features were more likely to make facial feature attribution, as compared to those
with darker skin tone and more prototypical facial features. In contrast, participants with
lighter skin tone and more prototypical facial features were more likely to make facial
feature attribution, as compared to those with lighter skin tone and less prototypical facial
features.

In contrast, the subjective measures of Black participants’ skin tone, facial
features, gender, and two-way interaction between skin tone and facial did not predict

their tendency to attribute ambiguously negative events to facial features.

77

Attribution to internal factors. See Table 24 for the results of interaction
attribution. The first analysis with objective measures of the Black participants’ skin tone
and facial features revealed a marginally significant main effect of skin tone, such that as
luminosity increases by one unit (i.e., as skin tone is rated lighter), Blacks’ tendency to
attribute ambiguously negative events to internal factors unrelated to their race increased
by 3.60, when holding all other variables constant.

The second analysis with the subjective measures of the Black participants’ skin
tone and facial features also revealed the significant main effect of skin tone, such that as
Whites’ subjective ratings of Black participants’ skin tone increase by one unit (i.e., as
skin tone is rated lighter), Blacks’ tendency to attribute negative events to internal factors
also increased by .25. However, this significant main effect was qualified by a marginally
significant two-way interaction between skin tone and facial features. Examination of the
pattern suggests that the relationship between facial features and internal attribution was
negative when Black participants had darker skin tone, but it was positive when
participants had lighter skin tone (see Figure 3). Thus, among Black participants with
darker skin tone, those with less prototypical facial features were more likely to attribute
negative events to internal factors unrelated to their race, as compared to those more
prototypical facial features. In contrast, among Black participants with lighter skin tone,
those with more prototypical facial features were more likely to make internal attributions,
as compared to those with less prototypical facial features.

Attribution to external factors. Neither the first model including the objective
measures of the Black participants’ skin tone and facial features nor the second model

including the subjective measures of the Black participants’ skin tone and facial features

78

predicted their tendency to attribute ambiguously negative events to external factors (see
Table 25 for the results).

Taken together, there were no consistent patterns to explain how Black
individuals’ skin tone and facial features influence their tendency to attribute
ambiguously negative events to certain causes. The objectively measured skin tone and
facial features were significant predictors in some cases, and the subjectively measured
skin tone and facial features were significant predictors in other cases. One thing that is
worth mentioning is that the effects of facial features when predicting skin tone and facial
features attributions were opposite from the predictions. Specifically, Black participants
with darker skin tone with less prototypical facial features were more likely to attribute
negative events to skin tone and facial features than Black participants with more
prototypical facial features. These unpredicted findings might be due to lack of statistical
power.

Experiences with Racial Discrimination

In addition to examining how one’s own skin tone and facial features affect Black
individuals’ cognitive tendency, the study also examined how their physical
characteristics influence their past experiences with racial discrimination. There are two
measures of experiences with racial discrimination: one for the experiences in the past
one year, one for the lifetime experiences with discrimination. The overall means for both
measures were fairly low, suggesting that Black participants do not report frequent
experiences with racial discrimination in general. The first analysis used the objective
measures of the Black participants’ skin tone and facial features to predict their

experiences with racial discrimination in the last year. See Table 26 for the summary of

79

the results. This analysis resulted in a marginally significant main effect of facial features.
When holding all other variables constant, as the average of standardized nose and lip
ratios increased by one unit, Blacks’ experiences with discrimination in the past one year
also increased by .23. In other words, Black individuals with more prototypical facial
features were more likely to experience racial discrimination in the past one year, as
compared to those with less prototypical facial features. The results changed with control
analysis including hostility (but not attractiveness or babyfacedness). More specifically,

in the model that included hostility as a control, the main effect of facial features was
significant (b = .29, SE = .13, ,8 = .25, t(82) = 2.13,p < .05).

The second model including the subjective measures of the Black participants’
skin tone and facial features showed that none of the main effects of skin tone, facial
features, and participant gender was significant. However, the analysis resulted in a
marginally significant two-way interaction between skin tone and facial features. Figure 4
shows the pattern of the interaction between participants’ skin tone and facial features.
Examination of the pattern suggest that, when Black individuals have darker skin tone,
those with more prototypical facial features were more likely to report experiencing racial
discrimination in the past one year, as compared to those with less prototypical facial
features. However, facial features seem to be unrelated to experiences of discrimination
for Black individuals with lighter skin tone.

The results changed with control analysis including hostility (but not
attractiveness or babyfacedness). The model that included hostility as a control resulted
in a significant main effect of facial features. The examination of coefficient suggest that

as rated prototypicality of Blacks’ facial features increases by one unit, their tendency to

80

report experiences of discrimination also increases by .42 (b = .42, SE = .19, ,6 = .26, t(82)
= 2.18, p < .05). This significant main effect was qualified by the significant main effect
of skin tone and facial features.

Next, the same sets of analyses were conduced on Black individuals’ lifetime
experiences with racial discrimination. See Tables 27 for the summary of the results. The
first model including the objective measures of participants’ skin tone and facial features
revealed that none of the main effects of skin tone, facial features, participant gender, or
two-way interaction between skin tone and facial features was significant. However, the
results changed with a control analysis including hostility (but not attractiveness and
babyfacedbess). More specifically, once hostility was entered into the model as a control
variable, the main effect of facial features became significant. Holding all other variables
constant, as the average standardized nose and lip ratios increases by one unit, Blacks’
tendency to report lifetime experiences with racial discrimination also increases by .31 (b
= .31, SE = .15, fl = .24, t(82) = 2.02,p < .05).

Similar results were found with the second sets of analyses with models including
the subjective measures of the Black participants’ skin tone and facial features. None of
the main predictor variables was significant when the model included only the main
effects of participants’ skin tone, facial features, and gender, and a two-way interaction
between skin tone and facial features. However, a control analysis including hostility (but
not attractiveness and babyfacedness) resulted in the different results. Specifically, once
hostility was entered into the model as a control, the main effect of facial features became
marginally significant. Examination of coefficients suggests that Black individuals with

more prototypical facial features were more likely to report experiencing racial

81

discrimination than those with less prototypical facial features (b = .3 7, SE = .22, fl = .20,
t(82) = 1.65, p = .10). In addition, the effect of hostility was significant (b = -.32, SE

= .14, ,B = -.27, t(82) = -2.32, p < .05), such that Black individuals seen as more hostile
were less likely to report experiences with racial discrimination in the past, as compared
to those seen as less hostile.

Taken together, as predicted, Black individuals with more prototypical facial
features were more likely to report experiencing racial discrimination in the past than
their counterparts with less prototypical facial features. However, inconsistent with the
prediction, Black individuals’ skin tone did not predict their past experiences with racial
discrimination. The absence of skin tone effects might be due to lack of Black male
participants.

Race-Based Rejection Sensitivity

Next, the effects of one’s own skin tone and facial features on Black participants’
sensitivity to racism were examined using race-based rejection sensitivity. Table 28
presents the results of Black individuals’ sensitivity to racism as a function of their skin
tone and facial features. Both analyses including the objective measures and the
subjective measures of the Black participants’ skin tone and facial features resulted in
marginally significant two-way interactions between skin tone and facial features.
Analyses of simple slopes of facial features at different levels of skin tone revealed the
same results for the both objective and subjective measures of participants’ physical
characteristics. Figures 5 and 6 show the nature of the interactions between skin tone and
facial features with objective and subjective measures, respectively. Specifically, among

the individuals with darker skin tone, those with less prototypical facial features are more

82

sensitive to racism, as compared to those with more prototypical facial features (B = -2.87,
SE = 1.95, t(81) = -1 .48,p = .14 for objective measures, and B = -4.13, SE = 2.89, t(81) =
-1.43, p = .16 for subjective measures). In contrast, among the individuals with lighter
skin tone, those with more prototypical facial features are more sensitive to racism, as
compared to those with less prototypical facial features (B = 2.11, SE = 1.73, t(81) = 1.22,
p = .23 for objective measures, and B = 2.25, SE = 2.19, t(81) = 1.03, p = .31 for
subjective measures).

The results of race-based rejection sensitivity was consistent with skin tone and
facial feature attributions, which are opposite from the predictions. Taken together, when
Black individuals have darker skin tone, those with more prototypical facial features are
less likely to be sensitive to racism and attribute negative events to skin tone and facial
features, as compared to those with more prototypical facial features. The results of facial
features effects were consistent when Blacks had lighter skin tone, such that those with
more prototypical facial features were more likely to be sensitive to racism and attribute
negative events to skin tone and facial features, as compared to those with less
prototypical facial features.

Other Criterion Variables

The same sets of analyses were conducted on other predictor variables measured
with the pre-laboratory questionnaire: racial identification, perception of discrimination,
overall self-esteem, social self-esteem, and appearance self-esteem. None of the models
showed evidence that Black participants’ skin tone or facial features predicted their
responses. The results of all analyses are presented in Tables 29 through 33.

Implicit Affective Reactions to Black Men

83

In addition to examining how one’s own physical characteristics influence
Blacks’ attribution tendencies, sensitivity to racism, and experiences with discrimination,
Study 4 also examined how Black targets’ physical characteristics influence Black
perceivers’ affective and cognitive responses toward them. Again, it is important to note
that half of the participants completed the sequential priming task with Black male targets
and the lexical decision task with Black female targets, and the other half completed the
lexical decision task with Black male targets and the sequential priming task with Black
female targets. Thus, the dfs in the following analyses, except the explicit liking section,
reflect the small sample size for these tasks.

Implicit negative attitudes. The following analyses examined whether Black
individuals’ affective reactions toward fellow male ingroup members were influenced by
target’s skin tone and facial features. A two-factor repeated ANOVA, treating target skin

tone (dark, light) and target facial features (more prototypical, less prototypical) as

. . . . . . . . 12
wrthrn-partrcrpants factors, was conducted on the facrlrtatron scores for negative words.

See Table 34 for the means and standard deviations13 for each prime category. The

analysis revealed that there was no main effect of target facial features, F (1, 39) = .90,
MSE = .003, p = .35. The main effect of target skin tone was marginally significant, F (1 ,
39) = 3.36, MSE = .004, p = .08, d = .14, such that Black participants responded faster to
negative words following light-skinned Black male primes (M = 27.28, SD = 106.95), as
compared to dark-skinned Black male primes (M = 9.61 , SD = 142.07). However, the
two-way interaction between target skin tone and facial features was also marginally
significant, F (1, 39) = 3.22, MSE = .003, p = .08. In the dark skin tone condition, Black

participants responded faster to negative words following Black men with more

84

prototypical facial features than Black men with less prototypical facial features. The
results suggest that Black participants felt more negatively toward Black men with darker
skin tone and more prototypical facial features than toward Black men with darker skin
tone and less prototypical facial features. However, it is important to note that the
differences between more or less prototypical target facial features are very small. In
contrast, in the light skin tone condition, Black participants responded faster to negative
words following Black men with less prototypical facial features than following Black
men with more prototypical facial features. These results indicate that Black participants
felt more negatively toward Black men with lighter skin tone and less prototypical facial
features than toward Black men with lighter skin tone and more prototypical facial
features.

Implicit positive attitudes. Next, participants’ responses to positive words
following male target pictures were examined using the same two-factor repeated
ANOVA. Table 35 presents the means and standard deviations for each prime category.
The analysis reveled that none of the main effects of skin tone (F (1 , 39) = .15, MSE
= .004, p = .70), facial features (F (1, 39) = .15, MSE = .002, p = .70), or the interaction
between skin tone and facial features (F (1 , 39) = .01, MSE = .004, p = .92) was
significant. Thus, there was no evidence that Black perceivers’ positive feelings toward
fellow ingroup members are influenced by targets’ skin tone and facial features.

Black vs. White control. In order to examine whether Black perceivers’
evaluations of Black men are different from the White control, facilitation scores for four
Black prime categories were averaged. First, a paired t-test on facilitation scores for the

negative words was conducted, and it reveled that there are no effect of target race. Thus,

85

there were no significant mean differences between facilitation scores for Black targets
(M= 18.44, SD = 122.82) and White targets (M= 20.86, SD = 113.12), t(39) =.36,p

= .72. The same analysis was conducted on facilitations scores for positive words and
revealed a marginally significant effect of target race, t(39) = .36, p = .09, d = .07. More
specifically, Black perceivers’ positive feelings were inhibited to a greater degree
following White targets (M = -28.96, SD = 124.60), as compared to following Black
targets (M = -l9.63, SD = 129.09).

Implicit Affective Reactions to Black Women

Implicit negative attitudes. Tables 36 presents the means and standard
deviations of facilitation scores for negative words in each prime category. Due to
computer malfunctions, the number of participants who were included in this section of
analyses was very low. Thus, it is important to keep in mind that the interpretation of the
results must been done with caution. A two-factor repeated ANOVA on the facilitation
scores for negative words revealed that there was no main effect of target skin tone (F (1 ,
24) = .07, MSE = .003, p = .80) nor the main effect of target facial features (F(1, 24)
= .32, MSE = .01, p = .58). The interaction between skin tone and facial features was not
significant either, F(1, 24) = .19, MSE = .01,p = .67.

Implicit positive attitudes. Next, participants’ responses to positive words
following female target pictures were examined using the same two-factor repeated
ANOVA (see Tables 37 for the means and standard deviations for each prime category).
The analysis reveled that neither the main effects of target skin tone (F (1, 24) = .01, MSE

= .01 , p = .93), nor the main effect of target facial features (F(1, 24) = .16, MSE = .01, p

86

.69) was significant. The interaction between skin tone and facial features (F (l, 24)

.60, MSE = .01, p = .45) was not significant either.

Black vs. White control. A paired t-test on facilitation scores for negative
words revealed that there was no effect of target race, t(24) = .08, p = .94, suggesting that
there were no mean differences in facilitation scores for the negative words between
Black targets (M = 90.44, SD = 105.03) and White targets (M = 91.73, SD = 152.45). The
same analysis was conducted for positive words and revealed that there were no mean
differences in facilitation scores for the positive words between Black targets (M = 76.59,
SD = 113.23) and White targets (M = 70.38, SD = 147.31), t(24) = .46, p = .65.

Explicit Affective Reactions to Fellow Ingroup Members

Liking toward Black men. Next, Black participants’ explicit liking toward
fellow ingroup members who vary in skin tone and facial features were examined. The
means and standard deviations of explicit liking toward Black men as a function of skin
tone and facial features are presented in Table 38. A two-factor repeated ANOVA,
treating target skin tone (dark, light) and target facial features (more prototypical, less
prototypical) as within-participants factors, was conducted to examine the effects of skin
tone and facial features on Black individuals’ explicit reports of liking toward male
ingroup members. The main effect of target skin tone was significant, F (1 , 82) = 12.62,
MSE = .12, p < .001, d = .15, such that Black participants liked light-skinned Black men
significantly more (M = 3.17, SD = .88) than dark-skinned Black men (M = 3.04, SD
= .92). In addition, the main effect of target facial features was also significant, F (1 , 82) =
7.29, MSE = .11, p < .01, d = .11, such that Black participants liked Black men with less

prototypical facial features significantly more (M = 3.15, SD = .90) than Black men with

87

more prototypical facial features (M = 3.06, SD = .91). The interaction between skin tone
and facial features was not significant, F (1, 82) = 1.20, MSE = .09, p = .28. These results
are consistent with those found for White perceivers.

Liking toward Black women. The means and standard deviations of explicit
liking toward Black women as a function of skin tone and facial features are presented in
Table 39. The same two-factor repeated ANOVA was conducted to examine the effects
of target skin tone and facial features on Black perceivers’ liking toward female ingroup
members. The main effect of target skin tone was significant, F (l, 82) = 5.85, MSE = .56,
p < .05, d = .19, such that Black participants liked light-skinned Black women
significantly more (M = 3.28, SD = 1.02) than dark-skinned Black women (M = 3.08, SD
= 1.11). The main effect of target facial features was not significant, F (l, 82) = .08, MSE
= .36, p = .78. The interaction between target skin tone and facial features was not
significant, F (l, 82) = .63, MSE = .39, p = .43. Again, this pattern of the results is
consistent with the results of White individuals’ explicit liking toward Black women.
Stereotype Activation among Blacks following Black Male Targets

In addition to examining Black perceivers’ affective reactions toward fellow
ingroup members who vary in skin tone and facial features, Study 4 also examined Black
perceivers’ cognitive reactions (i.e., stereotype activation) after looking fellow ingroup
members who vary in physical characteristics. Table 40 presents the means and standard
deviations of response latencies for the stereotypic and non-stereotypic words for each
prime category. A three-factor repeated ANOVA, treating target skin tone (dark, light),
target facial features (more prototypical, less prototypical), and word type (stereotypic,

non-stereotypic) as within-participants factors, was conducted on the response latencies.

88

The ANOVA revealed a significant main effect of word type, F (1, 36) = 6.23, MSE
= .003, p < .05, d = .10, such that participants responded significantly faster to stereotypic
words (M = 568.41, SD = 121.46) than to non-stereotypic words (M = 580.36, SD =
125.06). Neither the main effect of target skin tone (F(1, 36) = 1.06, MSE = .01, p = .31)
nor the main effect of target facial features (F (1, 36) = 1.09, MSE = .003, p = .30) was
significant. Furthermore, none of the two-way interactions between target skin tone and
target facial features (F (1 , 36) = 1.43, MSE = .004, p = .24), between target skin tone and
word type (F (1, 36) = .02, MSE = .01, p = .90), between target facial features and word
type (F (1, 36) = .22, MSE = .01, p = .64), and the three—way interaction between skin tone,
facial features, and word type (F (1, 36) = 1.09, MSE = .01, p = .30) was significant.

Next, whether the degree of stereotype activation was different between Black
male targets and the White control, a repeated ANOVA was conducted on response
latencies, treating race (Black, White) and word type (stereotype, non-stereotypic) as
within-participants factors. Again, response latencies for four Black prime categories
were averaged to compute response latencies for Black targets, regardless of target skin
tone and facial features. The analysis revealed that neither the main effect of target race
(F(1, 36) = 2.38, MSE = .004,p = .13) nor the main effect ofword type (F(1, 36) = 2.24,
MSE = .01, p = .14) was significant. The interaction between target race and word type
was not significant either, F(1, 36) = .002, MSE = .01,p = .97.
Stereotype Activation among Blacks following Black Female Targets

Table 42 presents the means and standard deviations of response latencies for the
stereotypic and non-stereotypic words for each prime category. Again, due to lack of

statistical power (i.e., small sample size due to computer malfunction), the results in this

89

section of analyses need to be interpreted with caution. The same three sets of analyses as
Black male targets were conducted for Black female targets. A three-factor repeated
ANOVA revealed that the main effect of target skin tone was not significant, F (l , 20) =
47, MSE = .01, p = .50. However, the main effect of target facial features was significant,
F(l , 20) = 5.59, MSE = .01, p < .05, d = .50, such that Black participants responded
significantly faster to all words following Black women with less prototypical facial
features (M = 565.82, SD = 196.56) than following those with more prototypical facial
features (M = 669.98, SD = 217.53). In addition, the main effect of word type was
marginally significant, F(1, 20) = 3.78, MSE = .01, p = .07, d = .08, such that Black
participants responded faster to non-stereotypic words (M = 665.94, SD = 289.93), as
compared to stereotypic words (M = 644.87, SD = 198.41).

The two-way interaction between target skin tone and word type also reached
marginal significance, F ( 1 , 20) = 3.61, MSE = .01, p = .07. The examination of the means
suggest that Black participants responded faster to stereotypic words following Black
female targets with darker skin tone (M = 637.31, SD = 190.59), as compared to
stereotypic words following Black female targets with lighter skin tone (M = 652.43, SD
= 206.23). In contrast, Black perceivers responded faster to non-stereotypic words
following Black female primes with lighter skin tone (M = 653.27, SD = 195.14), as
compared to Black female primes with darker skin tone (M = 678.61, SD = 236.23).
Neither the two-way interaction between target skin tone and target facial features (F (1 ,
20) = .70, MSE = .02, p = .41) nor between target facial feature and word type (F(1, 20) =

001, MSE = .01, p = .97) was significant.

90

However, they were qualified by a significant three-way interaction between
target skin tone, target facial features, and word type, F (1, 20) = 7.03, MSE = .01, p < .05.
However, again, due to extremely low power, the results of the three-way interaction
need to be interpreted with caution. Thus, instead of the conducting the simple interaction
and the simple main effects, the pattern of the means was examined. The examination of
the means suggest that the significant three-way interaction between target skin tone,
target facial features, and word type is driven by slower reaction time for the stereotypic
words following Black female primes with lighter skin tone and more prototypical facial
features and for non-stereotypic words following Black female primes with darker skin
tone and more prototypical facial features. More specifically, Black perceivers responded
faster to stereotypic words following light-skinned Black female primes with less
prototypical facial features, as compared to light-skinned Black female primes with more
prototypical facial features. However, the differences in response latencies between dark-
skinned Black female with more prototypical facial features and dark-skinned Black
female with less prototypical facial features were relatively small. In contrast, the patter
was opposite in the non-stereotypic word condition. The differences in response latencies
between light-skinned Black female primes with less prototypical facial features and
light-skinned Black female primes with more prototypical facial features were small.
However, the means suggest that Black perceivers responded faster to non-stereotypic
words following dark-skinned Black female primes with less prototypical facial features,
as compared to dark-skinned Black female primes with more prototypical facial features.

Next, it was examined whether the degree of stereotype activation following

Black targets is different from following White control. A repeated ANOVA, treating

91

target race (Black, White) and word type (stereotypic, non-stereotypic) as within-
participants factors, was conducted on response latencies. The analysis revealed that
neither the main effect of target race (F (l , 20) = .95, MSE = .01, p = .0834 nor the main
effect of word type(F (1, 20) = 2.08, MSE = .01, p = .17) was significant. The interaction
between target race and word type was not significant either, F (l , 20) = .003, MSE = .004,
p = .95.
Discussion

Study 4 examined how Black individuals’ skin tone and facial features
independently influence their own experiences in intergroup relations, as well as their
perceptions of and reactions to fellow ingroup members. More specifically, the first half
of Study 4 investigated how one’s own physical features influence Blacks’ tendency to
attribute negative events to racism, their experiences with discrimination, and their
sensitivity to racism. The second half of the study investigated how Blacks react, both
affectively and cognitively, to fellow ingroup members who vary in skin tone and facial
features, thus treating Blacks as perceivers.
Attributions for Ambiguous Negative Events

Black individuals who experience racial prejudice and discrimination throughout
their lives may become sensitive to discrimination and construe even ambiguous negative
events (that are not necessarily due to racism) in terms of their racial membership.
Previous research (as well as Study 1) has demonstrated that White individuals are likely
to perceive and treat Blacks with darker skin tone or with stronger Afrocentric features
more negatively that those with lighter skin tone or with weaker Afrocentric features.

Thus, it was predicted that Blacks with darker skin tone or more prototypical facial

92

features would be more likely to attribute ambiguous negative events to racism. Study 4
found that facial features are associated with Black individuals’ tendency to attribute
ambiguous negative events to racism. However, contrary to the prediction, Black
individuals with more prototypical facial features were less likely to make racism
attributions, as compared to Black individuals with less prototypical facial features. In
addition, Study 4 failed to find an association between skin tone and racism attributions.

Furthermore, this study found no consistency in the effects of Black participants’
own skin tone and facial features across the other attribution categories (i.e., attributions
to skin tone, facial features, internal factors, and external factors). For instance, facial
features were associated with skin tone attributions, such that Black individuals with
more prototypical facial features were less likely to attribute negative events to skin tone.
Skin tone predicted Black individuals’ tendency to attribute ambiguous negative events to
internal factors, such that individuals with darker skin tone were less likely to make
internal attributions. There are several reasons that may explain why no consistent
patterns emerged in the current findings.

First, there were not enough Black male participants in the current study. As
Studies 1 and 3 have shown, experiences with racial prejudice are often different for
Black men and Black women, with Black men experiencing more negative consequences
of having darker skin tone and more prototypical facial features. Therefore, the prediction,
which was that Black individuals with darker skin tone or more prototypical facial
features experience more prejudice and discrimination and thus become more sensitive to
racial discrimination, might be only be true for Black men. For Black women, there

should not be as strong of a link between dark skin/more prototypical facial features,

93

discrimination experiences, and racism attributions, because as suggested by Study 3,
Black women with darker skin tone are no more likely to experience prejudice than Black
women with lighter skin tone. Thus, failing to find the predicted results might be due to
the fact that there were only 12 Black male participants in the study and so the current
results reflect the experiences of Black women.

Second, the scenarios used in the attribution measure might not have been
sufficiently detailed and rich for the participants to imagine actually experiencing the
negative events. If this is the case, it would have been hard for them to guess how they
would have reacted to the situations. In addition, the measure might have been too
repetitive and introduced boredom among participants.

Finally, it might be possible that there are no effects of skin tone or facial features
on how Black individuals construe ambiguous negative events after all, or there might be
other factors that strongly influence Blacks’ tendency to attribute negative events to
racism, such as racial identification. Even if Black individuals had lighter skin tone or
less prototypical facial features, they may construe ambiguous negative events in terms of
their racial membership if they were strongly identified as Black.

Although the results suggest that the attribution measure might lack validity, there
was one additional interesting finding. Participant gender predicted skin tone attributions,
such that Black men were more likely to attribute negative events to skin tone than Black
women. This finding is consistent with the results of Whites’ implicit affective reactions,
which suggest that skin tone influences Whites’ negative feelings about Black men

whereas it influences Whites’ positive feelings about Black women. The current results

94

suggest that Black men may be aware of the negative consequences of having darker skin
tone.
Experience with and Sensitivity to Racism

Based on previous research and the results from Studies 1 through 3, it was
predicted that Black individuals (especially men) with darker skin tone or more
prototypical facial features would experience more racial discrimination and that they
would be more likely to be sensitive to racism than those with lighter skin tone or less
prototypical facial features. There was no evidence that skin tone predicts Black
individuals’ experience with and sensitivity to racism. The effects of facial features on
Blacks’ experience with racial discrimination in the past year were somewhat consistent
with the prediction, such that dark-skinned Blacks with more prototypical facial features
experienced more racial discrimination. In contrast, for the light-skinned Blacks, facial
features were not associated with the experience of racial discrimination. Black
individuals’ facial features also influenced their sensitivity to racism. However, the
direction of the effects was opposite from the prediction: Black individuals with more
prototypical facial features were less sensitive to racism. To summarize, even though
dark-skinned Blacks with more prototypical facial features reported experiencing more
racial discrimination, they were less likely to be sensitive to racism. This inconsistency
might be due to the fact that participants’ sensitivity to racism was assessed with a race-
based rejection sensitivity measure. The measure asks participants to indicate how
concerned/anxious they are that they will experience certain mistreatrnents because of

their race/ethnicity. When participants with more prototypical facial features respond to

95

these items, they might be thinking that “I am not experiencing these treatments because
of my race/ethnicity, but because of my appearance.”
The Effects of Black Targets’ Characteristics on Black Perceivers’ Attitudes toward
Them

Study 4 also examined whether Black individuals also have biases toward fellow
ingroup members who vary in skin tone and facial features. Black participants responded
significantly faster to negative words following Black men with lighter skin tone, as
compared to following Black men with darker skin tone, indicating that Black individuals
feel more negatively toward Black male targets with lighter skin tone than toward Black
male targets with darker skin tone. Furthermore, there were effects of facial features, but
only in the light skin condition, such that Black individuals felt more negatively toward
fellow ingroup members with light-skin and less prototypical features, as compared to
those with more prototypical facial features. Interestingly, this pattern of results is
somewhat opposite from that for White perceivers (Study 1). White perceivers showed
increased negativity toward Black men with darker skin tone and more prototypical facial
features, whereas Black perceivers showed increased negativity toward Black men with
lighter skin tone and less prototypical facial features. The results suggest that both Black
and White perceivers responded more negatively to someone who looks the most
different from them. This finding is somewhat inconsistent with previous research that
has shown that dark-skinned Blacks are perceived negatively by fellow Blacks (Averhart
& Bigler, 1997; Harvey etal., 2005; Maddox & Gray, 2002).

Just like White perceivers, there was no evidence that Black perceivers’ responses

to positive words were influenced by targets’ skin tone and facial features. Thus, targets’

96

skin tone seem to be associated with negativity, affecting how much Blacks dislike
ingroup members, not how much they like them. Contrary to the findings for Black male
targets, neither skin tone nor facial features influence Black individuals’ negative and
positive implicit attitudes toward female ingroup members.

Interestingly, the results of explicit liking were consistent with those for White
perceivers. More specifically, Black perceivers liked Black male targets with lighter skin
tone more than those with darker skin tone. They also liked Black male targets with less
prototypical facial features more than Black male targets with more prototypical facial
features. Likewise, Black perceivers liked Black female targets with lighter skin tone
more than Black female targets with darker skin tone (and there was no facial features
effects). Thus, the results of Blacks’ feelings about fellow ingroup members measured at
the explicit level were consistent with the previous findings showing that dark-skinned
Blacks are perceived negatively by fellow Blacks (Averhart & Bi gler, 1997; Harvey,
LaBeach, Pridgen, & Gocial, 2005; Maddox & Gray, 2002). So, there is clearly a
disassociation between attitudes assessed with implicit and explicit measures.

The Effects of Black Targets’ Characteristics on Stereotype Activation for Black
Perceivers

Study 4 also examined whether Blacks’ cognitive reactions toward their fellow
ingroup members (i.e., stereotype activation) were influenced by targets’ skin tone and
facial features. The results showed that Black individuals responded significantly faster
to stereotypic negative words following male target pictures, as compared to non-
stereotypic negative words, indicating that negative stereotypes associated with Black

men were activated following all Black prime categories, regardless of the degree of skin

97

tone and prototypicality of facial features. In addition, none of the four Black prime
categories was different from White controls. Unfortunately, the absence of mean
difference in reaction time to stereotypic words between Black primes and White controls
suggests that the lexical decision task created for the present studies may have failed to
capture stereotype activation associated with “Black.”

J ust like the lexical decision task for male targets, there was no mean difference in
reaction time to stereotypic words between Black female primes and White female
controls, suggesting that the task might not be valid. Therefore, the findings that Black
perceivers responded faster to stereotypic words following light-skinned Black women
with less prototypical facial features, as compared to light-skinned Black women with
more prototypical facial features, should be interpreted with caution. Again, there is little
evidence that the lexical decision task in the current study worked.

General Discussion

People in the same racial groups share sets of physical characteristics. However,
even within the same groups, there is real variation in physical appearance. Two
literatures have been developed to determine how perceivers respond to such differences
in physical characteristics of Black targets and why there are individual differences in the
experience of prejudice and discrimination within Blacks as a racial group. Research on
skin tone suggests that people rely on skin tone to make inferences of individuals’
attributes. Research on Afrocentric features suggests that other facial features, such as
eye color, shape of nose and lips, and texture of hair, in addition to skin tone information,
influence people’s perceptions of and reactions to targets. Although these two lines of

research argue that one type of physical characteristic may be more important than the

98

other when predicting within-group differences in the experience of prejudice and
discrimination, they have generally failed to separate these two sets of physical
characteristics. There were four overall goals of the present research: to examine 1)
whether there are effects of facial features on Whites’ perceptions of and reactions to
Black individuals above and beyond skin tone effects, 2) whether the effects of skin tone
and facial features on Whites’ perceptions differ between Black men and women as
targets, 3) whether the effects of these two types of physical features on Black perceivers
differ from their effects on White perceivers, and 4) what effects a Black person’s own
physical features have on his or her perceptions of and experiences with discrimination.
By manipulating facial features of target pictures, the current experiments successfully
separated facial features from skin tone and addressed the four questions mentioned
above.
Independent Effects of Skin Tone and Facial Features

Study 1 demonstrated that both skin tone and facial features independently
influence how White perceivers feel about Black men, at both implicit and explicit levels.
Because primary findings were two main effects, this suggests that the effects of skin
tone and facial features are additive, and that both skin tone and facial features are
important physical characteristics that affect people’s perceptions of and reactions to
Black men. The current results support the findings in previous Afrocentric features
research, by demonstrating that facial features in addition to skin tone influence people’s
perceptions of Blacks. However, the findings in the uncut research further demonstrate
that it is important to look at skin tone and facial features separately, instead of including

both skin tone and facial features under on variable (i.e., Afrocentric features), as skin

99

tone and facial features do not covary. In other words, some individuals have darker skin
tone with less prototypical facial features, and some individuals have lighter skin tone
with more prototypical facial features. Independent examination of the separate effects of
skin tone and facial features provides a clearer understanding of how White individuals
respond to Blacks with “mismatched physical characteristics.”

The Effect of Target Gender

Importantly, the results from Studies 1 and 3 further showed that the effects of
skin tone and facial features on White perceivers may be moderated by target gender.
When targets were Black men, both facial features and skin tone had negative effects on
White perceivers. On the other hand, when the targets were Black women, there was no
evidence of facial features effects. This suggests that White perceivers’ feelings about
Black women may be more strongly driven by skin tone than by facial features.
Moreover, the nature of the skin tone effect was different. Specifically, unlike Black male
targets, having lighter skin tone increased Whites’ positive feelings.

These results are somewhat consistent with the subordinate male target hypothesis
(Sidanius & Pratto, 1999; Pratto et al., 2006), but not with multiple-identity perspectives
(Buchanan & Fitzgerald, 2008; Buchanan & Ormerod, 2002; Fleming, 1983; King, 2005;
Landrine et al., 1995; McLeod & Owens, 2004; Rederstorff et al., 2007; Settles, 2006).
The multiple-identity perspective argues that Black women would experience prejudice
and discrimination to the same, or even greater, degree as Black men. However, Study 3
did not find evidence of negative consequences of having darker skin tone and more
prototypical facial features for Black women. If anything, Black women receive

favorable evaluations by White perceivers, depending on their skin tone. However, it is

100

important to note again that there were methodological limitations with Black female
target tasks due to lack of usable target stimuli. Thus, the findings and interpretations
presented here are preliminary, and replicating the results with more target stimuli is
critical.

The Effect of Perceiver Race

In addition to target gender, perceiver race also moderated the effects of physical
characteristics on people’s affective reactions to Black men and Black women at the
implicit level. Study 4 showed that Black men with lighter skin tone were evaluated more
negatively than were Black men with darker skin tone by fellow ingroup members. It is
important to note that this effect of skin tone is in the opposite direction from White.
perceivers, who evaluated Black men with darker skin tone more negatively than Black
men with lighter skin tone. Therefore, both Black and White perceivers negatively
evaluated someone who looks different from members of their own racial groups.

In contrast, when the targets were Black women, there was no evidence of the
effects of skin tone or facial features on Black perceivers’ automatic affective reactions.
Remember that White perceivers felt more positively toward Black women with lighter
skin tone. Strong conclusions regarding these differences cannot be drawn because of 1)
small sample of Black perceivers, 2) lack of Black female target stimuli, and 3) uneven
gender distribution in Black sample.

Interestingly, perceiver race did not moderate the effects of skin tone and facial
features on people’s affective reactions to Black men and Black women at the explicit
level. Both White and Black perceivers liked Black male targets with lighter skin tone or

less prototypical facial features more than Black male targets with darker skin tone or

101

more prototypical facial features. Likewise, both White and Black perceivers liked Black
female targets with lighter skin tone more than Black female targets with darker skin tone.
Furthermore, there was no evidence of facial features on White perceivers or Black
perceivers’ explicit liking toward Black women.
Implicit Attitudes-Explicit Attitudes Relation

The present results clearly demonstrate a disassociation between the implicit and
explicit attitudes among Black perceivers. The disassociation between Black perceivers’
automatic affective reactions and their explicit liking attitudes toward fellow ingroup
members may be due to their awareness of the negative consequences of having darker
skin tone and more prototypical facial features. Previous studies have shown that Black
individuals are highly aware of skin tone effects and that they even discriminate against
fellow ingroup members based on their skin tone (Davis, Daniels, & See, 1998; Harvey et
al., 2005; Maddox, 2004). It may also be the case that Black individuals are aware of the
negative consequences of more prototypical facial features as well, as Study 4
demonstrated that participants’ facial features predicted their experiences with racial
discrimination in the past year. As a result, Black perceivers may be concerned with
potential negative treatment they may receive when associating with individuals who
have darker skin tone or more prototypical facial features, even if they personally like the
target persons. Goffrnan (1963) argued that normal individuals can be stigmatized by
others as a function of their association with a marked individual or group (i.e., courtesy
stigma). Indeed, research has shown that mere association with someone who is
perceived negatively can result in being negatively evaluated (i.e., stigma by association;

Gallagher, Tait, McCologan, Dovey, & Holford, 2003; Goldstein & Johnson, 1997;

102

Neuberg, Smith, Hoffman, & Russel, 1994; Ostman & Kjellin, 2002). For instance,
Neuberg et al. (1994) found that people perceive a heterosexual man more negatively
when he was with his homosexual friend than when he was with his heterosexual fiiend.
Furthermore, several researchers have shown that mere proximity (e.g., sitting next to a
marked individual) can cause stigmatized perceptions of “normal” individuals (Hebl &
Mannix, 2003; Penny & Haddock, 2007) and that people are motivated to distance
themselves from a marked individual due to a concern over stigma by association (Swim,
Ferguson, & Hyers, 1999).

In contrast, there was no contradiction between White perceivers’ automatic,
implicit attitudes and their self-report, explicit attitudes. Researchers have shown that
White individuals are motivated not to appear racist because blatant, “old-fashion”
prejudiced attitudes and discriminatory behaviors became socially unacceptable (Crandall,
Eshleman, O’Brien, 2002; Dowden & Robinson, 1993; Gaertner & Dovidio, 1986;
Rokeach & Ball-Rokeach, 1989), and many are motivated to deliberately change their
attitudes to match socially accepted attitudes (Devine, 1989; Fazio et al., 1995;
Greenwald & Banaji, 1995; Nisbett & Wilson, 1977). If White perceivers were aware of
the biases associated with physical characteristics, at least those who are highly
egalitarian may have been motivated to inhibit their prejudiced attitudes toward Black
men with darker skin tone or more prototypical facial features on the explicit liking
measure. Therefore, the present results suggest that White perceivers may not be aware of
the effects of skin tone and facial features on their feelings about Black targets.

Black Person’s Own Physical Characteristics and Their Experiences with Racism

103

In addition to the examination of the effects of physical characteristics on
people’s affective reactions toward Black individuals, the present research also
investigated how one’s own physical characteristics influence their experiences with
intergroup relations. Study 4 examined how skin tone and facial features influence Black
individuals’ tendency to attribute ambiguous negative events to racism, experience with
racism, and sensitivity to racism. Only facial features predicted racism attribution.
Interestingly, the pattern of the effect was inconsistent with the prediction, such that
Black individuals with more prototypical facial features were less likely to make racism
attributions, as compared to Black individuals with less prototypical facial features. If
Black individuals with more prototypical facial features do not attribute negative events
to racism, what attributions do they make? Although facial features were also associated
with skin tone attributions (i.e., the more prototypical facial features, less likelihood of
making skin tone attribution), they were not associated with other types of attributions,
including facial features, internal, and external attributions. The lack of association may
simply indicate that Black individuals with more prototypical facial features (or with
darker skin tone) are likely to make certain attributions to the same degree as Black
individuals with less prototypical facial features (or with lighter skin tone). Alternatively,
the attribution measure may not have been a compelling methodology.

Why there was no effect of skin tone on Black individuals’ tendency to attribute
negative events to racism? Also, why were Black individuals with less prototypical facial
features more likely to make racism attributions, as compared to Black individuals with
more prototypical facial features? One reason might be because the majority of the

participants (85.7 %) were women. As shown in Studies 1 and 3, skin tone influences

104

Whites’ positive feelings about Black women, and facial features have no influence on
Whites’ affective reactions toward Black women. Because having darker skin tone
apparently does not increase the incidence of racial discrimination among Black women,
skin tone might not be associated with Black women’s tendency to attribute negative
events to racism or their sensitivity to racism. In fact, skin tone was not associated with
experiences with discrimination in the past year. If there were more Black men in the
current study, however, the results might have been different. Indeed, target gender
predicted skin tone attributions, such that Black men were more likely to attribute
negative events to skin tone, as compared to Black women, supporting the findings in the
present study that men’s skin tone influences Whites’ negativity whereas women’s skin
tone influences Whites’ positivity. If there were more Black men with different levels of
skin tone, it might have been possible to test whether Black men with darker skin tone
would attribute negative events to racism more than Black men with lighter skin tone.
However, due to the small number of Black male participants, the interactions between
target gender and skin tone or facial features could not be tested in the current regression
analyses.

One interesting finding in Study 4 is the discrepancy between Black individuals’
reports of discrimination experiences and their sensitivity to racism. Dark-skinned Black
individuals with more prototypical facial features, as compared to those with less
prototypical facial features, reported more incidents that they considered racial
discrimination, but they were less sensitive to racism. These findings seem inconsistent
with the previous research showing that individuals who experience greater amounts of

discrimination become more sensitive to racial issues and are more apt to interpret events

105

through the lenses of discrimination due to chronic activation of their social identity
(Jetten, Bransocombe, Schmitt, & Spears, 2001; Major, Quinton, & Schmader, 2003;
Operario & Fiske, 2001; Sellers & Shelton, 2003; Shelton & Sellers, 2000). However,
sensitivity to racism assessed in the current study is based on individual’s group
membership. Race-based rejection sensitivity was designed to assess people’s anxious
expectations of rejection specifically based on their racial group membership (Mendoza-
Denton et al., 2002). The measure asks participants to indicate how concerned/anxious
they are that they will experience certain mistreatments because of their race/ethnicity.
When participants with more prototypical facial features respond to these items, they
might be thinking that “I am not experiencing these treatments because of my
race/ethnicity, but because of my appearance.” Thus, instead of measuring their
sensitivity to racism, the current study should have assessed Black individuals’ sensitivity
to mistreatments in general using a general rejection sensitivity measure (Downey &
Feldman, 1996). However, again, the results should be replicated with more Black
participants, especially Black men, before making strong interpretations of these
seemingly inconsistent findings.
Limitations and Future Directions

Although the current research provides important insights into the relationship
between Blacks’ physical characteristics and Whites' and Blacks’ experiences with
intergroup relations, it is not without limitations. First, the lexical decision task used in
the present study suffered from low validity. The major difference between the original
LDT and the modified LDT used in the current study is that there was no baseline in the

current LDT. Although the stereotypic and non-stereotypic words used in the current

106

LDT were matched in terms of its negativity, other factors, such as word familiarity,
frequency, and number of letters, were not controlled. Therefore, it is hard to conclude
that the differences in reaction time between stereotypic and non-stereotypic words were
due to heightened accessibility of the words (i.e., stereotype activation) or due to other
extraneous factors.

In the original lexical decision task (Wittenbrink et al., 1997), there were baseline
trials, and participants’ response latencies to specific words following experimental
primes were compared to the same words following neutral primes (i.e., baseline). In
other words, dependent variables in the original task were facilitation scores (just like the
ones computed in the sequential priming task) rather than response latencies per se. This
way, the original lexical decision task eliminated differences in speed of word
recognition as a function of individual items. The reason why there were no baseline
trials in the current research was because of the concern over participants’ fatigue and
boredom. Even without the baseline trials, participants were required to complete 240
trials, which lasted about 20 minutes. If baseline was included, the task would have been
doubled, and participants likely would not have been able to stay focused for the entire
task.

In future studies, the LDT should include the baseline trials to control for many
factors that may influence participants’ reaction time. However, in order to avoid
participant fatigue, prime categories should be presented in a between-participants, not
within-participants, design. With a within—participants design, participants had to
complete 240 trials because there were five different prime categories. However, with a

between-participants design, there would be four different groups of participants: one

107

completing the task with dark-skinned Blacks with more prototypical facial features vs.

White controls, one completing the task with dark-skinned Blacks with less prototypical
facial features vs. White controls, another completing the task with light-skinned Blacks
with more prototypical facial features vs. White controls, and the last group completing

the task with light-skinned Blacks with less prototypical facial features vs. White controls.
If the task included only two prime categories, the number of trials would be cut down to

96.

Alternatively, the number of pictures in each prime category can be cut from eight
to four, in order to reduce the number of trials. In fact, variation across participants’
responses to eight pictures was relatively small (for negative words, means of eight
pictures within each prime category ranged from 27.30 to 41.40, and standard deviation
ranged from 11.64 to 16.38), suggesting that cutting the number of target pictures is a
reasonable solution to reduce the length of the task. If the number of trials could be
significantly reduced either by including only two prime categories or by reducing the
number of pictures, positive target words could also be included, just like the original
LDT. With positive words included in the task, future studies could exarrrine whether
darker skin tone and more prototypical facial features are associated with both negative
and positive stereotypes or only with negative stereotypes.

A second limitation of the current work is the lack of Black female target pictures
in the sequential priming and lexical decision tasks. There was only one female target
picture in each of the five prime categories. Although there were two sets of pictures
(thus resulting in two pictures for each prime category when sets were combined), there

still might be idiosyncratic effects of specific faces. Although attractiveness,

108

babyfacedness, and hostility did not differ between two light-skinned faces and two dark-
skinned faces, the results must be interpreted with caution. With only four pictures, the
results of inferential statistics may not be reliable. Notably, because additional Black
female pictures were collected in Study 4, these pictures can be included as target stimuli
in future research.

Third, the majority of the Black participants in Study 4 were Black women, and
there were only 12 Black men. As some of the findings in the current research
demonstrate, the experiences with prejudice are often different between Black men and
Black women, with Black men experiencing more prejudice than Black women. If Black
men and women are perceived and treated differently by others, their responses to
prejudice and discrimination also should be different. Therefore, it will be important for
future research to take into account participant gender when examining how one’s own
physical characteristics influence responses to and interpretations of negative events.

Fourth, there were substantial differences in the strength of skin tone
manipulation (dark vs. light) and the strength of the facial features manipulation (more vs.
less prototypical). As a result, the findings that the effect size is bigger for the skin tone
effect than for the facial features effects might be due to this methodological issue rather
than to actual differences in the effects of these constructs. Facial features were
manipulated to reflect one standard deviation above or below the mean, so that more
prototypical facial features and less prototypical facial features conditions were exactly
two standard deviations apart from one another. In contrast, the selection of skin tones
was based on darkest vs. lightest skin tones in the original picture pool, so that the

difference between dark skin tone and light skin tone conditions might have been bigger

109

than two standard deviations. Indeed, the average of standardized luminosity for 16 dark
skin tone faces was -1 .30 (individual standardized luminosity ranged between -1.78 and -
1.04), and that for 16 light skin tone faces was 1.64 (individual standardized luminosity
ranged between -1 .78 and -l .04), suggesting that the dark skin tone and light skin tone
conditions were about three standard deviations apart from one another. In future
research, it will be important to equate the strength of skin tone and facial features
manipulations.

Another limitation with the present research is the failure to collect White
participants’ gender information due to a computer malfunction. As the subordinate male
target hypothesis predicts, how White men affectively and cognitively react to Black men
may be different from how White women react to Black men. The subordinate male
target hypothesis not only argues that men are the primary targets of racial discrimination,
but also that men are the primary insti gators of racial discrimination (Pratto et al., 2006;
Sidanius & Pratto, 1999). If this is the case, White men, as compared to White women,
may express greater negativity toward Black men with darker skin tone and more
prototypical facial features. In future research, it will be important to examine whether
participant gender moderates the effects of physical characteristics on Whites’ affective
and cognitive reactions toward Black individuals.

The next step, after the limitations listed above are addressed, is to examine
underlying mechanisms that explain the relationships between physical characteristics
and people’s perceptions of and reactions to Black targets. Previous researchers who
study Afrocentric features have proposed several potential mechanisms. Some

researchers argue that strong Afrocentric features serve as a representation of “Black”

110

and activate this social category (Cloutier, Mason, & Macrae, 2005; Blair et al., 2002;
Maddox & Gray, 2002; Mason, Cloutier, & Macrae, 2006). Other researchers argue that
strong Afrocentric features are directly linked to group-related attributions through
repeated association (Blair et al., 2002; Livingston & Brewer, 2002; Meier, Robinson, &
Cole, 2004). This perspective asserts that Afrocentric features initially influence
perceptions through category activation. However, through repeated pairing of
Afrocentric features and stored associates of the category "Black," a direct association
between Afrocentric features and those associates becomes established. Finally, some
researchers argue that the negative effects of Afrocentric features on perceptions are due
to unfamiliarity (Livingston & Brewer, 2002; Zebrowitz, Bronstad, & Lee, 2007). It has
been shown that faces that are familiar, as opposed to unfamiliar, induce more positive
attitudes (DeBruinem 2002; Hill, Lewicki, Czyzewska, & Schuller, 1990, Peskin &
Newell, 2004) and are more preferred (Bomstein, 1993, Harnm, Baum, & Nikels, 1975;
Rhodes, Halberstadt, & Brajkovich, 2001 ). Because faces with strong Afrocentric
features, as compared to faces with weak Afrocentric features, are more unfamiliar to
White people, Whites’ negative perceptions of and reactions to Black men with stronger
Afrocentric features may reflect their responses to unfamiliarity.

Future research should also investigate how perceivers can control biases
associated with physical characteristics, once the underlying mechanisms have been
revealed. Previous research has shown that people can be educated about their automatic
biased attitudes toward certain social groups and that inhibition of prejudiced attitudes is
possible once people become aware of the biases, especially when they have sufficient

abilities and motivation to do so. For instance, researchers have found that judges who

111

were extensively trained to ensure race neutrality in sentencing were able to give roughly
equivalent sentences to Black and White offenders, given equivalent criminal histories
(Blair et al., 2004a). Inhibition of prejudiced attitudes is possible even with automatic
attitudes (Blair et al., 2004b; Monteith, Sherman, & Devine, 1998; Wyer, Sherman, &
Stroessner, 1998; 2000; also see Blair, 2002 for extensive review). Thus, it is important to
educate people about the biases associated with skin tone and facial features and to train
people to inhibit prejudiced attitudes, as the effects of skin tone and facial features can
have serious consequences in people’s lives, such as discriminatory criminal sentencing
(Blair et al., 2004a; Eberhardt et al., 2006) and shooting errors among police officers
(Kahn et al., 2008).

Future research should also investigate how targets with “disadvantageous”
physical characteristics can reduce the negative consequences of these characteristics.
Targets of social inequality are not passive agents---they are active participants in
intergroup relations and should have influence in determining the quality of their
interactions with members from other social groups (Swim & Stangor, 1998). Therefore,
examining what targets can do to reduce people’s negative feelings about them is as
important as examining what perceivers can do to reduce their biases toward certain
physical characteristics. When Brent Staples, whose essay “Black men and public
spaces” (Staples, 1986) caught people’s attention, walked down the street at night people
ran away from him from fear, thinking of him as a mugger or rapist, because he was a
young Black man. However, once he started whistling classical music while walking at
night, people were not as nervous when they walked by and saw him. This famous story

demonstrates that sending perceivers counter-stereotypic information can reduce negative

112

perceptions, possibly because perceivers individualize the targets or think of the targets
as exceptions who do not fit well to “Black” social category (Brewer, 1988; Fiske et al.,
1999)
Conclusion

The present studies demonstrated that skin tone and facial features independently
influence people’s perceptions of and reactions to Blacks and that the effects of these
visible physical characteristics are different depending on whether feelings are measured
at the implicit or explicit level, whether targets are Black men or Black women, and
whether perceivers are Whites or Blacks. There was relatively little evidence that Blacks'
own skin tone and facial features influence their reactions to negative events and
experiences with discrimination. As has been discussed, the sparse findings may reflect
lack of Black male participants with varied physical characteristics. The present studies
contribute to a better understanding of the effects of physical characteristics on people’s
perceptions of and reactions to Black individuals, by pointing to the benefits and
importance of identifying factors that explain who, when, and under what circumstances

targets are more likely to experience prejudice and discrimination.

113

Footnotes
1. Pictures of 89 White men and 43 White women were also collected for the selection of
control pictures.
2. Thirty five participants who rated Black women were the same people who categorized
the pictures into Black, White, or Racially Ambiguous.
3. There were two sets of pictures for each target gender because participants could not
be exposed to both high and low prototypical versions of the same face. Thus,
participants were randomly assigned to see one of the two sets.
4. One dark-skinned face was categorized as White by one participant, and the other
dark—skinned face was perceived to be racially ambiguous by one participant; one light-
skinned face was categorized as White by one participant and as racially ambiguous by
five participants, and the other light-skinned face was categorized as racially ambiguous
by nine participants.
5. Due to a computer malfunction, participants’ age and gender were not recorded.
Because there is no way to connect participants’ names to their data, only the gender ratio
could be coded using participants’ first names (women = 72.4%, men = 24.4%, and
gender neutral names = 3.2%).
6. Explicit liking toward Black targets was measured using a unidimensional item, rather
than measuring positive and negative attitudes toward targets separately, because people
are more comfortable indicating how much they like other people than indicating how
much they dislike other people. Especially in a racially charged situation, such as the
current studies, White participants would be very uncomfortable indicating how much

they dislike Black individuals. In fact, I assume that White participants would

114

deliberately change their feelings toward the Black targets to portray themselves as non-
prejudiced due to social desirability concerns, resulting in little variability in the explicit
negative attitudes toward Blacks.

7. Although the analysis was conducted using the log-transformed facilitation scores, the
results are presented in the original values. Means and standard deviations in log
transformed response latencies are presented in tables.

8. Because all analyses suggest that prime set does not influence the effects of skin tone
and facial features on Whites’ affective reactions to Black targets, prime set was droped
from the analyses of target gender effects.

9. One participant was excluded from the analyses of the sequential priming task for male
targets because she was talking during the task, and 20 participants (17 due to PC
malfirnction, four due to wrong categorization of target gender, and one due to random
factors) were excluded from the analyses of the lexical decision task for female targets.
10. Five participants (one due to PC malfunction and four due to random factors) were
excluded from the analyses of the lexical decision task for male targets, and 17
participants (10 due to PC malfunction, four due to random factors such as talking during
tasks, and three due to wrong categorization of target gender) were excluded from the
analyses of the sequential priming task for female targets.

11. A mixed models analysis was conducted, treating scenario as a within-participants
factor and facial features as a between-participants factor, was conducted for each of the
six attribution categories (racism, sexism, skin tone, and facial features, internal, external)
to examine whether participants’ responses to six different scenarios can be averaged

within each attribution category to compute attribution scores. The two-way interactions

115

between scenario and facial features were non-significant for the all six attribution
categories (all F s(5, 405) < 2.02, ps > .10), suggesting that participants’ responses to
different scenarios can be averaged over to create attribution scores for each attribution
category. Thus, each participants are associated with six attribution scores (racism
attribution score, sexism attribution score, skin tone attribution score, and facial features
attribution score, internal attribution score, external attribution score).

12. Prime set condition was excluded from all of the analyses because (1) there were only
a small number of participants in each condition, and (2) previous analyses with White
participants showed that prime set do not have any effects on the main results.

13. Standard deviations of Black participants were bigger than those of White
participants, indicating that there was more variability in responses to words among

Black, as compared to White participants.

116

Table 1
Means and standard deviations ofsubjective ratings and objective measures of Black

faces by race

 

 

 

Black Men Black Women

M SD M SD
Afrocentric Features 3.98 .40 3.46 .48
Attractiveness 2.01 .3 8 2.14 .54
Babyfacedness 2.28 .55 2.42 .56
Hostility 2.55 .47 2.43 .60
Lip Ratio .14 .02 .13 .02
Nose Ratio .32 .03 .31 .02
Luminosity .34 .09 .36 .06

 

Note. Units for luminosity is proportion and ranges fi'om 0.0 to 1.0.

117

o2 ”2.5.v

& 82865 1.... use ._o>o_ 3. v Q 05 8 02805:»? 828:2: .._ .o._ 3 od 88m momma 98 209285 $ biog—WEE Sm 3:5 $82

 

l .LNV 3.- gum... :.
l we. 2. 1N-
.. 3. nor
-- *ambr
h o w v m

8.

we.

NT

**_mr

*o_.

**er

**VM.

**_v.

**Nm.
**_mr

.LNH

@8583 .5
23m 302 .0
23m .5 .m
3:58: .4

mmocvooflznem .m

$8338.32 .N

8530a aficoootxx .—

 

=mE «was LQBQQVBSMKQ $23,435 $53.56 38 wMSBa wifimfig 9523 38239260

N 2an

118

om NZ .8. v

Q mBSEE .1. :5 ._o>o_ mo. v m 65 E oogocfigm 8:865 ... .o._ 3 od 80¢ moms: cg :omtomoa 3 bmmofisé 5.: BED .862

 

-- 2.- 2.- am.-
-- 8.- E.-
-- 2.
a e m 4

mm.

2.-

5.-

**nvr

5.

mo.-

2.-

*awmr

**~V.

*mmr

**mm.

**mv.

5.

No.

2.-

3:85:51— <-
0:an 082 .c
as: a: .m
55.8: .4
mmogooflxnem .m
mmo:o>:o€:< .N

magnum 05:83.3. ._

 

$59: «95m .8\ moo:\«oe~m\o mega-ems 958.3: 38 $2.52 miBmEg M208: 30236-269

m 2an

119

om is .So. v a 8:865 r... 2...
. S. v Q 833:5 a... ._o>o_ mo. v a. 05 “a oogomefi 8332: ... .o._ 3 06 Bow woman :5 :omtoaoa v: 368:5: :02 3E: .302

 

 

 

 

we. :3 am. am
mm. mm. E. Ni
O: on- mm. :.N
NE 13A on a?“
mm. 3. .8. mm.
8. Q. S. 3.
Be .1...me _- 8. 9..
n e S. S
25H 5% 52w:

 

Du. 2.0m
ow. end
cw. bod
vm. nod
we. mm.
8. v _.
mo. mm.
Qu- 2
0:8. :Em flax-ED

£38:

Boggflxfim

Baguette.

8:38.: 03:895..

28m 082

as: a:

58:58:.“

 

22338.6 38: Se}. 3%.:

as «as: Lo\mm§3m\~eo.53& m. 55 «ouEKo amaze-35 $88.50 33 9:283:33 mmctmamswah SSE has £223.53: 223323: .23»:

v 033-

120

m "\0 .3. v Q 3:365 _.. 6.. 9 od 80¢ mowg: 0:0 50.5080 mm 560582 :8 BED .0002

 

 

 

 

 

 

of of 6. SN 3. 8.... 0:38:
a. em. 8. SN 3. CS. 388830:
NI 9: 3. 80 E. 03. 383388...
40m 03 em. mam om. Sm 8.38: 00.885.
2: w _ .- 3. mm. 3. mm. as: 082
of w? 8. :. 8. E. as: a:
a: 1.3- So. 9.. mo. am. 28563
0 a 0m 2 00 2
0:09 :35 Hour—wmd DEC-H :U—m nae—0Q

 

00.0.0000 0000 0.0?- EMQ a;

«.000 00X§0500K~00N§0Q w. 0000.: «0022\0 000000000 09000.50 000 09000300.? 000000500 0000 000 #0000060 0000000.. .0000:

m 050%

121

.50. v :2 80020:: 3.... 0:02

 

 

 

 

 

 

 

3.: 1.82 we. 2 .m :0. . 8.0
0 a 00 :0 0:. 0
8:00:00 0200-.— m0:300m 30$
3020.388: 304 00289805 0:02
0.0 n >0 6:83
8.: 3.2.2 0. 00m 6. Sm
.0 : QM. 2 Q0. 2

 

 

00:300.; Eon-.2
30:50:05 000-—

82500”. 300 m
32:30:05 0:02

 

Awm HZV :02

5202030080

2200200885

 

m000\0§:00\20.:00\«00.0 000 000$.000:00K~0.:00\.m.0000 000.500 0.0.0.000 5000:2080-00- 0: m000000§0 0002

e 2.3

122

 

Awe:
wwo.

£50
0va

 

A2 3 A: 3 :03 so:
30. NS. £0. 05.
$3.0 $0.3 $0.00 @0000
$2 00% Ram 03».
33 :05 33 :05
3:02958000 092029500000

003000 300”—

0500& 300m

 

223

25 :00 £0:

250 :20 0.30

5.3
2

5.3

0.50m 00.500000; mo:

A0000: 008m 30%

 

000500\~0.C0\000 0000 0.0%. 0. 005 «00500 000.00% 0 00 0000.: 0>00M00 05 00\000000 00.000.02.005? 00000300 0.000003. 000 00002

n 2.3

123

 

A: 3
N8.

:38
R:

 

 

8:0 2:0 02.0 603
N8. $0. N8. $0.
8300 80000 80000 8300
002 8.: 008 8.8
33 :05 33 =05
021029300000 002—0295000052

008000 300.2

00308 3000

 

253

250 :30 £0:

25 =00 V.00

S00
2 Boom 00808000; woq

500

2 A0000: Boom 30m

 

000500\~0.~00\000 0000 030 0. 005 0003\0 00.0000\ 0 00 00003 090.00% 000 00\000000 00.000.00.005? 00000300 00000000 000 00002

w 030,—.

124

Table 9

Means and standard deviations of White perceivers ' explicit liking toward Black men as

a function of skin tone and facial features (Study 1)

 

 

 

 

 

Dark Skin Tone Light Skin Tone
Facial Features Facial Features
Prototypicality Prototypicality
High Low High Low
M 2.97 3.01 3.03 3.16
SD .70 .70 .74 .71

 

125

 

 

 

 

am: :20 80: am: am: 500
200 200 30.0 200 3.00 0 280 8080030 03
00.80 20.30 00.8: 80.00 @000 E00
0000 00000 0.300 00.000 $.30 E 0025 880 300
33 .00: 33 :05
520005980 0050289805
0003000 3000 0003000 3000
323 250 50 £0: 250 :20 0.00
00003 ofiboefim

 

0.000000 00.000\ 000 0000

0&0 0. 005 «00200 00000002 0 00 0000.: 0005000000000 000 0005000000 00 0000000000000 00.000.00.00 00000300 0.0000000 000 00002

a: 0E0H

126

 

 

 

:0: an: $03 803 0.03
9.00 000.0 3.00 0000 0000
£0.80 00.8: 000.80 $0.80 £003
000% 00.0% 00.000 00.000 2.0%

33 005 33 000:
00000330000; 00002990005
00.50000 3000 0003000 3000

223 88. 0&0 £0: 250 :20. 0:00

S00

2 0000m 008000000; mo:

S00
2

600—5 0000m 30m

 

00003 0_§000000m-:0z

 

00.0.80 2 200.0

127

Table 1 1

Means and standard deviations of White perceivers ' explicit liking toward Black men as

a function of skin tone andfacial features (Study 2)

 

 

 

 

Dark Skin Tone Light Skin Tone

Facial Features Facial Features
Prototypicality Prototypicality

High Low High Low

M 2.96 2.99 3.09 3.12
SD .75 .72 .73 .75

 

128

 

020
08.

$003
mm.~m

 

:03 80: am: 0.03
woo. owe. woo. woo.
20.00 20.: 0 £0.00: @000
3.00 0060 wmsv oodm
33 0000 33 00:0
300000500000 500009300000

00300 0 0000 0

00300 0 .060 0

 

0:55

250 030 £0:

250 :20 0:00

S00
2

S00

0.50m 00800000000. mod

80005 0000m 30,0

 

0000000\

00.00% 000 0000 030 0. 0000.: «0000\0 00000000 00 00.003 0>00M00 00 00\000000 00000000KK0 00000300 0.0000000 000 00002

.2 0300.

129

 

:0:
:0.

20.000
00.00

0:55

 

 

03.0 803 :30 $2.0
08. 000. N00. 000.
:08: 00.8: 0003 80.00
00.2 00.00 0000 00. a
33 0000 33 005
30000500000 320000300000

05000 0 3000

00300 0 3000

 

250 :20 £0:

250 30 0.00

500
0

S00

0000m 00800000000. mod

A0025 Boom 300

 

00.0000\

00.00% 000 0000 030 0. 00.00.: 0003\0 000000\ 0 00 00.00.: 030.0000. 00 00\000000 0000000030000 00000300 0.0000000 000 00002

2 0300.

130

Table 14

Means and standard deviations of White perceivers ' explicit liking toward Black women

as afunction of skin tone andfacial features (Study 3a)

 

 

 

 

 

Dark Skin Tone Light Skin Tone
Facial Features Facial Features
Prototypicality Prototypicality

High Low High Low

M 2.75 2.80 2.99 3.06

SD .96 .86 .91 .93

131

 

 

 

 

 

Gm: COS cm: 53 $3 55
33 SE 38 SE 83 S 28m 885%; we;
$3: 5.8 5 83m: 5.: 5 22m: 88
a .§ 358 3+8 cams 8.80 S @025 28m 5.x
33 :wE 33 :3:
bzmambowoi bzaegaowoi
magnum Evan— mohsaon— 32E
253 2:; 3m 5»: 88 5% x25
$5? 29508on

 

mmxSuQNEuuKhzc 86» :3“.

.4. 22:9: «95%? EEBSX a we v.39: Embomzmfiéot 35 meQESm 2: xQQBFdE mm=cm8x\e w=o2§>mw 3332an 38 was»:

2 2an

132

 

some 803 ES 8”: 3&3 5.8
:3. E8 2% 83. mayo 2 28m 8:833 m3

$20: :03: ago: 692: :93: 53

 

 

 

afimvc @336 5:30 563 81.3 E A825 Boom 39m
33 PE: 33 E:
bzmofiboaoi bzmembouoi
853$ 30mm 8::QO 30mm
255 25 :5 am: 25 5% En

 

€83 Baboobwméoz

 

.2329: S 2an

133

Table 16

Means and standard deviations of White perceivers ' explicit liking toward Black women

as a function of skin tone and facial features (Study 3b)

 

 

 

 

Dark Skin Tone Light Skin Tone
Facial Features Facial Features
Prototypicality Prototypicality
High Low High Low
M 2.69 2.82 2.92 2.91
SD .89 .98 .89 .92

 

134

2 n2 .5. v m 8:865 2. use .732 8

. v Q 05 as 8.805%? 8:865 ... .802

 

 

 

- :3.- Q: 8.- mm. 3.. mm; Samoa .5

- am. 2. 2.. *S. *8. woe-Sam e

-- a. 8. mm. .._ E. mmo=o>uo§< .m

-- :2. mm. mm. ”238m 0.5585 .v

-- 5.- we. 8:2 a: 082 o>< .m

-- 1.3. mused .830 Sam .N

-- b30583: .—

wm. we. Nm. 3. we. 9: mo. 3-

SN SN SN NM: 3. N?- R. E
n o w v m N _

 

$238.3 . $233953. £3: «93% LQKV-mEetg 33395. 2: @823 99:33-28 38 £233.:me 33.352 .28»:

C 053-

135

_ n n 2 .5. v E 82865 3.. 98 :32 mo. v a 2: “a 3:35:me 8:368 ... .802

 

- :6.- ..En- an. E. 8.- *8- £38?»
-- :8. La.- 2- 2. :. oomtname

-- :3.- 3- 2. mm. mmoco>coe:<.m

-- :8. :5. 8.- 858m one-.85. .v

-- m _. 2. mean a: 082 .o>< .m

-- :8. mag .28 .35 .N

-- bfiofigfi .—

 

 

mn. oo. 3. «m. mn- Om._ mo. QM.
Sim OWN m_.m 9mm 8.. 2.- VM. 3
N- o w v m N _

 

8.53.3 . mezefiawtem 3325\«935 3\ 836.29 33.3me 2: 38:8 atonfimtou 33 defies-3% @8333. .33»:

M: 2an

136

mm H 2 .5. v R 88065 .1. 0:0 ._0>o_ mo. v m 05 E oogomefi 80865 ... .802

 

 

 

-- 2mm. :8. 2. mo. .3. :2. *2. to... 8.058.085 £5885me can-05¢

-- 3.9.. 8. 8. 2. *mm. 2. .100 802335803 85502 8.0555200

-- .2. *2. .2. $3.. :2. :0... 330.2% .5085 82.0.82 .2

-- :8. 1.8. m _. in. *2. 8:352 E53 .0

-- *2. 2. 2. .00 8030508250

-- :3. :8. :2. 802:3. 8.380 00$ .0

-- 1%. 3.2. 80.532. 2.3 gm .m

-- :2. 8:33;. :2me .N

-- 803.52 8022 ._

8. 3. o; :1 02 v? :2 02 MS 00

RN 3: of: mm.»- Nm.m mi Sm 20 5. 3.
o w n 0 w v m N 2

 

0030.20.» 00.28.20 00x02 000 :0 M00000 0000020200 000 60000300 00000000 60002

2 030,—-

137

Table 20

The eflects of skin tone, facial features, and participants 'gender on attribution to racism

 

With Objective Measure of Skin Tone and Facial Features

 

 

b SE fl t
Intercept 4.20 . 16
Skin Tone .73 1.96 .05 .37
Facial Features -.27 .21 -.15 -1 .31
Gender .64 .47 .17 1.38
Skin x Face 2.51 2.68 .l l .94

R2: .04, F(4, 78) = .81,p = .53

 

With Subjective Measure of Skin Tone and Facial Features

 

 

b SE ,8 t
Intercept 4.21 . 1 6
Skin Tone .001 .1 l .001 .01
Facial Features -.67 .29 -.27 -2.31*
Gender .67 .44 .18 1.51
Skin x Face .25 .22 .13 1.12

R2: .08, F(4, 78) = 1.63,p = .18

 

Note. * indicates p < .05. df= 82.

138

Table 21

The effects of skin tone, facial features, and participants 'gender on attribution to sexism

 

With Objective Measure of Skin Tone and Facial Features

 

 

b SE ,B t
Intercept 3 .3 3 . 16
Skin Tone .26 1.95 .02 .13
Facial Features -.27 .21 -.15 -1.32
Gender -.50 .46 -.13 -1.09
Skin x Face 1.07 2.66 .05 .40

R2: .06, F(4, 78) = l.15,p = .34

 

With Subjective Measure of Skin Tone and Facial Features

 

 

b SE ,8 t
Intercept 3.34 1 6
Skin Tone .02 .1 1 .02 .14
Facial Features -.49 .29 -.20 -1.68T
Gender -.51 .44 -.14 -1.14
Skin x Face -.07 .22 -.04 -.31

R2= .08, F(4, 78)=1.57,p = .19

 

Note. 1' indicateSp < .10. df= 82.

139

Table 22

The effects of skin tone, facial features, and participants 'gender on attribution to skin

 

 

 

 

 

 

tone
With Objective Measure of Skin Tone and Facial Features
b SE E t
Intercept 3 .3 8 .1 7
Skin Tone -.01 2.08 .0001 —.004
Facial Features -.51 .22 -.26 -2.30*
Gender 1.22 .50 .29 2.46*
Skianace 3.19 2.85 .13 1.12
R2 = .12, F(4, 78) = 2.61,p < .05
With Subjective Measure of Skin Tone and Facial Features
b SE ,6 t
Intercept 3 .39 . 1 7
Skin Tone -.O3 .12 -.03 -.24
Facial Features -1.03 .30 -.38 -3.41**
Gender 1.28 .47 .31 2.75**
Skin x Face .37 .23 .17 1.61

R2: .18, F(4, 78) = 4.10,p < .01
Note. * indicates p < .05, and ** indicates p < .01. df= 82.

 

140

Table 23

The effects of skin tone, facial features, and participants ’gender on attribution to facial

 

 

 

features
With Objective Measure of Skin Tone and Facial Features

b SE )9 t
Intercept 2.49 . 1 6
Skin Tone -1.49 1.95 -.09 -.76
Facial Features -. 12 .21 -.07 -.59
Gender -.26 .46 -.07 -.55
Skin x Face 4.46 2.67 .19 1.671"

R2 = .07, F(4, 78) = 1.34, p = .25

 

With Subjective Measure of Skin Tone and Facial Features

 

 

b SE ,8 t
Intercept 2.56 .1 6
Skin Tone -.12 .12 -.12 -1.02
Facial Features -. 12 .30 -.05 -.41
Gender -.44 .46 -.12 -.97
Skin x Face .04 .23 .02 .16

R2 = .03, F(4, 78) = .49, p = .74

 

Note. '1' indicates p < .10. df= 82.

141

Table 24

The eflects of skin tone, facial features, and participants ’gender on internal attribution

 

With Objective Measure of Skin Tone and Facial Features

 

 

b SE ,6 t
Intercept 3.30 . 1 7
Skin Tone 3.60 2.00 .22 1.801’
Facial Features .03 .21 .01 .12
Gender .14 .47 .04 .30
Skin x Face .24 2.73 .01 .09

R2 = .05, F(4, 78) = .92, p = .46

 

With Subjective Measure of Skin Tone and Facial Features

 

 

b SE 19 t
Intercept 3.3 0 . 1 6
Skin Tone .25 .1 l .25 2.20*
Facial Features -.19 .29 -.08 -.66
Gender .29 .45 .08 .65
Skin x Face .40 .22 .20 1.801'

R2: .10, F(4, 78) = 2.09,p < .10

 

Note. T indicates p < .10, and * indicates p < .05. df= 82.

142

Table 25

The effects of skin tone, facial features, and participants 'gender on external attribution

 

With Objective Measure of Skin Tone and Facial Features

 

 

b SE 19 t
Intercept 3.30 . 1 8
Skin Tone 1.16 2.19 .07 .53
Facial Features .21 .23 .1 l .90
Gender .14 .52 .03 .27
Skin x Face .60 3.00 .02 .20

R2 = .02, F(4, 78) = .36, p = .84

 

With Subjective Measure of Skin Tone and Facial Features

 

 

b SE ,8 t
Intercept 3 .28 . 1 8
Skin Tone .12 .13 .11 .93
Facial Features .0001 .33 .0001 -.001
Gender .34 .50 .08 .67
Skin x Face .34 .25 .16 1.37

R2 = .04, F(4, 78) = .77, p = .55

 

Note. df= 82.

143

Table 26

The effects of skin tone, facial features, and participants 'gender on experiences with

racial discrimination in the past one year

 

With Objective Measure of Skin Tone and Facial Features

 

 

b SE ,8 t
Intercept 1 .85 . 1 0
Skin Tone -.89 1.25 -.09 -.71
Facial Features .23 .13 .20 1.70'1L
Gender .07 .30 .03 .23
Skin x Face —2.67 1.71 -.18 -1.56

R2 = .07, F(4, 78) = 1.54,p = .20

 

With Subjective Measure of Skin Tone and Facial Features

 

 

b SE ,8 t
Intercept 1 .82 .1 0
Skin Tone .01 .07 .01 .09
Facial Features .29 .19 . 18 1.52
Gender .17 .29 .07 .60
Skin x Face -.24 .14 -.19 -1.68T

R2 = .07, F(4, 78) = 1.33, p = .27

 

Note. ‘1' indicates p < .10. df= 82.

144

Table 27
The eflects of skin tone, facial features, and participants ’gender on lifetime experiences

with racial discrimination

 

With Objective Measure of Skin Tone and Facial Features

 

 

b SE ,B t
Intercept 2.35 .12
Skin Tone .27 1.43 .02 .19
Facial Features .24 .15 .19 1 .61
Gender .17 .34 .06 .49
Skin x Face -1.15 1.95 -.07 . -.59

R2 = .05, F(4, 78) = 1.04,p = .39

 

With Subjective Measure of Skin Tone and Facial Features

 

 

b SE 5 t
Intercept 2.34 . 1 2
Skin Tone .06 .08 .09 .76
Facial Features .23 .22 .13 1.05
Gender .27 .33 .10 .82
Skin x Face -.02 .16 -.Ol -.11

R2 = .03, F(4, 78) = .66, p = .62

 

Note. df= 82.

145

Table 28
The effects of skin tone, facial features, and participants 'gender on race-based rejection

sensitivity

 

With Objective Measure of Skin Tone and Facial Features

 

 

b SE ,B t
Intercept l 3.61 .99
Skin Tone 2.05 1 1.87 .02 .17
Facial Features -.38 1.26 -.04 -.30
Gender 3.51 2.82 .15 1.24
Skin x Face 30.20 16.23 .22 1861'

R2: .05, F(4, 78) = l.12,p = .35

 

With Subjective Measure of Skin Tone and Facial Features

 

 

b SE ,8 t
Intercept 14.00 .98
Skin Tone .09 .69 .02 .13
Facial Features -.94 1.78 -.06 -.53
Gender 3.03 2.74 .13 1.10
Skin x Face 2.34 1.35 .20 1.73'1'

R2 = .05, F(4, 78) = .95, p = .44

 

Note. ‘1' indicates p < .10. df= 82.

146

Table 29

The eflects of skin tone, facialfeatures, and participants 'gender on racial identification

 

With Objective Measure of Skin Tone and Facial Features

 

 

b SE ,6 t
Intercept 4.65 . 1 6
Skin Tone .53 1.97 .03 .27
Facial Features . 10 .21 .05 .46
Gender -.87 .47 -.20 -1.86T
Skin x Face .12 2.69 .01 .05

R2=.06,F(4,78)=1.15,p = .34

 

With Subjective Measure of Skin Tone and Facial Features

 

 

b SE [5' t
Intercept 4.65 . 16
Skin Tone .05 .1 l .05 .40
Facial Features .09 .29 .04 .30
Gender -.87 .45 -.23 -1 .921‘
Skin x Face -.19 .22 -.10 -.83

R2 = .06, F(4, 78) = 1.26, p = .30

 

Note. T indicatesp < .10. df= 82.

147

Table 30

The eflects of skin tone, facial features, and participants 'gender on perception of

discrimination

 

With Objective Measure of Skin Tone and Facial Features

 

 

b SE ,6 t
Intercept 2.92 . 19
Skin Tone -.10 2.24 -.01 -.04
Facial Features .05 .24 .03 .22
Gender -.20 .53 -.05 -.37
Skin x Face 2.55 3.06 .10 .83

R2 = .01, F(4, 78) = .28, p = .89

 

With Subjective Measure of Skin Tone and Facial Features

 

 

b SE ,6 t
Intercept 2 .96 .1 8
Skin Tone -.01 .13 -.01 -.O8
Facial Features .08 .34 .03 .23
Gender -.28 .52 -.07 -.54
Skin x Face .13 .25 .06 .53

R2= .01, F(4, 78) = .18,p= .95

 

Note. df= 82.

148

Table 31

The effects of skin tone, facial features, and participants ’gender on overall self-esteem

 

With Objective Measure of Skin Tone and Facial Features

 

 

b SE B t
Intercept 3.54 .08
Skin Tone -.83 .93 -.11 -.89
Facial Features .01 . 10 .02 .14
Gender .05 .22 .03 .23
Skin x Face -.50 1.28 -.05 -.39

R2 = .02, F(4, 78) = .30, p = .88

 

With Subjective Measure of Skin Tone and Facial Features

 

 

b SE E t
Intercept 3.52 .08
Skin Tone -.01 .05 -.02 -.21
Facial Features -.08 .14 -.O7 -.59
Gender . 16 .22 .09 .76
Skianace .13 .11 .15 1.24

R2 = .03, F(4, 78) = .51, p = .73

 

Note. df= 82.

149

Table 32

The effects of skin tone, facialfeatures, and participants 'gender on social self-esteem

 

With Objective Measure of Skin Tone and Facial Features

 

 

b SE 13 t
Intercept 3.50 . 1 0
Skin Tone -1.21 1.21 -.12 -1.01
Facial Features -.10 .13 -.09 -.78
Gender -.08 .29 -.04 -.29
Skin x Face -.73 1.65 -.05 -.44

R2 = .03, F(4, 78) = .52, p = .72

 

With Subjective Measure of Skin Tone and Facial Features

 

 

b SE ,6 t
Intercept 3 .48 . 1 0
Skin Tone -.04 .07 -.O7 -.59
Facial Features -.24 .18 -.16 -1.32
Gender .03 .28 .01 .12
Skianace .16 .14 .14 1.16

R2 = .04, F(4, 78) = .70, p = .60

 

Note. df= 82.

150

Table 33

The effects of skin tone, facial features, and participants 'gender on appearance self-

 

 

 

 

 

 

esteem
With Objective Measure of Skin Tone and Facial Features
b SE ,8 t
Intercept 3.42 .09
Skin Tone .90 1.13 .10 .80
Facial Features .12 .12 .12 1.01
Gender .16 .27 .08 .61
Skin x Face -.05 1.54 -.003 -.03
R2 = .03, F(4, 78) = .64,p = .63
With Subjective Measure of Skin Tone and Facial Features
b SE ,6 t
Intercept 3.39 .09
Skin Tone .1 1 .07 .19 1.63
Facial Features -.10 .17 -.07 -.58
Gender .34 .26 .16 1.32
Skianace .13 .13 .12 .99

R2 = .05, F(4, 78) = 1.08, p = .37

 

Note. df= 82.

151

 

2m:
coo.

5.2:
0de

 

 

0:03

3.3 an:
Kb. 03.
08.8: $2; :
mmém 8.3
33 :02:
320293805

00500..— 32¢

 

A~o_0 Am0_0
mvo. So.
Amfimm: Swim:
No.0 am;—
33 :00:
bzaflguoaoi

8:th 360m

 

88. :05 £0:

250 5%. V.00

5.3
E

58
2

200m 0080820; we;

A0025 200m 30m

 

00.0000X

20.00500 0000 030 . 000N000 0~0E 0003\0 00.0000\ 0 00 0000.0 00.00M00 0S 00\000000 00.000.02.005? 000.000.0000 00000000 000 0000:

0m 030%

152

 

at:
for

83%:
03a-

 

 

A8: 2:: 32.: a2:
So. So. 38. 88.
83v: GEN: 5.2: 5.3:
8.2- 32- 8.2. was-
33 :3: 33 FE:
bzmommboaopm bzwflmbowoi

838m Eon”—

838m :3on

 

223

2:: Em Em:

25? Em v.25

E:
E

5:

Boom concommcfih we:

38:: Boom and

 

8.3“ch

NEUQXNEQ was 2.3? . @332 33: $35.? «5.293%: .3 $39: 9::on 2: memxoow 2233335? 38.2633: 333:8“. 33 .33:

mm 2an

153

 

:2:
v2.

A38:
93

 

Ahm_: Amn_: Av¢_: Ao¢_:
mo_. ~o_. ac_. mm_.
48.2: 62;: 5.8: 35::
8.8 3% 3.8 $8
33 fix 33 @z
5:813:8on 528E388;

233m 36mm

833m .3on

 

2:5,

2:; :5 E:

2.8. 5% VEa

an:
E

8:

Boom :oEpOmmcmF mod

38:: Boom and

 

9.933%

3.83533 NEE 23m . “.338 m35m\«u£m\o =EB§\G an $39: méBMm: m5 ameSB. =0:S:Bu\\o $523.33: 3333:. 35 33:

cm 2an

154

 

3::
m2.

2 05.:
on?

253

 

 

Am5_: Ammo: A_o~: Avw_:
om_. mm_. ov_. ov_.
:93: 3.0.2: 3.3%: 5.2:
8.? 9.3 11: $2
33 PE: 33 :3:
35853585 bzoofiboooi

Qua—mom amoun—

thmom Emomh

 

253 5% Em:

25.3 5m BS

5:

5.:

Boom oogommcfirr wou—

AoomEV Boom Box

 

woLEoo\

NSoobozo ozoo 2.3}. . flows“ o\oso\«oo\m\o =o.:o§\o no £29: o>.:.2oo. of xo\moxoow 2233.835? 3.233363 gonzo: .33 2:8:

mm oEoH

155

Table 38

Means and standard deviations of Black perceivers ’ explicit liking toward Black male

targets as a function of skin tone and facial features

 

 

 

 

Dark Skin Tone Light Skin Tone
Facial Features Facial Features
Prototypicality Prototypicality
High Low High Low
M 3.01 3.07 3.10 3.24
(SD) (.93) (.91) (.89) (.88)

 

156

Table 39

Means and standard deviations of Black perceivers ' explicit liking toward Black female

targets as a function of skin tone and facial features

 

 

 

 

 

 

Dark Skin Tone Light Skin Tone
Facial Features Facial Features
Prototypicality Prototypicality
High Low High Low
M 3.05 3.12 3.30 3.27
(SD) (1.09) (1.12) (.98) (1.06)

 

157

 

 

 

 

5.3 A8: so: 53 53 5.3
88 a? 23 £3 :3 E 28m 8:58“: we;
96$: :Nd : 6%: z :03: 22:: R5
23% 2E? 35% 3.5 83% E 6025 28m Ex
33 Ex 33 $5
bzaflmfiouem tax—«29380.5
mocawom Eon”— mocamom 38m
223 25 Em Em: 2:; :5 0:8
2:95 Baboflfim

 

EEcQBBthctu was 5.3}.

.fimeS £3: «comic =o:u==\o me $39: oEbouSBéo: has mmbomgmu. 2t xeXmfiotfiS mwzommmLKQ «£83336 33332 35» was:

ow 2an

158

 

ate
3?

Ewfi :
vodom

953

A23 9:3

 

2nd ammo
zoom: 93%:
mimnm omdwm
33 3?:
3:853085

8:58“— 30$

 

35$ A8: 5.3

:3 23 E 28m 3582833
:33 $92: 5.8

2.0% 31% E acméeoom Ex
33 E:

bzmflnboaoi

mpuawonm fimmowm

 

25 5%. 2w:

25 Em £3

 

€83 meoobuméoz

 

APEooV ow 2an

159

 

 

 

 

$3 $03 6.0.3 A33 9&3 5.9
83 $3 3% 2% 83 E 28m 8.895%; mod
33m : 692$ 6&3: $35 A38: 55
3&8 38¢ 2.38 NEE 2.5% E A825 28m Em
33 :3: 33 En
bzmofibouoi 5:8:5805
magnum Eon”— mohzmom 30mm
323 2:; :5 E»: 25 5% x55
$53 2938me

 

358\~39th6 28“ 5%,. SmeS

m~cEm\«uc\m\o =e:u§\.3 ,3 $39.: uabomxmfiéoa has 238me m3 gogxmmtzfis 3:934:ng naotafimfi 33.35% 38 .38»:

I» 2an

160

 

 

 

5.3.3 GNS ans GMS 603 5,3
m 3% SE 3% :10 $33 2 28m 898353 33
93.3: :33: 835 A33 a 833$ 53
.380 3.23 3.5. 8.33 3.2 E E 33.5 88m 33
33 .35 33 :3:
bzaofisgofi bzaofibsoi
8:53..— .Eomm 8528.3 303
223 25 :5 £3 2.8 =5 33

 

38>» ofiboeofiméoz

 

.5353 S 29$

16]

Figure 1. Subjectively measured Black participants’ skin tone moderates the effect of
subjectively measured participant facial features on their tendency to attribute ambiguous

negative events to skin tone.

 

   

 

 

 

 

 

4.5 -
—9—— Darker skin tone (-1 SD)

0 4‘0 ‘ ------I Lighter skin tone (+ISD)
5 3.5 .
[-
.5 3.0 l
m
8 2.5 '1
5
g 2.0 .
:9.
E 1.5 4
<

1.0 .

0.5 .

0.0

Less prototypical More prototypical
facial features facial features
(-ISD) (+ISD)

162

Figure 2. Objectively measured Black participants’ skin tone moderates the effect of
objectively measured participant facial features on their tendency to attribute ambiguous

negative events to facial features.

 

 

 

 

 

 

 

3.5 -
3 3.0 r
I-
B
3 2.5 4
tr.
3
g 2.0 ~
tr.
e
2 1.5 i
e
'5
:3 1-0 ‘ +Darker skin tone (-ISD)
g 0.5 4 ------I Lighter skin tone (+ISD)

0.0 .

Less prototypical More prototypical
facial features facial features
(-ISD) (+ISD)

163

Figure 3. Subjectively measured Black participants’ skin tone moderates the effect of

subjectively measured participant facial features on their tendency to attribute ambiguous

negative events to internal factors.

Attribution to Internal Factors

4.5 ‘
4.0‘
3.5 *
3.04
2.5 i
2.0 ‘
1.5 ‘
1.0 *
0.5 ‘

 

 

 

—o— Darker skin tone (-1 SD)
---I- - - Lighter skin tone (+ISD)

 

 

 

 

0.0

Less prototypical More prototypical
facial features facial features
(-1 SD) (+1 30)

164

Figure 4. Subjectively measured Black participants’ skin tone moderates the effect of
subjectively measured participant facial features on their experiences with racial

discrimination in the past 1 year.

 

 

 

 

 

 

 

a 2.5l

e

'5

a

g 2.0‘

8

E 1.5‘

.2

C)

a

a:

5 1.0 ‘

3

3 —O— Darker skin tone (-1 SD)
{—3 0-5 ---I---Lighterskintone(+ISD)
8.

31

“J 0.0

Less More
prototypical prototypical
facial features facial features

165

Figure 5. Objective] y measured Black participants’ skin tone moderates the effect of

objectively measured participant facial features on their sensitivity to racism.

 

 

 

 

 

 

 

18.0]
g 16.01
'5 14.0 4
:1
‘2 12 0 .
a .
.o
g 10.0 <
52 8.0 ~
'8
g 6.0 * .
m —6— Darker skin tone (-1 SD)
é 4'0 1 ---I--- Lighter skin tone (+ISD)
9‘ 2.0 ~
0.0
Less More
prototypical prototypical
facial features facial features

166

Figure 6. Subjectively measured Black participants’ skin tone moderates the effect of

subjectively measured participant facial features on their sensitivity to racism.

 

 

 

 

 

 

 

18.0”
3:16.04
IE
.2: 14.0 .
m
:i
a“; 12.0«
a
{-3- 10.0«
8
'3‘ 8.0 .
a:
B 6.0 ~
g 4 0 —+— Darker skin tone (-1 SD)
3 ' -- -I- - - Lighter skin tone (+ISD)
a 2.0 *

0.0

Less More
prototypical prototypical
facial features facial features

167

Appendix A: Objective measures of facial features
Figure 1. Objective measurements of face width, face length, nose width, and lip

thickness.

 

168

Appendix B: An example of outlined pictures

Figure 11. An example of outlined pictures.

 

 

.m" Mm“?-
‘1'
r!" , »
if
.0" -
1’
I .
.' l
a. .9. .
f“ x -. ~45?‘
'5" ‘.y 3 x. ‘ 3;
W' “a
‘W'tf-itaar'
2": . e
r,» ,.
J’". '
.H" . ‘
,.-.r
o—‘J' " ‘
‘fi‘x‘,

 

 

 

 

169

Appendix C: The final 64 Black male target pictures

Figure 111. Black male target pictures used in the main studies.

Setl:
Dark Skin with High Prototypical Facial Features

 

170

Figure 111(cont’d).

Set 2:
Dark Skin with High Prototypical Facial Features

 

171

Appendix D: The final 8 Black female target pictures

Figure IV. Black female target pictures used in the main studies.

Dark Skin with High Prototypical Facial Features
Set 1 Set 2

 

Dark Skin with Low Prototypical Facial Features
Set 1 Set 2

 

Light Skin with High Prototypical Facial Features
Set 1 Set 2

 

Light Skin with Low Prototypical Facial Features
Set 1 Set 2

   

172

Appendix E: A design of phase 4 in the sequential priming task

Eight sets of matched five faces

 

 

 

 

 

 

1

 

 

 

 

 

 

__Block 1 $0;le M2 M91
DM1+Good1 DM1+Good2 DM1+Bad1 DM1+Bad2

LMl + Good] LMl + Good2 LMl + Badl LMl + Bad2

DLl + Good] DLl + Good2 DLl + Badl DLl + Bad2 ‘i\
LL1 + Good] LL1 + Good2 LL1 + Badl LL1 + Bad2 g?
W1 + Good] W1 + Good2 W1 + Badl W1 + Bad2 g.
DMZ + Good2 DMZ + Good3 DMZ + Bad2 DMZ + Bad3 ‘6
LM2 + Good2 LM2 + Good3 LM2 + Bad2 LM2 + Bad3 g
DL2 + Good2 DL2 + Good3 DL2 + Bad2 DL2 + Bad3 5°
LL2 + Good2 LL2 + Good3 LL2 + Bad2 LL2 + Bad3 91
W2 + Good2 W2 + Good3 W2 + gag W2 + m g.
DM3 + Good3 DM3 + Good4 DM3 + Bad3 DM3 + Bad4 8.
LM3 + Good3 LM3 + Good4 LM3 + Bad3 LM3 + Bad4 37
DL3 + Good3 DL3 + Good4 DL3 + Bad3 DL3 + Bad4 G
LL3 + Good3 LL3 + Good4 LL3 + Bad3 LL3 + Bad4 9",”
W3 + Good3 W3 + Good4 W3 + Bad3 W3 + Bad4 8
DM4 + Good4 DM4 + Good5 DM4 + Bad4 DM4 + Bad5 g,
LM4 + Good4 LM4 + Good5 LM4 + Bad4 LM4 + Bad5 3
DL4 + Good4 DL4 + Good5 DL4 + Bad4 DL4 + Bad5 0
LL4 + Good4 LL4 + Good5 LL4 + Bad4 LL4 + Bad5 é."
W4 + Good4 W4 + Good5 W4 + Bad4 W4 + Bag g
DMS + Bad5 DMS + Bad6 DMS + Good5 DMS + G00d6 &
LM5 + Bad5 LM5 + Bad6 LM5 + Good5 LM5 + Good6 c,
DL5 + Bad5 DLS + Bad6 DLS + Good5 DL5 + G00d6 ‘<
LLS + Bad5 LL5 + Bad6 LLS + Good5 LLS + G00d6 §
W5 + Bad5 W5 + Bad6 W5 + Good5 W5 + Good6 g
DM6 + Bad 6 DM6 + Bad7 DM6 + Good6 DM6 + Good7 a
LM6 + Bad6 LM6 + Bad7 LM6 + GOOd6 LM6 + Good7 g
DL6 + Bad6 DL6 + Bad7 DL6 + Good6 DL6 + Good7 g
LL6 + Bad6 LL6 + Bad7 LL6 + G00d6 LL6 + Good7 0»
W6 + Bad6 W6 + Bad7 W6 + GOOd6 W6 + Good7 §
DM7 + Bad7 DM7 + Bad8 DM7 + Good7 DM7 + Good8 g
LM7 + Bad7 LM7 + Bad8 LM7 + Good7 LM7 + Good8 .5;
DL7 + Bad7 DL7 + Bad8 DL7 + Good7 DL7 + Good8 a
LL7 + Bad7 LL7 + Bad8 LL7 + G00d7 LL7 + Good8 5"
W7 + Bad7 W7 + Bad8 W7 + Good7 W7 + Good8 E.
DM8 + Bad8 DM8 + Badl DM8 + Good8 DM8 + Goodl a
LM8 + Bad8 LM8 + Badl LM8 + Good8 LM8 + Goodl g
DL8 + Bad8 DL8 + Badl DL8 + Good8 DL8 + Good] °
LL8 + Bad8 LL8 + Badl LL8 + Good8 LL8 + Good]

W8 + Bad8 W8 + Badl W8 + Good8 W8 + Good]

Note. DM indicates “Dark-skinned More prototypical facial features,” LM indicates “Light-

skinned More prototypical facial features,” DL indicates “Dark-skinned Less prototypical facial

features,” LL indicates “Light-skinned Less prototypical facial features,” and W in indicates
“White control.”

173

Appendix F: Selection of the target words for the lexical decision task

A list of 36 target words (12 negative words associated with Black male
stereotypes, six negative words associated with Black female stereotypes, 12 negative
words not associated with Black male stereotypes, and six negative words not associated
with Black female stereotypes) was created for the lexical decision task. It was important
to ensure that the 18 stereotypic words and 18 non-stereotypic words were matched in
terms negativity, because a control prime condition (which could serve as a measure of
baseline reaction times) was not included. In addition, 36 non-words were matched with
each of the 36 real words in terms of the number of letters were created.

Sixty-eight White participants were first asked to rate 55 adjectives in terms of
negativity. More specifically, participants were asked to rate each of the 55 words on a
scale ranging from 1 (Not at All Negative) to 7 (Extremely Negative). Next, participants
were asked to rate the same 55 words in terms of stereotypicality of Blacks. Of 68
participants, 35 participants were asked to indicate Whether an adjective was stereotypic
of Black men, and the other 33 participants were asked to rate Whether an adjective was
stereotypic of Black women, using a scale ranging from 1 (Not at All Stereotypic of
African American Men/W omen) to 7 (Extremely Stereotypic of African American
Men/W omen). Importantly, participants were instructed to make their judgments based
on stereotypes that are held by other people in general, but not necessarily based on their
personal beliefs. Based on the negativity and stereotypicality ratings, 36 target words (12
negative words associated with Black male stereotypes, six negative words associated

with Black female stereotypes, 12 negative words not associated with Black male

174

stereotypes, and six negative words not associated with Black female stereotypes) were

selected (see the list below).

 

 

 

 

 

 

 

 

 

 

For Black Male Targets

Stereotypic Non-Words Non-Stereotypic Non-Words
CREEPY TWINDE ANNOYING WROURSTE
DANGEROUS PHAUGHNED ARROGANT WHARPHTE
IGNORANT WROOSSED BORING KREIND
INCOMPETENT SCKWAUGHLDE CONDESCENDING SCHOUGHCKINGS
LAZY UPC E GREEDY MYSSED
MEAN LAWD IRRITATING PHLEIGHLDE
NOISY GEAZE JUDGMENTAL SQUOUGHLED
POOR OADE SELFISH FEIGHLD
RECKLESS BREIGNED SMELLY PAISTE
THREATENING SC KWOUGHLED STUCKUP SCKWEAZ
UNTRUSTWORTHY SC KREETHECHED UGLY SUKT
VIOLENT FAUSSED WEAK CECT

For Black Female Targets

Stereotypic Non-Words Non-Stereotypic Non-Words
ANNOYING WROURSTE EXPLOITATIVE CHEIGHNDGING
IRRITATING PHLEIGHLDE SELF ISH F EIGHLD
NOISY GEAZE SHELTERED SCKWOARLD
NOSEY VAMPT SLOPPY PHIGHZ
RUDE C EXT SLOW HEZE
STUBBORN KOURGNED STUCKUP SCKWEAZ

 

In order to examine whether the 12 words selected as stereotypic of Black men
were significantly different from the 12 words selected as non-stereotypic of Black men
in terms of stereotypicality of Black men, a paired-samples t-test was conducted on
stereotypicality ratings. The result showed that two sets of words were significantly
different from one another, t(34) = 10.21, p < .001, d = 1.68. Next, a paired-sample t-test
was conducted to examine whether these two sets of words were similar in terms
negativity. The test revealed that there was no mean difference in the degree of negativity
between the two sets, t(67) = .02, p = .99.

The same tests were conducted on words for Black female targets. A paired-
sarnples t-test showed that six words selected as stereotypic of Black women were
perceived to be significantly more stereotypic of Black women than six words selected as
non-stereotypic of Black women, t(32) = 8.96, p < .001, d = 1.21. In addition, a paired-
samples t-test on negativity ratings showed that these two sets of words did not differ in
negativity, t(67) = .60, p = .55. In sum, 18 words selected as stereotypic of Blacks and 18
words selected as non-stereotypic of Blacks only differ in stereotypicality, but not in
negativity, indicating that the selection of 36 words was successful. The means and

standard deviations for each rating are presented in the table below.

176

Table I
Means dflerences in stereotypicality and negativity ratings between stereotypic words

and non-stereotypic words within each gender

 

 

 

 

 

Black Male Targets Black Female Targets
. Non- . Non-
$3203: 1c Stereotypic Staegtgprc Stereotypic
Word Words

Stereotypicality M 4.86 2.82 4.79 3.32
SD 1.46 .90 1.40 1.00

Negativity M 5.19 5.19 4.80 4.76
SD .78 .77 .74 .76

 

177

Appendix G: A list of pre-Iaboratory questionnaires
A. Group Identification
1. Overall, my racial group membership has very little to do with how I feel about
myself.
2. The racial group I belong to is an important reflection of who I am.
3. The racial group I belong to is unimportant to my sense of what kind of person
I am.
4. In general, belonging to my racial group is an important part of my self-image.
B. Self-Esteem
1. I feel that I am a person of worth, at least on an equal basis with others.
2. I feel that I have a number of good qualities.
3. All in all, I am inclined to think I am a failure.
4. I am able to do things as well as most people.
5. I feel that I do not have much to be proud of.

6. I take a positive attitude toward myself.

\I

. On the whole, I am satisfied with myself.

00

. I wish I could have more respect for myself.

\0

. I certainly feel useless at times.

10. At times, I think I am no good at all.

1 l. I am worried about Whether I am regarded as a success or a failure.
12. I feel self-conscious.

13. I feel displeased with myself.

14. I am worried about what other people think of me.

178

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

I am worried about looking foolish.

I feel inferior to others at this moment.

I feel concerned about the impression I am making.

I feel confident about my abilities.

I feel frustrated or rattled by my performance.

I feel that I am having trouble understanding things I read.
I feel as smart as others.

I feel like I am not doing well.

I feel confident that I understand things.

I feel I have less scholastic ability right now than others.
I feel satisfied with the way my body looks right now.

I feel that other respect and admire me.

I am dissatisfied with my weight.

I feel good about myself.

I am pleased with my appearance right now.

I feel unattractive.

C. Past Experience with Discrimination

1. How many times have you been treated unfairly by teachers and professors

because you are Black?

1) How many times in the past year?
2) How many times in your entire life?

3) How stressful was this for you?

179

2. How many times have you been treated unfairly by your employers, bosses,
and supervisors because you are Black?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
3. How many times have you been treated unfairly by your coworkers, fellow
students, and colleagues because you are Black?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
4. How many times have you been treated unfairly by people in service jobs (store
clerks, waiters, bartenders, bank tellers and others) because you are Black?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
5. How many times have you been treated unfairly by strangers because you are
Black?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
6. How many times have you been treated unfairly by people in helping jobs
(doctors, nurses, psychiatrists, case workers, dentists, school counselors,

therapists, social workers and others) because you are Black?

180

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
7. How many times have you been treated unfairly by neighbors because you are
Black?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
8. How many times have you been treated unfairly by institutions (schools,
universities, law firms, the police, the courts, the Department of Social Services,
the Unemployment Office and others) because you are Black?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
9. How many times have you been treated unfairly by people that you thought
were your friends because you are Black?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
10. How many times have you been accused or suspected of doing something
wrong (such as stealing, cheating, not doing your share of the work, or breaking
the law) because you are Black?

1) How many times in the past year?

181

2) How many times in your entire life?

3) How stressful was this for you?
1 I. How many times have people misunderstood your intentions and motives
because you are Black?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
12. How many times did you want to tell someone off for being racist but didn 't
say anything?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
13. How many times have you been really angry about something racist that was
done to you?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
14. How many times were you forced to take drastic steps (such as filing a
grievance, filing a lawsuit, quitting your job, moving away, and other actions) to
deal with some racist thing that was done to you?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressfiil was this for you?

182

15. How many times have you been called a racist name like n__, coon, jungle
bunny or other names?

I) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
16. How many times have you gotten into an argument or a fight about something
racist that was done to you or done to somebody else?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
17. How many times have you been made fun ofi picked on, pushed, shoved, hit,
or threatened with harm because you are Black?

1) How many times in the past year?

2) How many times in your entire life?

3) How stressful was this for you?
18. How diflerent would your life be now if you HAD NOT BEEN treated in a
racist and unfair way

1) In the past year?

2) In your entire life?

D. Perception of Discrimination

1. I experience discrimination because of my race.
2. I feel that I am discriminated against because of my race.

3. I feel like I am personally a victim of society because of my race.

183

4. I consider myself a person who is deprived of opportunities that available to
others because of my race.
5. I personally have been a victim of racial discrimination.

E. Demographic Information
1. Age
2. Gender (Male, Female)
3. Race
4. Year in school (Freshman, Sophomore, Junior, Senior)
5. Number of same-race close friends participants have on the MSU campus
6. Number of different-race close friends participants have on the MSU campus
7. Number of same-race close fiiends participants have outside the MSU campus
8. Number of different-race close friends participants have outside the MSU
campus
9. Family’s socioeconomic status (Extremely Poor, Poor, Average, Wealthy,
Extremely Wealthy)

10. Current overall GPA

184

Appendix H: Selection of attribution items and scenarios for the attribution measure in
Study 3

The attribution measure created for Study 3 originally consisted of 10 different
scenarios and 10 different attribution items for each scenario (see the list of scenarios at
the end of this appendix). Because this measure required participants to provide 100
responses, there were potential issues with participants’ fatigue or boredom. In order to
test whether the measure could be shortened, internal consistency Within 10 scenarios as
well as within 10 attribution items were computed, based on the responses provided by
non-Black ethnic minorities, for looking for redundancy within scenarios and items.

F ifiy-seven non-Black ethnic minority participants (Asian American = 54.4 %,
Latino American = 29.8 %, Native American = 3.5 %, Multiracial/Other = 1.8 %, and
International Student = 10.5 %; women = 52.6 %; M age = 20.04, SD = 2.05 years old)
were recruited for Pilot Study 3. Ideally, participants in Pilot Study 3 would be Blacks.
However, those Black individuals participated in Pilot Study 3 would not be eligible to
take part in Study 3. Due to the limited number of available Black participants in the
subject pool, I decided not to recruit Black individuals for the pilot study and maximize
the number of Black participants for the main study.

Participants were asked to read 10 scenarios (most of which were obtained from
Branscombe et al.’s study (1999) that examined the perception of discrimination among
Black individuals). Each scenario described a negative experience in different situations
that could be due to racial discrimination or to other causes. Participants were asked to
imagine that they have experienced these ambiguous negative events and then speculate

why these negative events have happened to them.

185

More specifically, participants were provided with 10 attribution categories: 1)
internal non-discrimination attributions (e.g., their personality, mood, abilities, or
intelligence), 2) external nondiscriminatory attributions (e. g., interaction partners’
personality, mood, abilities, or intelligence), 3) situational attributions (e.g., social norm,
lack of opportunity, rule, or regulation), 4) internal racial discrimination attributions (e.g.,
their racial group membership), 5) external racial discrimination attributions (e. g.,
interaction partners’ prejudiced attitudes toward their racial group), 6) internal gender
discrimination attributions (e.g., their gender group membership), 7) external gender
discrimination attributions (e. g., interaction partners’ prejudiced attitudes toward their
gender group), 8) skin tone attribution (e. g., their skin tone), 9) facial feature attribution
(e. g., their facial appearance), and 10) other (participants were asked to specify the
attribution). Some of these attribution categories were derived from King’s study (2005).
For each attribution item, participants were asked to rate their perception of the cause of
the experience, using a scale ranging from 1 (Very Unlikely the Cause of the Experience)
to 7 (Very Likely the Cause of the Experience).

Means and standard deviations for each of the nine attribution items within each
scenario are presented in Table II at the end of this appendix. First, internal consistency
among the 9 attribution items Within each scenario was examined. As shown in Table III
at the end of this appendix, as were high for all scenarios, indicating that those who
attributed ambiguous negative events to one cause were also likely to attribute the same
events to other causes (e. g., participants who attributed their failure to get a job to their
racial membership also attributed the failure to their personal characteristics that are

unrelated to racial membership).

186

Because internal consistency was high within each scenario, several attribution
items were eliminated to shorten the measure, so that participants would be less
susceptible to fatigue and boredom. More specifically, the external racism and external
sexism attribution items (i.e., due to my partner's racism/sexism attitudes) were
eliminated, because these items were highly correlated with the internal racism and
internal sexism attribution items (i.e., due to my racial/ gender membership; see Tables IV
for correlations across the 10 scenarios), suggesting that participants did not differentially
responded to the internal vs. external discrimination attribution items. In addition, the
partner's personality and situation attribution items were combined into a single
"external" attribution item because both are not under one’s own control. Finally, “other”
attribution, which participants need to specify, was drop from the list because only a few
participants actually came up with “other” potential causes. This brought the number of
attribution items for each scenario from ten to six.

Next, internal consistency within each attribution category across the 10 scenario
was computed. As shown in Tables V at the end of this appendix, as were high across all
attribution categories, indicating that those who attributed ambiguous negative events to
one cause in one scenario were also likely to attribute negative events to the same cause
in different scenarios (e. g., participants who attributed their failure to get a job to their
racial membership also attributed their negative experience at a fancy restaurant to their
racial membership). Again, in order to prevent potential fatigue, four scenarios with the
lowest item-total correlation values based on the internal racial discrimination item were
eliminated from the original list of 10 items (i.e., scenarios 5, 8 ,9, and 10), resulting in

the total of six scenarios. Taken together, the attribution measure for Study 3 consisted of

187

six different scenarios that described ambiguous negative events, and each scenario could

be attributed to six different causes.

I88

The original list of ambiguous negative scenarios for the attribution task

1. Imagine that you went to a "fancy" restaurant. Your server seemed to be taking care of
all the other customers except you. You were the last person whose order was taken.

2. Imagine that you applied for a job that you believed you were qualified for. After the
interview you learned that you didn't get the job.

3. Imagined that you parked your car at a parking meter and it has just expired. You
arrived back at the car just as an officer was writing up a ticket. You tried to persuade the
officer not to give you the ticket, after all you were there and the meter just expired. The
officer gave you the ticket anyway.

4. Imagine that you went to look at an apartment for rent. The manager of the building
refused to show it to you, saying that it has already been rented.

5. Imagine that you were attracted to a particular White man/woman and asked that
person out for a date and was turned down.

6. Imagine that you had to fill out some government forms in order to apply for a loan
that was important to you. You went to one office and they send you to another, then you
went there and were sent somewhere else. No one seemed to be really willing to help you
out.

7. Imagine that you were driving a few miles over the speed limit and the police pull you
over. You received a ticket for the maximum amount allowable.

8. Imagine that you wanted to join a social organization. You were told that they are not
taking any new members at this time.

9. Imagine that your boss told you that you were not performing your job as well as

others doing that job.

189

10. Imagine that you went to Chicago a few weeks ago. You tried to catch a cab but
several empty cabs passed by you, even thought you were pretty sure that the drivers saw

you.

190

 

 

 

 

 

191

5.: 5.: 5.: 5.: 5.: so: 5.: 2w: 5.: 5.: 3m:
in com 45 Em 25m 25 Ba 3m 3m 43 2 833a
5.: 5.: 5.: 5.: 5.: 8w: 5.: 5.: 5.: 5.: Se
2% cam 43“ No.4 a; 63 m3 :1“ «mm m: E $qu 650
5.: 5.: 5.: 5.: 5.: 5.: 5.: 5.: 5.: 5.: 3.:
men m3 35 5a 23 SN 63 85 SN MS 2 Seam =2:me
5.: 5.: 5.: 5.: 5.: 5.: 5.: . 5.: 5.: 5.: 3,:
3% mod 85 com 55 55. Ba 93 43 m2 3 828m 383%
5.: 84.: 5.: AS: 5.: $5.: 36.: 5.: 5.: 5.: 3:
9% w a .N 35 35 NE 3m 35 N5 oz 85 E mass; 368
5.: 5.: 5.: 5.: 5.: 5.: 5.: 5.: 5.: 5.: 3.:
3m 82 o: 63. Sm Sm £5. 52 mam SM 3. 685 5m
86.: 5.: 6m: 2w: 5.: 5.: so: 5.: 5.: 5.: 3.:
mom 3m Rm 6? 8a 33. :3 £5 8.4 :3. E 3825 see
5. : 5.: 5.: 5.: 5.: 5.: 5.: 5.: 5.: 5.: 5.:
63 com 2: tum SN w _ .N 85 E .85 new E 53% 3835
:5. : 5.: 5. : so: so. : 5.: 5.: em: 26.: 5. : 5:
5.4 SN mom M3. 3m 91.. 3m 23 43 Rm 3 828m. 3895
2 o w 5 e m 4 m N _
ogcoow cousnwfixx

 

3.238% : N mee «one agent»: 2223.22: 2:: mSXQ 92333: $3382 :38 gem:

= 2an

 

av. mm. 3. no. on. on. mm. mm. 3. mm. :2852m

am. 3. mm. on. 0:. mm. on. we. mm. 8. 858m 850
3. mm. 3. mm. on. 3. :0. mm. 3.. mm. Emmxom 8:835
No. 5. 2.. .w. ow. R. mm. K. mm. mo. E203“ 8:82..“—
Su. mm. :u 8. cm. mm. ov. No. 0:. Q. 858$ 868:
on. we. 3.. E. 2.. mm. ow. E. 5. mo. 0:8. :3m
am. am. mm. 3. 5:. am. vm. am. mm. 58. 852:: 850
3. Eu ow. on. we. v.8: on. 3.. Vm. G. Eflxom 8E8:—
oo. mm. cm. on. S. w». on. mm. mm. 0:. 8203: 850::

 

228.230 30%-:8: 38880

 

$38. $38. 8:8. 8:8. $38. $38. $38. 8:8. 8:8. :33. 2:388:29

 

S o m N. o w v m N 2
2.8:oom 2.8:on 28:8m 2.8:on 2.8:8m 38:03 2.8:on 2.8:on 2.8:8m 2.8:on

 

3.2::an : N mfi\o :93 53.3» 9.26: 22392:: :5: 2: 86.8: 83:33:: 3525

E 038:.

192

 

Ahm: Ahm: Anm: Ahm: Ann: Aom: Aom: Aom: Aom: Aom: ¢<:
3.9V. **mw. 31.0. 1.00. 1&0. *hN. 3&0. .13.. “Iain 1.0m. acts—0800 :Ombwom

 

 

 

82me
Ahm: Anm: 28m: Amm: Ann: 28m: 25m: Amm: Amm: Anm: o<:
sine. .1.on “.350. 1&9 "1.3.. from. “numb. 3.3.. 1.0m. .Ibm. :OEEPCOU :Ombwom
E203:
o_ o w b o n v m N _

Ogaoom OCNCOOW OCNGOOW OCMCDOW Ogcoom Oran—Dom OEGDOW Ofifioom Gina—Dom OEGOOW

 

”0.262990 QN m<~ thkUG 2.29% 20339226 :31wa“. \BgsquN 3:3 NGEWEK» 323 Shrugs NGQKNNRN 323 N32k®u2m QNNENQ :OCQNWKKOU

>~ 2:8...

193

 

NV. 2.. G. Vm. me. mm. mm. o0. no. 2 2.8:on

on. S. we. mo. 3. mo. wv. mm. mm. m ogcoom
ac. mm. 3. No. ac. mm. om. oo. on. w otwcoom
on. on. mm. on. me. w». 6. mm. on. n 082.com
we. on. 3. ac. mm. 3. mm. 3. me. o ogcoom
S. K. co. on. 2‘. 2.. No. ow. co. m 28:on
mm. E. on. 2.. 3. mm. 6. cm. mm. v otmcoom
2.. on. mm. on. 3. R. 8. on. E. m oficoom
oo. on. me. we. K. mu. 3.. mm. S. m 2355
3. 3. mm. B. E. on. .3. E. K. _ oracoom

 

coca—8.50 309-88— cowootou

 

68%. $33. £33. £33. 89%. 838. 95$. :33. $33. 33.5885

 

coca—Em 558m Emmxom Emmomm $580... 25,—. E 358.: Emiom Emfimm
. . 550 :25me $5me :28; .xm 550 3585 EEBE

 

2:»: 20.23.53 E 233 fig 2.233 8.23289. Q N 2: $6.33 AutmEmzou 358$

> 033.

194

Appendix I: Race-Based Rejection SenSitivity
1. Imagine that you are in class one day, and the professor asks a particularly difficult
question. A few people, including yourself, raise their hands to answer the question.
1) How concerned/anxious would you be that the professor might not choose you
because of your race/ethnicity?
2) I would expect that the professor might not choose me because of my
race/ethnicity.
2. Imagine that you are in a pharmacy, trying to pick out a few items. While you’re
looking at the different brands, you notice one of the store clerks glancing your way.
1) How concerned lanxious would you be that the clerk might be looking at you
because of your race/ethnicity?
2) I would expect that the clerk might continue to look at me because of my
race/ ethnicity.
3. Imagine you have just completed a job interview over the telephone. You are in good
spirits because the interviewer seemed enthusiastic about your application. Several days
later you complete a second interview in person. Your interviewer informs you that they
will let you know about their decision soon.
1) How concerned/anxious would you be that you might not be hired because of
your race/ ethnicity?
2) I would expect that I might not be hired because of my race/ethnicity.
4. It’s late at night and you are driving down a country road you’re not familiar with.
Luckily, there is a 24-hour 7-11 just ahead, so you stop there and head up to the counter

to ask the young woman for directions.

195

1) How concerned/anxious would you be that she might not help you because of

your race/ethnicity?

2) I would expect that the woman might not help me because of my race/ethnicity.
5. Imagine that a new school counselor is selecting students for a summer scholarship
fund that you really want. The counselor has only one scholarship lefi and you are one of
several students that is eligible for this scholarship.

I) How concerned/anxious would you be that the counselor might not choose

you because of your race/ethnicity?

2) I would expect that he might not select me because of my race/ethnicity.
6. Imagine you have just finished shopping, and you are leaving the store carrying several
bags. It’s closing time, and several people are filing out of the store at once. Suddenly,
the alarm begins to sound, and a security guard comes over to investigate.

1) How concerned/anxious would you be that the guard might stop you because

of your race/ethnicity?

2) I would expect that the guard might stop me because of my race/ethnicity.
7. Imagine you are riding the bus one day. The bus is full except for two seats, one of
which is next to you. As the bus comes to the next stop, you notice a woman getting on
the bus.

1) How concemed/anxious would you be that she might avoid sitting next to you

because of your race/ethnicity?

2) I would expect that she might not sit next to me because of my race/ethnicity
8. Imagine that you are in a restaurant, trying to get the attention of your waitress. A lot

of other people are trying to get her attention as well.

196

1) How concerned/anxious would you be that she might not attend you right
away because of your race/ethnicity?

2) I would expect that she might not attend to me right away because of my

race/ ethnicity
9. Imagine you’re driving down the street, and there is a police barricade just ahead. The
police officers are randomly pulling people over to check drivers’ licenses and
registrations.
1) How concemed/anxious would you be that an officer might pull you over
because of your race/ethnicity?
2) I would expect that the officers might stop me because of my race/ethnicity.
10. Imagine that it’s the second day of your new class. The teacher assigned a writing
sample yesterday and today the teacher announces that she has finished correcting the
papers. You wait for your paper to be returned.
1) How concerned/anxious would you be that you might receive a lower grade
than others because of your race/ethnicity?
2) I would expect to receive a lower grade than others because of my
race/ ethnicity.
11. Imagine that you are standing in line for the ATM machine, and you notice the
woman at the machine glances back while she’s getting her money.
I) How concemed/anxious would you be that she might be suspicious of you

because of your race/ethnicity?

2) I would expect that she might be suspicious of me because of my race/ethnicity.

197

12. Imagine you’re at a pay phone on a street corner. You have to make a call, but you
don’t have change. You decide to go into a store and ask for change for your bill.
I) How concerned/anxious would you be that the cashier might not give you

change because of your race/ethnicity?

2) I would expect that the cashier might not give me change because of my

race/ethnicity.

198

.muofioo
885%: No 2: macaw €28 “macaw: 2: mo 28 go 228
“macaw 2: go 28 £ 830 women“ 2: 882?» made“ on 53 50>

hO-OU £28 858%“
I No 32: 9. 860 Homem—
uwmh NP 2: 8388 EB 50%

lllllllllllnllll
mam-ulmllauuluunul
EUIDUIuIDIDDIDm
[Dun-all

.30—co aim owmbiw mo 8388a .x. 33m:

 

   
       

    

 

who—8 aim omega“ mo moEmem a. vacuum“?

|

199

REFERENCES

Anderson, C., & Cromwell, R. L. (1977). “Black is beautiful" and the color preferences
of Afro-American youth. Journal of Negro Education, 46, 76-88.

Ashbum-Nardo, L., Volis, C. 1., & Monteith, M. J. (2001). Implicit associations as the
seeds of intergroup bias: How easily do they take root? Journal of Personality and Social
Psychology, 81, 789-799.

Averhart, C. J ., & Bigler, R. S. (1997). Shades of meaning: Skin tone, racial attitudes,
and constructive memory in African American children. Journal of Experimental Child
Psychology, 67, 363-388.

Banerj i, S. (2006). Darker-skinned Black Job Applicants Face More Obstacles. Diverse
Issues in Higher Education, 23, 20.

Bargh, J. A. (1994). The four horsemen of automaticity: Awareness, intention, efficiency,
and control in social cognition. In R. S., Wyer & T. K. Srull (Eds), Handbook of social
cognition (2“d ed., pp. 140). Hillsdale, NJ: Erlbau.

Blackwell, J. E., (1985). The Black Community: Diversity and Unity. New York: Dodd,
Mead.

Blair, 1. V. (2002). The malleability of automatic stereotypes and prejudice. Personality
and Social Psychology Review, 6, 242-261.

Blair, 1. V. (2006). The efficient use of race and Afrocentric features in inverted faces.
Social Cognition, 24, 563-579.

Blair, 1. 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, 1. V., Judd, C. M., & Chapleau, K. M. (2004a). The influence of Afrocentric facial
features in criminal sentencing. Psychological Science, 15, 674-679.

Blair, 1. V., Judd, C. M., & Fallman, J. L. (2004b). The automaticity of race and
Afrocentric facial features in social judgments. Journal of Personality and Social
Psychology, 87, 763-778.

Blair, 1. 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
and Social Psychology, 83, 5-25.

200

Bond, S., & Cash, T. F. (1992). Black beauty: Skin color and body images among
African American college women. Journal of Applied Social Psychology, 22, 874-888.

Bornstein, R. F. (1993). Mere exposure effects with outgroup stimuli. In D. M. Mackie &
D. L. Hamilton (Eds) Aflect, cognition, and stereotyping: Interactive processes in group
perception (pp. 195-211). San Diego, CA: Academic Press.

Boyle, E. (1970). Biological patterns in hypertension by race, sex, body weight, and skin
color. Journal of the American Medical Association, 213, 1637-1643.

Bradbum, N. H. (1969). The structure of psychological well-being. Oxford, England:
Aldine.

Branscombe, N. R., Schmitt, M. T., & Harvey, R. D. (1999). Perceiving pervasive
discrimination among African Americans: Implications for group identification and well-
being. Journal of Personality and Social Psychology, 77, 135-149.

Brewer, M. B., (1979). In-group bias in the minimal intergroup situation: A cognitive-
motivational analysis. Psychological Bulletin, 86, 307-324.

Brewer, M. B. (1988). A dual process model of impression formation. In T. Srull &
R.Wyer (Eds.). Advances in Social Cognition. Vol. 1, Earlbaum.

Brewer, M. B. (1999). The psychology of prejudice: Ingroup love or outgroup hate?
Journal of Social Issues, 55, 429-444.

Brown, T. D., Dane, F. C., & Durham, M. D. (1998). Perception of race and ethnicity.
Journal of Social Behavior & Personality, 13, 295-306.

Brown, E., & Perrett, D. I. (1993). What gives a face its gender? Perception, 22, 829-840.
Buchanan, N. T., & Fitzgerald, L. F. (2008). Effects of racial and sexual harassment on
work and the psychological well-being of African American women. Journal of

Occupational Health Psychology, 13, 137-151.

Buchanan, N. T., & Ormerod, A. J. (2002). Racialized sexual harassment in the lives of
Afiican American women. Women & T heory, 25, 107-124.

Burton, A. M., Bruce, V., & Dench, N. (1993). What’s the difference between men and
women? Evidence from facial measurement. Perception, 22, 153-176.

Cloutier, J ., Mason, M. F., & Macrae, C. N. (2005). The perceptual determinants of

person construal: Reopening the social-cognitive toolbox. Journal of Personality and
Social Psychology, 88, 885-894.

201

Crandall, C.S., Eshleman, A., & O’Brien, L. (2002). Social Norms and the Expression
and Suppression of Prejudice: The Struggle for Intemalization. Journal of Personality
and Social Psychology, 82, 359—378.

Coard, S. I., Breland, A. M., & Raskin, P. (2001). Perceptions of and preferences for skin
color, Black racial identity, and self-esteem among African Americans. Journal of
Applied Social Psychology, 31, 2256-2274.

Davis, K. B., Daniels, M., & See, L. A. (1998). The psychological effects of skin color
on African Americans’ self-esteem. Journal of Human Behavior in the Social
Environment, 1, 63-90.

DeBruine, L. (2002). Facial resemblance enhances trust. Proceedings of the Royal
Society of London, 269, 1307-1312.

Deregowski, J. B., Ellis, H. D., & Shepherd, J. W. (1975). Descriptions of White and
Black faces by White and Black subjects. International Journal of Psychology, 10, 119-
123.

Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled
components. Journal of Personality and Social Psychology, 56, 5-18.

Diamond, R., & Carey, S. (1986). Why faces are and are not special: An effect of
expertise. Journal of Experimental Psychology: General, 115, 107-117.

Diener, E., & Emmons, R. A., (1984). The independence of positive and negative affect.
Journal of Personality and Social Psychology, 4 7, 1105-1 117.

Dixon, T. L., & Maddox, K. B. (2005). Skin tone, crime news, and social reality
judgments: Priming the stereotype of the dark and dangerous Black criminal. Journal of
Applied Social Psychology, 35, 1555-1570.

Dowden, S., & Robinson, J. P. (1993). Age and cohort differences in American racial
attitudes: The generational replacement hypothesis revisited. In P. M. Sniderman, P. E.
Tetlock, & E. G. Carmines (Eds.), Prejudice, politics, and the American dilemma (pp.
86—103). Stanford, CA: Stanford University Press.

Downey, G., & Feldman, S. (1996). Implications of rejection sensitivity for intimate
relationships. Journal of Personality and Social Psychology, 70, 1327—1343.

Dressler, W. W. (1991). Social class, skin color, and arterial blood pressure in two
societies. Ethnicity & Disease, 1, 60-77.

Eberhardt, J. L., Davies, P. G., Purdie-Vaughns, V. J ., & Johnson, S. L. (2006). Looking

deathworthy: Perceived stereotypicality of Black defendants predicts capital-sentencing
outcomes. Psychological Science, I 7, 383-386.

202

Eberhardt, J. L., Goff, P. A., Purdie, V. J ., & Davies, P. G. (2004). Seeing Black: Race,
crime, and visual processing. Journal of Personality and Social Psychology, 8 7, 876-893.

Fazio, R. 11., Jackson, J. R., Dunton, B. C., & Williams, C. J. (1995). Variability in
automatic activation as an unobtrusive measure of racial attitudes: A bona fide pipeline?
Journal of Personality and Social Psychologv, 69, 1013-1027.

Fiske, S. T., Lin, M. H., & Neuberg, S. L. (1999). The Continuum Model: Ten years later. In
S. Chaiken & Y. Trope (Eds) Dual process theories in social psychology (pp. 231-254). New
York: Guilford.

Fleming, J. (1983). Black women in Black and White college environments: The making
of a matriarch. Journal of Social Issues, 39, 41 -54.

Frazier, E. F. (1966). The Negro Family in the United States. Chicago: University of
Chicago Press.

Gaertner, S. L., & Dovidio, J. F. (1986). The aversive form of racism. In J. F. Dovidio &
S. L. Gaertner (Eds), Prejudice, discrimination, and racism (pp. 61-89). New York:
Academic Press.

Gallagher, 8., Tait, T. J ., McCologan, B., Dovey, T. M., & Halford, J. C. G. (2003).
Negative judgments about male associates of obese females. Obesity Research, 11, Al 18-
A119.

Goffinan, E. (1963). Stigma: Notes on the management of spoiled identity. Englewood
Cliffs, NJ: Prentice-Hall.

Goldstein, S. B., & Johnson, V. A. (1997). Stigma by association: Perceptions of the
dating partners of college students with physical disabilities. Basic and Applied Social
Psychology, 19, 495-504.

Goldstein, M. D., & Strube, M. J. (1994). Independence revisited: The relation between
positive and negative affect in a naturalistic setting. Personality and Social Psychology
Bulletin, 20, 57-64.

Goshen-Gottstein, Y., & Ganel, T. (2000). Repetition priming for familiar and unfamiliar
faces in a sex-judgment task: Evidence for a common route for the processing of sex and

identity. Journal of Experimental Psychology: Learning, Memory, and Cognition, 26,
1198-1214.

Gossett, Thomas F. (1963). Race: The History of an Idea in America. Dallas: Southern
Methodist University Press.

Greenwald, A.G., & Banaji, MR. (1995). Implicit social cognition: Attitudes, self-esteem,
and stereotypes. Psychological Review, 102, 4—27.

203

Haley, H., Sidanius, J ., Lowery, B., & Neil, M. (2004). The interactive nature of sex and
race discrimination: A social dominance perspective. In G. Philogene (Ed.), Racial
identity in context: The legacy of Kenneth B. Clark. Decade of behavior (pp. 149-160).
Washington, DC: American Psychological Association.

Hall, R. E. (1992). Bias among African-Americans regarding skin color: Implications for
social work practice. Research on Social Work Practice, 2, 479-486.

Hall, R. E. (2003). Skin color as post-colonial hierarchy: A global strategy for conflict
resolution. The Journal of Psychology, 13 7, 41 -53.

Hall, R. E. (2005). From the psychology of race to the issues of skin color for people of
African descent. Journal of Applied Social Psychology, 35, 1958-1967.

Hamm, N. H., Baum, M. R., & Nikels, K. W. (1975). Effects of race and exposure on
judgments of interpersonal favorability. Journal of Experimental Social Psychology, 11,
14-24.

Harburg, E. Gleibermann, L. Roeper, P., Schork, M. A., & Schull, W. J. (1978). Sking
color, ethnicity, and blood pressure I: Detroit Blacks. American Journal of Public Health,
68, 1177-1183.

Harvey, R. D., LaBeach, N., Pridgen, E., & Gocial, T. M. (2005). The intragroup
stigmatization of skin tone among Black Americans. Journal of Black Psychology, 31,
237-253.

Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale for
measuring state self-esteem. Journal of Personality and Social Psychology, 60, 895-910.

Hebl, M. R., & Mannix, L. M. (2003). The weight of obesity in evaluating others: A mere
proximity effect. Personality and Social Psychology Bulletin, 29, 28-38.

Hernandez, T. K. (2006). Bringing Clarity to Race Relations in Brazil. Diverse: Issues in
Higher Education, 23, 85.

Hill, M. E. (2000). Color differences in the socioeconomic status of Afi'ican American
men: Results of a longitudinal study. Social Forces, 78, 1437-1460.

Hill, M. E. (2002). Skin color and the perception of attractiveness among African
Americans: Does gender make a difference? Social Psychology Quarterly, 65, 77-91.

Hill, T., Lewicki, P., Czyzewska, M., & Schuller, G. (1990). The role of learned

inferential encoding rules in the perception of faces: Effects of nonconscious self-
perpetuation of a bias. Journal of Experimental Social Psychology, 26, 350-371.

204

Hochschild, J. L. (2007). The Skin Color Paradox and the American Racial Order. Social
Forces, 86, 643-670.

Hughes, M., & Hertel, B. R. (1990). The significance of color remains: A study of life
chances, mate selection, and ethnic consciousness among Black Americans. Social
Forces, 68, 1 105-1120.

Hunter, M. L. (2002). “If you’re light you’re alright”: Light skin color as social capital
for women of color. Gender and Society, 16, 175-193.

Jetten, J ., Branscombe, N.R., Schmitt, M.T., & Spears, R. (2001). Rebels with a cause:
Group identification as a response to perceived discrimination from the mainstream.
Personality and Social Psychology Bulletin, 27, 1204-1213.

Kahn, K. B., Davies, P. G., Eberhardt, J. L., & Correll, J. (2008, February). Not all Black
men are treated equal: Stereotypicality and shooter bias [Poster]. Poster presented at the
annual conference of Society for Personality and Social Psychology, Albuquerque, NM.

Keith, V. M., & Herring, C. (1991). Skin tone and stratification in the Black community.
The American Journal of Sociology, 97, 760-778.

King, K. R. (2005). Why is discrimination stressful? The mediating role of cognitive
appraisal. Cultural Diversity and Ethnic Minority Psychology, 11, 202-212.

Klonoff, E. A., & Landrine, H. (1999). Cross-validation of the Schedule of Racist Events.
Journal of Black Psychology, 25, 231-254.

Klonoff, E. A., & Landrine, H. (2000). Is skin color a marker for racial discrimination?

Explaining the skin color-hypertension relationship. Journal of Behavioral Medicine, 23,
329-338.

Kunda, Z. (1999). Social Cognition: Making sense of people. Cambridge, MA: The MIT
Press.

Landrine, H., & Klonoff, E. A. (1996). The Schedule of Racist Events: A measure of
racial discrimination and a study of its negative physical and mental health consequences.
Journal of Black Psychology, 22, 144-168.

Landrine, H., Klonoff, E. A., Alcaraz, R., Scott, J ., & Wilkins, P. (1995). Multiple
variables in discrimination. In B. Lott & D. Maluso (Eds), The social psychology of
interpersonal discrimination (pp. 183-224). New York: NY: Guilford Press.

Leder, H., & Bruce, V. (2000). When inverted faces are recognized: The role of

configural information in face recognition. Quarterly Journal of Experimental
Psychology, 53A, 513-536.

205

Livingston, R. W., & Brewer, M. B. (2002). What are we really priming? Cue-based
versus Category-based processing of facial stimuli. Journal of Personality and Social
Psychology, 82, 5-18.

Luhtanen, R., & Crocker, J. (1992). A collective self-esteem scale: Self-evaluation of
one’s social identity. Personality and Social Psychology Bulletin, 18, 302-318.

Macrae, C. N., & Martin, D. (2007). A boy primed sue: Feature-based processing and
person construal. European Journal of Social Psychology, Special Issues: F amiliarity
impacts person perception, 3 7, 793-805.

Maddox, K. B. (2004). Perspectives on racial phenotypicality bias. Personality and
Social Psychology Review, 8, 383-401.

Maddox, K. B., & Chase, S. G. (2004). Manipulating subcategory salience: Exploring the
link between skin tone and social perception of Blacks. European Journal of Social
Psychology, 34, 533-546.

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.

Major, B., & O’Brien, LT. (2005). The social psychology of stigma. Annual Review of
Psychology, 56, 393-421.

Major, B., Quinton, W.J., & McCoy, SK. (2002). Antecedents and consequences of
attributions to discrimination: Theoretical and empirical advances. The Advances in
Experimental Social Psychology, 34, 251-330.

Major, B., Quinton, W.J., & Schmader, T. (2003). Attributions to discrimination and self
esteem: Impact of group identification and situational ambiguity. Journal of
Experimental Social Psychology, 39, 220-231.

Martin, D., & Macrae, C. N. (2007). A face with a cue: Exploring the inevitability of
person categorization. European Journal of Social Psychology, Special Issues:
Familiarity impacts person perception, 3 7, 806-816.

Mason, M. F ., Cloutier, J ., & Macrae, C. N. (2006). On construing others: Category and
stereotype activation from facial cues. Social Cognition, 24. 540-562.

McLeod, J. D., & Owens, T. J. (2004). Psychological well-being in the early life course:

variations by socioeconomic status, gender, and race/ethnicity. Social Psychology
Quarterly, 67, 257-278.

206

Meier, B. P., Robinson, M. D., & Clore, G. L. (2004). Why good guys wear White:
Automatic inferences about stimulus valence based on brightness. Psychological Science,
15, 82-87.

Mendoza-Demon, R., Downey, G., Purdie, V., Davis, A., & Pietrazk, J. (2002).
Sensitivity to race-based rejection: Implications for Afiican-American students' college
experience. Journal of Personality and Social Psychology, 83, 896-918.

Minear, M. & Park, D.C.(2004). A lifespan database of adult facial stimuli. Behavior
Research Methods, Instruments, & Computers. 36, 630-633.
https://pal.utdallasedu/facedb/request/login

Montalvo, F. F. (2004). Surviving race: Skin color and the socialization and acculturation
of Latinas. Journal of Ethnic & Cultural Diversity in Social Work, 13, 25-43.

Monteith, M. J ., Sherman, J. W., & Devince, P. G. (1998). Suppression as a stereotype
control strategy. Personality and Social Psychology Review, 2, 63-82.

Morales, M. C. (2008). The ethnic niche as an economic pathway for the dark skinned:
Labor market incorporation of Latina/o workers. Hispanic Journal of Behavioral
Sciences, 30, 280-298.

Mukhtar, A. (2007). The crisis of identity in northern Sudan: The dilemma of a Black
people with a White culture. In Carolyn Fluehr-Lobban and Kharyssa Rhodes (Eds),
Race and Identity in the Nile Valley (pp. 213—24). Trenton, NJ: Red Sea Press.

Nisbett, R.E., & Wilson, TD. (1977). Telling more than we can know: Verbal reports on
mental processes. Psychological Review, 84, 231—259.

Neuberg, S. L., Smith, D. M., Hoffman, J. C., & Russel, F. J. (1994). When we observe
stigmatized and “normal” individuals interacting: Stigma by association. Personality and
Social Psychology Bulletin, 20, 196-209.

Oliver, M. B., Jackson 11, R. L., Moses, N. N., & Dangerfield, C. L. (2004). The face of
crime: Viewers’ memory of race-related facial features of individuals pictured in the
news. Journal of Communication, 54, 88-104.

Operario, D., & Fiske, ST. (2001). Ethnic identity moderates perceptions of prejudice:
Judgments of personal versus group discrimination and subtle versus blatant bias.
Personality and Social Psychology Bulletin, 2 7, 550-561.

Ostman, M., & Kjellin, L. (2002). Stigma by association: Psychological factors in
relatives of people with mental illness. The British Journal of Psychiatry, 181, 494-498.

207

 

Otten, S., & Wentura, D. (1999). About the impact of automaticity in the minimal group
paradigm: Evidence from affective priming tasks. European Journal of Social
Psychology, 29, 1049- 1 071.

Penny, H., & Haddock, G. (2007). Anti-fat prejudice among children: The “mere
proximity” effect in 5-10 year olds. Journal of Experimental Social Psychology, 43, 678-
683.

Perdue, C. W., Dovidio, J. F ., Gurtman, M. B., & Tyler, R. B. (1990). Us and them:
Social categorization and the process of intergroup bias. Journal of Personality and
Social Psychology, 59, 475-486.

Peskin, M., & Newell, F. N. (2004). Familiarity breeds attraction: Effect of exposure on
the attractiveness of typical and distinctive faces. Perception, 33, 147-157.

Porter, C. P. (1991). Social reasons for skin tone preferences of Black school-age
children. American Journal of Orthopsychiatry, 61 , 149-154.

Pratto, F., Sidanius, J ., & Levin, S. (2006). Social dominance theory and the dynamics of
intergroup relations: Taking stock and looking forward. European Journal of Social
Psychology, 1 7, 271-320.

Rederstorff, J. C., Buchanan, N. T., & Settles, 1. H. (2007). The moderating roles of race
and gender-role attitudes in the relationship between sexual harassment and
psychological well-being. Psychology of Women Quarterly, 31, 50-61.

Rhodes, G., Halberstadt, J ., & Brajkovich, G. (2001). Generalization of mere exposure
effects to averaged composite faces. Social Cognition, 19, 57-70.

Rokeach, M., & Ball-Rokeach, S. J. (1989). Stability and change in American value
priorities, 1968—1981. American Psychologist, 44, 775—784.

Rondilla, J. L., & Spickard, P. R. (2007). Is lighter better?: Skin-tone discrimination
among Asian Americans. Lanham, MD.: Rowman & Littlefield Pub Inc.

Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton
University Press.

Ross, L. E. (1997). Mate selection preferences among African American college students.
Journal of Black Studies, 27, 554-569.

Russell, Kathy, Midge Wilson, and Ronald Hall. (1992). The Color Complex: The
Politics of Skin Color Among African Americans. New York: Anchor.

Sellers, R. M., & Shelton, J. N. (2003). The role of racial identity in perceived racial
discrimination. Journal of Personality and Social Psychology, 84, 1079-1092.

208

Settles, I. H. (2006). Use of an intersectional framework to understand Black women’s
racial and gender identities. Sex Roles, 54, 589-601.

Shelton, J .N., & Richeson, J.A. (2006). Interracial interactions: A relational approach. In
M.P. Zanna (Ed.), Advances in Experimental Social Psychology (pp. 121-181). New
York, NY: Academic Press.

Shelton, J .N., & Sellers, RM. (2000). Situational stability and variability in Afiican
American racial identity. Journal of Black Psychology, 26, 27-50.

Sidanius, J ., Levin, S., Liu, J ., & Pratto, F. (2000). Social dominance orientation, anti-
egalitarianism and the political psychology of gender: An extension and cross-cultural
replication. European Journal of Social Psychology, 30, 41-67.

Sidanius, J ., & Pratto, F. (1999). Social dominance: An intergroup theory of social
hierarchy and oppression. New York: Cambridge University Press.

Sorce, J. F. (1979). The role of physiognomy in the development of racial awareness.
Journal of Genetic Psychology, 134, 33-41.

Staples, B. (1986). Black men and public spaces. Retrieved from
http://www.ghc.edu/faculty/llerych/BlackMen.pdf on April 20, 2010.

Sweet, E., McDade, T. W., Kiefe, C. 1., & Liu, K. (2007). Relationships between skin
color, income, and blood, pressure among Afiican Americans in the CARDIA study.
American Journal of Public Health, 97, 2253-2259.

Swim, J. K., Ferguson, N. J ., & Hyers, L. L. (1999). Avoiding stigma by association:
Subtle prejudice against lesbians in the form of social distancing. Basic and Applied
Social Psychology, 21, 61-68.

Swim, J .K., & Stangor, C. (1998). Prejudice: The target ’5 perspective. San Diego:
Academic Press.

Tajfel, H., Billing, M., Bundy, R., & Flament, C. (1971). Social categorization and
intergroup behavior. European Journal of Social Psychology, 1, 149-178.

Thompson, M. S., & Keith, V. M. (2001). The blacker the berry: Gender, skin tone, self-
esteem, and self-efficacy. Gender & Society, 15, 336-357.

Uhlmann, E., Dasgupta, N., Elgueta, A., Greenwald, A. G., & Swanson, J. (2002).
Subgroup prejudice based on skin color among Hispanics in the United States and Latin
America. Social Cognition, 20, 198-225.

Valentine, T. (1988). Upside-down faces: A review of the effect of inversion upon face
recognition. British Journal of Psychology, 79, 471-491.

209

Valentine, T. (1991). A unified account of the effects of distinctiveness, inversion, and
race in face recognition. Quarterly Journal of Experimental Psychology, 43A, 161-204.

Wade, T. J. (1996). The relationships between skin color and self-perceived global,
physical, and sexual attractiveness, and self-esteem for African Americans. Journal of
Black Psychologv, 22, 358-373.

Wade, T. J ., & Bielitz, S. (2005). The differential effect of skin color on attractiveness,
personality evaluation, and perceived life success of Afiican Americans. Journal of Black
Psychology, 31, 215-236.

Wade, T. J., Romano, M. J., & Blue, L. (2004). The effect of African American skin
color on hiring preferences. Journal of Applied Social Psychology, 34, 2550-2558.

Warr, P. B., Barter, J ., & Brownbridge, G. (1983). On the independence of positive and
negative affect. Journal of Personality and Social Psychology, 44, 644-651.

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief
measures of positive and negative affect: The PANAS scale. Journal of Personality and
Social Psychologv, 54, 1063-1070.

Wittenbrink, B., Judd, C. M., & Park, B. (1997). Evidence for racial prejudice at the
implicit level and its relationship with questionnaire measures. Journal of Personality
and Social Psychology, 72, 262-274.

Wyer, N. A., Sherman, J. W., & Stroessner, S. J. (1998). The spontaneous suppression of
racial stereotype. Social Cognition, 16, 340-352.

Wyer, N. A., Sherman, J. W., & Stroessner, S. J. (2000). The role of motivation and

ability in controlling the consequences of stereotype supression. Personality and Social
Psychology Bulletin, 26, 13-25.

Zebrowitz, L. A., Bronstad, P. M., & Lee, H. K. (2007). The contribution of face
familiarity to ingroup favoritism and stereotyping. Social Cognition, 25, 306-338.

210

Y LIB

TATEUNNERsn RARms
| l lllllllllllllllllllllll
3 5621

 

”11111111

1293 030