USF Libraries
USF Digital Collections

Perceptions of weight status

MISSING IMAGE

Material Information

Title:
Perceptions of weight status the effects of target features (fatmuscularity level, gender, ethnicity) and rater features (ethnicity and gender)
Physical Description:
Book
Language:
English
Creator:
Yanover, Tovah
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Weight bias
Social norms
Gender
Ethnicity
BMI
Dissertations, Academic -- Psychology -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Previous research has explored self-perception of weight and has established that women tend to overestimate their own weight while men tend to underestimate. New research has also begun to examine parental perceptions of their children's weight and has indicated that parents tend to be fairly inaccurate, particularly when it comes to recognizing overweight in their own children. No research has focused on the way in which we perceive the weight of the many other individuals we encounter on a daily basis. The present study was designed to investigate the way in which the weight of others is rated and the factors that affect the way in which these ratings are made. Undergraduate male (N = 140) and female (N = 193) students viewed a series of slides depicting male and female figures of varying levels of muscularity and adiposity. The race of the figures was also varied. Each figure was presented once in each racial category (Caucasian, Hispanic, and African American).Participants then filled out questionnaires assessing potential covariates: trait levels of body dissatisfaction, thin-ideal internalization, muscularity dissatisfaction, proximate social norms, appearance comparison, and social desirability. BMI was calculated from self-reported height and weight. The effects of target race, rater race, and rater gender on ratings were examined. Results indicated that the race of the figure affected the ratings given to the figure, though consistent patterns of influence were not identified. Males consistently rated the weight of the figures higher than females and African American raters consistently assigned lower weight ratings than did Caucasian raters. The analyses failed to identify consistent covariates of these effects. Results also provided tentative support for the hypothesis that, given two figures equal in adiposity, raters will provide a lower weight rating to the figure with more muscularity.Exploratory analyses also examined health and attractiveness ratings. The findings are discussed in the context of research on self-perception and the way in which the trends in perception of others differ from the trends seen in self-perception. Study limitations are discussed and possibilities for future research are offered.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2009.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Tovah Yanover.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 275 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002063945
oclc - 558713309
usfldc doi - E14-SFE0003029
usfldc handle - e14.3029
System ID:
SFS0027346:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200397Ka 4500
controlfield tag 001 002063945
005 20100318121939.0
007 cr mnu|||uuuuu
008 100318s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003029
035
(OCoLC)558713309
040
FHM
c FHM
049
FHMM
090
BF121 (Online)
1 100
Yanover, Tovah.
0 245
Perceptions of weight status :
b the effects of target features (fat/muscularity level, gender, ethnicity) and rater features (ethnicity and gender)
h [electronic resource] /
by Tovah Yanover.
260
[Tampa, Fla] :
University of South Florida,
2009.
500
Title from PDF of title page.
Document formatted into pages; contains 275 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2009.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
520
ABSTRACT: Previous research has explored self-perception of weight and has established that women tend to overestimate their own weight while men tend to underestimate. New research has also begun to examine parental perceptions of their children's weight and has indicated that parents tend to be fairly inaccurate, particularly when it comes to recognizing overweight in their own children. No research has focused on the way in which we perceive the weight of the many other individuals we encounter on a daily basis. The present study was designed to investigate the way in which the weight of others is rated and the factors that affect the way in which these ratings are made. Undergraduate male (N = 140) and female (N = 193) students viewed a series of slides depicting male and female figures of varying levels of muscularity and adiposity. The race of the figures was also varied. Each figure was presented once in each racial category (Caucasian, Hispanic, and African American).Participants then filled out questionnaires assessing potential covariates: trait levels of body dissatisfaction, thin-ideal internalization, muscularity dissatisfaction, proximate social norms, appearance comparison, and social desirability. BMI was calculated from self-reported height and weight. The effects of target race, rater race, and rater gender on ratings were examined. Results indicated that the race of the figure affected the ratings given to the figure, though consistent patterns of influence were not identified. Males consistently rated the weight of the figures higher than females and African American raters consistently assigned lower weight ratings than did Caucasian raters. The analyses failed to identify consistent covariates of these effects. Results also provided tentative support for the hypothesis that, given two figures equal in adiposity, raters will provide a lower weight rating to the figure with more muscularity.Exploratory analyses also examined health and attractiveness ratings. The findings are discussed in the context of research on self-perception and the way in which the trends in perception of others differ from the trends seen in self-perception. Study limitations are discussed and possibilities for future research are offered.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: J. Kevin Thompson, Ph.D.
653
Weight bias
Social norms
Gender
Ethnicity
BMI
690
Dissertations, Academic
z USF
x Psychology
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.3029



PAGE 1

Perceptions of Weight Status: The Effects of Target Features (Fat/Muscularity Level, Gender, Ethnicity) and Rater Features (Ethnicity and Gender) by Tovah Yanover A dissertation submitted in partial fulfillment Of the requirements for the degr ee of Doctor of Philosophy Department of Psychology College of Arts and Sciences University of South Florida Major Professor: J. Kevin Thompson, Ph.D. Jamie Lyn Goldenberg, Ph.D. Vicky Phares, Ph.D. Jon Rottenberg, Ph.D. Joe Vandello, Ph.D. Date of Ap proval: May 8 2008. Keywords: weight bias, social norms, gender, ethnicity, BMI Copyright 2009, Tovah Yanover

PAGE 2

Dedication I dedicate this dissertation to my family: to my mother and father who have supported me tirelessly and always encouraged me to value learning and education, and to my husband who is a constant source of encouragement and comfort.

PAGE 3

Acknowledgements I would like to thank Dr. J. Kevin Thompson for his time, effort, and patience in supervising this project as well as for the opportu nity to explore an area of interest to me. I thank him also for his gracious acceptance of a me as a new member of his research team. I would also like to thank my committee, Dr. Jamie Goldenberg, Dr. Vicky Phares, Dr. Jon Rottenberg, Dr. Joe Vandello, and Dr. Brent Small as chairman, for giving their time and for offering their valuable insights. Thank you to the entire Body Image Research Group for welcoming me into the lab and for offering their valuable time and input in the preparation of this research

PAGE 4

Note to Reader: In its original form, Appendix A contains colored versions of the stimuli used in the research described herein. The stimuli are colored in "light cool brown" in Adobe Photoshop. The color is not integral to the understanding of the resea rch but the interested reader is directed to the original dissertation on file in the library of the University of South Florida in Tampa, Florida.

PAGE 5

ii Table of Contents List of Tables ................................ ................................ ................................ ................ vi Abstract ................................ ................................ ................................ ...................... viii Introduction ................................ ................................ ................................ ................... 1 Self Perception Research ................................ ................................ ............................. 3 Weight Categorization of "Others" ................................ ................................ .............. 4 Psychological Covariates: Body Shape and Size Dissatisfaction, Mu scularity Dissatisfaction, Appearance Comparison, Social Norms, and Internalization of Appearance Ideals ................................ ................................ ................................ ....... 7 Body Dissatisfaction ................................ ................................ ................................ 8 Appearance Comparison ................................ ................................ ........................ 10 Social Norm s ................................ ................................ ................................ ......... 11 Overview of Current Study ................................ ................................ ........................ 13 Hypotheses ................................ ................................ ................................ ............ 14 Exploratory Questions ................................ ................................ ........................... 14 Pilot Study ................................ ................................ ................................ ................. 14 Method ................................ ................................ ................................ .................. 15 Participants ................................ ................................ ................................ ........ 15 Measures ................................ ................................ ................................ ........... 15 Stimuli (Target Figures) ................................ ................................ ................. 15 Figure Ratings ................................ ................................ ............................... 16 Proximate Social Environment Rating Scal e ................................ .................. 16 Demographic Information ................................ ................................ .............. 17 Focus Groups ................................ ................................ ................................ 17 Procedure ................................ ................................ ................................ ........... 17 Results ................................ ................................ ................................ ................... 17

PAGE 6

iii Stimuli ................................ ................................ ................................ ............... 17 Proximate Social Environment Rating Scale ................................ ...................... 19 Focus Groups ................................ ................................ ................................ ..... 19 Summary of Changes ................................ ................................ ............................. 21 Method ................................ ................................ ................................ ......................... 22 Participants ................................ ................................ ................................ ................ 22 Measures ................................ ................................ ................................ ................... 23 Stimuli and Stimulus Rating Form ................................ ................................ ......... 23 Distraction Task ................................ ................................ ................................ .... 23 Appearance Comparison ................................ ................................ ........................ 24 Social Desirability ................................ ................................ ................................ 24 Social Norms Measures ................................ ................................ ......................... 24 Body Shape and Size Dissatisfaction ................................ ................................ ..... 25 Muscularity Dissatisfaction ................................ ................................ .................... 25 Demographic Information ................................ ................................ ...................... 25 Procedure ................................ ................................ ................................ .................. 26 Data Analysis ................................ ................................ ................................ ............ 26 Preliminary Analyses ................................ ................................ ............................. 26 Weight Ratings ................................ ................................ ................................ ...... 27 Adiposity and Muscularity ................................ ................................ ..................... 28 Additional Analyses ................................ ................................ .............................. 28 Results ................................ ................................ ................................ .......................... 29 Descriptive Information ................................ ................................ ............................. 29 Preliminary Analyses ................................ ................................ ................................ 30 Initial Weight Analyses ................................ ................................ ............................. 34 Female Figures ................................ ................................ ................................ ...... 34 Male Figures ................................ ................................ ................................ .......... 37 Follow Up Weight Analyses ................................ ................................ ...................... 40 Repeated Measures Effects: Three Way Interactions ................................ ............. 40 Repeated Meas ures Effects: Two Way Interactions ................................ ............... 41

PAGE 7

iv Repeated Measures Effects: Main Effects of Target Race Female Stimuli ........... 41 Repeated Measures Effects: Main Effects of Target Race Male Stimuli .............. 42 Between Subjects Effects: Rater Gender X Rater Race Interaction ........................ 43 Between Subjects Effects: Main Effect of Rater Gender Female Stimuli ............. 44 Between Subjects Effects: Main Effe ct of Rater Gender Male Stimuli ................ 44 Between Subjects Effects: Main Effect of Rater Race Female Stimuli ................ 45 Between Subjects Effects: Main Effect of Rater Race Male Stimuli .................... 46 Between Subjects Effects: Significant Covariates ................................ .................. 46 Adiposity and Muscularity Analyses ................................ ................................ ......... 47 Additional Analyses: Health and Attractiveness Data ................................ ................ 53 Discussi on ................................ ................................ ................................ .................... 54 Weight Analyses ................................ ................................ ................................ ....... 54 Adiposity and Muscularity ................................ ................................ ......................... 56 Covariate Analyses ................................ ................................ ................................ .... 56 Health and Attractiveness Analyses ................................ ................................ ........... 64 Limitations ................................ ................................ ................................ ................ 65 Future Research ................................ ................................ ................................ ......... 66 References ................................ ................................ ................................ .................... 68 Appendices ................................ ................................ ................................ ................... 79 Appendix A: Pilot Materials ................................ ................................ ...................... 80 Appen dix B: Instructions to Participants ................................ ................................ .... 86 Appendix C: Target Rating Items ................................ ................................ .............. 87 Appendix D: Distraction Task ................................ ................................ .................. 88 Appendix E: Marlowe Crown Social Desirability Scale ................................ ............. 89 Appendix F: Physical Appearance Comparison Scale ................................ ................ 91 Appendix G: Sociocultural Attitudes Towards Appearance Questionnaire 3 Internalization General subscale ................................ ................................ ............. 92 Appendi x H: Proximate Social Environment Rating Scale ................................ ......... 93 Appendix I: Eating Disorder Inventory 3 Body Dissatisfaction Subscale ................. 96 Appendix J: Drive for Muscularity Scale ................................ ................................ ... 97

PAGE 8

v Appendix K: Demographic Information ................................ ................................ ..... 98 Appendix L: Means and Standard Deviations, F p and 2 values for Weight Analyses ................................ ................................ ................................ ....... 99 Appendix M: Significant F p and 2 values with Means and Standard Deviations for Health Analyses ................................ ................................ ............... 239 Appendix N: Significant F p and 2 values with Means and Standard Deviations for Attractiveness Analyses ................................ ................................ .... 249 Appendix O: Additional Analyses Results ................................ ............................... 264 Health Analyses: Three Way Interactions ................................ ............................ 264 Health Analyses: Covariate X Target Race Interactions ................................ ....... 26 4 Health Analyses: Target Race X Rater Gender Interactions ................................ .. 264 Health Analyses: Target Race X Rater Race Interactions ................................ ..... 264 Health Analyses: Main Effects of Target Race Female Stimuli ......................... 265 Health Analyses: Main Effects of Target Race M ale Stimuli ............................. 266 Health Analyses: Between Subjects Interactions ................................ .................. 267 Health Analyses: Main Effects of Rater Gender ................................ ................... 267 Health Analyses: Main Effects of Rater Race ................................ ....................... 268 Health Analyses: Significant Covariates ................................ .............................. 269 Attractiveness Analyses: Three Way Interactions ................................ ................ 269 Attractiveness Analyses: Covariate X Target Race Interactions ........................... 269 Attractiveness Analyses: Target Race X Rater Gender Interactions ...................... 269 Attractiveness Analyses: Target Race X Rater Race Interactions ......................... 270 Attractiveness Analyses: Main Effec ts of Target Race Female Stimuli .............. 270 Attractiveness Analyses: Main Effects of Target Race Male Stimuli ................. 272 Attractiveness Analyses: Between Subjects Interactions ................................ ...... 272 Attractiveness Analyses: Main Effects of Rater Gender ................................ ....... 272 Attractiveness Analyses: Main Effects of Rater Race ................................ ........... 273 Attractiveness Analyses: Significant Covariates ................................ ................... 274 About The Author ................................ ................................ ........................... End Page

PAGE 9

vi List of Tables Table 1. Means for health, weight, and attractiveness of target figures ........................... 18 Table 2. Demographic characteristics. ................................ ................................ ........... 22 Table 3. Means and standard deviations for covariates ................................ ................... 29 Table 4. Correlations among covariates ................................ ................................ ......... 30 Table 5. Means, standard errors, F p and partial 2 values for gender effects ............... 31 Table 6. Means, standard errors, F p and partial 2 values for race effects ................... 32 Table 7. Means, standard errors, F p and partial 2 values for interaction effects ......... 33 Table 8. Correlations between mean target weight ratings and MCSDS scores .............. 34 Table 9. Means and standard deviations of weight ratings for pairs of figures ................ 47 Table 10. Paired sample t tests ................................ ................................ ...................... 50 Table 11. Effects of covariate variables on significance ................................ ................. 59 Table L.1. Repeated measures effects: Three way interactions (target race X rater race X rater gender) ................................ ................................ ....... 100 Table L.2. Repeated measures effects: Two way interactions (target race X covariate) ................................ ................................ .............................. 136 Table L.3. Repeated measures effects: Two way interactions (target race X rater gender) ................................ ................................ .......................... 142 Table L.4. Repeated mea sures effects: Two way interactions (target race X rater race) ................................ ................................ .............................. 164 Table L.5. Repeated measures effects: Main effects of target race ............................... 186 Table L.6. Between subjects effects: Rater gender X rater race interaction .................. 194 Table L.7. Between subjects effects: Main effects of rater gender ................................ 215 Table L.8. Between subjects effects: Main effects of rater race ................................ .... 224 Table L .9. Between subjects effects: Covariate effects ................................ ................ 232

PAGE 10

vii Table M.1. Repeated measures effects: Significant two way interactions (target race X covariate) ................................ ................................ .............................. 239 Table M.2. Repeated measures effects: Sign ificant two way interactions (target race X rater gender) ................................ ................................ .......................... 240 Table M.3. Repeated measures effects: Significant two way interactions (target race X rater race) ................................ ................................ .............................. 241 Table M.4. Repeated measures e ffects: Significant main effects of target race ............. 243 Table M.5. Between subjects effects: Significant main effects of rater gender ............. 245 Table M.6. Between subjects effects: Significant main effects of rater race ................. 246 Table M.7. Between subjects effects: Significant covariate effects .............................. 248 Table N.1. Repeated measures effects: Significant two way interactions (target race X covariate) ................................ ................................ .............................. 249 Table N.2. Repeated measures effects: Significant two way interactions (target race X rater gender) ................................ ................................ .......................... 250 Table N.3. Repeated measures effects: Significant two way interactions (target race X rater rac e) ................................ ................................ .............................. 252 Table N.4. Repeated measures effects: Significant main effects of target race .............. 256 Table N.5. Between subjects effects: Significant main effects of rater gender .............. 258 Table N.6. Between subjects effects: Significant main effects of rater race .................. 260 Table N.7. Between subjects effects: Significant covariate effects ............................... 262

PAGE 11

viii Perceptions of Weight Status: The Effe cts of Target Features (Fat/Muscularity Level, Gender, Ethnicity) and Rater Features (Ethnicity and Gender) Tovah Yanover ABSTRACT Previous research has explored self perception of weight and has established that women tend to overestimate their own weight while men tend to underestimate. New research has also begun to examine parental perceptions of their children's weight and has indicated that parents tend to be fairly inaccurate, particularly when it comes to recognizing overweight in their own children No research has focused on the way in which we perceive the weight of the many other individuals we encounter on a daily basis. The present study was designed to investigate the way in which the weight of others is rated and the factors that affect the w ay in which these ratings are made. Undergraduate male (N = 140) and female (N = 193) students viewed a series of slides depicting male and female figures of varying levels of muscularity and adiposity. The race of the figures was also varied. Each figure was presented once in each racial category (Caucasian, Hispanic, and African American). Participants then filled out questionnaires assessing potential covariates: trait levels of body dissatisfaction, thin ideal internalization, muscularity dissatisfactio n, proximate social norms, appearance comparison, and social desirability. BMI was calculated from self reported height and weight. The effects of target race, rater race, and rater gender on ratings were examined. Results indicated that the race of the fi gure affected the ratings given to the figure, though consistent patterns of influence were not identified. Males consistently rated the weight of the figures higher than females and African American raters consistently assigned lower weight ratings than d id Caucasian raters. The analyses failed to identify consistent

PAGE 12

ix covariates of these effects. Results also provided tentative support for the hypothesis that, given two figures equal in adiposity, raters will provide a lower weight rating to the figure with more muscularity. Exploratory analyses also examined health and attractiveness ratings. The findings are discussed in the context of research on self perception and the way in which the trends in perception of others differ from the trends seen in self pe rception. Study limitations are discussed and possibilities for future research are offered.

PAGE 13

1 Introduction In recent years, a number of variables have been examined as potential risk factors for eating disorders and obesity (Thompson, 2004; Thompson, Heinbe rg, Altabe, & Tantleff Dunn, 1999) including genetic influences, social factors (media images), and interpersonal experiences (peer and parental pressures). In addition, researchers have recently begun to explore and evaluate the potential importance of we ight status ratings, for one's own body or that of other individuals, as an important variable that may have treatment or preventive implications. The findings from this research indicate that individuals may have poor accuracy when it comes to estimating both their own weight and the weight of others, often assigning an average weight status to someone who, by prevailing objective standards (e.g., BMI), is overweight or obese. This study will make use of a novel design for the investigation of weight categ orization assignments made by individuals who differ in ethnicity and gender for a target image that varies on gender, ethnicity, and fat/muscularity body composition. First, a review of the general area of self perception research is offered to frame the current methodology, and then the emerging area of research on weight status categorization is reviewed. Potential covariates will then be examined, followed by an outline of the specific methodology. The study of weight categorization of others has severa l implications for the fields of eating disorders and obesity. For instance, if there exists a tendency to underestimate the weight of others, this will lead to inaccurate social comparisons with peers and others in one's environment. The majority of adult s are either overweight or obese, making an elevated weight status the norm. Therefore, overweight may be perceived as normal or average. One might feel satisfied with one's own weight compared to others based on inaccurate perceptions and therefore fail t o recognize a potential weight problem. If a weight problem goes unrecognized it will also likely go untreated. Second, inaccurate

PAGE 14

2 perceptions of the weight of others may, in fact, be an unidentified risk factor for the maintenance of weight related pathol ogy in specific subgroups. Obesity rates among African Americans are higher than among Caucasians (Racette, Deusinger, & Deusinger, 2003) but studies have shown that African Americans perceive themselves as normal weight or less than their actual weight wi th greater frequency than Caucasians (Bhuiyan, Gustat, Srinivasan, & Berenson, 2003; Desmond, Price, Hallinan, & Smith, 1989; Paeratakul, White, Williamson, Ryan, & Bray, 2002; Rand & Kuldau, 1990). African Americans also report fewer weight concerns than do Caucasians (Kemper, Sargent, Drane, Valois, & Hussey, 1994; Neumark Sztainer et al., 2002) and African American men report a preference for larger body size in African American women ( Greenberger & LaPorte, 1996; Rosen et al., 1993 ). Conceivably, when a n overweight African American female makes efforts to eat more healthfully or to engage in greater levels of physical activity, these efforts may go unsupported because the perception in the community is that she does not need to lose weight. Unsupported e fforts could, in the long, run, result in abandonment or failure of the efforts. Conversely, a Caucasian female whose objective status is underweight likely meets the thin ideal transmitte d in the media and, therefore, the perception may be that she is jus t right. Close friends or significant others might support her underweight status rather than encouraging appropriate weight gain efforts. Weight and eating related pathologies can range from extreme restriction of eating, excessive exercise, and binging and purging, to overeating and a lack of physical activity (Thompson, 2004). The eating disorders of anorexia nervosa and bulimia nervosa lie at one end of this continuum. Anorexia nervosa is characterized by intense fear of fatness, refusal to maintain a healthy weight, and distorted body image (American Psychiatric Association, 2000). Disturbance in body image is also a core feature of bulimia nervosa along with recurrent episodes of binging and vomiting or other compensatory behaviours such as excessive exercise, laxative and diuretic use, or fasting (American Psychiatric Association, 2000). Eating disorders represent a serious problem with sequelae including psychiatric comorbidity (Fichter & Quadflieg, 1999; Sullivan, Bulik, Carter, & Joyce, 1996), and high rates of morbidity and mortality (Reijonen, Pratt, Patel, & Greydanus, 2003).

PAGE 15

3 At the other extreme lies the problem of obesity. Obesity rates are staggering and they continue to climb. Defined as an excess of body fat (Dehghan, Akhtar Danesh & Mercha nt, 2005), obesity increases the risk for a multitude of health problems including diabetes mellitus, cardiovascular disease, and all cause mortality (National Heart, Lung and Blood Institute, 1998). A body mass index (BMI), a ratio of weight to height squ ared (Field, Barnoya, & Colditz, 2002), above 30 defines obesity (Devlin, Yanovski, & Wilson, 2000; Flegal, Carroll, Kuczmarski, & Johnson, 1998; National Heart, Lung, and Blood Institute, 1998) and a BMI between 25.0 and 29.9 defines overweight (Ogden et al., 2006) in adults. Children at or above the 95th percentile of BMI for age are labeled overweight and children who fall between the 85th and 95th percentiles of BMI for age are labeled at risk for overweight (Flegal, Wei, & Ogden, 2002; Himes & Dietz, 1 994). In the United States in 2003 2004, 34.8% of children aged 2 19 years were overweight or at risk for overweight and 64.5% of adults aged 20 years and up were overweight or obese (Ogden et al., 2006), making the problem of obesity and overweight a majo r public health problem. Self Perception Research A great deal of research has focused on the way in which adults and adolescents rate or categorize their own weight. These studies ask adolescents and adults to assign themselves to a weight or BMI category and then those category assignments are compared to the objective BMI of the participants. These studies have revealed consistent trends in adult and adolescent weight self perception. Females tend to overestimate their weight status and males tend to und erestimate their weight status (Chang & Christakis, 2001; Chang & Christakis, 2003; Gray, 1988; McCreary, 2002; Pritchard, King & Czajka Narins, 1994; Viner et al., 2006; Wardle & Johnson, 2002). Even some objectively underweight females place themselves i n the overweight category (Kaplan, Busner, & Pollack, 1988). The typical gender patterns of over and underestimation of weight status hold up cross culturally in Korean (Kim & Kim, 2001), Chinese (Xie et al., 2006), Bahraini (Al Sendi et al., 2004), and T aiwanese (Page, Lee, & Miao, 2005) adolescents. The finding that more women than men rate themselves as overweight holds up in 22 countries worldwide (Wardle, Haase, & Steptoe, 2006). Results among African

PAGE 16

4 American smokers (Lee et al., 2005) and community and agricultural worker Latino Americans (Hubert, Snider, & Winkelby, 2005) are consistent with trends from other studies. Most disconcerting is that patterns of reporting in pediatricians echo those seen in other adult studies (Perrin, Flower, & Ammerman, 2005). Interestingly, in a study of the association between religiosity and weight perception, Kim (2007) found that Jewish women were more likely to overestimate their weight than were women of other religious backgrounds. Studies of weight perception ha ve often relied on figure ratings rather than verbal categories. In these studies, participants are shown a series of line drawings of people that vary in adiposity and are asked to make a number of ratings, most frequently their current perceived body siz e and the body size that they consider ideal (Yanover & Thompson, 2009). These studies tend to reveal similar self perception trends in that women tend to choose an ideal figure that is much smaller than their current perceived body size (e.g., Barnett, Ke el, & Conoscenti, 2001; Safir, Flaisher Kellner, & Rosenmann, 2005) while men express a preference for a more muscular figure (e.g., Kowner, 2004; Olivardia, Pope, Borwiecki, & Cohane, 2004; Pope et al., 2000; Thompson & Cafri, 2007; Yang, Gray, & Pope, 20 05). Importantly, self ideal discrepancy is associated with higher levels of body dissatisfaction and eating disturbance (Heinberg, 1996; Thompson, 1990). Weight Categorization of "Others" A relatively new area in the field of weight perception involves th e investigation of classification accuracy when it comes to rating someone else. The term "others" is, of necessity and desirability, quite broad. Throughout daily life, many people are encountered including family, peers, strangers, and many others, and i t is very likely that judgments of these individuals, known or unfamiliar, are immediately made implicitly on multiple dimensions including weight. Research in this field is scant, though some intriguing research has examined the way in which parents rate their children's weight. The methodology of these studies is relatively straightforward. Parents are asked to rate whether their children are underweight, just right, or overweight and their ratings are compared with the children's objective weight status The research reveals a consistent pattern. Reliably, some parents are inaccurate. Importantly, the inaccuracy is

PAGE 17

5 generally such that they underestimate the weight of an overweight child (Akerman, Williams, & Meunier, 2007; Baughcum, Chamberlin, Deeks, Po wers, & Whitaker, 2000; Carnell, Edwards, Croker, Boniface, & Wardle, 2005; Etelson, Brand, Patrick, & Shirali, 2003; Fisher, Fraser, & Alexander, 2006; Jackson, Strauss, Lee, & Hunter, 1990; Jeffery, Voss, Metcalf, Alba, & Wilkin, 2006; Maynard, Galuska, Blanck, & Serdula, 2003; Wing, Epstein, & Neff, 1980). Two recent studies illustrate this phenomenon well. Carnell et al. (2005) recruited children between the ages of 3 and 5 years in state funded primary schools in London, England. Nearly three quarter s of their sample was Caucasian and almost 95% of parents who responded were mothers. They found that only 1.9% of overweight children and 17.1% of obese children were rated as overweight. No parent placed his or her child in the "very overweight" category Skelton, Busey, and Havens (2006) examined inner city African American children between the ages of 10 and 20 years. This study differed from many previous studies because it asked children to rate their own weight status in addition to having parents r ate the weight status of the children. Results showed that of the 52 overweight children in the sample, 67% felt that they were of normal weight and 77% felt that their weight was healthy, although the likelihood of perceiving one's weight as normal decrea sed with increasing objective weight status. Parental ratings of child weight and health mirrored those seen in the children. Of those parents whose child was overweight or at risk for overweight, 68% thought their child's weight was normal and 80% thought it was healthy. Furthermore, 28% of the parents in this study felt that being heavier was "good for your health." Research also suggests that parents of overweight children may not be worried about their child's current weight (Campbell, Williams, Hampton & Wake, 2006; Jain, Sherman, Chamberlin, Carter, Powers, & Whitaker, 2001; Wake, Salmon, Water, Wright, & Hesketh, 2002) or health status (Young Hyman, Herman, Scott, & Schlundt, 2000). In one study (Jain et al., 2006), low income African American mother s of preschool age children were fairly accurate at assessing child weight; ten out of 15 mothers of overweight children labeled their children as either a little or very overweight. However,

PAGE 18

6 of these mothers, only two were concerned about their child's cu rrent weight and five about their child's future weight. T o date, the findings on parental ratings of their children's weight are relatively consistent in indicating a lack of weight awareness. Importantly, research indicates strongly that o verweight chil dren are likely to become overweight adults (Reilly et al., 2003) therefore the lack of accurate recognition by parents of their children's weight status might conceivably lead them to deny their children's weight problem and perhaps, even deter them fro m seek ing treatment. To date, the area of research on weight categorization accuracy of others has been confined entirely to parents' ratings of their offspring. However, m any "others" are encountered on a daily basis The present study seeks to understan d how others' weight is rated and the factors that affect those ratings. The arguments presented at the start of this paper indicate that race and ethnicity may play a role in the view that we take of others. There are several factors that are important i n the way that others are rated. These factors fall into two groups: rater features and target features, where target refers to the individual to be rated. Rater features such as gender and ethnicity are likely to affect the ratings made of others. In self perception studies, men and women tend to have opposite biases in their ratings and men, in general, tend to exhibit less body dissatisfaction than women (Safir et al., 1995). Members of different racial and ethnic groups also tend to exhibit different pa tterns of self perception with African Americans, for example, indicating greater weight related satisfaction than Caucasians ( Parker, Nichter, Vuckovic, Sims, & Ritenbaugh, 1995 ). It is likely that the gender and ethnic patterns will also extend to rating s of others. For example, African Americans may allow greater latitude before placing a target into the overweight category than do Caucasians because African Americans tend to experience less body dissatisfaction than do Caucasians (Grabe & Hyde, 2006; Wi ldes, Emery, & Simons, 2001 ). Rater gender and race/ethnicity are also likely to interact. For instance, African American men may be more likely to underestimate the weight of overweight women due to their preferences for larger women (Greenberger & LaPort e, 1996; Rosen et al., 1993). Finally, rater BMI will also likely have an effect on ratings. Gray (1977)

PAGE 19

7 found that underweight individuals were more likely to overestimate their weight while overweight individuals were more likely to underestimate their w eight. In other words, those individuals at the extremes of the distribution were more likely to normalize their weight. Furthermore, it is likely that the features of the target to be rated will have an effect on the ratings. Gender and race/ethnicity of the target are key factors. A woman may be placed in a heavier weight category than a man of the same proportions because of the pressure for thinness placed on women in our society. African American and Hispanic female targets may receive ratings indicat ing that they are heavier than Caucasian female figures of the same objective size. Muscularity and body fat of the target are also likely to play a role. Past studies have tended to focus solely on adiposity but the proposed study also seeks to investigat e the effects of muscularity. Using figures that vary along both of these dimensions will allow for the exploration of the question of whether two figures with equal body fat are perceived differently if they have different amounts of musculature. In sum, a wealth of research findings is suggestive of the possible effects of rater (gender, ethnicity) and target (gender, ethnicity, fat/muscularity composition) characteristics in weight status ratings. In addition, extant research suggests the possible covar iate effects of other dispositional factors. These variables will now be reviewed. Psychological Covariates: Body Shape and Size Dissatisfaction, Muscularity Dissatisfaction, Appearance Comparison, Social Norms, and Internalization of Appearance Ideals In addition to the factors of gender and ethnicity, several psychological factors likely affect the way in which the weight of others is judged. Body shape and size dissatisfaction and muscularity dissatisfaction shape the way that individuals view themselves and likely contribute to the way that others are viewed. Appearance comparison refers to the tendency to compare aspects of one's physical appearance to some external standard, usually another individual. These comparisons affect the way one feels about o ne's body and could play a role in the perception of others. Social norms,

PAGE 20

8 too, affect attitudes and behaviour and likely play a role in the way that others are perceived. These factors will now be examined, in turn. Body Dissatisfaction Body dissatisfacti on is conceptualized in many ways. Past research has often focused on weight and shape dissatisfaction, which is more common in women and found to be normative in the population (Rodin et al., 1984). More recent research has acknowledged that men, too, exp erience body dissatisfaction but in ways that may differ from women. Men more commonly experience dissatisfaction with their degree of musculature and strength (e.g., Thompson & Cafri, 2007). In a qualitative study of male body image, Ridgeway and Tylka (2 005) identified five domains of muscularity to which men aspire. These are definition, large size, big but not too big, strength, and athleticism. Overall, their results showed that men desire a tall, lean, muscular, body that looks athletic and strong. In particular, men tended to focus their concern on the arms, the chest, and the abdominal region. Exposure to images of muscular male models has been found to decrease body satisfaction in college men (Lorenzen, Grieve, & Thomas, 2004) indicating that body dissatisfaction in males likely stems from a desire to emulate the muscular ideal (Humphreys & Paxton, 2004). Exposure to the muscular ideal in media has been linked to muscularity concerns as well as dietary supplement use to build muscle (Hatoum & Belle, 2004). Media exposure to the muscular ideal was also associated with a higher value placed on thinness in women in this study. Davis, Karvinen, and McCreary (2005) examined personality correlates of the drive for muscularity in men, hypothesizing that th ey would be similar to the correlates of drive for thinness in women. Neuroticism, perfectionism, fitness orientation, and appearance orientation all predicted drive for muscularity. Given the link between drive for muscularity and inappropriate weight con trol practices such as steroid use and excessive exercise, these personality factors may also pose a risk for such behaviours. Olivardia et al. (2004) also found that a phenomenon known as muscle belittlement, the degree to which participants feel that the y are less muscular than they actually are, is related to depression, body dissatisfaction, and eating disturbance. Jones and Crawford (2005) conducted a structural equation modeling study and found that there are two

PAGE 21

9 distinct and independent paths to body dissatisfaction in adolescent males, one via weight concerns and the other via muscularity concerns. Those males with a higher BMI were susceptible to body dissatisfaction as a function of their excess weight. Males with a lower BMI, on the other hand, ex perienced dissatisfaction as a function of the desire to be larger and more muscular. In women, elevated levels of body dissatisfaction have been associated with dieting, eating disordered symptoms, and negative affect (Heinberg, 1996; Stice, 2001; Thompso n, Heinberg, Altabe, & Tantleff Dunn, 1999). Body dissatisfaction is also a key factor in many theories of eating disturbance, among them the tripartite influence model (Shroff & Thompson, 2006) and the dual pathway model of bulimic symptomatology (Stice, Nemeroff, & Shaw, 1996). In the tripartite influence model (Shroff & Thompson, 2006) peers, parents, and media are thought to send messages that, when internalized, lead to body dissatisfaction, which leads to bulimic symptoms. In the dual pathway model (S tice et al., 1996) body dissatisfaction is thought to contribute to negative affect and restricting behaviour, which, in turn, combine to produce bulimic symptoms. A key feature of body dissatisfaction in both males and females appears to be a mispercepti on of the self, as evidenced in the self perception studies discussed above. Inaccurate perception of the self is also a core diagnostic criterion in both anorexia nervosa and bulimia nervosa (American Psychiatric Association, 2004). In the studies of self perception of weight in adolescents both perception of overweight status (Pritchard et al., 1997) and of a weight problem (Kim & Kim, 2001) were predictive of negative self esteem. Among Japanese adults, body esteem was negatively related to self ideal di screpancy in the domains of weight, body shape, and muscularity in men and women. One question that has yet to be answered is if body dissatisfaction, whether the dissatisfaction is with shape, size, or muscularity, affects how we perceive others. It is p ossible that dissatisfaction with the self could extend to others such that women would overestimate the weight of others and men would underestimate. It is also possible that the converse is true; dissatisfaction with one's own body could lead to idealiza tion of others' bodies, leading women to underestimate and men to overestimate the weight status of another.

PAGE 22

10 Appearance Comparison Festinger (1954) first put forward social comparison theory (SCT). According to SCT, individuals compare themselves to other s to form assessments of their own status on a dimension. When these comparisons are made to people who are doing less well than themselves, the comparison is deemed "downward" and when the comparison is made to another who is doing better, the comparison is termed "upward" (Fiske, 2004). Both types of social comparisons can lead to either negative or positive affective outcomes, depending on the motivation and characteristics of the individual making the comparison (Buunk, Collins, Taylor, VanYperen, & Dak of, 1990). Research has shown that upward comparisons to thin models increase body dissatisfaction (Engeln Maddox, 2005) and predict the presence of eating disorder symptoms (Corning, Krumm, & Smitham, 2006). Tiggeman and McGill (2004) found that social co mparison to models' bodies or body parts increased negative mood and body dissatisfaction. The effect of image type on mood and body dissatisfaction was mediated by the amount of social comparison reported. Krones, Stice, Batres, and Orjada (2005) also fou nd that in vivo exposure to a thin ideal confederate increases body dissatisfaction. Social comparison has been found to be a predictor of body dissatisfaction in women, even when controlling for self esteem and level of obesity (Stormer & Thompson, 1996). In adolescent males, social comparison was related to negative outcomes including increased body dissatisfaction and inappropriate weight gain practices. In adolescent females, social comparison was also related to increased body dissatisfaction, and to i nappropriate weight loss practices (Morrison, Kalin, & Morrison, 2004). Social comparison also prospectively predicts changes in body dissatisfaction in adolescents (Jones, 2004). Females are also more likely than males to engage in universalistic social c omparison in the domain of appearance (Morrison et al., 2004). Through increased body dissatisfaction, appearance comparison may affect the way one perceives others' weight. Given the presence of other participants and a female experimenter in the room whi le ratings were being made, it is possible that appearance comparison could have

PAGE 23

11 occurred during the study. The degree to which participants usually engage in appearance comparison was evaluated and examined as a covariate in the present study. Social Norm s Perceived norms can affect behaviour (Bergrstrom & Neighbours, n.d.). Social norms theory has set forth two types of norms. Descriptive norms involve perceptions of what is popular and injunctive norms involve perceptions of what is typically approved or disapproved, that is, what one should do (Cialdini, 2003). Injunctive body norms in our society tend to be promulgated by the media, who pronounce that females must be thin and males must be muscular. Peers and parents also play a role in shaping our perc eptions of injunctive norms by making comments about how we should look (van den Berg, Thompson, & Obremski Brandon, 2002). Most individuals are unable to live up to these unrealistic norms and, in many, this leads to some level of body dissatisfaction, pa rticularly among those who internalize or "buy into" the thin ideal and feel the need to emulate it (Thompson & Stice, 2001). Research has shown that girls are more likely than boys to perceive higher weight and dieting concerns among family and friends (T hompson et al., 1999). Ratings of perceived weight in adults, therefore, may reflect their recognition that they depart from what is considered normative. Research has shown that the individuals in one's immediate environment affect how one is viewed by o thers. One study by Halford and colleagues (BBC News, 2003) digitally manipulated a prom photo so that the same attractive male was seen beside either a thin or heavy well dressed prom date. College women's descriptions of the gentleman in the picture, bas ed on a negative adjective rating scale, were more negative when the woman in the photograph was heavy. The norm group one considers relevant may also affect one's own self view. One study presented women with body image feedback stating that they differed (i.e., were either larger or smaller) than either 1) women at their own college or 2) the United States population (Heinberg & Thompson, 1992). Only the feedback that one differed from one's own college population led to increased body image anxiety and g eneral distress ratings. A study of Bahraini adolescents (Al Sendi et al., 2004), found that 75% of boys thought their friends would consider them underweight though only 11% were objectively underweight, indicating an

PAGE 24

12 awareness of a social norm for muscul arity or, at least, a larger body size in males. Such findings indicate that not only are distal societal norms important in one's view of weight, so too are proximate social norms. The individuals in one's immediate environment likely shape, to some exten t, how one views the weight of another. It is possible that if one's social network is comprised primarily of overweight and/or obese individuals one might come to see excess weight as normative or average and, therefore, underestimate the weight status of another individual. In the body image field, measurement tends to focus on the cultural norms of thinness and muscularity while putting less emphasis on the more proximate influences in an individual's life. In a fascinating study of social networks, Chr istakis and Fowler (2007) found that the likelihood of becoming obese increased as close friends became obese, particularly among male friendships. The same held true of spousal relationships. One might conclude that some more proximate social factor is ha ving an effect on weight in these instances. A barometer of immediate or proximate weight norms could be obtained by having individuals indicate their perceptions of those closest to them including immediate family members and friends. Research has indicat ed that, even in adolescence, girls are more likely to compare themselves to friends as opposed to more distal peers, and to peers as opposed to family members (Schutz, Paxton, & Wertheim, 2002). When the tripartite influence model was tested on adolescent s in a structural equation modeling study (Shroff & Thompson, 2006), the path from parental influence to other body image and eating disorder symptom variables was not significant. It may be assumed that the likelihood of using family members as a source o f comparison diminishes further during the college years since many individuals move away from home at this time and have even less exposure to family members. In this study, therefore, the focus will be on a social network of peers. Cultural and societal norms are not to be neglected, however, when considering the influence of social norms on an individual. Thin ideal internalization likely plays an important role in the development of body dissatisfaction based on social norms. Terry and Hogg (1996) foun d that perceived norms were predictive of sun protective behaviour intentions only in those individuals who strongly identified with a behaviour relevant

PAGE 25

13 reference group. The parallel in body image would be that individuals who strongly identify with, or i nternalize, the importance of thinness disseminated in Western culture would be more at risk for body dissatisfaction and inappropriate weight loss (for women) or weight gain (for men) strategies. Thin ideal internalization is considered a causal risk fact or for body dissatisfaction and bulimic symptoms (Thompson & Stice, 2001). Internalization of norms has a stronger relationship to body image than does awareness of norms (Cafri, Yamamiya, Brannick, & Thompson, 2005) and is likely a key factor in the effec ts of norms on ratings of weight status. Individuals who internalize the cultural standards of thinness and appearance are likely to show the common patterns of over and underestimation in their self perceptions of weight. When it comes to rating others, it is hypothesized that individuals who strongly internalize cultural norms are likely to apply those norms to others as well and will carry the patterns of over and underestimation into their ratings of others. It is possible, therefore, to assess social norms at multiple levels. Internalization measures assess the cultural norms in society at a broad level, assumed to be common for most individuals in the culture. Ratings of peers, on the other hand, assess a more proximal social network norm, assumed to vary by individual. Thus, there is available evidence that at least five psychological variables may affect judgments of other individuals' body sizes: body shape and weight dissatisfaction, muscularity dissatisfaction, appearance comparison, proximate s ocial norms (body sizes of those in one's immediate environment), and internalization of appearance ideals. Overview of Current Study The current study was designed to investigate how individuals categorize the weight status of other individuals. Both rate r and target characteristics were examined, along with several dispositional measures as potential covariates of ratings. The lack of any previous empirical work in this area of research renders the presentation of well supported hypotheses untenable. Howe ver, as discussed throughout the introduction, a review of related work in the field of body image (in particular, the self perception literature) and obesity suggests the possibility that rater gender and ethnicity may influence weight category assignment s. Additionally, it is possible that target features (gender, ethnicity) may also interact with rater features.

PAGE 26

14 Hypotheses Based on speculations from related literature, the hypotheses below are presented as initial, exploratory hypotheses: 1. Target feature s ( gender and ethnicity) will interact with rater features (gender and ethnicity) to affect target weight ratings. For instance, African American men may place African American women in a lower (less heavy) weight categorization than Caucasian men. 2. The bod y fat and the muscularity of the target will interact to influence ratings. Targets with greater muscularity but equal body fat will receive a lower (less heavy) weight categorization. Exploratory Questions 3. Social norms, appearance comparison, body dissat isfaction, weight dissatisfaction and rater BMI will act as covariates. Individuals with a higher BMI and those higher in appearance comparison will assign lower weight categorizations, whereas those high in body dissatisfaction, muscularity dissatisfactio n, or social norms, will assign higher weight categorizations. Pilot Study Prior to the main study, a pilot study was conducted in which the slide rating task and a new, exploratory measure were tested. The purposes of the pilot study were to 1) test the n ew measure, the Proximal Social Environment Rating Scale (PSERS) to insure the instructions were clear, 2) to establish the number of friends to be rated using the new PSERS in the main study, 3) test the procedures for the slide ratings to ensure that the slides were being rated as intended, and 4) determine the optimal number of stimuli but consider respondent burden in the total number retained.

PAGE 27

15 Method Participants Twenty nine undergraduates from the University of South Florida participated in the pilot study. One case was deleted because the participant failed to indicate gender. The final sample consisted of nine males and 19 females. There were three Caucasian males and 12 Caucasian females, three Hispanic males and two Hispanic females, and three Afr ican American males and five African American females in the sample. Mean age of the participants was 22.24 years ( SD = 5.60) with a range of 18 44 years. The mean BMI of the sample was 22.24 ( SD = 4.13), which falls in the normal weight range. The maximum BMI in the sample was 32.38 and the minimum was 18.29. Measures Please see Appendix A for the pilot stimuli, the target rating items, the instructions read to participants, the PSERS instructions, and the focus group questions. Please note that the stimul i are lettered for ease of identification. Participants did not see these letters during slide presentations. Stimuli (Target Figures) The stimuli were culled from the Somatomorphic Matrix (Gruber, Pope, Borowiecki, & Cohane, 1999). The Somatomorphic Matri x consists of 100 figures of each gender arranged in a 10 X 10 matrix with the figures varying along the dimensions of muscularity and adiposity. The figures have known fat free mass indices and body fat percentages, which were determined by photographing individuals with known measurements and having a graphic artist converting these photographs into line drawings (Gruber et al., 1999). This assessment instrument has rapidly become the standard in the body image field for the assessment of both fat and mus cularity dimensions (e.g., Cafri & Thompson, 2007). For the pilot study, thirty figures were chosen, fifteen male and fifteen female. Each figure was presented in three different racial/ethnic categories for a total of 90 targets. Race/ethnicity was indica ted on the bottom of the slide. The racial/ethnic categories were Caucasian, African American, and Hispanic. The initial fifteen figures of

PAGE 28

16 each gender were chosen as follows: The three thinnest figures were chosen from the lowest body fat quartile of the somatomorphic matrix at a low, medium, and high level of muscularity. The four thinnest figures of the somatomorphic matrix were not chosen because they do not resemble what is usually seen on a daily basis and are, therefore, low in ecological validity. T he remaining 12 figures represented the three upper quartiles of body fat of the somatomorphic matrix representing the lowest, highest, and two intermediate levels of muscularity. The figures were altered from their usual presentation in the following way s: 1) the figures were colored in to make them more credible in any of the racial categories. The figures were colored in "light cool brown" in Adobe Photoshop. Due to inconsistencies in projector color, the color was altered for presentation in some class rooms to make the appearance of the slides consistent across presentations. 2) The male figures' bathing suits were colored black. 3) The heads of the female figures were removed and replaced with the heads of the male figures to remove any suggestion of r ace that could come from the hairstyles of the female figures The male figures' heads were simply circles with ears so the exchange did not render the female figure masculine in appearance (see Appendix A). Figure Ratings Participants were asked to rate e ach figure's health, weight, and attractiveness on a seven point Likert type scale. The health and attractiveness items were presented as distractor items to reduce the focus on weight. Some participants were also asked to estimate the weight of the target given the height (for males 5'10", for females 5'4"). Proximate Social Environment Rating Scale An exploratory social norms measure was created for the purposes of this study called the Proximate Social Environment Rating Scale (PSERS). Participants were asked to rate the body shape and size of peers with whom they spend the most time using a subset of figures from the Somatomorphic Matrix (Gruber et al., 1999). The figures in this scale can be found in Appendix H. In the pilot study, participants were ask ed to rate

PAGE 29

17 the three peers with whom they spend the most time and the ten peers with whom they spend the most time, in counterbalanced order. Demographic Information Participants provided demographic information including age, race/ethnicity, height, weig ht, and year in school. Body mass index (BMI) was calculated using self reported weight and height with the standard formula: [weight in pounds/(height in inches) 2 ] X 703. See Appendix K for the specific demographic items. Focus Groups After the presentati on of the slides, participants were asked a number of focus questions. The questions were designed to elicit any aspects of the rating task that participants found problematic. Procedure Participants enrolled in the study and scheduled an appointment via the USF Sona system. The study was conducted in a group setting in classrooms equipped with a computer and Proxima projector. Participants viewed the slides and provided ratings and then completed the PSERS. Finally, the participants responded to the focus group questions. Participants were then debriefed and awarded three extra credit points. Results Stimuli Means for each figure were calculated, collapsing across the three presentations of each slide (see Table 1). Mean weight ratings seemed to increas e w ith increasing muscularity and adiposity but there did seem to be a tendency for participants to stick to the middle of the scale, particularly for the figures in the middle. In other words, participants had a tendency to provide a rating between 3 and 5 on a seven point scale, particularly for the figures intermediate in adiposity and muscularity. This tendency led to a clustering of means around the middle of the scale, making it difficult to detect differences among them. Because the means of several fi gures were very close, six male

PAGE 30

18 and six female figures were eliminated. The reduction in the number of stimuli will also help to reduce respondent burden. The eliminated figures were those in the three upper quartiles of body fat of the somatomorphic matri x representing the two intermediate levels of muscularity (for males figures T, U, X, Y, BB, and CC; for females figures E, F I, J, M, and N). Table 1. Means for health, weight, and attractiveness of target figures Figure Weight Mean Rating Health Mean R ating Attractiveness Mean Rating A 4.00 5.37 2.83 B 3.87 5.31 3.01 C 2.35 2.98 2.23 D 4.46 5.41 2.91 E 4.11 5.24 4.00 F 3.93 5.05 4.06 G 3.89 4.80 4.52 H 4.44 5.17 3.71 I 4.44 4.84 4.25 J 4.23 4.74 4.22 K 4.44 4.57 4.20 L 4.82 4.51 3.36 M 4.90 4.17 3.64 N 5.16 3.66 3.40 O 5.29 3.21 2.89 P 4.30 5.90 4.99 Q 4.01 5.47 4.87 R 3.44 4.56 3.90 S 4.47 5.74 4.75 T 4.18 5.75 4.94 U 3.93 5.29 4.60 V 3.67 4.11 3.44

PAGE 31

19 Figure Weight Mean Rating Health Mean R ating Attractiveness Mean Rating W 4.73 5.28 4.26 X 4.41 5.28 4.29 Y 4.50 4.61 3.70 Z 4.40 3.86 3.13 AA 5.49 3. 25 2.69 BB 5.57 3.11 2.53 CC 5.79 2.67 2.31 DD 5.51 2.99 2.43 Proximate Social Environment Rating Scale The mean adiposity and mean muscularity were calculated for each participant's ratings of three friends and ten friends. Mean adiposity for three fr iends ( M = 4.96, SD = 1.71) was significantly correlated with mean adiposity for ten friends ( M = 3.42, SD = 2.19; r = .60, p < .01). Mean muscularity for three friends ( M = 4.96, SD = 2.19) was significantly correlated with mean muscularity for ten friend s ( M = 3.24, SD = 1.32; r = .73, p < .01). Given these high correlations and the desire to keep participant burden to a minimum, it was determined that ratings of three friends would be a sufficient measure for the PSERS. Focus Groups Several of the partic ipants reported that they either did not notice or did not attend to the race/ethnicity labels on the slides. Of those who did, the majority reported that they found the figures credible although a small minority felt that they had to use their imagination To increase the likelihood that participants would attend to the labels, the instructions for the main study were changed so that the participants' attention was explicitly directed to the labels. Some p articipants reported that they felt they were watch ing the same few slides over and over again. Some participants also complained about slide quality, citing "fuzziness" as a factor that made the ratings difficult. M ost participants reported that they were able to focus throughout but there were several pa rticipants who reported that they

PAGE 32

20 were becoming tired or restless by the end of the presentation. Another reported that the presentation was "boring" and "repetitive." As noted above, t he number of slides was reduced to help reduce participant burden. Elim inating the slides at the intermediate levels of muscularity was also designed to make the slides more distinct and distinguishable. The quality of the slides was improved. Some participants complained that having friends in the room distracted them from c ompleting the task. The experimenter also noticed some participants responding to cell phone pages and text messages during the study. The instructions were modified for the main study to reduce the likelihood of these distractions. Participants were told that if they disrupted the study in any way or were found using cell phones, they would be asked to leave and not receive credit for participating in the study. They were also directed not to make any noise or speak during the study. During the pilot study the majority (57%) of the participants received the information regarding the height of the target to be rated. The other participants did not. During the focus groups, attempts were made to understand how knowing or not knowing this information affected participant's ratings. One participant who did not know the target's height reported that they assumed the target was "average." Another reported that they looked at how the target "fit in the frame" of the slide. Several of those who did know said that t hey did not attend to it or they forgot about it or did not factor it into their decisions. A minority reported that it had affected their ratings and that they would have rated the targets differently had they not known the height. Based on these findings the decision was made to eliminate the item giving participants the target's height. It was thought that it would be best to allow the participants to establish their own metric for height. Many of the participants had no difficulty with the PSERS instr uctions. However, there were several participants who indicated that the instructions did not clearly explain how the measure was to be filled out. Based on the questions asked by participants during the focus group, the PSERS instructions were modified. T he modifications were intended to make it easier for participants to understand how the scale was to be filled out.

PAGE 33

21 Summary of Changes Based on the results of the pilot study, several changes were made. The instructions read to participants were altered to place a higher premium on silence and focus during the rating task. The participants' attention was also directed to the race/ethnicity labels on the slides and they were explicitly asked to take these into account in their ratings. Please see Appendix B for the altered instructions. The number of ratings to be made was reduced from 90 to 54. The slide quality was improved. The number of friends to be rated in the PSERS was set at three. The instructions on the PSERS were altered to increase clarity. Final ly, because of the responses of the participants during the focus groups, the item stating the height of the figure was removed.

PAGE 34

22 Method Participants Participants were 342 undergraduates from the University of South Florida. Two students did not comple te the study and three failed to provide essential demographic information. Two participants indicated a mixed racial background and were therefore ineligible. One did not complete the slide ratings and one final participant incorrectly completed the rati ng task. These nine individuals were eliminated. The final sample consisted of 333 undergraduates, 140 males and 193 females. The participants ranged in age from 18 to 46 years with a mean of 21.43 years ( SD = 2.18). Thirty nine percent of the sample self identified as Caucasian, 31% as Hispanic, and 30% as African American or Black. The breakdown of the sample by gender and race is presented in Table 2. The mean BMI of the sample was 24.86, which falls at the top end of the normal weight range. BMI breakdo wn by gender and race is also presented in Table 2. Table 2. Demographic characteristics. BMI Gender N M SD African American Male 34 26.23 4.64 Female 66 26.12 6.74 Hispanic Male 51 26.06 4.48 Female 53 23.56 3.73 Caucasian Male 55 25.99 5.16 F emale 74 22.37 3.72

PAGE 35

23 Measures Stimuli and Stimulus Rating Form The stimuli were 18 figures, nine male and nine female, from the Somatomorphic Matrix (Gruber, Pope, Borowiecki, & Cohane, 1999). The Somatomorphic Matrix consists of 100 figures of each gender arranged in a 10 X 10 matrix with the figures varying along the dimensions of muscularity and adiposity. The figures have known fat free mass indices and body fat percentages, which were determined by photographing individuals with known measurements and having a graphic artist converting these photographs into line drawings (Gruber et al., 1999). This assessment instrument has rapidly become the standard in the body image field for the assessment of both fat and muscularity dimensions (e.g., Cafri & Thomp son, 2007). Each figure was presented three times to participants, once in each of the race/ethnicity categories, for a total of 54 targets. Target race/ethnicity was indicated on the bottom of the slide. The racial/ethnic categories were Caucasian, Africa n American, and Hispanic. Participants were asked to rate each figure's health, weight, and attractiveness on a seven point Likert type scale. The health and attractiveness items were presented as distractor items to reduce the focus on weight. The target rating items can be found in Appendix C. To reduce fatigue effects and order effects, four random orders of the slides were created and one order was randomly selected for each group of participants. Distraction Task After the ratings, participants complet ed a distraction task (see Appendix D). Research has shown that a brief (5 8 minutes), externally focused, active task will return experimentally induced dysphoric moods to baseline (Lyubomirsky & Nolen Hoeksema, 1993, 1995; Morrow & Nolen Hoeksema, 1990). A similar procedure was used in the present study to wash out any negative affect induced as a result of the rating task. Participants were asked to spend 5 10 minutes thinking about the countries of the world. They were asked to compile a list of ten loc ations they have heard about but have never visited and to indicate how the media portrays these destinations.

PAGE 36

24 Appearance Comparison The Physical Appearance Comparison Scale (PACS; Thompson, Heinberg, & Tantleff, 1991; Appendix E) was used to assess the t endency to compare oneself to others in various domains of physical appearance. The PACS is a 5 item scale that uses a five point Likert type scale ranging from "Never" to "Always." It has demonstrated adequate internal reliability and test retest reliab ility as well as moderate convergent validity with measures of body image dissatisfaction, eating disturbance, and self esteem (Thompson, Heinberg, & Tantleff, 1991). After reverse coding item 4, the responses to the items were summed to create a scale sco re (possible total = 25). Internal consistency in the present sample was good (Alpha = .74). Social Desirability Social desirability was assessed using the Marlowe Crown S ocial Desirability Scale (MCSDS; Crown & Marlowe, 1964; Appendix F ). The MCSDS is a 3 3 item measure of individuals' approach to self and socially evaluative situations and the meanings of such situations for them. It uses a true false response format. Internal consistency of this scale has been found to be good (Cronbach's alpha = .88) an d one month test retest reliability in a sample of undergraduates is high ( r = .88). The scale score was created by reverse scoring all items keyed false and then by counting the number of "true" responses (possible total = 33). Internal consistency in the present sample was very good (Alpha = .78). Social Norms Measures Two measures of social norms were used. The first was the Sociocultural Attitudes Towards Appearance Questionnaire 3 (SATAQ; Thompson, van den Berg, Roehrig, Guarda, & Heinberg, 2004; App endix G) Internalization General subscale. This subscale is designed to assess trait levels of thin ideal internalization, specifically from media messages. It is assumed that those with higher levels of internalization are those who hold more strongly t o the current cultural norms of thinness. Ratings are made on a five point Likert scale ranging from "Definitely Agree" to "Definitely Disagree." The Internalization General subscale has shown excellent reliability (Cronbach's alpha

PAGE 37

25 = .96; Thompson et a l., 2004). Responses were summed to create a total score (possible total = 45). Internal consistency for the present sample was excellent (Alpha = .93). A second, exploratory social norms measure was created for the purposes of this study called the Proxim ate Social Environment Rating Scale (PSERS). Participants were asked to rate the body shape and size of the three peers with whom they spend the most time using a subset of figures from the Somatomorphic Matrix (Gruber et al., 1999). The figures in this sc ale can be found in Appendix H. The mean adiposity of the participant's peer group served as a measure of the individual's proximal social network. Body Shape and Size Dissatisfaction Body dissatisfaction was assessed using the Eating Disorder Inventory 3 Body Dissatisfaction subscale (EDI BD; Garner, 2004; Appendix I), a ten item scale that assesses overall satisfaction with weight related body sites. The EDI BD shows good internal consistency in clinical samples of adults and adolescents in both the U nited States and internationally (all alphas > .9). The test retest reliability is also very good (r = 0.95). After reverse coding items 3, 4, 5, 7, and 9, responses were summed to create a total score (possible total = 54). Internal consistency in the pre sent sample was excellent (Alpha = .89). Muscularity Dissatisfaction Dissatisfaction with one's muscular appearance was assessed with the Drive for Muscularity Scale (DMS; McCreary & Sasse, 2000; Appendix J). The DMS consists of 15 items on a six point Lik ert type scale. The Likert scale is in reverse (from 1 "Always" to 6 "Never") and the items are all reverse coded before they are scored. The DMS shows good internal consistency with alphas ranging between .85 and .91 (McCreary, 2006). Test retest correlat ions in a sample of college men were also high (all r s > .84; Cafri & Thompson, 2004). Responses were summed to create a total score (possible total = 90). Internal consistency in the present sample was very good (Alpha = .89). Demographic Information Part icipants provided demographic information including age, race/ethnicity, height, weight, and year in school. Body mass index (BMI) was calculated using self

PAGE 38

26 reported weight and height with the standard formula: [weight in pounds/(height in inches) 2 ] X 703. See Appendix K for the specific demographic items. Procedure Participants enrolled in the study and scheduled an appointment via the USF Sona system. Near the end of the subject enrollment phase, African American males were specifically targeted via recru itment efforts. Classes in which extra credit points were offered were visited and an announcement was made about the study. Eligible participants were given the opportunity to sign up in class or they were provided with the study number to sign up at a la ter date. The study was conducted in a group setting in classrooms equipped with a computer and Proxima projector. First, participants provided informed consent and then viewed the two slide presentations in one of four randomly selected orders, with a fi ve minute break in between. Each slide was presented to the group for 5 seconds followed by a black screen for 15 seconds. Participants were asked to rate the figure during the time the black screen was presented. Once all 54 figures were rated, participan ts were asked to complete the distraction task. After five minutes, they were told they could move on to the questionnaires after they had completed the distraction task. The questionnaires were presented in the following order: EDI BD, SATAQ, DMS, PSERS, PACS, MCSDS. After completing the questionnaires, participants were debriefed. Two extra credit points were awarded after participation. Data Analysis Preliminary Analyses Before beginning the primary analyses, the internal consistency reliability of each variable was computed. The data were checked for outliers using a standardized score of +/ 3 to establish outlier status. The covariate variable distributions were also checked for normality and were transformed if necessary. Age could not be normalized an d no transformation was applied. The SATAQ, MCSDS, and PSERS did not require transformations. A square root transformation was applied to the EDI BD, DMS, and PACS. An inverse transformation was applied to BMI. The correlations among the

PAGE 39

27 covariate variable s were computed. Next, the groups were examined for differences on age, BMI, and each of the potential covariate variables with a 2 (gender) X 3 (ethnicity) ANOVA. LSD post hoc tests were used. To further verify that social desirability was not influencing responses, the correlation between the MCSDS and the mean weight rating for each target was calculated. Weight Ratings First, three omnibus ANOVAs were calculated comparing groups of three figures with a common level of muscularity or adiposity. These tes ts were 3 (figure) X 2 (rater gender: male, female) X 3 (rater race/ethnicity: Caucasian, African American, Hispanic) X 3 (target race/ethnicity: Caucasian, African American, Hispan ic) repeated measures ANOVAs. A p value of .01 was used for statistical sig nificance for these analyses due to the large number of tests. After the omnibus ANOVAs, follow up analyses were conducted to examine the individual figures. For the purposes of these analyses, a Bonferroni correction was applied to guard against an elevat ed Type 1 error rate. Because there were 18 initial ANOVAs, the p value of .10 was divided by 18 to yield a required significant p value of .006. Slightly higher p values, between .006 and .01, were considered a trend towards significance. For each of the eighteen figures, a 2 (rater gender: male, female) X 3 (rater race/ethnicity: Caucasian, African American, Hispanic) X 3 (target race/ethnicity: Caucasian, African American, Hispanic) repeated measures ANOVA wa s conducted on the target weight ratings. Targ et gender was not entered into the analyses because the figures were not directly comparable across genders. If differences were found across gender, it would not be clear whether these were, in fact, due to gender or to the different body compositions of the male and female figures. Dispositional variables were entered as c ovariates if they showed significant between groups differences in the preliminary analyses. Each covariate was entered in a separate ANCOVA. Due to the large number of covariates, ente ring them all in a single analysis would likely reduce power to detect any significant effects. Greenhouse Geisser values are reported for all repeated measures effects. LSD post hoc tests were used.

PAGE 40

28 Adiposity and Muscularity Dependent t tests were carried out to see if increasing muscularity affected the ratings of figures whose body fat composition remain ed unchanged. A mean for each figure was computed collapsed across the three presentations. Separate analyses were conducted for the male and female slid es. These analyses were also conducted separately for each of the race/gender groups in the study. Because of the large number of tests, a p value of .01 was used to establish significance. Additional Analyses The same analytic procedure described above fo r the weight ratings was undertaken again for the distractor items asking about the health and attractiveness of the figures

PAGE 41

29 Results Descriptive Information One outlier was detected on the DMS. All analyses were run with and without this outlier value. T he results were unchanged. All results are presented with the outlier in the data. Presentation orders for the figures were randomly selected for each experimental group. Forty six percent of participants saw Order 1, 17% saw Order 2, 18% saw Order 3, and 19% saw order 4. On the distractor task, 96.1% of the sample complied with the instructions and listed five countries. Five individuals (1.5%) listed only four countries, 6 individuals (1.8%) listed three countries and two individuals (0.6%) listed only t wo countries. Scale means and standard deviations are presented in Table 3. Table 3. Means and standard deviations for covariates Scale Score Range M SD EDI Body Dissatisfaction 9 54 27.46 10.04 SATAQ Internalization 9 45 26.50 9.20 DMS 15 90 37.09 13.94 PSERS 1 10 5.19 1.70 PACS 5 25 14.35 3.99 MCSDS 0 33 16.41 5.28 BMI 24.86 5.12

PAGE 42

30 Preliminary Analyses The correlations among the covariates were all low enough to merit entering each covariate separately into the analysis rather than forming a c omposite (see Table 4). Table 4. Correlations among covariates Scale 1. 2. 3. 4. 5. 6. 7. 1. EDI BD 2. SATAQ .26** 3. DMS .05 .20** 4. PSERS .12* .04 .09 5. PACS .32** .46** .13* .01 6. MCSDS .24** .24** .13* .08 .29** 7. BMI .34** .08 .10 .12* .00 .07 p < .05, ** p < .01 A 2 (gender) X 3 (ethnicity) ANOVA tested for participant differences on the dispositional measures. All relevant means are presented in Tables 5 through 7. Group differences were found for BMI for both gender ( F (1,326) = 22.00, p < .001, partial 2 = .06) and race ( F (2,326) = 4.34, p < .05, partial 2 = .03). There was a strong trend for an interaction as well ( F (2,326) = 3.03, p = .05, partial 2 = .02). Group scores on the EDI BD also differed for gender ( F (1,327) = 34.37, p < .001, partial 2 = .11) and race ( F (2,327) = 8.74, p < .001, partial 2 = .05) but the interaction was not significant ( F (2,327) = 2.43, ns partial 2 = .02). For the SATAQ, group differences were found for gen der ( F (1,325) = 7.33, p < .01, partial 2 = .02) and race ( F (2,325) = 15.67, p < .001, partial 2 = .09) and there was a significant interaction ( F (2,325) = 4.69, p < .05, partial 2 = .03). This same pattern was found for the DMS for gender ( F (1,323) = 14 8.72, p < .001, partial 2 = .32), race ( F (2,323) = 4.27, p < .05, partial 2 = .03), and the interaction effect ( F (2,323) = 3.87, p < .02, partial 2 = .02). There were group differences for gender ( F (1,326) = 10.94, p < .01, partial 2 = .03) and race ( F (2,326) = 13.66, p < .001, partial 2 = .08) for the PACS but the interaction was not significant ( F (2,326) = 0.10, ns

PAGE 43

31 partial 2 < .01). These scales, which demonstrated significant between group differences, were used as covariates in all subsequent ana lyses. No group differences were found on age for gender ( F (1,327) = 0.43, ns partial 2 < .01), or race ( F (2,327) = 1.07, ns partial 2 = .01). The interaction was also not significant ( F (2,327) = 1.74, ns partial 2 = .01). The PSERS also failed to y ield significant group differences for gender ( F (1,299) = 0.33, ns partial 2 < .01) or race ( F (2,299) = 0.66, ns partial 2 < .01) and the interaction effect was nonsignificant ( F (2,299) = 0.82, ns partial 2 < .01). There were also no group difference s on the MCSDS for gender ( F (1,322) = 0.18, ns partial 2 < .01) or race ( F (2,322) = 2.60, ns partial 2 = .02) and the interaction, too, was not significant ( F (2,322) = 1.34, ns partial 2 = .01). These scales, because they did not demonstrate between groups differences, were not used as covariates in any of the subsequent analyses. Table 5. Means, standard errors, F p and partial 2 values for gender effects Group Variable Males M (SD) Females M (SD) F p and partial 2 values Age 21.67 (4.35) 2 1.26 (4.06) F (1,327) = 0.43, ns partial 2 < .01 BMI 26.07 (4.77) a 23.98 (5.20) b F (1,326) = 22.00, p < .001, partial 2 = .06 EDI BD 23.97 (10.04) a 30.00 (9.28) b F (1,327) = 34.37, p < .001, partial 2 = .11 SATAQ 25.11 (8.71) a 27.50 (9.44) b F (1,325) = 7.33, p < .01, partial 2 = .02 DMS 46.27 (13.32) a 30.38 (10.02) b F (1,323) = 148.72, p < .001, partial 2 = .32 PSERS 5.28 (1.78) 5.13 (1.66) F (1,299) = 0.33, ns partial 2 < .01 PACS 13.63 (4.16) a 14.87 (3.79) b F (1,326) = 10.94, p < .01, partial 2 = .03 MCSDS 16.40 (5.08) 16.42 (5.30) F (1,322) = 0.18, ns partial 2 < .01 Note: Superscripts indicate means that differ significantly.

PAGE 44

32 Table 6. Means, standard errors, F p and partial 2 values for race effects Group Variable Caucasian M (SD) Hispa nic M (SD) African American M (SD) F p and partial 2 values Age 21.78 (4.34) 21.21 (4.37) 21.21 (3.76) F (2,327) = 1.07, ns partial 2 = .01 BMI 23.91 (4.73) a 24.78 (4.29) 26.16 (6.10) b F (2,326) = 4.34, p < .05, partial 2 = .03 EDI BD 29.13 (10.12) a 27.65 (9.68) a 25.10 (9.95) b F (2,327) = 8.74, p < .001, partial 2 = .05 SATAQ 29.18 (9.19) a 27.13 (8.23) a 22.39 (8.82) b F (2,325) = 15.67, p < .001, partial 2 = .09 DMS 37.25 (13.68) 40.31 (15.70) a 33.50 (11.33) b F (2,323) = 4.27, p < .05, partial 2 = 03 PSERS 5.03 (1.77) 5.30 (1.59) 5.28 (1.73) F (2,299) = 0.66, ns partial 2 < .01 PACS 15.69 (4.00) a 13.76 (3.70) b 13.21 (3.80) b F (2,326) = 13.66, p < .001, partial 2 = .08 MCSDS 16.06 (5.14) 16.02 (5.06) 17.25 (5.60) F (2,322) = 2.60, ns partial 2 = .02 Note: Superscripts indicate means that differ significantly. Subscripts indicate means that differ significantly.

PAGE 45

33 Table 7. Means, standard errors, F p and partial 2 values for interaction effects Group Caucasian Hispanic African Ameri can Variable Male M (SD) Female M (SD) Male M (SD) Female M (SD) Male M (SD) Female M (SD) F p and partial 2 values Age 22.64 (5.31) 21.15 (3.34) 21.02 (2.82) 21.40 (5.49) 21.09 (4.33) 21.27 (3.46) F (2,327) = 1.74, ns partial 2 = .01 BMI 25.99 (5.16) a 22. 37 (3.71) b 26.06 (4.48) c 23.56 (3.73) d 26.23 (4.64) 26.12 (6.74) F (2,326) = 3.03, p = .05, partial 2 = .02 EDI BD 25.49 (10.69) 31.83 (8.82) 25.69 (9.79) 29.55 (9.26) 18.94 (7.60) 28.27 (9.56) F (2,327) = 2.43, ns partial 2 = .02 SATAQ 25.43 (8.65) a 31 .92 (8.64) b 27.24 (8.71) 27.01 (7.82) 21.30 (7.74) 22.94 (9.32) F (2,325) = 4.69, p < .05, partial 2 = .03 DMS 44.67 (14.03) a 31.65 (10.42) b 51.28 (13.94) c 29.96 (8.70) d 41.47 (8.05) e 29.26 (10.55) f F (2,323) = 3.87, p < .05, partial 2 = .02 PSERS 5.16 ( 1.82) 4.94 (1.74) 5.51 (1.75) 5.11 (1.44) 5.11 (1.71) 5.37 (1.75) F (2,299) = 0.82, ns partial 2 < .01 PACS 15.02 (4.17) 16.19 (3.82) 13.06 (3.95) 14.43 (3.35) 12.14 (3.89) 13.72 (3.77) F (2,326) = 0.10, ns partial 2 < .01 MCSDS 15.60 (4.87) 16.40 (5.3 5) 16.00 (4.84) 16.06 (5.32) 18.33 (5.42) 16.71 (5.66) F (2,322) = 1.34, ns partial 2 = .01 Note: Superscripts indicate means that differ significantly. Subscripts indicate means that differ significantly.

PAGE 46

34 The MCSDS was correlated with the mean target w eight rating for each figure. These correlations were small and only four were significant. The significant correlations were small and represent very little variance accounted for. The MCSDS did not appear to have a large influence on the weight ratings. Table 8. Correlations between mean target weight ratings and MCSDS scores Figure MCSDS A .02 B .02 C .05 D .11* G .11* H .11* K .07 L .06 O .02 P .02 Q .08 R .06 S .07 V .04 W .03 Z .11* AA .07 DD .02 p < .05 Initial We ight Analyses Female Figures Three sets of three female figures were compared because they shared a common level of adiposity or muscularity. First, figures A, B, and C, all at the lowest level of

PAGE 47

35 adiposity but varying in muscularity, were compared. There was a significant interaction of target race and muscularity ( F (4,1178) = 13.17, p < .001, partial 2 = .04). Post hoc LSD tests of the interaction revealed that for figure A, the highest weight rating was assigned when the target was African American ( M = 3.90, SD = 0 .05) with the Caucasian target ( M = 3.89, SD = 0 .05) receiving an intermediate rating and the Hispanic target ( M = 3.74, SD = 0 .05) receiving the lowest rating. For figure B, the highest weight rating was assigned when the target was Caucasian ( M = 3.63, SD = 0 .05) with the Hispanic target ( M = 3.62, SD = 0 .05) receiving an intermediate rating and the African American target ( M = 3.61, SD = 0 .05) receiving the lowest rating. For figure C, the highest weight rating was assigned when the target w as Hispanic ( M = 2.58, SD = 0 .06) with the Caucasian target ( M = 2.35, SD = 0. 06) receiving an intermediate rating and the African American target ( M = 2.21, SD = 0 .06) receiving the lowest rating. This analysis also revealed a significant effect of muscu larity ( F (2,458) = 506.76, p < .001, partial 2 = .68), which must be qualified by the significant muscularity by target race interaction. Post hoc LSD comparisons revealed that figure A ( M = 3.85, SD = 0 .04), the figure with the highest muscularity, recei ved the highest rating. Figure B ( M = 3.62, SD = 0 .04) was intermediate and figure C, the figure with the lowest muscularity, received the lowest rating ( M = 2.38, SD = 0 .05). All three means differed significantly. There was also a significant main effect of gender in these analyses ( F (2,311) = 22.68, p < .01, partial 2 = .02). Post hoc LSD tests revealed that, overall, females ( M = 3.37, SD = 0 .04) provided higher ratings than did males ( M = 3.19, SD = 0 .05). Next, figures D, H, and L, all at the highest level of muscularity but varying in adiposity were compared. There was a significant interaction between adiposity and target race ( F (4,1151) = 6.00, p < .001, partial 2 = .02). Post hoc examination of the means showed that for figure D, the highest wei ght rating was assigned when the target was Caucasian ( M = 4.24, SD = 0 .04) with the Hispanic target ( M = 4.15, SD = 0 .04) receiving an intermediate rating and the African American target ( M = 4.11, SD = 0 .04) receiving the lowest rating. For figure H, the highest weight rating was assigned when the target was Caucasian ( M = 4.49, SD = 0 .04) with the Hispanic target ( M = 4.39, SD = 0 .04) receiving an intermediate rating and the African American target ( M = 4.29, SD =

PAGE 48

36 0 .04) receiving the lowest rating. For f igure L, the highest weight rating was assigned when the target was Caucasian ( M = 5.10, SD = 0 .04) with the African American target ( M = 5.04, SD = 0 .04) receiving an intermediate rating and the Hispanic target ( M = 4.82, SD = 0 .04) receiving the lowest r ating. For this set of figures, there was a main effect of adiposity ( F (2,546) = 329.45, p < .001, partial 2 = .51), which must be qualified by the significant adiposity by target race interaction. Post hoc LSD tests revealed that figure L ( M = 4.99, SD = 0 .03), the figure with the highest level of adiposity, received the highest weight rating. Figure H ( M = 4.39, SD = 0 .03) was intermediate and figure D ( M = 4.17, SD = 0 .03), the figure with the lowest level of adiposity, received the lowest weight rating All three means differed significantly. The analysis also revealed a significant main effect of target race ( F (2,603) =22.25, p < .001, partial 2 = .07), which must also be qualified by the significant adiposity by target race interaction. Post hoc test s revealed that when the targets were presented as Caucasian ( M = 4.61, SD = 0 .03) and African American ( M = 4.48, SD = 0 .03), they did not differ significantly. Both, however, received significantly higher weight ratings than did the targets presented as Hispanic ( M = 4.45, SD = 0 .03). Finally, figures G, K, and O, all at the lowest level of adiposity but varying in muscularity were compared. There was a significant interaction between adiposity and target race ( F (4,1180) = 11.13, p < .001, partial 2 = 0 03). Post hoc examination of the means showed that for Figure G, the highest weight rating was assigned when the target was African American ( M = 3.78, SD = 0 .04) with the Hispanic target ( M = 3.77, SD = 0 .04) receiving an intermediate rating and the Cauca sian target ( M = 3.74, SD = 0 .04) receiving the lowest rating. For figure K, the highest weight rating was assigned when the target was African American ( M = 4.49, SD = 0 .04) with the Hispanic target ( M = 4.46, SD = 0 .03) receiving an intermediate rating a nd the Caucasian target ( M = 4.37, SD = 0 .04) receiving the lowest rating. For figure O, the highest weight rating was assigned when the target was Caucasian ( M = 5.40, SD = 0 .05) with the African American target ( M = 5.08, SD = 0 .05) receiving an intermed iate rating and the Hispanic target ( M = 5.32, SD = 0 .05) receiving the lowest rating. For this set of figures, there was a main effect of adiposity ( F (2,540) = 772.28, p < .001, partial 2 = .71), which must be qualified by the

PAGE 49

37 significant adiposity by ta rget race interaction. Post hoc LSD tests revealed that figure O ( M = 5.26, SD = 0 .04), the figure with the highest level of adiposity, received the highest weight rating. Figure K ( M = 4.44, SD = 0 .03) was intermediate and figure G ( M = 3.77, SD = 0 .03), the figure with the lowest level of adiposity, received the lowest weight rating. All three means differed significantly. The analysis also revealed a significant main effect of target race ( F (2,616) = 6.32, p < .01, partial 2 = .02), which must also be q ualified by the significant adiposity by target race interaction. Post hoc tests revealed that when the targets were presented as Caucasian ( M = 4.50, SD = 0 .03) and African American ( M = 4.53, SD = 0 .02), they did not differ significantly. Both, however, received significantly higher weight ratings than did the targets presented as Hispanic ( M = 4.44, SD = 0 .03). There was also a significant effect of race in these analyses ( F (2,314) =8.09, p < .001, partial 2 = .05). Post hoc tests revealed that Caucasia n ( M = 4.56, SD = 0 .03) and Hispanic ( M = 4.54, SD = 0 .04) raters did not differ but both provided significantly higher ratings than did African American raters ( M = 4.37, SD = 0 .04). Male Figures Three sets of three male figures were compared because they shared a common level of adiposity or muscularity. First, figures P, Q, and R, all at the lowest level of adiposity but varying in muscularity, were compared. There was a significant interaction of target race and muscularity ( F (4,1076) = 5.18, p < .01, p artial 2 = .02). Post hoc LSD tests of the interaction revealed that for figure P, the highest weight rating was assigned when the target was Caucasian ( M = 4.25, SD = 0 .03) with the African American target ( M = 4.18, SD = 0 .03) receiving an intermediate rating and the Hispanic target ( M = 4.10, SD = 0 .03) receiving the lowest rating. For figure Q, the highest weight rating was assigned when the target was Caucasian ( M = 4.05, SD = 0 .03) with the African American target ( M = 4.01, SD = 0 .03) receiving an i ntermediate rating and the Hispanic target ( M = 3.99, SD = 0 .02) receiving the lowest rating. For figure R, the highest weight rating was assigned when the target was Caucasian ( M = 3.45, SD = 0 .04) with the Hispanic target ( M = 3.42, SD = 0 .05) receiving an intermediate rating and the African American target ( M = 3.26, SD = 0 .05) receiving the lowest rating. This analysis also revealed a significant effect of muscularity ( F (2,568) = 318.86, p < .001, partial 2 = .50),

PAGE 50

38 which must be qualified by the signi ficant muscularity by target race interaction. Post hoc LSD comparisons revealed that figure P ( M = 4.18, SD = 0 .02), the figure with the highest muscularity, received the highest rating. Figure Q ( M = 4.02, SD = 0 .02) was intermediate and figure R, the fi gure with the lowest muscularity, received the lowest rating ( M = 3.38, SD = 0 .03). All three means differed significantly. The analysis also revealed a significant main effect of target race ( F (2,616) = 8.79, p < .001, partial 2 = .03), which must also b e qualified by the significant adiposity by target race interaction. Post hoc tests revealed that when the targets were presented as Caucasian ( M = 3.92, SD = 0 .02) and Hispanic ( M = 3.84, SD = 0 .02), they did not differ significantly. Both, however, recei ved significantly higher weight ratings than did the targets presented as African American ( M = 3.82, SD = 0 .02). There was also a significant main effect of gender in these analyses ( F (2,316) = 9.23, p < .01, partial 2 = .03). Post hoc LSD tests revealed that, overall, females ( M = 3.81, SD = 0 .02) provided lower ratings than did males ( M = 3.91, SD = 0 .02). Next, figures S, W, and AA, all at the highest level of muscularity but varying in adiposity were compared. There was a significant interaction of ta rget race and adiposity ( F (4,1153) = 4.31, p < .01, partial 2 = .01). Post hoc LSD tests of the interaction revealed that for figure S, the highest weight rating was assigned when the target was African American ( M = 4.40, SD = 0 .04) with the Hispanic tar get ( M = 4.38, SD = 0 .04) receiving an intermediate rating and the Caucasian target ( M = 4.36, SD = 0 .04) receiving the lowest rating. For figure W, the highest weight rating was assigned when the target was African American ( M = 3.69, SD = 0 .05) with the Caucasian target ( M = 3.65, SD = 0 .05) receiving an intermediate rating and the Hispanic target ( M = 3.62, SD = 0 .05) receiving the lowest rating. For figure AA, the highest weight rating was assigned when the target was Hispanic ( M = 4.51, SD = 0 .04). The Caucasian ( M = 2.35, SD = 0 .04) and African American ( M = 2.21, SD = 0 .04) targets did not differ significantly but both received significantly lower ratings than did the Hispanic target. This analysis also revealed a significant effect of adiposity ( F (2 ,437) = 217.34, p < .001, partial 2 = .42), which must be qualified by the significant adiposity by target race interaction. Post hoc LSD comparisons revealed that figure AA ( M = 4.40, SD = 0 .03), the figure with the

PAGE 51

39 highest adiposity, received the highes t rating. Figure S ( M = 4.38, SD = 0 .03), the figure with the lowest adiposity, was intermediate and figure W received the lowest rating ( M = 3.66, SD = 0 .04) All three means differed significantly. There was also a significant main effect of gender in the se analyses ( F (2,306) = 13.74, p < .001, partial 2 = .04). Post hoc LSD tests revealed that, overall, females ( M = 4.07, SD = 0 .03) provided lower ratings than did males ( M = 4.22 SD = 0 .03). Finally, there was also a significant effect of race in these a nalyses ( F (2,306) = 6.68, p < .01, partial 2 = .04). Post hoc tests revealed that Caucasian ( M = 4.21, SD = 0 .03) and Hispanic ( M = 4.20, SD = 0 .04) raters did not differ but both provided significantly higher ratings than did African American raters ( M = 4.03, SD = 0 .04). Finally, figures V, Z, and DD, all at the lowest level of adiposity but varying in muscularity were compared. There was a significant interaction between adiposity and target race ( F (4,1227) = 15.91, p < .001, partial 2 = .05). Post hoc examination of the means showed that for Figure V, the highest weight rating was assigned when the target was African American ( M = 3.69, SD = 0 .05) with the Caucasian target ( M = 3.64, SD = 0 .04) receiving an intermediate rating and the Hispanic target ( M = 3.62, SD = 0 .05) receiving the lowest rating. For figure Z, the highest weight rating was assigned when the target was Hispanic ( M = 4.80, SD = .04) with the Caucasian target ( M = 4.62, SD = 0 .05) receiving an intermediate rating and the African Americ an target ( M = 4.53, SD = 0 .05) receiving the lowest rating. For figure DD, the highest weight rating was assigned when the target was Caucasian ( M = 5.95, SD = 0 .05) with the Hispanic target ( M = 5.84, SD = 0 .05) receiving an intermediate rating and the A frican American target ( M = 5.49, SD = 0 .05) receiving the lowest rating. For this set of figures, there was a main effect of adiposity ( F (2,606) = 1031.01, p < .001, partial 2 = .76), which must be qualified by the significant adiposity by target race in teraction. Post hoc LSD tests revealed that figure DD ( M = 5.76, SD = 0 .04), the figure with the highest level of adiposity, received the highest weight rating. Figure Z ( M = 4.65, SD = 0 .04) was intermediate and figure V ( M = 3.65, SD = 0 .03), the figure with the lowest level of adiposity, received the lowest weight rating. All three means differed significantly. The analysis also revealed a significant main effect of target race ( F (2,628) =21.18, p < .001, partial 2 = .06), which

PAGE 52

40 must also be qualified b y the significant adiposity by target race interaction. Post hoc tests revealed that when the targets were presented as Caucasian ( M = 4.74, SD = 0 .03) and Hispanic ( M = 4.75, SD = 0 .03), they did not differ significantly. Both, however, received significa ntly higher weight ratings than did the targets presented as African American ( M = 4.57, SD = 0 .03). Finally, there was also a significant effect of race in these analyses ( F (2,319) = 15.08, p < .001, partial 2 = .09). Post hoc tests revealed that Caucasi an ( M = 4.77, SD = 0 .04) and Hispanic ( M = 4.79, SD = 0 .04) raters did not differ but both provided significantly higher ratings than did African American raters ( M = 4.50, SD = 0 .04). Follow Up Weight Analyses Individual figures were examined next. For th e sake of parsimony, only significant effects are discussed below. All means, standard deviations and standard errors, F values, p values, and partial 2 values for all of the analyses for each figure are presented in Appendix L. Please refer to Appendix A to match the figures to their letter labels. The findings for the ANOVA are presented first, followed by the ANCOVA. Significant effects related to target race (the repeated factor) are discussed first, because of their relevance for the hypotheses. Repea ted Measures Effects: Three Way Interactions There were no significant three way interactions between target race, rater race, and rater gender. One male figure, figure P, showed a trend towards significance for this interaction effect ( F (4,611) = 3.54, p = .008, partial 2 = .02 ). This trend remained when each of the covariates was entered into the equation. For each target race, the gender means were examined at each level of participant race. These post hoc LSD analyses revealed that when the target figu re was Hispanic, mean ratings differed for Hispanic male ( M = 4.22, SD = 0 .73) and female ( M = 3.98, SD = 0 .50) raters such that Hispanic males gave a higher weight rating. Additionally, when the target was Caucasian, mean ratings differed for African Amer ican male ( M = 4.37, SD = 0 .67) and female ( M = 4.05, SD = 0 .35) raters such that African American females gave a lower weight rating. When BMI was covaried, one additional difference was found. In this analysis, when the target

PAGE 53

41 was Hispanic, African Amer ican male ( adjusted M = 3.96, SE = 0 .09) and female ( adjusted M = 4.21, SE = 0 .07) raters differed significantly, with African American males giving a lower weight rating. When the PACS was covaried, the difference between the mean weigh t ratings for Hisp anic male and female raters was no longer significant. None of the results for the female figures support ed a trend towards significance Figure D ( F (4,640) = 3.22, p = .013, partial 2 = .02 ) was closest to reaching significance. Repeated Measures Effects : Two Way Interactions Target race did not significantly interact with rater race or rater gender for any of the figures. The only significant two way interaction at the repeated measures level was with covariates. For figure DD, the male figure with the h ighest level of body fat and lowest level of muscularity, BMI significantly interacted with target race ( F (2, 619) = 10.27, p < .001 partial 2 = .03 ). Repeated Measures Effects: Main Effects of Target Race Female Stimuli There was a main effect of targ et race for Figure A ( F (2, 634) = 6.56, p = .002, partial 2 = .02). For figure A, the rating of the Caucasian target ( M = 3.88, SD = 0 .86) did not differ significantly from the rating for the Hispanic target ( M = 3.74, SD = 0 .82). Both of these ratings, h owever, were significantly lower than the rating given to the African American target ( M = 3.91, SD = 0 .83). This significant effect disappeared when each of the covariates was entered into the equation. Figure D also demonstrated a main effect for target race ( F (2, 638) = 18.36, p < .001 partial 2 = .05 ) For this figure, the rating given to the African American target ( M = 2.11, SD = 1.13) was significantly lower than the rating given to the Caucasian target ( M = 2.37, SD = 1.16), which was significantly lower than the rating given to the Hispanic target ( M = 2.59, SD = 1.09). All three means were significantly different in this case. This effect remained significant only when the SATAQ was entered as a covariate. The effect disappeared when each of the o ther covariates was entered into the equation. Figure H, too, demonstrated a main effect of target race ( F (2, 618) = 10.34, p < .001 partial 2 = .03 ). All three means differed significantly with the rating given to the African American target ( M = 4.27, S D = 0 .63)

PAGE 54

42 lowest and the rating given to the Caucasian target ( M = 4.49, SD = 0 .75) highest. The rating given to the Hispanic target was intermediate ( M = 4.39, SD = 0 .63). This significant main effect disappeared when each of the covariates was entered in to the equation. The two female figures with the highest level of adiposity also showed a main effect of target race. For figure L ( F (2, 634) = 17.31, p < .001 partial 2 = .05), the mean ratings given to the Caucasian target ( M = 5.10, SD = 0 .77) and the African American target ( M = 5.04, SD = 0 .70) did not differ significantly. However, both were significantly higher than the rating assigned to the Hispanic target ( M = 4.81, SD = 0 .75). This effect no longer reached significance when each of the covariat es was entered into the equation. For figure O ( F (2,6 46) = 17.45, p < .001 partial 2 = .05) the ratings demonstrated the same pattern wherein the mean ratings assigned to the Caucasian ( M = 5.39, SD = 0 .79 ) and African American ( M = 5.29, SD = 0 .81) targ ets did not differ significantly but both were higher than that given to the Hispanic target ( M = 5.10, SD = 0 .90 ). Once again, this significant effect disappeared when each of the covariates was entered into the equation. Repeated Measures Effects: Main E ffects of Target Race Male Stimuli For the male figures, there was a significant main effect of target race for figure P ( F (2, 611) = 6.74, p = .002, partial 2 = .02), however this main effect must be qualified by the trend towards a significant three wa y interaction. Nonetheless, post hoc LSD tests revealed that the mean ratings assigned to the Hispanic ( M = 4.10, SD = 0 .59 ) and African American ( M = 4.18, SD = 0 .61 ) targets did not differ significantly while both were significantly lower than the mean r ating given to the Caucasian target ( M = 4.23, SD = 0 .54 ). This significant effect disappeared when each of the covariates was entered into the equation. Figure R also displayed a significant main effect of target race ( F (2, 619) = 7.12, p = .001, partial 2 = .02). For this figure, post hoc LSD tests revealed that the mean rating given to the Caucasian target ( M = 3.44, SD = 0 .78 ) and the Hispanic target ( M = 3.43, SD = 0 .78 ) did not differ significantly but they were both higher than the rating assigned to the African American target ( M = 3.27, SD = 0 .80 ). The main effect of target race was no longer significant when each of the covariates was entered into the equation.

PAGE 55

43 Figure W also showed a significant main effect of target race ( F (2, 600) = 12.83, p < .00 1 partial 2 = .04). For this figure, the mean rating given to the Caucasian target ( M = 4.30, SD = 0 .71 ) did not differ significantly from the mean rating assigned to the African American target ( M = 4.38, SD = 0 .61 ). Both of these mean ratings were sign ificantly lower than that assigned to the Hispanic target ( M = 4.51, SD = 0 .61 ). This effect was no longer significant when each of the covariates was entered into the analysis. For figure Z, the main effect of target race ( F (2, 642) = 12.84, p < .001 part ial 2 = .04) was such that the mean ratings assigned to the Caucasian ( M = 4.61, SD = 0 .87) and African American ( M = 4.5 4, SD = 0 .87) targets did not differ significantly. However, the mean rating for the Hispanic target ( M = 4.77, SD = 0 .79) was signifi cantly higher than for the other two targets. This effect remained significant when the EDI BD was entered as a covariate but failed to reach significance when each of the other covariates was entered into the model. As with the female targets, there was a significant effect of target race for both of the male targets at the highest level of adiposity. For figure AA ( F (2, 646) = 36.20, p < .001 partial 2 = .10), the post hoc tests revealed that the mean ratings for the Caucasian ( M = 5.95, SD = 0 88) and Hispanic ( M = 5.86, SD = 0 .85 ) targets were not significantly different but were both higher than the rating for the African American target ( M = 5.50, SD = 0 .87 ). This significant effect disappeared when each of the covariates was entered into the model, with the exception of the SATAQ. There remained a significant effect of target race when the SATAQ was entered into the equation. For figure DD, the main effect of target race was nonsignificant with no covariates in the model ( F (2, 618) = 0.01, p = .901, p artial 2 < .01 ). Only when BMI was entered into the analysis did the effect become significant ( F (2, 619) = 9.95, p < .001 partial 2 = .03). Despite the fact that the effect was significant, pairwise comparisons indicated that there were no significant differences among the adjusted means for the Caucasian ( adjusted M = 5.63, SE = 0 .05), Hispanic ( adjusted M = 5.64, SE = 0 .05), and African American ( adjusted M = 5.61, SE = 0 .05) targets. Between Subjects Effects: Rater Gender X Rater Race Interaction Onl y one figure, figure Q, displayed a trend towards a two way interaction between rater race and rater gender. This trend was the strongest when the SATAQ was

PAGE 56

44 entered as a covariate ( F (2, 323) = 4.72, p = .010, partial 2 = .03 ). Pairwise comparisons of gende r for each racial group indicated that African American females ( adjusted M = 4.10, SE = 0 .05 ) gave lower weight ratings to figure Q than did African American males ( adjusted M = 3.95, SE = 0 .03 ). Between Subjects Effects: Main Effect of Rater Gender Fem ale Stimuli Several female figures displayed a main effect of rater gender collapsed across target race for each figure. For figure B ( F (1, 322) = 9.25, p = .003, partial 2 = .03), post hoc tests revealed that females ( M = 3.74, SD = 0 .58 ) gave a higher we ight rating to the target than did males ( M = 3.50, SD = 0 .78). This significant main effect of gender was found when the EDI BD, the PACS, and the SATAQ were covaried but not when BMI or the DMS were covaried. For figure G ( F (1, 323) = 10.65, p = .001, par tial 2 = .03), the post hoc tests revealed that males ( M = 3.88, SD = 0 .44 ) assigned a higher weight rating than did females ( M = 3.67, SD = 0 .52 ). This significant main effect of gender disappeared when the DMS was entered as a covariate but it remained significant when each of the other covariates were entered into the equation. For figure H ( F (1, 323) = 8.44, p = .004, partial 2 = .03), post hoc tests once again revealed that males ( M = 4.48, SD = 0 .52 ) assigned a higher weight rating to the figure than did females ( M = 4.31, SD = 0 .43 ). This effect failed to reach significance when the DMS was covaried and only trended towards significance when the EDI BD was covaried but remained significant with each of the other covariates in the model. Between Subje cts Effects: Main Effect of Rater Gender Male Stimuli A main effect of rater gender collapsed across target race for each figure was also found for several male figures. For figure R, the main effect of rater gender was not significant when there were no covariates in the model ( F (1, 320) = 5.24, p = .023, partial 2 = .02 ) and reached significance only when the DMS was entered as covariate ( F (1, 315) = 10.56, p = .001, partial 2 = .03). Pairwise comparisons indicated that males ( adjusted M = 3.51, SE = 0 06) assigned a higher weight rating to the target than did females ( adjusted M = 3.26, SE = 0 .05). For figure V, there was no significant main effect of rater gender without covariates in the equation ( F (1, 323) = 3.17, p = .076,

PAGE 57

45 partial 2 = .01) and there was a trend towards significance when the DMS was entered as a covariate ( F (1, 318) = 7.24, p = .008, partial 2 = .02 ). Post hoc LSD tests indicated that males ( adjusted M = 3.76, SE = 0 .06) gave a higher weight rating than did females ( adjusted M = 3.54, SE = 0 .05). For figure W, the main effect of rater gender was significant with no covariates in the model ( F (1, 317) = 8.87, p = .003, partial 2 = .03 ). Post hoc tests revealed that males ( M = 4.48, SD = 0 .53) gave higher weight ratings than did females ( M = 4.32, SD = 0 .43). This effect remained significant when the SATAQ was entered as a covariate and showed a trend towards significance when both BMI and the EDI BD were covaried. It failed to reach significance when the DMS and the PACS were covaried. Fo r figure Z, there was no main effect of rater gender when there were no covariates in the equation ( F (1, 325) = 6.29, p = .013, partial 2 = .02). There was a strong trend towards a significant effect when both the EDI BD ( F (1, 324) = 6.91, p = .009, partial 2 = .02 ) and the SATAQ ( F (1, 322) = 7.43, p = .007, partial 2 = .02 ) were entered as covariate s In both cases, males ( adjusted M = 4.75, SE = 0 .06) assigned a higher weight rating than did females ( adjusted M = 4.55, SE = 0 .05). Between Subjects Effects : Main Effect of Rater Race Female Stimuli Three female figures showed a main effect of rater race collapsed across target race for each figure. Figure K showed this significant effect with no covariates entered ( F (2, 320) = 6.59, p = .002, partial 2 = 04). Post hoc LSD tests revealed that Caucasian raters ( M = 4.54, SD = 0 .44 ) gave higher weight ratings than did African American raters ( M = 4.33, SD = 0 .39 ). This effect remained significant with each covariate entered into the model with the exception o f the SATAQ. The effect failed to reach significance when the SATAQ was entered as a covariate. For figure L, the main effect of target race was also significant with no covariates in the model ( F (2, 321) = 5.42, p = .005, partial 2 = .04). Post hoc tests revealed that the mean rating assigned by Caucasian ( M = 5.06, SD = 0 .56 ) and Hispanic ( M = 5.05, SD = 0 .51 ) raters did not differ significantly but both were higher than ratings provided by African American raters ( M = 4.81, SD = 0 .55 ). This effect remain ed significant with the EDI BD entered as a covariate. When BMI and the DMS were covaried, there was a trend for the effect to reach significance. The effect failed to reach significance when the PACS and the SATAQ were entered as covariates.

PAGE 58

46 Figure O also displayed a significant main effect of rater race with no covariates in the model ( F (2, 325) = 5.53, p = .004, partial 2 = .03). Pairwise comparisons revealed that the mean rating assigned by Caucasian ( M = 5.34, SD = 0 .57 ) and Hispanic ( M = 5.34, SD = 0 60 ) raters did not differ significantly but both were higher than ratings provided by African American raters ( M = 5.08, SD = 0 .68 ). This effect remained significant when the DMS was covaried. When the EDI BD was covaried, there was a trend for the effect to reach significance. The significant effect disappeared when BMI, the PACS, and the SATAQ were entered as covariates. Between Subjects Effects: Main Effect of Rater Race Male Stimuli Two male figures also showed a main effect of rater race collapsed ac ross target race for each figure. For figure V, the effect was significant with no covariates in the model ( F (2, 323) = 10.32, p < .001 partial 2 = .06) and remained significant with each of the covariates entered. Post hoc tests revealed that that the me an rating assigned by Caucasian ( M = 3.72, SD = 0 .53 ) and Hispanic ( M = 3.78, SD = 0 .54 ) raters did not differ significantly but both were higher than ratings provided by African American raters ( M = 3.42, SD = 0 .71 ). For figure Z, the main effect was also significant with no covariates in the model ( F (2, 325) = 8.67, p < .00 1 partial 2 = .05) and remained significant with each of the covariates entered. Post hoc tests revealed that that the mean rating assigned by Caucasian ( M = 4.69, SD = 0 .64 ) and Hispa nic ( M = 4.79, SD = 0 .52 ) raters did not differ significantly but both were higher than ratings provided by African American raters ( M = 4.41, SD = 0 .74 ). Between Subjects Effects: Significant Covariates There was only one significant covariate in all of t he analyses performed. For figure B, BMI was a significant covariate ( F (1,320) = 10.73, p = .001, partial 2 = .03). There was a strong trend for BMI to be a significa nt covariate for figure A ( F (1, 320) = 7.22, p = .008, partial 2 = .02).

PAGE 59

47 Adiposity and Mu scularity Analyses Means and standard deviations for the overall weight ratings of the figures examined are presented in Table 8. Table 9 presents the results of the t tests for the full sample and each subgroup. Table 9. Means and standard deviations of weight ratings for pairs of figures Figure Mean Standard Deviation Full Sample L 4.99 0 .55 O 5.27 0 .62 H 4.38 0 .48 K 4.45 0 .43 D 4.16 0 .54 G 3.76 0 .49 AA 5.78 0 .63 DD 5.62 0 .68 W 4.39 0 .47 Z 4.64 0 .66 S 4.37 0 .55 V 3.65 0 .61 Caucasian Males L 5.10 0 .50 O 5.38 0 .59 H 4.54 0 .45 K 4.56 0 .43 D 4.23 0 .71 G 3.93 0 .44 AA 5.88 0 .61 DD 5.80 0 .72 W 4.52 0 .49

PAGE 60

48 Figure Mean Standard Deviation Z 4.91 0 .54 S 4.59 0 .68 V 3.87 0 .56 Caucasian Females L 5.03 0 .59 O 5.31 0 .56 H 4.39 0 .48 K 4.53 0 .46 D 4.15 0 .44 G 3.71 0 .41 AA 5.81 0 .61 DD 5.60 0 .65 W 4.36 0 .39 Z 4.53 0 .68 S 4.30 0 .42 V 3.61 0 .49 Hispanic Males L 5.08 0 .59 O 5.37 0 .69 H 4.46 0 .58 K 4.52 0 .45 D 4.22 0 .68 G 3.90 0 .50 AA 5.69 0 .63 DD 5.75 0 .61 W 4.44 0 .55 Z 4.78 0 .54 S 4.44 0 .62

PAGE 61

49 Figure Mean Standard Deviation V 3.82 0 .51 Hi spanic Females L 5.03 0 .42 O 5.30 0 .51 H 4.29 0 .41 K 4.36 0 .42 D 4.07 0 .50 G 3.72 0 .49 AA 5.93 0 .53 DD 5.66 0 .58 W 4.34 0 .43 Z 4.78 0 .50 S 4.38 0 .50 V 3.77 0 .58 African American Males L 4.92 0 .51 O 5.16 0 .60 H 4.39 0 .53 K 4.34 0 .43 D 4.17 0 .52 G 3.73 0 .31 AA 5.64 0 .56 DD 5.53 0 .58 W 4.49 0 .57 Z 4.51 0 .76 S 4.39 0 .63 V 3.44 0 .53 African American Females

PAGE 62

50 Figure Mean Standard Deviation L 4.78 0 .56 O 5.07 0 .72 H 4.25 0 .41 K 4.33 0 .37 D 4.11 0 .37 G 3.61 0 .61 AA 5.67 0 .73 DD 5.41 0 .80 W 4.24 0 .43 Z 4.38 0 .73 S 4.22 0 .48 V 3.40 0 .78 Table 10. P aired sample t tests Group t df, and p values Figures L and O Full sample t (324) = 8.03, p < .001 Caucasian Males t (51) = 2.85, p = .006 Caucasian Females t (73) = 4.29, p < .001 Hispanic Males t (49) = 2.99, p = .004 Hispanic Females t (52) = 3.57, p = .001 African American Males t (30) = 2.20, p = 035 African American Females t (64) = 3.53, p = .001 Figures H and K Full sample t (321) = 2.19, p = .030 Caucasian Males t (51) = 0.28, p = .785 Caucasian Females t (69) = 2.03, p = 046

PAGE 63

51 Group t df, and p values Hispanic Males t (48) = 0 .63, p = .535 Hispanic Females t (52) = 1.05, p = .297 African American Males t (32) = 0.66, p = .516 African American Females t (64) = 1.22, p = .226 Figures D and G Full sample t (324) = 10.81, p < .001 Caucasian Males t (54) = 2.98, p = 004 Caucasian Females t (71) = 7.42, p < .001 Hispanic Males t (48) = 2.82, p = .007 Hispanic Females t (51) = 3.88, p < .001 African American Males t (32) = 3.95, p < .00 1 African American Females t (63) = 6.12, p < .001 Figures AA and DD Full sample t (319) = 4.34, p < .001 Caucasian Males t (50) = 0.76, p = .450 Caucasian Females t (71) = 2.91, p = .005 Hispanic Males t (49) = 0.73, p = 468 Hispanic Females t (51) = 3.17, p < .001 African American Males t (29 ) = 1.31, p = .202 African American Females t (64) = 3.02, p = .004 Figures W and Z Full sample t (320) = 5.83, p < .001 Caucasian Males t (51) = 4.04, p < .001 Caucasian Females t (72) = 1.92, p = .059 Hispanic Males t (48) = 3.10, p = .003 Hispanic Females t (50) = 4.92, p < .001 African American Males t (32) = 0.14, p = .892 African American Females t (62) = 8.03, p = .232

PAGE 64

52 Group t df, and p values Figures S and V Full sample t (319) = 15.76, p < .001 Caucasian Males t (51) = 5.97, p < .001 Caucasian Females t (71) = 9.1 2, p < .001 Hispanic Males t (47) = 5.32, p < .001 Hispanic Females t (52) = 5.99, p < .001 African American Males t (29) = 5.58, p < .001 African American Females t (64) = 6.92, p < .001 For figures L and O, the two female figures with the largest degre e of adiposity, t tests revealed significant differences for the full sample and all subgroups with the exception of African American males. In all cases, the mean rating for the weight of figure L was lower than the figure for rating O, which is consisten t with the hypothesis that a figure with greater muscularity will be assigned a lower weight rating than a figure with equal adiposity but a lower level of muscularity. For figures H and K, the female figures with an intermediate level of adiposity, no sig nificant differences in weight ratings were found in any of the groups examined. For figures D and G, the female figures with the lower level of adiposity, t tests revealed significant differences among weight ratings for the full sample and for each of the subgroups. In this case, the ratings for figure G, the figure with the lower level of muscularity, were lower than those for figure D, the figure with greater muscularity. These findings do not seem to support the hypothesis described above. For the ma le figures with the highest level of adiposity, figures AA and DD, significant differences in weight ratings were detected in the full sample, among Caucasian females, Hispanic females, and African American females. In those groups where a significant diff erence was detected, the ratings for figure AA, the figure with the higher level of muscularity, were lower than the ratings for figure DD, the figure with a lower level of muscularity. These findings support the hypothesis presented above. No differences were detected in any of the male subgroups.

PAGE 65

53 For figures W and Z, the male figures with an intermediate level of adiposity, t tests revealed significant differences in weight ratings in the full sample, Caucasian males, Hispanic males, and Hispanic females In the groups where differences were detected, the direction of the effect was such that the figure with the higher level of muscularity, figure W, was given a lower weight rating than figure Z, the figure with the lower level of muscularity. Again, thes e findings support the above mentioned hypothesis. No significant differences were found among Caucasian females or African American males or females. For figures S and V, the male figures with the lower level of adiposity, significant differences in weigh t ratings were found in the full sample and in all subgroups. In this case, the ratings for figure V, the figure with the lower level of muscularity, were lower than those for figure S, the figure with greater muscularity. These findings do not support the hypothesis above. Additional Analyses: Health and Attractiveness Data F values, p values, and partial 2 values for the analyses of the health data are presented in Appendix M. The same information for the attractiveness data is presented in Appendix N. A ll significant results are presented. These results are discussed in Appendix O.

PAGE 66

54 Discussion This study was designed to investigate the effect of target race, rater race, and rater gender on perceptions of weight status. It was hypothesized that these feat ures would interact to affect weight ratings assigned to male and female targets of various racial groups. Secondarily, the effects of muscularity and adiposity on weight ratings were examined. It was hypothesized that two targets with equal adiposity but different levels of muscularity would receive different weight ratings such that the figure with greater muscularity would be rated as less heavy. Finally, an exploratory aspect of this study was to examine whether social norms, appearance comparison, body and muscularity dissatisfaction, and rater BMI would act as covariates for the target weight ratings. Weight Analyses The first hypothesis regarding the interaction of target and rater features was largely unsupported. Only one of the male figures showed a trend towards a three way interaction between target and rater features. There were also no two way interactions between target race and rater gender or rater race. Only one male figure showed a trend towards an interaction between rater gender and rate r race. The findings did, on the other hand, support main effects of target race and of rater gender and rater race on target weight ratings. For the individual female figures, there was a main effect of target race for each of the figures with the highes t level of muscularity as well as the figure with the highest level of adiposity and the lowest level of muscularity. For the individual male figures, the trend was less consistent. The four figures with the highest and next to highest levels of adiposity showed main effects of target race as did the two figures with the lowest level of adiposity at the highest and lowest levels of muscularity in that category. There was no consistent trend as to which race had the highest and which the lowest rating. The r ace of

PAGE 67

55 the target did appear to be an important factor in the rating of weight, but it did not appear to have a consistent effect on raters. Perhaps the race of the target interacts with the muscularity or adiposity of the target. The initial weight analys es show an interaction between either muscularity or adiposity and target race. This possibility should be considered more fully in future research to perhaps help clarify the effects of target race on weight ratings. A main effect of rater gender collapse d across levels of target race was found for three female figures and one male figure. One male figure showed a strong trend and three others showed either a significant effect or a strong trend only when covariates were entered. In all but one case, male raters gave higher weight ratings than did female raters. These findings are opposite to what is seen in the self perception literature where females tend to overestimate and males underestimate their weight ( Chang & Christakis, 2001; Chang & Christakis, 2 003; Gray, 1988; McCreary, 2002; Pritchard, King & Czajka Narins, 1994; Wardle & Johnson, 2002 ). One possibility may be that both males and females idealize the bodies of others. Research has shown that females are more likely to experience weight and shap e dissatisfaction ( Rodin et al., 1984 ), wishing that they could be thinner, while men are more likely to experience muscularity dissatisfaction ( Humphreys & Paxton, 2004; Thompson & Cafri, 2007 ), wishing they could be larger and more muscular. The ratings made of others' weight may indicate that they see others as closer to their ideals than they are themselves. A main effect of rater race collapsed across levels of target race was also found for several figures. In every case, the ratings given by Caucasi an raters were higher than those given by African American raters. In all but one case, the mean ratings given by Hispanic raters did not differ from those given by Caucasian raters. These findings seem consistent with the literature on weight acceptance i n African Americans. Overall, African American individuals seem more accepting of larger body sizes. African American individuals have less body dissatisfaction at larger sizes ( Kemper, Sargent, Drane, Valois, & Hussey, 1994; Neumark Sztainer et al., 2002 ) than do Caucasian individuals. Additionally, African American men report that they prefer larger women as mates ( Greenberger & LaPorte, 1996; Rosen et al., 1993 ). The findings of the present

PAGE 68

56 study indicate that these preferences may actually be based in a difference of perception. African American individuals may not see large body sizes as large as do Caucasian individuals. At the very least, their threshold for categorizing overweight may differ from that of Caucasian and perhaps Hispanic individuals. Ad iposity and Muscularity It was hypothesized that when the figures are equal in adiposity but differ in muscularity, a lower, less heavy weight rating w ould be given to the figure with the higher level of muscularity. Findings were mixed. The majority of th e analyses supported the hypothesis. However, for one pair of figures there were no significant differences between weight ratings and, for two pairs of figures, the difference, when significant was in the opposite direction such that the figure with the l ower level of adiposity received a lower weight rating. Visual inspection of the means for the nonsignificant effects indicat ed that, in all but one instance, the differences were in the hypothesized direction but were not large enough to produce a signifi cant effect. It is unclear why, for the pair of male figures and the pair of female figu r es with the lowest level of adiposity that the mean differences were not in the hypothesized direction. It is, perhaps, a feature of the stimuli used as targets. Becau se each of the somatomorphic matrix figures is based on a different individual, the two figures being compared are not, in fact, equal in every way with the exception of muscularity. They are close but, in the case of the pairs of figures with the lower le vel of adiposity, perhaps the differences were great enough that the raters noticed. It could also be that the effects of muscularity and adiposity interact such that their effects on weight ratings differ at the opposite ends of the weight spectrum. Covar iate Analyses Each covariate was included because research supports the idea that these dispositional factors affect weight self perception and, by extension, would likely affect perception of others. Those at the extremes of the BMI continuum are most lik ely to normalize their own weight when asked to categorize themselves (Gray, 1977). It was expected that the BMI of the rater would also affect perception of the target such that individuals with higher BMI would provide lower, less heavy weight ratings.

PAGE 69

57 Body shape and size dissatisfaction and muscularity dissatisfaction were also included as covariates. Body dissatisfaction plays a key role in etiological theories of eating disturbance (Shroff & Thompson, 2006; Stice, Nemeroff, & Shaw, 1996) and is a nece ssary criterion for the major forms of eating disorder (American Psychiatric Association, 2000). Body and muscularity dissatisfaction also appear to play an important role in the misperception of the self. In adolescents negative self esteem is related to perception of overweight status (Pritchard et al., 1997) and perception of a weight problem (Kim & Kim, 2001). This research attempted to explore the influence of body dissatisfaction in the perception of others. It was hypothesized that those high in body dissatisfaction would provide higher, heavier weight ratings of the targets. Appearance comparison was also included as a covariate. Social comparison theory was first proposed by Festinger (1954). The theory proposes that there are affective consequences of comparing oneself to another ( Buunk, Collins, Taylor, VanYperen, & Dakof, 1990). The direction of the comparison made, be it upward or downward, as well as the characteristics of the individual making the comparison, determine what the affective conseq uence will be. The tendency to compare oneself to others by necessity involves making a judgment of the other. Appearance comparison, therefore, might act as a covariate of the ratings made of the weight of others. It was hypothesized that those high in a ppearance comparison would provide lower ratings of the targets' weight. The final covariate of interest was social norms. Two measures of social norms were included in the present study. A new, exploratory social norms measure was created for the purpose s of this study. This measure, the Proximate Social Environment Rating Scale (PSERS) was developed with the goal of assessing social norms at a more proximate level than is usually done. Measure s of social norms tend to focus at the level of the culture o r subculture, as does the other measure of social norms used in the present study, the Sociocultural Attitudes Towards Appearance Questionnaire 3 (SATAQ; Thompson, van den Berg, Roehrig, Guarda, & Heinberg, 2004). The Internalization General subscale o f the SATAQ used in the present study assesses internalization of the thin ideal as perpetuated by the Western media. The PSERS, on the other hand, is

PAGE 70

58 designed to examine the influence of those closest to the participant by having the participant choose a figure that most resembles the body shape of each of his or her three closest friends. Research has found that those in one's social network have an effect on one's weight. Christakis and Fowler (2007) examined friendship and spousal networks and found tha t when the weight of a friend or a spouse increased, so did the weight of the proband. In our study no group differences were found for the PSERS and it was, therefore, not used as a covariate in any of the subsequent analyses. However, this scale is in it s infancy and should not be rejected out of hand. The mean adiposity of the participant's friend network was used as the PSERS score in this study. Perhaps a more sophisticated scoring system that incorporates information on both muscularity and adiposity might be developed for the PSERS. This measure taps a new and potentially important aspect of weight related social norms and should be further examined in future research. The SATAQ did show group differences and was, therefore, included as a covariate i n the weight rating analyses. Internalization of norms has a stronger relationship to body image than does awareness of norms (Cafri, Yamamiya, Brannick, & Thompson, 2005). Those who most enthusiastically buy into the cultural ideal of thinness are also mo st likely those who display the patterns of over and underestimation repeatedly seen in studies of weight self perception. By extension, idealization of thinness could affect the way that others are viewed. It was hypothesized that those higher in social norms would provide higher ratings of the target. There were some interesting findings with the covariates In virtually all cases, entering the covariates into the analyses rendered a significant effect nonsignificant. Once participants were equated on th e covariate, the effect was no longer significant, indicating that variability on the covariate had an effect on the ratings made. As expected, including such a diverse participant sample led to group differences on a variety of the dispositional variables ; entering these measures as covariates led to changes in significant effects, usually rendering previous effects nonsignificant. Table 10 shows which effects become nonsignificant for each covariate as well as those effects which became significant after a covariate was entered into the analysis and those that remained significant after the

PAGE 71

59 covariates were entered. The general trend was for each of the covariates to render nonsignificant the main effect of target race. Rater BMI, body dissatisfaction, musc ularity dissatisfaction, appearance comparison, and thin ideal internalization all seem to play an important role. The effects of the covariates on the main effects of rater gender and rater race are less consistent, but no less important. At times, the si gnificant effects remain significant, indicating that the covariates do not play a role. At other times, the covariates increase the p value to yield a trend or a significant effect where one was not found without covariates in the model. This particular h appening was, however, rare. Finally, at times, the covariates reduce the effect to a trend or to nonsignificance indicating that the covariates do play a role in these effects. Table 11. Effects of covariate variables on significance Figure Effect and Cov ariates Effect of Covariate A Main effect of target race BMI rendered nonsignificant EDI BD rendered nonsignificant DMS rendered nonsignificant PACS rendered nonsignificant SATAQ rendered nonsignificant B Main effect of rater gender B MI rendered nonsignificant EDI BD unchanged DMS rendered nonsignificant PACS unchanged SATAQ unchanged C Main effect of target race BMI rendered nonsignificant EDI BD rendered nonsignificant DMS rendered nonsignificant PACS render ed nonsignificant

PAGE 72

60 Figure Effect and Cov ariates Effect of Covariate SATAQ unchanged G Main effect of rater gender BMI unchanged EDI BD unchanged DMS rendered nonsignificant PACS unchanged SATAQ unchanged H Main effect of target race BMI rendered nonsignificant EDI BD rendered n onsignificant DMS rendered nonsignificant PACS rendered nonsignificant SATAQ rendered nonsignificant Main effect of rater gender BMI unchanged EDI BD reduced to a trend towards significance DMS rendered nonsignificant PACS unchanged SATAQ unchanged K Main effect of rater race BMI unchanged EDI BD unchanged DMS unchanged PACS unchanged SATAQ rendered nonsignificant L Main effect of target race BMI rendered nonsignificant EDI BD rendered nonsignificant

PAGE 73

61 Figure Effect and Cov ariates Effect of Covariate DM S rendered nonsignificant PACS rendered nonsignificant SATAQ rendered nonsignificant Main effect of rater race BMI reduced to a trend towards significance EDI BD unchanged DMS reduced to a trend towards significance PACS rendered nonsi gnificant SATAQ rendered nonsignificant O Main effect of target race BMI rendered nonsignificant EDI BD rendered nonsignificant DMS rendered nonsignificant PACS rendered nonsignificant SATAQ rendered nonsignificant Main effect of rate r race BMI rendered nonsignificant EDI BD reduced to a trend towards significance DMS unchanged PACS rendered nonsignificant SATAQ rendered nonsignificant P Three way interaction BMI unchanged EDI BD unchanged DMS unchanged P ACS unchanged SATAQ unchanged Main effect of target race

PAGE 74

62 Figure Effect and Cov ariates Effect of Covariate BMI rendered nonsignificant EDI BD rendered nonsignificant DMS rendered nonsignificant PACS rendered nonsignificant SATAQ rendered nonsignificant R Main effect of target race BMI rendered nonsignificant EDI BD rendered nonsignificant DMS rendered nonsignificant PACS rendered nonsignificant SATAQ rendered nonsignificant Main effect of rater gender DMS became significant S Main effect of rater gender BMI rendered nonsignificant from a trend EDI BD rendered nonsignificant from a trend DMS rendered nonsignificant from a trend PACS rendered nonsignificant from a trend SATAQ rendered nonsignificant from a trend V Main effect of rater gender DM S increased to a trend towards significance Main effect of rater race BMI unchanged EDI BD unchanged DMS unchanged PACS unchanged SATAQ unchanged

PAGE 75

63 Figure Effect and Cov ariates Effect of Covariate W Main effect of target race BMI reduced to a trend towards significance EDI BD reduced to a trend towards significance DMS rendered nonsignificant PACS rendered nonsignificant SATAQ unchanged Main effect of rater gender BMI rendered nonsignificant EDI BD rendered nonsignificant DMS rendered nonsignificant PACS rendered nonsignificant SATAQ rendered nonsignificant Z Main effect of target race BMI rendered nonsignificant DMS rendered nonsignificant PACS rendered nonsignificant SATAQ rendered nonsignificant Main effect of rater gender EDI BD increased to a trend towards significance SATAQ increased to a trend towards significance Main effect of rater race BMI unchanged EDI BD unchanged DMS unchanged PACS unchanged SATAQ unchanged AA Main effect of target race BMI ren dered nonsignificant

PAGE 76

64 Figure Effect and Cov ariates Effect of Covariate EDI BD rendered nonsignificant DMS rendered nonsignificant PACS rendered nonsignificant SATAQ unchanged DD Main effect of target race BMI became significant In sum, both the ANOVA and ANCOVA analyses provide valuable information about the nature of weight ratings made for targets that vary in gender and ethnicity by raters who also vary in gender and ethnicity. The ANOVA analyses provide a picture of the main effects while the ANCOVA analyses provide suggestions about the possible sources of some of the participant differences. Although none of the dispositional traits was consistently significant in the analyses, the pattern of results changed depending on which covariate was entered into the analysis. These sources o f variation provide a starting point for future research. Health and Attractiveness Analyses Two distractor items were included in the rating task to reduce the emphasis on weight. These distractor items asked participants to rate the health and attractiv eness of the target. Although there were no hypotheses regarding these items, they were analyzed in the hopes that they would yield some interesting findings. In fact, the analyses of the health and attractiveness ratings did reveal some interesting trends As with the weight analyses, the health and attractiveness analyses failed to yield any significant three way interactions between target race, rater gender, and rater race. There were, however, several two way interactions. The most robust of these int eractions occurred between target race and rater race. For the health data, the interaction effects occurred only for male figures. Regardless of the race of the target, Caucasians provided higher ratings of health than did African Americans. For the attra ctiveness interactions, there was also a tendency for Caucasians to give higher attractiveness ratings than

PAGE 77

65 African Americans. There were some in group preferences with African Americans rating figure H as more attractive when the target was African Americ an and Hispanics rating figure R more attractive when the target was Hispanic. These findings are consistent with in group preferences wherein individuals tend to rate members of their own group more positively than members of a different or out group (Fis ke, 2004). The main effects of rater gender and rater race also showed interesting trends. For health ratings, males provided higher ratings than females in all cases while, for attractiveness ratings, females assigned higher ratings than males in all cas es. For the attractiveness ratings, African American raters provided higher ratings than Caucasian and Hispanic raters for female figures and lower ratings than Caucasian a nd Hispanic raters for male figures. Interestingly, health and attractiveness ratin gs for figure AA, the male figure with the highest level of adiposity and muscularity, coincide exactly. That is, Caucasians rated the figure as least healthy and least attractive. African Americans found the figure healthiest and most attractive. Hispanic raters were intermediate. In this figure, at least, there is evidence to indicate that health and attractiveness are correlated in the eye of the beholder. Caucasian and Hispanic raters also rated this figure as heavier than did African American raters, p erhaps indicating that higher weight may be associated with poorer health and lower attractiveness for these racial groups. Some other interesting contrasts occurred. For figure W, Caucasian and African American raters did not differ in their ratings of we ight or health. Hispanic raters gave the highest ratings of weight and the lowest ratings of health. These findings are in line with what was found for figure AA. Higher weight ratings occurred with lower health ratings. For figure C, the female figure wit h the lowest level of adiposity and muscularity, health and attractiveness ratings coincided exactly, as they did for figure AA. Caucasian and African American raters did not differ in their ratings, but both groups gave lower ratings than did Hispanic rat ers. Limitations There are some important limitations to note in this study. First, only undergraduates were sampled. While the highest age in the sample was 46 years, the

PAGE 78

66 results may not generalize to adult and other non student samples. Second, while the sample was quite diverse, the small group size for some of the subgroups, particularly African American males, may have made it difficult to detect effects in some instances. Third, the weight ratings in this study were made on a relative scale. Therefore some level of value judgment may be implied in the rating that was made. It was established that social desirability was not correlated with the dependent measures, nor was it the source of group differences. However, there may still have been some level of implicit bias present in the ratings that were made. An objective scale, such as having raters estimate the weight of the target in pounds, could help to resolve this issue somewhat. In addition, one must consider the inherent limitations in the self r eport data. In the present study, it is of particular importance to note that both BMI and the rating of one's proximate social network were self reported. Such measurement may introduce a similar bias as the rating task itself. Because these measurements are limited to the perceptions of the raters, on e cannot be certain that they match the objective measurements. Finally, because the data are cross sectional and correlational or quasi experimental, causal inferences cannot be drawn. Future Research The pr esent study has revealed that the race of the target as well as the race and gender of the rater do play in important role in the way that the weight of others is perceived. These features also play a role in the perception of health and attractiveness. Fu ture research should continue to investigate the role that these factors play in the perception of weight. Rather than use stick figures, future research might benefit from the use of more realistic stimuli such as photographs altered with a graphics prog ram to produce different body shapes and sizes and target races while controlling for the target's appearance. Replication and extension of the current study with more realistic figures may help to clarify the role that target race, rater race, and rater g ender play in the perception of weight status. Future research shou ld also consider obtaining objec tive as well as subjective weight ratings. Further extending this line of research, future studies could investigate the way perception of weight causes the rater to act towards the target. This line of research may ultimately help us to identify an as yet unidentified

PAGE 79

67 interpersonal or cultural risk factor for weight related pathology such as eating disorders and obesity.

PAGE 80

68 References Akerman, A., Williams, M. E ., & Meunier, J. (2007). Perception versus reality: An exploration of children's measured body mass in relation to caregivers' estimates. Journal of Health Psychology, 12, 871 882. Al Sendi, A. M., Shetty, P., & Musaiger, A. O. (2004). Body weight percept ion among Bahraini adolescents. Child Care, Health & Development, 30, 369 376. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (4th edition, text revision) Washington, DC: Author; 2000. Barnett, H. L., Keel., P. K., & Conoscenti, L. M. (2002). Body type preferences in Asian and Caucasian college students. Sex Roles, 45, 867 878. Baughcum, A. E., Chamberlin, L. A., Deeks, C. M., Powers S. W., & Whitaker, R. C. (2000). Maternal perceptions of overweight preschool child ren. Pediatrics, 106, 1380 1386. BBC News. (2003, October 23). Stigma' of having obese friends. Retrieved January 21, 2007, from http://news.bbc.co.uk/2/hi/health/3207087.stm Bhuiyan, A. R., Gustat, J., Srinivasan, S. R., & Berenson, G. S. (2003). Diffe rences in body shape representations among young adults from a biracial (black white), semirural community. American Journal of Epidemiology, 158, 792 797. Buunk, B. P. Collins, R. L., Taylor, S. E., VanYperen, N. W., & Dakof, G. A. (1990). The aff ective consequences of social comparison: Either direction has its ups and downs. Journal of Personality and Social Psychology, 39, 1238 1249. Cafri, G., & Thompson, J. K. (2004). Evaluating the convergence of muscle appearance attitude measures. Assessmen t, 11 224 229.

PAGE 81

69 Cafri, G., & Thompson, J. K. (2007). Measurement of the muscular ideal. In J. K. Thompson and G. Cafri (Eds.), The muscular ideal: Psychological, social, and medical perspectives (pp.107 120) Washington, DC: American Psychological Associa tion. Cafri, G., Yamamiya, Y., Brannick, M., & Thompson, J. K. (2005). The influence of sociocultural factors on body image: A meta analysis. Clinical Psychology: Science and Practice, 12, 421 433. Campbell, M. W. C., Williams, J., Hampton, A., & Wake, M. (2004). Maternal concern and perceptions of overweight in Australian preschool aged children. Medical Journal of Australia, 184, 274 277. Carnell, S., Edwards, C., Croker, H., Boniface, D., & Wardle, J. (2005). Parental perceptions of overweight in 3 5 y olds. International Journal of Obesity, 29, 353 355. Chang, V. W., & Christakis, N. A. (2001). Extent and determinants of discrepancy between self evaluations of weight status and clinical standards. Journal of General Internal Medicine, 16, 538 543. Chang V. W., & Christakis, N. A. (2003). Self perception of weight appropriateness in the United States. American Journal of Preventive Medicine, 24, 332 339. Christakis, N., & Fowler, J. H. (2007). The spread of obesity in a large social network over 32 years New England Journal of Medicine, 357, 370 379. Cialdini, R. B. (2003). Crafting normative messages to protect the environment. Current Directions in Psychological Science, 12 105 109. Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155 15 9. Corning, A F., Krumm, A. J., & Smitham, L. A. (2006). Differential social comparison processes in women with and without eating disorder symptoms. Journal of Counseling Psychology, 53, 338 349. Crown, D. P., & Marlowe, D. (1964). The appearance motive: Studies in evaluative dependence. New York: Wiley. Davis, C., Karvinen, K., & McCreary, D. R. (2005). Personality correlates of a drive for muscularity in young men. Personality and Individual Differences, 39, 349 359.

PAGE 82

70 Dehghan, M., Akhtar Danesh, N., & Me rchant, A. T. (2005). Childhood obesity, prevalence and prevention. Nutrition Journal, 4. doi:10.1186/1475 2891 4 24 Desmond, S. M., Price, J. H., Hallinan, C., & Smith, D. (1989). Black and white adolescents' perceptions of their weight. Journal of School Health, 59, 353 358. Devlin, M. J., Yanovski, S. Z., & Wilson, G. T. (2000). Obesity: What mental health professionals need to know. American Journal of Psychiatry, 157, 854 866. Engeln Maddox, R. (2005). Cognitive responses to idealized media images of women: The relationship of social comparison and critical processing to body image disturbance in college women. Journal of Social and Clinical Psychology, 24, 1114 1138. Etelson, D., Brand, D. A., Patrick, P. A., & Shirali, A. (2003). Childhood obesity: Do parents recognize this health risk? Obesity Research, 11, 1362 1368. Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7 117 140. Fichter, M. M., & Quadflieg, N. (1999). Six year course and outcome of anorexia nervosa. I nternational Journal of Eating Disorders, 26, 359 385. Field, A. E., Barnoya, J., & Colditz, G. A. (2002). Epidemiology and health consequences of obesity. In T. A. Wadden & A. J. Stunkard (Eds.), Handbook of Obesity Treatment (pp.3 18). New York: Guilf ord Press. Fisher, L., Fraser, J., & Alexander, C. (2006). Caregivers' inability to identify childhood adiposity: A cross sectional survey of rural children and their caregivers' attitudes. Australian Journal of Rural Health, 14, 56 61. Fiske, S. T. (2004) Social beings: A core motives approach to social psychology Hoboken, NJ: John Wiley & Sons. Flegal, K. M., Carroll, M. D., Kuczmarski, R. J., & Johnson, C. L. (1998). Overweight and obesity in the United States: Prevalence and trends, 1960 1994. Inter national Journal of Obesity, 22, 39 47.

PAGE 83

71 Flegal, K. M., Wei R., & Ogden, C. (2002). Weight for stature compared with body mass index for age growth charts for the United States from the Centers for Disease Control and Prevention. American Journal of Clini cal Nutrition, 75, 761 766. Garner, D. M. (2004). Eating disorder inventory 3: Professional manual. Odessa, FL: Psychological Assessment Resources. Grabe, S. & Hyde, J. S. (2006). Ethnicity and body dissatisfaction among women in the United States: A met a analysis. Psychological Bulletin, 132, 622 640. Gray, S. H. (1977). Social aspects of body image: Perception of normalcy of weight and affect of college undergraduates. Perceptual and Motor Skills, 45, 1035 1040. Greenberg, D. R., & LaPorte, D. J. (1996) Racial differences in body type preferences of men for women. International Journal of Eating Disorders, 19, 275 278. Gruber, A. J., Pope, H. G., Borowiecki, J., & Cohane, G. (1999). The development of the somatomorphic matrix: A bi axial instrument for measuring body image in men and women. In T. S. Olds, J. Dollman, & K. I. Norton (Eds.), Kinanthropometry VI (pp.217 231). Sydney : International Society for the Advancement of Kinanthropometry. Hatoum, I J., & Belle, D. (2004). Mag and abs: Media consump tion and bodily concerns in men. Sex Roles, 51, 397 407. Heinberg, L. J. (1996). Theories of body image disturbance: Perceptual, developmental, and sociocultural factors. In J. K. Thompson (Ed.), Body Image, Eating Disorders, and Obesity: An Integrative Guide for Assessment and Treatment (pp.83 107). Washington, DC: American Psychological Association. Heinberg, L. J. & Thompson, J. K. (1992). Social comparison: Gender, target importance ratings, and relation to body image disturbance. Journal of Social B ehavior and Personality, 7, 335 344. Himes, J. H., & Dietz, W. H. (1994). Guidelines for overweight in adolescent preventive services Recommendations from an expert committee. American Journal of Clinical Nutrition, 59, 307 316.

PAGE 84

72 Hubert, H. B., Snider, J. & Winkelby, M. A. (2005). Health status, health behaviours, and acculturation factors associated with overweight and obesity in Latinos from a community and agricultural labor camp survey. Preventive Medicine, 40, 642 651. Humphreys, P., & Paxton, S. J. (2004). Impact of exposure to idealised male images on adolescent boys' body image. Body Image, 1, 253 266. Jackson, J., Strauss, C. C., Lee, A. A., & Hunter, K. (1990). Parents' accuracy in estimating child weight status. Addictive Behaviors, 15, 65 68. Jain, A., Sherman, S. N., Chamberlin, L. A., Carter, Y., Powers, S. W., & Whitaker, R. C. (2001). Why don't low income mothers worry about their preschoolers being overweight? Pediatrics, 107, 1138 1146. Jeffery, A. N., Voss, L. D., Metcalf, B. S., Alba, S ., & Wilkin, T. J. (2005). Parents' awareness of overweight in themselves and their children: Cross sectional study within a cohort (EarlyBird 21). British Medical Journal, 330, 23 24. Jones, D. C. & Crawford, J. K. (2005). Adolescent boys and body image : Weight and muscularity concerns as dual pathways to body dissatisfaction. Journal of Youth and Adolescence, 34, 629 636. Kaplan, S. L., Busner, J., & Pollack, S. (1988). Perceived weight, actual weight, and depressive symptoms in a general adolescent sampl e. International Journal of Eating Disorders, 7, 107 113. Kemper, K. A., Sargent, R. G., Drane, J. W., Valois, R F., & Hussey, J. R. (1994). Black and white females' perceptions of ideal body size and social norms. Obesity Research, 2, 117 126. Kim, K. J. (2007). Religion, weight perception, and weight control behavior. Eating Behavior, 8, 121 131. Kim, O., & Kim, K. (2001). Body weight, self esteem, and depression in Korean female adolescents. Adolescence, 36, 315 322. Kowner, R. (2004). When ideals are t oo "far off": Physical self ideal discrepancy and body dissatisfaction in Japan. Genetic, Social, and General Psychology Monograph, 130, 333 361.

PAGE 85

73 Krones, P. G., Stice, E., Batres, C., & Orjada, K. (2005). In vivo social comparison to a thin ideal promotes body dissatisfaction: A randomized experiment. International Journal of Eating Disorders, 38, 134 142. Lee, R. M., Harris, K. J., Catley, D., Shostrom, V., Choi, S., Mayo, M. S., Okuyemi, K., Kaur, H., & Ahluwalia, J. S. (2005). Factors associated with BM I, weight perceptions and trying to lose weight in African American smokers. Journal of the National Medical Association, 97, 53 61. Lorenzen, L. A., Grieve, F G., & Thomas, A. (2004). Exposure to male models decreases men's body satisfaction. Sex Roles, 51, 743 748. Lyubomirsky, S. & Nolen Hoeksema, S. (1993). Self perpetuating properties of dysphoric rumination. Journal of Personality and Social Psychology, 65, 330 349. Maynard, L. M., Galuska, D. A., Blanck, H. M., & Serdula, M. K. (2003). Maternal pe rceptions of weight status of children. Pediatrics, 111, 1226 1231. McCreary, D. R. (2002). Gender and age differences in the relationship between body mass index and perceived weight: Exploring the paradox. International Journal of Men's Health, 1, 31 42. McCreary, D. R. (2006). Drive for Muscularity Scale. Retrieved January 17, 2007 from http://spartan.ac.brocku.ca/~dmccrear/muscularity.html. McCreary, D. R., & Sasse, D. K. (2000). An exploration of the drive for muscularity in adolescent boys and girls. Journal of American College Health, 48 297 304. Morrison, T. G., Kalin, R., & Morrison, M. A. (2004). Body image evaluation and body image investment among adolescents: A test of sociocultural and social comparison theories. Adolescence, 39, 571 592. Morr ow, J. & Nolen Hoeksema, S. (1990). Effects of response to depression on the remediation of depressive affect. Journal of Personality and Social Psychology, 58 519 527.

PAGE 86

74 National Heart, Lung and Blood Institute. (1998). Clinical guidelines for the ident ification, evaluation, and treatment of overweight and obesity in adults: The evidence report [Electronic version]. Bethesda (MD): National Heart, Lung and Blood Institute. Retrieved November 11, 2003 from http://www.nhlbi.nih.gov/guidelines/obesity/ob_gdlns.htm Neumark Sztainer, D. Croll, J., Story, M., Hannan, P. J., French, S. A., & Perry, C. (2002). Ethnic/ racial differences in weight related concerns and behaviors among adolesc ent girls and boys: Findings from Project EAT. Journal of Psychsomatic Research, 53, 963 974. Ogden, C. L., Carroll, M. D., Curtin, L. R., McDowell, M. A., Tabak, C. J., & Flegal, K. M. (2006). Prevalence of overweight and obesity in the United States, 199 9 2004. Journal of the American Medical Association, 295, 1549 1555. Olivardia, R., Pope, H. G., Borowiecki, J. J., & Cohane, G. H. (2004). Biceps and body image: The relationship between muscularity, self esteem, depression, and eating disorder symptoms. Psychology of Men & Masculinity, 5, 112 120. Paeratakul, S., White, M. A., Williamson, D. A., Ryan, D. H., & Bray, G. A., (2002). Sex, race/ethnicity, socioeconomic status, and BMI in relation to self perception of overweight. Obesity research, 10, 345 350 Page, R. M., Lee, C. M., & Miao, N. F (2005). Self perception of body weight among high school students in Taipei, Taiwan. International Journal of Adolescent Medicine and Health, 17, 123 136. Parker, S., Nichter, M., Vuckovic, N., Sims, C., & Ritenbaug h, C. (1995). Body image and weight concerns among African American and white adolescent females: Differences that make a difference. Human organization, 54, 103 114. Perrin, E. M., Flower, K. B., & Ammerman, A. S. (2005). Pediatricians' own weight: Self p erception, misclassification, and ease of counseling. Obesity Research, 13, 326 332. Pope, H. G., Gruber, A. J., Mangweth, B., Bureau, B., deCol, C., Jouvent, R., & Hudson, J. I. (2000). Body image perception among men in three countries. American Journal of Psychiatry, 157, 1297 1301.

PAGE 87

75 Pritchard, M. E., King, S. L., & Czajka Narins, D. M. (1997). Adolescent body mass indices and self perception. Adolescence, 32, 863 880. R acette, S. B., Deusinger, S. S. & Deusinger, R. H. (2003). Obesity: Overview of preva lence, etiology, and treatment. Physical Therapy, 83, 276 288. Rand, C. S. W., & Kuldau, J. M. (1990). The epidemiology of obesity and self defined weight problem in the general population: Gender, race, age, and social class. International Journal of Eati ng Disorders, 9, 329 343. Reijonen, J. H., Pratt, H. D., Patel, D. R., & Greydanus, D. E. (2003). Eating disorders in the adolescent population: An overview. Journal of Adolescent Research, 18, 209 222. Reilly, J. J., Methven, E., McDowell, Z. C., Hackin g, B., Alexander, D., Stewart, L., & Kelnar, C. J. H. (2003) Health consequences of obesity. Archives of Disease in Childhood, 88, 748 752. Ridgeway, R. T., & Tylka, T. L. (2005). College men's perception of ideal body composition and shape. Psychology of Men & Masculinity, 6, 209 220. Rodin, J., Silberstein, L., & Striegel Moore, R. (1984). Women and weight: A normative discontent. Nebraska Symposium on Motivation, 32, 267 307. Rosen, E. F ., Brown, A., Braden, R. A., Dorsett, H. W., Franklin, D. N., Garlin gton, R. A., et al. (1993). African American males prefer a larger female body silhouette than do Whites. Bulletin of the Psychonomic Society, 31, 599 601. Safir, M. P., Flaisher Kellner, S., & Rosenma n, A. (2005). When gender differences surpass cultural differences in personal satisfaction with body shape in Israeli college students. Sex Roles, 52, 369 378. Schutz, H. K. Paxton, S. J., & Wertheim, E. H. (2002). Investigation of body comparison among adolescent girls. Journal of Applied Social Psychology, 32 1906 1937. Shroff, H., & Thompson, J. K. (2006). The tripartite influence model of body image and eating disturbance: A replication with adolescent girls. Body Image, 3, 17 26. Skelton, J. A., Busey, S. L., & Havens, P. L. (2006). Weight and health st atus if inner city African American children: Perceptions of children and their parents. Body Image, 3, 289 293.

PAGE 88

76 Stice, E. (2001). A prospective test of the dual pathway model of bulimic pathology: Mediating effects of dieting and negative affect. Journal of Abnormal Psychology, 110, 124 135. Stice, E., Nemeroff, C. & Shaw, H. E. (1996). Test of the dual pathway model of bulimia nervosa: Evidence for dietary restraint and affect regulation mechanisms. Journal of Social and Clinical Psychology, 15, 340 363. Stormer, S. M., & Thompson, J. K. (1996). Explanations of body image disturbance: A test of maturational status, negative verbal commentary, social comparison, and sociocultural hypotheses. International Journal of Eating Disorders, 19, 193 202. Sullivan, P. F ., Bulik, C. M., Carter, F A., & Joyce, P. R. (1996). Correlates of severity in bulimia nervosa. International Journal of Eating Disorders, 20, 239 251. Terry, D. J., & Hogg, M. A. (1996). Group norms and the attitude behavior relationship: A role for group identification. Personality and Social Psychology Bulletin, 22, 776 793. Thompson, J. K. (1990). Body image disturbance: Assessment and treatment. Elmsford, NJ: Pergammon. Thompson, J. K. (2004). Handbook of Eating Disorders and Obesity. Hoboke n, New Jersey: John Wiley & Sons, Inc. Thompson, J. K., & Cafri, G. (2007). The muscular ideal: Psychological, social, and medical perspectives. Washington, DC: American Psychological Association. Thompson, J. K., Heinberg, L. J., Altabe, M., & Tantleff Du nn, S. (1999). Exacting beauty: Theory, assessment, and treatment of body image disturbance. Washington, DC : American Psychological Association. Thompson, J. K., Heinberg, L., & Tantleff, S. (1991). The Physical Appearance Comparison Scale. The Behavi or Therapist, 14, 174. Thompson, J. K., & Stice, E. (2001). Thin ideal internalization: Mounting evidence for a new risk factor for body image disturbance and eating pathology. Current Directions in Psychological Science, 10, 181 183.

PAGE 89

77 Thompson, J. K., van den Berg, P., Roehrig, M., Guarda, A. S., & Heinberg, L. J. (2004). The Sociocultural Attitudes Towards Appearance Sacle 3 (SATAQ 3): Development and validation. International Journal of Eating Disorders, 35, 293 304. Tiggeman, M., & McGill, B. (2004). Th e role of social comparison in the effect of magazine advertisements on women's mood and body dissatisfaction. Journal of Social and Clinical Psychology, 23, 23 44. van den Berg, P., Thompson, J. K., & Obremski Brandon, K. (2002). The tripartite influence model of body image and eating disturbance: A covariance structure modeling investigation testing the mediational role of appearance comparison. Journal of Psychosomatic Research, 53, 1007 1020. Viner, R. M., Haines, M. M., Taylor, S. J. C., Head, J., Booy R., & Stansfeld, S. (2006). Body mass, weight control behaviors, weight perception and emotional well being in a multiethnic sample of early adolescents. International Journal of Obesity, 30, 1514 1521. Wake, M., Salmon, L., Waters, E., Wright, M., & Hes keth, K. (2002). Parent reported health status of overweight and obese Australian primary school children: A cross sectional population survey. International Journal of Obesity, 26, 717 724. Wardle, J., Haase, A. M., & Steptoe, A. (2006). Body image and we ight control in young adults: International comparisons in university students from 22 countries. International Journal of Obesity, 30, 644 651. Wardle, J., & Johnson, F (2002). Weight and dieting, Examining levels of weight concern in British adults. Int ernational Journal of Obesity, 26, 1144 1149. Wildes, J. E., Emery, R. E., & Simons, A. D. (2001). The roles of ethnicity and culture in the development of eating disturbance and body dissatisfaction: A meta analytic review. Clinical Psychology R eview, 21, 521 551. Wing, R. R., Epstein, L. H., & Neff, D. (1980). Accuracy of parents' reports of height and weight. Journal of Behavioral Assessment, 2, 105 110.

PAGE 90

78 Xie, B., Chou, C. P., Spruijt Metz, D., Reynolds, K., Clark, F ., Palmer, P. H., Gallaher, P., Guo, Q ., & Johnson, C. A. (2006). Weight perception and weight related sociocultural and behavioural factors in Chinese adolescents. Preventive Medicine, 42, 229 234. Yang, C. J., Gray, P., & Pope, H. G. (2005). Male body image in Taiwan versus the west: Yangga ng Zhiqi meets the Adonis complex. American Journal of Psychiatry, 162, 263 269. Yanover, T., & Thompson, J. K. ( 2009 ). Assessment of body image in children and adolescents. In L. Smolak & J.K. Thompson (Eds.), Body i mage, eating disorders, and obesity in youth (second ed., pp. 177 192), Washington, DC: American Psychological Association. Young Hyman, D., Herman, L. J., Scott, D. L., & Schlunt, D. G. (2000). Care giver perception of children's obesity related health risk: A study of African American familie s. Obesity Research, 8, 241 248.

PAGE 91

79 Appendices

PAGE 92

80 Appendix A: Pilot Materials Stimuli; Male:

PAGE 93

81 Appendix A (Continued) Stimuli; Female:

PAGE 94

82 Appendix A (Continued) Target Rating Items: Please rate the individual in the previous slide on the following dimension s. 1. How healthy is the person you just saw? 1 ----------2 ----------3 ----------4 ----------5 ----------6 ----------7 Very Unhealthy Moderately Unhealthy Moderately Healthy Very Healthy 2. How would you classify the weight of the person you just saw ? 1 ----------2 ----------3 ----------4 ----------5 ----------6 ----------7 Very Underweight Underweight Overweight Very Overweight 3. How attractive is the person you just saw? 1 ----------2 ----------3 ----------4 ----------5 ----------6 ---------7 Very Un a ttractive Moderately Unattractive Moderately Attractive Very Attractive 4. Given that the man/woman you just saw is 5'10"/5'4", estimate his weight in pounds. _____________________ lbs

PAGE 95

83 Appendix A (Continued) Instructions to Participants : In just a moment, you will be asked to watch the screen at the front of the room. A series of images will appear briefly on the screen. Between each image, there will be a black screen. While the black screen is up, please complete the ratings on the fi gure you just saw in your test booklet. Please make sure that the number on the slide matches the number in your rating booklet. Don't think about the ratings too long. Mark down your first instinct. You will only have a few seconds before the next image a ppears on the screen. Also, it is very important that you remain completely quiet while viewing the images and completing your ratings. Please do not speak or make any noise while completing these tasks.

PAGE 96

84 Appendix A (Continued) Proximate Social Environment Rating Scale Instructions: Please think of the three/ten friends with whom you spend the most time. For each friend, please choose the scale that depicts the appropriate gender. Next, circle the figure on the scale that you feel most closely resembles th eir body size and shape. Please use a separate sheet for each friend.

PAGE 97

85 Appendix A (Continued) Focus Group Items: Items Regarding Slides 1. Did you find the figures credible as African American? Caucasian? Hispanic? 2. How did you feel at the end of the ratin gs? Were you fatigued? Did you feel that there were too many or were you able to focus throughout? 3. Did the slides all seem different to you or did you feel like there were some that were the same? 4. Was there anything about the procedure in today's study t hat you found difficult or troubling? Did you have enough time between slides to complete the ratings? Did you have long enough to see the slide? 5. Did the final item asking you to guess the person's weight affect the way you made the ratings? Did knowing t he person's height change the way you looked at the figure? OR Would you have liked to know the person's height? How did you deal with not knowing ? 6. Did having other people in the room distract you from the task or were you able to focus throughout? Was t here anything that distracted your attention from the task at hand? Items Regarding Social Norms Scale 1. Were the instructions clear? How did you think you were to fill out the scale?

PAGE 98

86 Appendix B: Instructions to Participants At this time, please turn off and put away all cell phones. It is very important that you remain completely quiet during this experiment. Please don't speak or make any noise. Anyone found talking, whispering, answering their cell phone, sending text messages, or otherwise disrupting t he experiment will be asked to leave and will not receive credit for participating in this experiment. In just a moment, you will be asked to watch the screen at the front of the room. A series of images will appear briefly on the screen. Between each imag e, there will be a black screen. While the black screen is up, please complete the ratings of the figure you just saw in your test booklet. Please make sure that the number on the slide matches the number in your rating booklet and complete all three quest ions for each slide. Please note that there are two slides per page so for slide 1 you will complete to the line, slide two below the line and then turn the page and so on. Don't think about the ratings too long. Mark down your first instinct. You will onl y have a few seconds before the next image appears on the screen. For each slide, please pay attention to all of the relevant information while making your rating. The slides will differ in gender and in their physical characteristics. It is also important to note that each slide is labeled at the bottom with information concerning the ethnicity of the person in the slide. Again, please consider all of this information while making your ratings.

PAGE 99

87 Appendix C: Target Rating Items Please rate the individual in the previous slide on the following dimensions. 1. How healthy is the person you just saw? 1 ----------2 ----------3 ----------4 ----------5 ----------6 ----------7 Very Unhealthy Moderately Unhealthy Moderately Healthy Very Healthy 2. How would you classify the weight of the person you just saw? 1 ----------2 ----------3 ----------4 ----------5 ----------6 ----------7 Underweight Normal Weight Overweight Obese 3. How attractive is the person you just saw? 1 ----------2 ---------3 ----------4 ----------5 ----------6 ----------7 Very Unattractive Moderately Unattractive Moderately Attractive Very Attractive

PAGE 100

88 Appendix D: Distraction Task Now, take about 5 10 minutes to think about countries you have learned about through the m edia but have never been to. After giving it some thought, imagine 5 countries you have read about or heard about through the media but that you have not yet been to. Please take your time with this. Think about these places and visualize yourself on vacation in each of them. What would it be like? What would you be doing there? What would you see? What sensations would you feel? To help you with this exercise, write these countries in the space below. Also, provide a brief description of each de stination, what you would like to do and see there, and how the media has described this destination. Travel Destination Activities/Sights/Feelings There Media Description 1. 2. 3. 4. 5

PAGE 101

89 Appendix E: Marlowe Crown Social Desirability Sc ale Listed below are a number of statements concerning personal attitudes and traits. Read each item and using the scale below, decide whether the statement is true or false as it pertains to you personally. True False 1 2 True False 1. Before voting I thoroughly investigate the qualifications of all the candidates. 1 2 2. I never hesitate to go out of my way to help someone in trouble. 1 2 3. It is sometimes hard for me to go on with my work if I am not encouraged. 1 2 4. I have never intensely disliked anyone. 1 2 5. On occasion I have had doubts about my ability to succeed in life. 1 2 6. I sometimes feel resentful when I don't get my way. 1 2 7. I am always careful about my manner of dress. 1 2 8. My table manners at home are as good as when I eat out in a restaurant. 1 2 9. If I could get into a movie without paying and be sure I was not seen, I would probably do it. 1 2 10. On a few occasions, I have given up doing something because I thought too little of my ability. 1 2 11. I like to gossip at times. 1 2 12. There have been times when I felt like rebelling against people in authority even though I knew they were right. 1 2 13. No matter who I'm talking to, I'm always a good listener. 1 2 1 4. I can re member "playing sick" to get out of something. 1 2 15. There have been occasions when I took advantage of someone. 1 2

PAGE 102

90 Appendix E (Continued) 16. I'm always willing to admit it when I make a mistake. 1 2 17. I always try to practice what I preach. 1 2 18. I don't find it particularly difficult to get along with loud mouthed, obnoxious people. 1 2 19. I sometimes try to get even, rather than forgive and forget. 1 2 20. When I don't know something I don't at all mind admitting it. 1 2 21. I a m always courteous, even to people who are disagreeable. 1 2 22. At times I have really insisted on having things my own way. 1 2 23. There have been occasions when I felt like smashing things. 1 2 24. I would never think of letting someone else be punished for my wrongdoings. 1 2 25. I never resent being asked to return a favor. 1 2 26. I have never been irked when people expressed ideas very different from my own. 1 2 27. I never make a long trip without checking the safety of my car. 1 2 28 There have been times when I was quite jealous of the good fortune of others. 1 2 29. I have almost never felt the urge to tell someone off. 1 2 30. I am sometimes irritated by people who ask favors of me. 1 2 31. I have never felt that I was pun ished without cause. 1 2 32. I sometimes think when people have a misfortune they only got what they deserved. 1 2 33. I have never deliberately said something that hurt someone's feelings. 1 2

PAGE 103

91 Appendix F : Physical Appearance Comparison Scale Using t he scale below, please circle the number that best matches your agreement with the following statements. Never Seldom Sometimes Often Always 1 2 3 4 5 1. At parties or other social events, I compare my physical appearance to the physical appea rance of others. 1 2 3 4 5 2. The best way for a person to know if they are overweight or underweight is to compare their figure to the figure of others. 1 2 3 4 5 3. At parties or other social events, I compare how I am dressed to how other people are dressed 1 2 3 4 5 4. Comparing your "looks" to the "looks" of others is a bad way to determine if you are attractive or unattractive. 1 2 3 4 5 5. In social situations, I sometimes compare my figure to the figures of other people. 1 2 3 4 5

PAGE 104

92 Appendi x G: Sociocultural Attitudes Towards Appearance Questionnaire 3 Internalization General subscale Using the scale below, please write the number that best matches your agreement with the following statements. Definitely disagree Mostly disagree Neither agree nor disagree Mostly agree Definitely agree 1 2 3 4 5 1. _____ I would like my body to look like the people who are on TV. 2. _____ I compare my body to the bodies of TV and movie stars. 3. _____ I would like my body to look like the models who appear in magazines. 4. _____ I compare my appearance to the appearance of TV and movie stars. 5. _____ I would like my body to look like the people who are in movie s. 6. _____ I compare my body to the bodies of people who appear in magazines. 7. _____ I wish I looked like the models in music videos. 8. _____ I compare my appearance to the appearance of people i n magazines. 9. _____ I try to look like the people on TV.

PAGE 105

93 Appendix H: Proximate Social Environment Rating Scale Please think of the three friends with whom you spend the most time. For each friend, please choose the scale that depic ts the appropriate gender. Next, circle the figure on the scale that you feel most closely resembles their body size and shape. Please use a separate sheet for each friend. Circle ONLY a male OR a female for each friend, NOT both.

PAGE 106

94 Appendix H (Continued)

PAGE 107

95 Appendix H (Continued)

PAGE 108

96 Appendix I: Eating Disorder Inventory 3 Body Dissatisfaction Subscale The items below ask about your attitudes, feelings, and behavior. Some of the items relate to food or eating. Other items ask about your feelings about yourse lf. For each item, decide if the item is true about you. Circle the letter that corresponds to your rating. Respond to all of the items, making sure that you circle the letter for the rating that is true. Always Usually Often Sometimes Rarely Never A U O S R N 1. 1. I think that my stomach is too big. A U O S R N 2. I think that that my thighs are too large. A U O S R N 3. I think that my stomach is just the right size. A U O S R N 4. I feel satisfied with the shape of my body. A U O S R N 5. I like the shape of my buttocks. A U O S R N 6. I think my hips are too big. A U O S R N 7. I think that my thighs are just the right size. A U O S R N 8. I think my buttocks are too large. A U O S R N 9. I think that my hips are just th e right size. A U O S R N

PAGE 109

97 Appendix J: Drive for Muscularity Scale Please read each item carefully then, for each one, circle the number that best applies to you. 1 2 3 4 5 6 Always Very Often Often Sometimes Rarely Never 1. I wish that I were more m uscular. 1 2 3 4 5 6 2. I lift weights to build up muscle. 1 2 3 4 5 6 3. I use protein or energy supplements. 1 2 3 4 5 6 4. I drink weight gain or protein shakes. 1 2 3 4 5 6 5. I try to consume as many calories as I can in a day. 1 2 3 4 5 6 6. I f eel guilty if I miss a weight training session. 1 2 3 4 5 6 7. I think I would feel more confident if I had more muscle mass. 1 2 3 4 5 6 8. Other people think I work out with weights too often. 1 2 3 4 5 6 9. I think that I would look better if I gaine d 10 pounds in bulk. 1 2 3 4 5 6 10. I think about taking anabolic steroids. 1 2 3 4 5 6 11. I think that I would feel stronger if I gained a little more muscle mass. 1 2 3 4 5 6 12. I think that my weight training schedule interferes with other aspects of my life. 1 2 3 4 5 6 13. I think that my arms are not muscular enough. 1 2 3 4 5 6 14. I think that my chest is not muscular enough. 1 2 3 4 5 6 15. I think that my legs are not muscular enough. 1 2 3 4 5 6

PAGE 110

98 Appendix K: Demographic Information Pl ease provide the following information accurately and honestly. It is very important that the information is correct. Please remember that this study is anonymous and your name will not appear anywhere on these forms. Age : ______ Year in school : _____ Fr eshman _____ Sophomore _____ Junior ______ Senior Major : ____________________________ Race (Please choose one): ______ African American ______ Caucasian ______Asian American ______Native American ______Pacific Islander ______ Other (Please specify): __________________________ Ethnicity (Please choose one): ______Hispanic ______Non Hispanic Weight in pounds : ___________ Height : ___________

PAGE 111

99 Appendix L: Means and Standard Deviations, F p and 2 values for Weight Analyses General Notes: 1. Find ings in bold are significant at the p < .006 level. Italicized findings represent a trend towards significance (.006 < p < .01). Significant pairwise differences are indicated for values up to and including p = .014. 2. For analyses without covariates, raw me ans are presented with standard deviations. Adjusted means are presented for all ANCOVAs with adjusted standard errors. 3. Superscripts denote means that differ significantly from each other. Subscripts denote means that differ significantly from each other.

PAGE 112

100 Table L.1. Repeated measures effects: Three way interactions (target race X rater race X rater gender) Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure A No covariates Caucasian 3.89 (1.03) 3.88 ( 0 .84) 3.67 (1.01) 3.98 ( 0 .64) 3.99 ( 0 .94) 3.90 ( 0 .70) Hispanic 3.67 ( 0 .91) 3.75 ( 0 .86) 3.47 (1.14) 3.87 ( 0 39) 3.90 ( 0 .80) 3.83 ( 0 .62) African American 4.02 (1.02) 3.92 ( 0 .68) 3.57 (1.01) 3.94 ( 0 .57) 4.04 ( 0 .92) 3.99 ( 0 .75) F (4,633) = 0.22, p = .926, partial 2 < .01 BMI Caucasian 3.94 ( 0 .12) 3.87 ( 0 .10) 3.69 ( 0 .12) 3.99 ( 0 .12) 4.00 ( 0 .15) 3.91 ( 0 .12) Hispanic 3.72 ( 0 .11) 3.69 ( 0 .10) 3.51 ( 0 .12) 3.84 ( 0 .11) 3.97 ( 0 .14) 3.85 ( 0 .10) African American 4.05 ( 0 .11) 3.83 ( 0 .10) 3.61 ( 0 .12) 3.91 ( 0 .11) 4.09 ( 0 .14) 4.02 ( 0 .10) F (4,630) = 0.28, p = .886, partial 2 < .01 EDI BD Caucasian 3.91 ( 0 .12) 3.93 ( 0 .10) 3.66 ( 0 .12) 4.01 ( 0 .12) 3.94 ( 0 .15) 3.91 ( 0 .10) Hispanic 3.67 ( 0 .11) 3.79 ( 0 .10) 3.46 ( 0 .12) 3.89 ( 0 .11) 3.83 ( 0 .15) 3.83 ( 0 .10) African American 3.99 ( 0 .11) 3.96 ( 0 .10) 3.55 ( 0 .11) 3.97 ( 0 .11) 3.94 ( 0 .15) 4.00 ( 0 .10) F (4,632) = 0.22, p = .928, partial 2 < .01

PAGE 113

101 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 3.92 ( 0 .12) 3.90 ( 0 .10) 3.66 ( 0 .13) 4.00 ( 0 .12) 3.99 ( 0 .13) 3.91 ( 0 .11) Hispanic 3.72 ( 0 .12) 3.73 ( 0 .10) 3.50 ( 0 .13) 3.84 ( 0 .10) 3.91 ( 0 .14) 3.84 ( 0 .11) African American 4.01 ( 0 .12) 3.91 ( 0 .10) 3 .55 ( 0 .13) 3.95 ( 0 .10) 4.04 ( 0 .14) 4.00 ( 0 .11) F (4,624) = 0.20, p = .939, partial 2 < .01 PACS Caucasian 3.93 ( 0 .12) 3.93 ( 0 .10) 3.65 ( 0 .12) 4.01 ( 0 .12) 3.95 ( 0 .15) 3.90 (.10) Hispanic 3.70 ( 0 .11) 3.77 ( 0 .10) 3.46 ( 0 .12) 3.87 ( 0 .11) 3.88 ( 0 .1 4) 3.85 (0. 10) African American 4.04 ( 0 .11) 3.95 ( 0 .10) 3.54 ( 0 .12) 3.95 ( 0 .11) 3.99 ( 0 .14) 3.98 ( 0 .10) F (4,630) = 0.20, p = .936, partial 2 < .01 SATAQ Caucasian 3.93 ( 0 .12) 3.89 ( 0 .10) 3.67 ( 0 .12) 4.00 ( 0 .11) 4.09 ( 0 .15) 3.91 ( 0 .10) Hisp anic 3.67 ( 0 .11) 3.83 ( 0 .10) 3.48 ( 0 .11) 3.88 ( 0 .11) 3.82 ( 0 .14) 3.78 ( 0 .10) African American 4.00 ( 0 .11) 3.91 ( 0 .10) 3.57 ( 0 .11) 3.94 ( 0 .11) 4.13 ( 0 .14) 3.99 ( 0. 10) F (4,629) = 0.24, p = .911, partial 2 < .01

PAGE 114

102 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure B No covariates Caucasian 3.55 ( 0 .83) 3.78 ( 0 .76) 3.41 (1.17) 3.70 ( 0 .61) 3.57 (1.05) 3.75 ( 0 .69) Hispanic 3.53 ( 0 .98) 3.62 ( 0 .82) 3.47 (1.01) 3.79 ( 0 .63) 3.61 ( 0 .68) 3.73 ( 0 .71) African American 3.42 ( 0 .90) 3.78 ( 0 .53) 3.40 ( 0 .95) 3.74 ( 0 .81) 3.52 ( 0 .97) 3.75 ( 0 .90) F (4,635) = 0.39, p = .817, partial 2 < .01 BMI Caucasian 3.56 ( 0 .11) 3.75 ( 0 .10) 3.48 ( 0 .12) 3.69 ( 0 .12) 3.62 ( 0 .15) 3.75 ( 0 .11) Hispanic 3.57 ( 0 .11) 3.53 ( 0 .10) 3.53 ( 0 .12) 3.76 ( 0 .11) 3.69 ( 0 .15) 3.77 ( 0 .10) African American 3.46 ( 0 .11) 3.70 ( 0 .11) 3.45 ( 0 .12) 3.70 ( 0 .11) 3.59 ( 0 .15) 3.77 ( 0 .10) F (4,634) = 0.44, p = .776, partial 2 < .01 EDI BD Caucasian 3.53 ( 0 .11) 3.81 ( 0 .10) 3.45 ( 0 .12) 3.71 ( 0 .12) 3.53 ( 0 .16) 3.75 ( 0 .11) Hispanic 3.52 ( 0 .11) 3.64 ( 0 .10) 3.47 ( 0 .12) 3.80 ( 0 .11) 3.57 ( 0 .15) 3.75 ( 0 .10) African American 3.40 ( 0 .11) 3.81 ( 0 .10) 3.87 ( 0 .12) 3.75 ( 0 .12) 3.45 ( 0 .15) 3.75 ( 0 .10) F (4,634) = 0.40, p = .812, partial 2 < .01

PAGE 115

103 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 3.53 ( 0 .12) 3.79 ( 0 .10) 3.48 ( 0 .13) 3.72 ( 0 .12) 3.57 ( 0 .15) 3.75 ( 0 .11) Hispanic 3.52 ( 0 .12) 3.62 ( 0 .10) 3.65 ( 0 .13) 3.80 ( 0 .12) 3.61 ( 0 .15) 3.72 ( 0 .11) African American 3.44 ( 0 .12) 3.77 ( 0 .10) 3.44 ( 0 .13) 3.71 ( 0 .12) 3.53 ( 0 .15) 3.68 ( 0 .11) F (4,629) = 0.55, p = .701, partial 2 < .01 PACS Caucasian 3.55 ( 0 .11) 3.78 ( 0 .10) 3.46 ( 0 .12) 3.80 ( 0 .11) 3.58 ( 0 .15) 3.74 ( 0 .11) Hispanic 3.52 ( 0 .11) 3.61 ( 0 .10) 3.49 ( 0 .12) 3.79 ( 0 .11) 3.63 ( 0 .15) 3.74 ( 0 .10) African American 3.43 ( 0 .11) 3.81 ( 0 .10) 3.38 ( 0 .12) 3.74 ( 0 .12) 3.48 ( 0 .15) 3.73 ( 0 .11) F (4,632) = 0.37, p = .828, partial 2 < .01 SATAQ Ca ucasian 3.52 ( 0 .12) 3.80 ( 0 .10) 3.46 ( 0 .12) 3.70 ( 0 .12) 3.56 ( 0 .15) 3.74 ( 0 .11) Hispanic 3.52 ( 0 .11) 3.61 ( 0 .10) 3.48 ( 0 .12) 3.79 ( 0 .11) 3.61 ( 0 .15) 3.75 ( 0 .11) African American 3.42 ( 0 .11) 3.80 ( 0 .10) 3.40 ( 0 .12) 3.74 ( 0 .12) 3.55 ( 0 .15) 3.73 ( 0 .11) F (4,629) = 0.55, p = .698, partial 2 < .01

PAGE 116

104 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure C No covariates Caucasian 2.33 (1.18) 2.60 (1.13) 2.08 (1.06) 2.42 (1.10) 2.19 (1.10) 2.45 (1.29) Hispanic 2.58 (1.17) 2.78 (1.05) 2.41 (1.06) 2.72 (1.03) 2.49 (1.17) 2.49 (1.12) African American 2.02 (1.11) 2.38 (1.22) 1.94 (1.20) 2.40 (1.01) 2.11 (1.09) 2.28 (1.08) F (4,638) = 0.15, p = .963, partial 2 < .01 BMI Caucasian 2.33 ( 0 .16) 2.63 ( 0 .14) 2.06 ( 0 .17) 2.42 ( 0 .16) 2.20 ( 0 .21) 2.45 ( 0 .14) Hispanic 2.63 ( 0 .15) 2.73 ( 0 .13) 2.42 ( 0 .16) 2.70 ( 0 .15) 2.52 ( 0 .20) 2.51 ( 0 .14) African American 2.04 ( 0 .15) 2.37 ( 0 .14) 1.96 ( 0 .16) 2.37 ( 0 .14) 2.17 ( 0 .20) 2.30 ( 0 .14) F (4,633) = 0.13, p = .971, partial 2 < .01 EDI BD Caucasian 2.33 ( 0 .16) 2.62 ( 0 .14) 2.06 ( 0 .17) 2.42 ( 0 .1 6) 2.20 ( 0 .21) 2.45 ( 0 .14) Hispanic 2.61 ( 0 .15) 2.79 ( 0 .13) 2.39 ( 0 .16) 2.72 ( 0 .15) 2.45 ( 0 .20) 2.49 ( 0 .14) African American 2.01 ( 0 .16) 2.41 ( 0 .14) 1.93 ( 0 .16) 2.41 ( 0 .16) 2.07 ( 0 .21) 2.28 ( 0 .14) F (4,636) = 0.13, p = .972, partial 2 < .01

PAGE 117

105 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 2.31 ( 0 .16) 2.63 ( 0 .14) 2.05 ( 0 .19) 2.43 ( 0 .16) 2.18 ( 0 .20) 2.40 ( 0 .15) Hispanic 2.63 ( 0 .16) 2.76 ( 0 .13) 2.46 ( 0 .18) 2.70 ( 0 .16) 2.48 ( 0 .19) 2.43 ( 0 .14) African American 2.01 ( 0 .16) 2.41 ( 0 .14 ) 1.97 ( 0 .18) 2.39 ( 0 .16) 2.11 ( 0 .20) 2.24 ( 0 .15) F (4,627) = 0.08, p = .988, partial 2 < .01 PACS Caucasian 2.33 ( 0 .16) 2.63 ( 0 .14) 2.06 ( 0 .17) 2.42 ( 0 .16) 2.20 ( 0 .20) 2.42 ( 0 .14) Hispanic 2.61 ( 0 .15) 2.78 ( 0 .13) 2.40 ( 0 .16) 2.72 ( 0 .15) 2.47 ( 0 .19) 2.47 ( 0 .14) African American 2.02 ( 0 .15) 2.43 ( 0 .14) 1.93 ( 0 .17) 2.40 ( 0 .16) 2.90 ( 0 .20) 2.27 ( 0 .14) F (4,634) = 0.14, p = .969, partial 2 < .01 SATAQ Caucasian 2.34 ( 0 .16) 2.55 ( 0 .14) 2.05 ( 0 .17) 2.41 ( 0 .16) 2.31 ( 0 .21) 2.50 ( 0 .15) Hispanic 2.61 ( 0 .15) 2.76 ( 0 .14) 2.39 ( 0 .16) 2.72 ( 0 .15) 2.54 ( 0 .20) 2.51 ( 0 .14) African American 2.02 ( 0 .16) 2.34 ( 0 .14) 1.92 ( 0 .16) 2.40 ( 0 .16) 2.22 ( 0 .20) 2.33 ( 0 .14) F (4,632) = 0.13, p = .971, partial 2 < .01

PAGE 118

106 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure D No covariates Caucasian 4.31 ( 0 .94) 4.28 ( 0 .63) 4.35 ( 0 .74) 4.1 7 ( 0 .55) 4.34 ( 0 .73) a 3.98 ( 0 .44) b Hispanic 4.18 ( 0 .75) 4.04 ( 0 .71) 4.28 ( 0 .86) 4.02 ( 0 .70) 4.02 ( 0 .94) 4.22 ( 0 .57) African American 4.20 ( 0 .97) 4.10 ( 0 .48) 4.06 ( 0 .82) 4.02 ( 0 .60) 4.16 ( 0 .73) 4.14 ( 0 .54) F (4,640) = 3.22, p = .013, partial 2 = .02 BMI Caucasian 4.33 ( 0 .09) 4.24 ( 0 .08) 4.34 ( 0 .10) 4.16 ( 0 .09) 4.37 ( 0 .12) 3.99 ( 0 .08) Hispanic 4.18 ( 0 .10) 4.08 ( 0 .09) 4.29 ( 0 .11) 4.02 ( 0 .10) 4.02 ( 0 .13) 4.23 ( 0 .09) African American 4.20 ( 0 .09) 4. 10 ( 0 .08) 4.06 ( 0 .10) 4.02 ( 0 .10) 4.16 ( 0 .12) 4.13 ( 0 .09) F (4,635) = 2.98, p = .019, partial 2 = .02 EDI BD Caucasian 4.30 ( 0 .09) 4.30 ( 0 .08) 4.32 ( 0 .10) 4.19 ( 0 .09) 4.29 ( 0 .12) a 3.98 ( 0 .08) b Hispanic 4.17 ( 0 .10) 4.10 ( 0 .09) 4.28 ( 0 .10) 4.03 ( 0 .10) 3.97 (.13) 4.23 ( 0 .09) African American 4.19 ( 0 .09) 4.03 ( 0 .08) 4.06 ( 0 .10) 4.03 ( 0 .10) 4.13 ( 0 .12) 4.13 ( 0 .09) F (4,638) = 3.17, p = .014, partial 2 = .02

PAGE 119

107 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 4.31 ( 0 .09) 4.26 ( 0 .08) 4.32 ( 0 .11) 4.17 ( 0 .10) 4.34 ( 0 .12) 4.01 ( 0 .09) Hispanic 4.15 ( 0 .10) 4.10 ( 0 .09) 4.22 ( 0 .12) 4.04 ( 0 .10) 4.00 ( 0 .13) 4.26 ( 0 .10) African American 4.21 ( 0 .10) 4.09 ( 0 .08) 4.08 ( 0 .11) 4.01 ( 0 .10) 4.17 ( 0 .08) 4.11 ( 0 .09) F (4,627) = 2.72, p = .029, partial 2 = .02 PACS Caucasian 4.31 ( 0 .09) 4.28 ( 0 .08) 4.32 ( 0 .10) 4.17 ( 0 .09) 4.33 ( 0 .12) a 3.98 ( 0 .09) b Hispanic 4.18 ( 0 .10) 4.09 ( 0 .09) 4.28 ( 0 .11) 4.02 ( 0 .10) 4.01 ( 0 .13) 4.23 ( 0 .09) African American 4.21 ( 0 .09) 4.11 ( 0 .08) 4.05 ( 0 .10) 4.02 ( 0 .10) 4.14 ( 0 .12) 4.13 ( 0 .09) F (4,636) = 3.23, p = .012, partial 2 = .02 SATAQ Caucasian 4.30 ( 0 .09) 4.25 ( 0 .08) 4.32 ( 0 .10) 4.17 ( 0 .09) 4.38 ( 0 .12) a 3.99 ( 0 .09) b Hispanic 4.16 ( 0 .10) 4.13 ( 0 .09) 4.29 ( 0 .10) 4.03 ( 0 .10) 3.97 ( 0 .13) 4.19 ( 0 .09) African American 4.18 ( 0 .10) 4.10 ( 0 .12) 4.06 ( 0 .10) 4.02 ( 0 .10) 4.16 ( 0 .12) 4.13 ( 0 .09) F (4,634) = 3.18, p = .014, partial 2 = .02

PAGE 120

108 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure G No covariates Caucasian 3.93 ( 0 .54) 3.77 ( 0 .59) 3.94 ( 0 .54) 3.72 ( 0 .57) 3.59 ( 0 .61) 3.57 ( 0 .76) Hispanic 3.95 ( 0 .59) 3.67 ( 0 .65) 4.00 ( 0 45) 3.72 ( 0 .69) 3.81 ( 0 .39) 3.49 ( 0 .85) African American 3.95 ( 0 .68) 3.64 ( 0 .59) 3.77 ( 0 .81) 3.75 ( 0 .65) 3.81 ( 0 .52) 3.80 ( 0 .77) F (4,617) = 1.44, p = .220, partial 2 = .01 BMI Caucasian 3.92 ( 0 .08) 3.78 ( 0 .09) 3.94 ( 0 .09) 3.72 ( 0 .09) 3.57 ( 0 .11) 3.57 ( 0 .08) Hispanic 3.94 ( 0 .09) 3.72 ( 0 .09) 3.99 ( 0 .09) 3.72 (0 .09) 3.81 ( 0 .12) 3.48 ( 0 .08) African American 3.95 ( 0 .09) 3.75 ( 0 .08) 3.77 ( 0 .10) 3.75 ( 0 .10) 3.80 ( 0 .12) 3.79 ( 0 .09) F (4,611) = 1.71, p = .037, partial 2 = .01 EDI BD Caucasian 3.93 ( 0 .08) 3.76 ( 0 .07) 3.94 ( 0 .09) 3.71 ( 0 .09) 3.60 ( 0 .11) 3.58 ( 0 .08) Hispanic 3.95 ( 0 .09) 3.65 ( 0 .08) 4.01 ( 0 .09) 3.70 ( 0 .09) 3.84 ( 0 .12) 3.48 ( 0 .08) African American 3.94 ( 0 .09) 3.65 ( 0 .08) 3.96 ( 0 .10) 3.76 ( 0 .10) 3.77 ( 0 .13) 3.80 ( 0 .09) F (4,614) = 1.43, p = .224, partial 2 = .01

PAGE 121

109 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 3.92 ( 0 .09) 3.77 ( 0 .08) 3.93 ( 0 .10) 3.72 ( 0 .09) 3.59 ( 0 .11) 3.57 ( 0 .08) Hispanic 3.93 ( 0 .09) 3.68 ( 0 .08) 3.97 ( 0 .10) 3.73 ( 0 .09) 3.79 ( 0 .11) 3.53 ( 0 .08) African American 3.96 ( 0 .10) 3 .63 ( 0 .08) 3.77 ( 0 .11) 3.74 ( 0 .10) 3.81 ( 0 .12) 3.79 ( 0 .09) F (4,604) = 1.17, p = .324, partial 2 = .01 PACS Caucasian 3.92 ( 0 .08) 3.74 ( 0 .07) 3.96 ( 0 .09) 3.71 ( 0 .09) 3.62 ( 0 .11) 3.58 ( 0 .08) Hispanic 3.94 ( 0 .09) 3.65 ( 0 .08) 4.02 ( 0 .09) 3.71 ( 0 09) 3.83 ( 0 .11) 3.48 ( 0 .08) African American 3.95 ( 0 .09) 3.64 ( 0 .08) 3.76 ( 0 .10) 3.75 ( 0 .10) 3.79 ( 0 .12) 3.79 ( 0 .09) F (4,613) = 1.52, p = .198, partial 2 = .01 SATAQ Caucasian 3.93 ( 0 .08) 3.74 ( 0 .08) 3.94 ( 0 .09) 3.71 ( 0 .09) 3.63 ( 0 .11) 3.59 ( 0 .08) Hispanic 3.95 ( 0 .09) 3.64 ( 0 .08) 4.00 ( 0 .09) 3.71 ( 0 .09) 3.83 ( 0 .12) 3.50 ( 0 .08) African American 3.97 ( 0 .09) 3.62 ( 0 .08) 3.76 ( 0 .10) 3.75 ( 0 .10) 3.82 ( 0 .12) 3.80 ( 0 .09) F (4,611) = 1.39, p = .236, partial 2 = .01

PAGE 122

110 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure H No covariates Caucasian 4.76 ( 0 .84) 4.46 ( 0 .60) 4.57 ( 0 .81) 4.40 ( 0 .72) 4.43 ( 0 .87) 4.34 ( 0 .72) Hispanic 4.56 ( 0 .60) 4.38 ( 0 .64) 4.44 ( 0 .70) 4.32 ( 0 .55) 4.31 ( 0 .52) 4.33 ( 0 .68) African American 4.35 ( 0 .62) 4.33 ( 0 .67) 4.35 ( 0 .87) 4.16 ( 0 .46) 4.40 ( 0 .63) 4.10 ( 0 .44) F (4,618) = 1.78, p = .135, partial 2 = .01 BMI Caucasian 4.77 ( 0 .10) 4.43 ( 0 .09) 4.59 ( 0 .11) 4.39 ( 0 .10) 4.45 ( 0 .13) 4.35 ( 0 .09) Hispanic 4.57 ( 0 .09) 4.35 ( 0 .08) 4.45 ( 0 .09) 4.31 ( 0 .09) 4.33 ( 0 .11) 4.33 ( 0 .08) African American 4.46 ( 0 .09) 4.31 ( 0 .08) 4.37 ( 0 .11) 4.14 ( 0 .09) 4.42 ( 0 .11) 4.11 ( 0 .08) F (4,614) = 1.70, p = .152, partial 2 = .01 EDI BD Caucasian 4.76 ( 0 .10) 4.46 ( 0 .09) 4.58 ( 0 .11) 4.40 ( 0 .10) 4.42 ( 0 .13) 4.34 ( 0 .09) Hispanic 4.56 ( 0 .09) 4.38 ( 0 .08) 4.44 ( 0 .09) 4.33 ( 0 .09) 4.28 ( 0 .11) 4.33 ( 0 .08) African American 4.35 ( 0 .09) 4.33 ( 0 .08) 4.36 ( 0 .09) 4.15 ( 0 .09) 4.41 ( 0 .11) 4.10 ( 0 .08) F (4,616) = 1.82, p = .126, partial 2 = .01

PAGE 123

111 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 4.73 ( 0 .10) 4.47 ( 0 .11) 4.52 ( 0 .12) 4.43 ( 0 .10) 4.40 ( 0 .13) 4.38 ( 0 .10) Hispanic 4.56 ( 0 .09) 4.37 ( 0 .08) 4.45 ( 0 .10) 4.32 ( 0 .09) 4.31 ( 0 .10) 4.34 ( 0 .08) African American 4.32 ( 0 .09) 4.36 ( 0 .08) 4.32 ( 0 .10) 4.18 ( 0 .08) 4.38 ( 0 .11) 4.13 ( 0 .08) F (4,609) = 1.83, p = .125, part ial 2 = .01 PACS Caucasian 4.76 ( 0 .10) 4.45 ( 0 .09) 4.58 ( 0 .11) 4.40 ( 0 .10) 4.43 ( 0 .13) 4.35 ( 0 .09) Hispanic 4.57 ( 0 .09) 4.37 ( 0 .08) 4.44 ( 0 .09) 4.32 ( 0 .09) 4.31 ( 0 .11) 4.32 ( 0 .08) African American 4.33 ( 0 .08) 4.30 ( 0 .08) 4.39 ( 0 .09) 4.15 ( 0 .09) 4.44 ( 0 .11) 4.11 ( 0 .08) F (4,613) = 1.83, p = .124, partial 2 = .01 SATAQ SATAQ Caucasian 4.77 ( 0 .10) 4.41 ( 0 .09) 4.57 ( 0 .11) 4.39 ( 0 .10) 4.47 ( 0 .13) 4.37 ( 0 .09) Hispanic 4.55 ( 0 .09) 4.38 ( 0 .09) 4.44 ( 0 .09) 4.32 ( 0 .09) 4.31 ( 0 .11) 4.32 ( 0 .08) African American 4.36 ( 0 .09) 4.31 ( 0 .08) 4.36 ( 0 .09) 4,15 ( 0 .09) 4.43 ( 0 .11) 4.12 ( 0 .08) F (4,612) = 1.60, p = .175, partial 2 = .01

PAGE 124

112 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure K No covariates Caucasian 4.49 ( 0 .60) 4.54 ( 0 .60) 4.47 ( 0 .61) 4.30 ( 0 .64) 4.20 ( 0 .73) 4.30 ( 0 .63) Hispanic 4.52 ( 0 .63) 4.50 ( 0 .56) 4.50 ( 0 .58) 4.36 ( 0 .62) 4.45 ( 0 .56) 4.39 ( 0 .52) African American 4.73 ( 0 .66) 4.56 ( 0 .67) 4.57 ( 0 .64) 4.34 ( 0 .54) 4.34 ( 0 .56) 4.30 ( 0 .47) F (4,640) = 0.78, p = .539, partial 2 = .01 BMI Caucasian 4.47 ( 0 .09) 4.51 ( 0 .08) 4.49 ( 0 .09) 4.29 ( 0 .09) 4.22 ( 0 .11) 4.31 ( 0 .08) Hispanic 4.50 ( 0 .08) 4.49 ( 0 .07) 4.51 ( 0 .08) 4.36 ( 0 .08) 4.46 ( 0 .10) 4.38 ( 0 .07) African American 4.75 ( 0 .08) 4.53 ( 0 .07) 4.58 ( 0 .09) 4.42 ( 0 .08) 4.38 ( 0 .11) 4.31 ( 0 .07) F (4,636) = 0.77, p = .543, partial 2 = .01 EDI BD Caucasian 4.46 ( 0 .09) 4.53 ( 0 .08) 4.48 ( 0 .09) 4.30 ( 0 .09) 4.21 (0. 12) 4.30 ( 0 .08) Hispanic 4.51 ( 0 .08) 4.48 ( 0 .07) 4.51 ( 0 .08) 4.35 ( 0 .08) 4.50 ( 0 .11) 4.38 ( 0 .07) African American 4.73 ( 0 .08) 4.56 ( 0 .08) 4.56 ( 0 .11) 4.43 ( 0 .08) 4.35 ( 0 .11) 4.30 ( 0 .07) F (4,638) = 0.79, p = .534, partial 2 = .01

PAGE 125

113 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 4.48 ( 0 .09) 4.53 ( 0 .08) 4.53 ( 0 .10) 4.29 ( 0 .09) 4.21 ( 0 .11) 4.27 ( 0 .08) Hispanic 4.52 ( 0 .08) 4.84 ( 0 .07) 4.55 ( 0 .09) 4.34 ( 0 .08) 4.48 ( 0 .07) 4.35 ( 0 .08) African American 4.74 ( 0 .09) 455 ( 0 .07) 4.55 ( 0 .10) 4.43 ( 0 .08) 4.55 ( 0 .07) 4.28 ( 0 .08) F (4,629) = 1.02, p = .397, partial 2 = .01 PACS Caucasian 4.46 ( 0 .09) 4.53 ( 0 .08) 4.48 ( 0 .09) 4.30 ( 0 .09) 4.20 ( 0 .11) 4.29 ( 0 .08) Hispanic 4.50 ( 0 .08) 4.50 ( 0 .07) 4.50 ( 0 .08) 4 .36 ( 0 .08) 4.46 ( 0 .10) 4.37 ( 0 .07) African American 4.73 ( 0 .08) 4.55 ( 0 .07) 4.57 ( 0 .09) 4.43 ( 0 .08) 4.36 ( 0 .11) 4.29 ( 0 .08) F (4,636) = 0.76, p = .550, partial 2 = .01 SATAQ Caucasian 4.47 ( 0 .09) 4.52 ( 0 .08) 4.48 ( 0 .09) 4.30 ( 0 .09) 4.28 ( 0 .11) 4.31 ( 0 .08) Hispanic 4.49 ( 0 .08) 4.48 ( 0 .07) 4.50 ( 0 .08) 4.36 ( 0 .08) 4.50 ( 0 .10) 4.39 ( 0 .07) African American 4.73 ( 0 .08) 4.51 ( 0 .07) 4.55 ( 0 .08) 4.43 ( 0 .08) 4.41 ( 0 .11) 4.33 ( 0 .08) F (4,634) = 0.86, p = .490, par tial 2 = .01

PAGE 126

114 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure L No covariates Caucasian 5.26 ( 0 .76) 5.18 ( 0 .75) 5.12 ( 0 .74) 5.11 ( 0 .75) 5.05 ( 0 .81) 4.90 ( 0 .78) Hispanic 4.94 ( 0 .64) 4.81 ( 0 .77) 4.88 ( 0 .86) 4.83 ( 0 .62) 4.81 ( 0 .70) 4.61 ( 0 .83) African American 5.15 ( 0 .62) 5.10 ( 0 .71) 5.24 ( 0 .62) 5.13 ( 0 .65) 4.81 ( 0 .70) 4.77 ( 0 .77) F (4,634) = 0.32, p = .865, partial 2 < .01 BMI Caucasian 5.25 ( 0 .11) 5.13 ( 0 .09) 5.14 ( 0 .11) 5.09 ( 0 .11) 5.08 ( 0 .14) 4.91 ( 0 .10) Hispanic 4.94 ( 0 .10) 4.82 ( 0 .09) 4.88 ( 0 .11) 4.83 ( 0 .09) 4.82 ( 0 .14) 4.64 ( 0 .09) African American 5.14 ( 0 .10) 5.08 ( 0 .08) 5.24 ( 0 .10) 5.13 ( 0 .09) 4.80 ( 0 .12) 4.80 ( 0 .09) F (4,631) = 0.31, p = .870, partial 2 < .01 EDI BD Caucasian 5.22 ( 0 .11) 5.20 ( 0 .09) 5.11 ( 0 .11) 5.12 ( 0 .11) 5.01 ( 0 .14) 4.90 ( 0 .10) Hispanic 4.94 ( 0 .10) 4.84 ( 0 .10) 4.88 ( 0 .11) 4.84 ( 0 .10) 4.81 ( 0 .14) 4.64 ( 0 .09) African American 5.14 ( 0 .10) 5.09 ( 0 .08) 5.24 ( 0 .10) 5.13 ( 0 .09) 4.80 ( 0 .13) 4.79 ( 0 .09) F (4,632) = 0.27, p = .898, partial 2 < .01

PAGE 127

115 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 5.23 ( 0 .11) 5.19 ( 0 .09) 5.13 ( 0 .12) 5.11 (. 0 11) 5.05 ( 0 .14) 4.93 ( 0 .10) Hispanic 4.94 ( 0 .11) 4.81 ( 0 .09) 4.88 ( 0 .12) 4.83 ( 0 .11) 4.83 ( 0 .13) 4.61 ( 0 .10) African American 5.13 ( 0 .10) 5.10 ( 0 .08) 5.23 ( 0 .10) 5.14 ( 0 .09) 4.79 ( 0 .12) 4.85 ( 0 .09) F (4,622) = 0.49, p = .74 1, partial 2 < .01 PACS Caucasian 5.22 ( 0 .11) 5.15 ( 0 .09) 5.13 ( 0 .11) 5.11 ( 0 .11) 5.07 ( 0 .14) 4.92 ( 0 .10) Hispanic 4.94 ( 0 .10) 4.80 ( 0 .09) 4.89 ( 0 .13) 4.83 ( 0 .10) 4.84 ( 0 .13) 4.63 ( 0 .09) African American 5.13 ( 0 .10) 5.08 ( 0 .08) 4.82 ( 0 .12) 5.13 ( 0 .09) 4.82 ( 0 .12) 4.81 ( 0 .09) F (4,631) = 0.36, p = .836, partial 2 < .01 SATAQ Caucasian 5.24 ( 0 .11) 5.14 ( 0 .09) 5.11 ( 0 .11) 5.11 ( 0 .11) 5.08 ( 0 .14) 4.93 ( 0 .10) Hispanic 4.95 ( 0 .10) 4.79 ( 0 .09) 4.88 ( 0 .11) 4.83 ( 0 .10) 4.84 ( 0 .14) 4.66 ( 0 .09) African American 5.14 ( 0 .09 ) 5.05 ( 0 .08) 5.23 ( 0 .10) 5.13 ( 0 .09) 4.83 ( 0 .12) 4.82 ( 0 .90) F (4,630) = 0.31, p = .870, partial 2 < .01

PAGE 128

116 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure O No covariates Caucasian 5.64 ( 0 .65) 5.42 (1.02) 5.45 ( 0 .76) 5.34 ( 0 .52) 5.34 ( 0 .53) 5.17 ( 0 .86) Hispanic 5.04 (1.12) 5.20 ( 0 .70) 5.30 ( 0 .86) 5.21 ( 0 .79) 4.83 (1.02) 4.92 ( 0 .88) African American 5.51 ( 0 .77) 5.30 ( 0 .79) 5.37 ( 0 .85) 5.34 ( 0 .76) 5.16 ( 0 .70) 5.08 ( 0 .91) F (4,646) = 0.91, p = .457, partial 2 = .01 BMI Caucasian 5.65 ( 0 .11) 5.38 ( 0 .09) 5.46 ( 0 .11) 5.34 ( 0 .11) 5.38 ( 0 .12) 5.19 ( 0 .10) Hispanic 5.03 ( 0 .12) 5.21 ( 0 .11) 5.29 ( 0 .13) 5.21 ( 0 .11) 4.85 ( 0 .16) 4.91 ( 0 .11) African American 5.51 ( 0 .11) 5.29 ( 0 .10) 5.38 ( 0 .12) 5.34 ( 0 .11) 5.21 ( 0 .14) 5.09 ( 0 .10) F (4,642) = 0.9 5, p = .433, partial 2 = .01 EDI BD Caucasian 5.63 ( 0 .11) 5.43 ( 0 .09) 5.44 ( 0 .11) 5.36 ( 0 .11) 5.33 ( 0 .14) 5.18 ( 0 .10) Hispanic 5.04 ( 0 .12) 5.19 ( 0 .11) 5.31 ( 0 .13) 5.20 ( 0 .12) 4.86 ( 0 .16) 4.91 ( 0 .11) African American 5.52 ( 0 .11) 5.28 ( 0 .10) 5.39 ( 0 .12) 5.33 ( 0 .11) 5.20 ( 0 .15) 5.08 ( 0 .10) F (4, 644) = 0.88, p = .477, partial 2 = .01

PAGE 129

117 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 5.63 ( 0 .11) 5.43 ( 0 .11) 5.42 ( 0 .12) 5.36 ( 0 .11) 5.35 ( 0 .14) 5.20 ( 0 .10) Hispanic 5.05 ( 0 .12) 5.20 ( 0 .13) 5.34 ( 0 .14) 5.20 ( 0 .13) 4.84 ( 0 .16) 4.92 ( 0 .12) African American 5.51 ( 0 .11) 5.30 ( 0 .10) 5.39 ( 0 .13) 5.34 ( 0 .12) 5.17 ( 0 .14) 5.09 ( 0 .11) F (4,636) = 1.05, p = .382, partial 2 = .01 PACS Caucasian 5.63 ( 0 .11) 5.39 ( 0 .09) 5.46 ( 0 .11) 5.36 ( 0 .09) 5.39 ( 0 .14) 5.19 ( 0 .10) Hispanic 5.02 ( 0 .11) 5.16 ( 0 .11) 5.33 ( 0 .12) 5.16 ( 0 .11) 4.88 ( 0 .16) 4.91 ( 0 .11) African American 5.50 ( 0 .11) 5.26 ( 0 .10) 5.41 ( 0 .12) 5.26 ( 0 .10) 5.21 ( 0 .14) 5.10 ( 0 .10) F (4,642) = 0.91, p = .457, partial 2 = .01 SATAQ Caucasian 5.64 ( 0 .11) 5.38 ( 0 .10) 5.44 ( 0 .11) 5.36 ( 0 .11) 5.40 ( 0 .14) 5.20 ( 0 .10) Hispanic 5.04 ( 0 .12) 5.19 ( 0 .11) 5.30 ( 0 .13) 5.21 ( 0 .12) 4.84 ( 0 .16) 4.93 ( 0 .11) African American 5.52 ( 0 .11) 5.20 ( 0 .10) 5.37 ( 0 .11) 5.33 ( 0 .11) 5.26 ( 0 .14) 5.15 ( 0 .10) F (4,639) = 1.19, p = .3 15, partial 2 = .01

PAGE 130

118 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure P No covariates Caucasian 4.36 ( 0 .70) 4.24 ( 0 .49) 4.24 ( 0 .59) 4.17 ( 0 .38) 4.37 ( 0 .67) c 4.05 ( 0 .35) d Hispanic 4.22 ( 0 .57) 4.15 ( 0 .46) 4.22 (0. 73) a 3.98 ( 0 .50) b 4.07 ( 0 .80) 3.98 ( 0 .50) African American 4.36 ( 0 .62) 4.19 ( 0 .54) 3.98 ( 0 .77) 4.19 ( 0 .44) 4 .22 ( 0 .62) 4.10 ( 0 .63) F(4,611) = 3.54, p = .008, partial 2 = .02 BMI Caucasian 4.36 ( 0 .07) 4.23 ( 0 .07) 4.24 ( 0 .08) 4.18 ( 0 .08) 4.38 ( 0 .09) e 4.05 ( 0 .07) f Hispanic 4.22 ( 0 .08) 4.14 ( 0 .07) 4.23 ( 0 .08) a 3.98 ( 0 .08) b 4.08 ( 0 .10) 3.98 ( 0 .07) Af rican American 4.35 ( 0 .08) 4.20 ( 0 .07) 3.96 ( 0 .09) c 4.21 ( 0 .07) d 4.21 ( 0 .11) 4.09 ( 0 .07) F(4,608) = 3.48, p = .009, partial 2 = .02 EDI BD Caucasian 4.36 ( 0 .07) 4.24 ( 0 .06) 4.24 ( 0 .08) 4.18 ( 0 .08) 4.36 ( 0 .10) c 4.05 ( 0 .07) c Hispanic 4.22 ( 0 .08 ) 4.16 ( 0 .07) 4.22 ( 0 .08) a 3.98 ( 0 .08) b 4.06 ( 0 .11) 3.98 ( 0 .07) African American 4.37 ( 0 .08) 4.15 ( 0 .07) 3.99 ( 0 .09) 4.17 ( 0 .08) 4.26 ( 0 .11) 4.09 ( 0 .07) F(4,609) = 3.68, p = .007, partial 2 = .02

PAGE 131

119 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 4.39 ( 0 .07) 4.22 ( 0 .06) 4.27 ( 0 .08) 4.16 ( 0 .08) 4.38 ( 0 .09) c 4.03 ( 0 .07) d Hispanic 4.24 ( 0 .08) 4.13 ( 0 .07) 4.27 ( 0 .09) a 3.96 ( 0 .08) b 4.09 ( 0 .10) 3.95 ( 0 .08) African American 4.37 ( 0 .08) 4.17 ( 0 .07) 3.99 ( 0 .10) 4.19 ( 0 .09) 4.22 ( 0 .10) 4.08 ( 0 .08) F (4,600) = 3.73, p = .006, partial 2 = .02 PACS Caucasian 4.37 ( 0 .07 ) 4.24 ( 0 .06) 4.24 ( 0 .08) 4.18 ( 0 .08) 4.36 ( 0 .09) a 4.05 ( 0 .07) b Hispanic 4.23 ( 0 .08) 4.18 ( 0 .08) 4.20 ( 0 .08) 3.98 ( 0 .08) 4.04 ( 0 .10) 3.97 ( 0 .07) African American 4.36 ( 0 .08) 4.16 ( 0 .07) 3.99 ( 0 .09) 4.20 ( 0 .0 8) 4.23 ( 0 .11) 4.10 ( 0 .08) F(4,608) = 3.49, p = .009, partial 2 = .02 SATAQ Caucasian 4.35 ( 0 .07) 4.24 ( 0 .08) 4.24 ( 0 .08) 4.18 ( 0 .08) 4.37 ( 0 .08) c 4.05 ( 0 .07) d Hispanic 4.19 ( 0 .08) 4.15 ( 0 .07) 4.22 ( 0 .08) a 3.98 ( 0 .08) b 4.22 ( 0 .08) 3.98 ( 0 .07) African American 4.35 ( 0 .08) 4.17 ( 0 .07) 3.98 ( 0 .09) 4.20 ( 0 .08) 4.23 ( 0 .09) 4.10 ( 0 .08) F(4,607) = 3.65, p = .007, partial 2 = .02

PAGE 132

120 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure Q No covariates Caucasian 4.07 ( 0 .50) 3.95 ( 0 .33) 4.06 ( 0 .24) 4.13 ( 0 .59) 4.14 ( 0 .42) 3.93 ( 0 .50) H ispanic 4.06 ( 0 .65) 4.00 ( 0 .29) 3.88 ( 0 .52) 4.04 ( 0 .34) 4.02 ( 0 .43) 3.95 ( 0 .33) African American 3.96 ( 0 .64) 3.97 ( 0 .29) 4.00 ( 0 .29) 4.04 ( 0 .27) 4.10 ( 0 .63) 3.96 ( 0 .40) F (4,651) = 0.49, p = .744, partial 2 < .01 BMI Caucasian 4.07 ( 0 .06) 3.94 ( 0 .05) 4.06 ( 0 .06) 4.13 ( 0 .06) 4.11 ( 0 .08) 3.93 ( 0 .06) Hispanic 4.06 ( 0 .06) 3.99 ( 0 .05) 3.89 ( 0 .06) 4.03 ( 0 .06) 4.02 ( 0 .08) 3.95 ( 0 .05) African American 3.96 ( 0 .06) 3.98 ( 0 .05) 4.00 ( 0 .06) 4.04 ( 0 .06) 4.11 ( 0 .08) 3.96 ( 0 .05) F (4,647) = 0.51, p = .732, partial 2 < .01 EDI BD Caucas ian 4.07 ( 0 .06) 3.95 ( 0 .05) 4.05 ( 0 .06) 4.14 ( 0 .06) 4.12 ( 0 .08) 3.93 ( 0 .06) Hispanic 4.05 ( 0 .06) 4.01 ( 0 .05) 3.88 ( 0 .06) 4.04 ( 0 .06) 3.99 ( 0 .08) 3.95 ( 0 .05) African American 3.96 ( 0 .06) 3.97 ( 0 .05) 4.00 ( 0 .0 6) 4.04 ( 0 .06) 4.11 ( 0 .08) 3.96 ( 0 .05) F (4,649) = 0.47, p = .756, partial 2 < .01

PAGE 133

121 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 4.07 ( 0 .06) 3.95 ( 0 .05) 4.05 ( 0 .07) 4.14 ( 0 .06) 4.13 ( 0 .08) 3.93 ( 0 .06) Hispanic 4.04 ( 0 .06) 4.01 ( 0 .05) 3.89 ( 0 .07) 4.05 ( 0 .06) 4.00 ( 0 .07) 3.97 ( 0 .06) African American 3.95 ( 0 .06) 3.98 ( 0 .05) 3.95 ( 0 .07) 4.05 ( 0 .06) 4.10 ( 0 .08) 3.98 ( 0 .06) F (4,642) = 0.34, p = .850, partial 2 < .01 PACS Caucasian 4.08 ( 0 .06) 3.97 ( 0 .05) 4.04 ( 0 .06) 4.13 ( 0 .06) 4.11 ( 0 .08) 3.92 ( 0 .06) Hispanic 4.06 ( 0 .06) 4.02 ( 0 .05) 3.87 ( 0 .06) 4.04 ( 0 .06) 3.99 ( 0 .08) 3.94 ( 0 .05) African American 3.97 ( 0 .06) 3.98 ( 0 .05) 4.00 ( 0 .06) 4.04 ( 0 .06) 4.10 ( 0 .08) 3.96 ( 0 .05) F (4,647) = 0.50, p = .734, partial 2 < .01 SATAQ Caucasian 4.07 ( 0 .06) 3.95 ( 0 .06) 4.06 ( 0 .06) 4.13 ( 0 .06) 4.13 ( 0 .08) 3.93 ( 0 .0 6) Hispanic 4.10 ( 0 .06) 3.99 ( 0 .05) 3.88 ( 0 .06) 4.04 ( 0 .06) 4.03 ( 0 .08) 3.96 ( 0 .05) African American 3.95 ( 0 .06) 3.95 ( 0 .05) 4.00 ( 0 .06) 4.04 ( 0 .06) 4.13 ( 0 .08) 3.98 ( 0 .05) F (4,646) = 0.72, p = .580, partial 2 < .01

PAGE 134

122 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure R No covariates Caucasian 3.51 ( 0 .74) 3.37 ( 0 .69) 3.58 ( 0 .73) 3.49 ( 0 .70) 3.60 ( 0 .92) 3.22 ( 0 .89) Hispanic 3.62 ( 0 .71) 3.34 ( 0 .76) 3.55 ( 0 .64) 3.34 ( 0 .62) 3.25 ( 0 .92) 3.43 (1.00) African American 3.35 ( 0 .62) 3.26 ( 0 .88) 3.4 1 ( 0 .73) 3.17 ( 0 .65) 3.22 ( 0 .93) 3.21 ( 0 .92) F (4,619) = 2.68, p = .033, partial 2 = .02 BMI Caucasian 3.52 ( 0 .11) 3.39 ( 0 .10) 3.58 ( 0 .11) 3.48 ( 0 .11) 3.59 ( 0 .14) 3.21 ( 0 .10) Hispanic 3.63 ( 0 .11) 3.34 ( 0 .10) 3.58 ( 0 .11) 3.33 ( 0 .11) 3.22 ( 0 .14) 3.43 ( 0 .10) African American 3.35 ( 0 .11) 3.27 ( 0 .10) 3.41 ( 0 .11) 3.18 ( 0 .11) 3.22 ( 0 .14) 3.21 ( 0 .10) F (4,615) = 2.61, p = .036, partial 2 = .02 EDI BD Caucasian 3.53 ( 0 .11) 3.36 ( 0 .09) 3.59 ( 0 .11) 3.47 ( 0 .11) 3.65 ( 0 .14) 3.21 ( 0 .10) Hispa nic 3.62 ( 0 .11) 3.35 ( 0 .09) 3.57 ( 0 .11) 3.34 ( 0 .11) 3.20 ( 0 .14) 3.44 ( 0 .10) African American 3.35 ( 0 .11) 3.27 ( 0 .10) 3.42 ( 0 .11) 3.18 ( 0 .11) 3.19 ( 0 .14) 3.22 ( 0 .10) F (4,618) = 3.03, p = .018, partial 2 = .02

PAGE 135

123 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 3.56 ( 0 .11) 3.36 ( 0 .09) 3.65 ( 0 .12) 3.44 ( 0 .11) 3.63 ( 0 .14) 3.1 6 ( 0 .10) Hispanic 3.66 ( 0 .11) 3.30 ( 0 .09) 3.62 ( 0 .12) 3.30 ( 0 .11) 3.27 ( 0 .13) 3.34 ( 0 .10) African American 3.42 ( 0 .11) 3.20 ( 0 .10) 3.55 ( 0 .12) 3.11 ( 0 .11) 3.26 ( 0 .14) 3.12 ( 0 .11) F (4,613) = 2.46, p = .046, partial 2 = .02 PACS Caucasian 3. 52 ( 0 .11) 3.40 ( 0 .09) 3.57 ( 0 .11) 3.48 ( 0 .11) 3.59 ( 0 .14) 3.20 ( 0 .10) Hispanic 3.65 ( 0 .11) 3.37 ( 0 .09) 3.55 ( 0 .11) 3.33 ( 0 .11) 3.20 ( 0 .14) 3.42 ( 0 .10) African American 3.36 ( 0 .11) 3.29 ( 0 .10) 3.40 ( 0 .11) 3.1 8 ( 0 .11) 3.19 ( 0 .14) 3.21 ( 0 .10) F (4,615) = 2.74, p = .030, partial 2 = .02 SATAQ Caucasian 3.53 ( 0 .11) 3.40 ( 0 .10) 3.58 ( 0 .11) 3.48 ( 0 .11) 3.67 ( 0 .14) c 3.21 ( 0 .10) d Hispanic 3.68 ( 0 .11) a 3.33 ( 0 .09) b 3.58 ( 0 .11) 3.33 ( 0 .11) 3.29 ( 0 .14) 3.44 ( 0 .10) African American 3.34 ( 0 .11) 3.26 ( 0 .10) 3.42 ( 0 .11) 3.17 ( 0 .11) 3.20 ( 0 .14) 3.21 ( 0 .10) F (4,607) = 3.36, p = .011, partial 2 = .02

PAGE 136

124 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure S No covariates Caucasian 4.52 ( 0 .77) 4.28 ( 0 .54) 4.50 ( 0 .71) 4.40 ( 0 .57) 4.43 ( 0 .80) 4.14 ( 0 .59) Hispanic 4.60 ( 0 .69) 4.35 ( 0 .61) 4.48 ( 0 .81) 4.34 ( 0 .59) 4.28 ( 0 .77) 4.24 ( 0 .67) African American 4.60 ( 0 .85) 4.30 ( 0 .49) 4.38 ( 0 .67) 4.42 ( 0 .66) 4.49 ( 0 .76) 4.30 ( 0 .58) F (4,633) = 1.08, p = .368, partial 2 = 01 BMI Caucasian 4.54 ( 0 .09) 4.28 ( 0 .08) 4.46 ( 0 .09) 4.40 ( 0 .09) 4.35 ( 0 .12) 4.14 ( 0 .08) Hispanic 4.61 ( 0 .10) 4.35 ( 0 .08) 4.48 ( 0 .10) 4.34 ( 0 .10) 4.29 ( 0 .12) 4.23 ( 0 .09) African American 4.62 ( 0 .09) 4.25 ( 0 .08) 4.38 ( 0 .10) 4.41 ( 0 .09) 4.50 ( 0 .12) 4.30 ( 0 .08) F (4,629) = 1.00, p = .408, partial 2 = .01 EDI BD Caucasian 4.53 ( 0 .09) 4.29 ( 0 .08) 4.46 ( 0 .09) 4.40 ( 0 .09) 4.36 ( 0 .12) 4.15 ( 0 .08) Hispanic 4.62 ( 0 .10) 4.34 ( 0 .08) 4.48 ( 0 .10) 4.34 ( 0 .10) 4.29 ( 0 .13) 4.23 ( 0 .08) African American 4.60 ( 0 .09) 4.29 ( 0 .08) 4.3 7 ( 0 .10) 4.43 (.09) 4.43 ( 0 .12) 4.30 ( 0 .08) F (4,631) = 0.92, p = .453, partial 2 = .01

PAGE 137

125 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 4.58 ( 0 .09) 4.26 ( 0 .08) 4.54 ( 0 .11) 4.36 ( 0 .09) 4.40 ( 0 .12) 4.11 ( 0 .09) Hispanic 4.64 ( 0 .10) 4.34 ( 0 .08) 4.51 ( 0 .11) 4.32 ( 0 .09) 4.30 ( 0 .12) 4.27 ( 0 .09) African American 4.61 ( 0 .09) 4.27 ( 0 .08) 4.33 ( 0 .11) 4.42 ( 0 .09) 4.47 ( 0 .12) 4.32 ( 0 .09) F (4,624) = 1.65, p = .159, partial 2 = .01 PACS Caucasian 4.55 ( 0 .09) 4.30 ( 0 .08) 4.44 ( 0 .09) 4.40 ( 0 .09) 4.34 ( 0 .12) 4.14 ( 0 .09) Hispanic 4.27 ( 0 .10) 4.37 ( 0 .08) 4.46 ( 0 .10) 4.34 ( 0 .09) 4.25 ( 0 .12) 4.23 ( 0 .09) African American 4.63 ( 0 .09) 4.30 ( 0 .08) 4.35 ( 0 .10) 4.42 ( 0 .09) 4.42 (.12) 4.29 ( 0 .08) F (4,629) = 1.09, p = .362, partial 2 = .01 SATAQ Caucasian 4.51 ( 0 .09) 4.28 ( 0 .08) 4.46 ( 0 .09) 4.40 ( 0 .09) 4.38 ( 0 .12) 4.14 ( 0 .08) Hispanic 4.59 ( 0 .10) 4.36 ( 0 .08) 4.48 ( 0 .10) 4.34 ( 0 .09) 4.27 ( 0 .13) 4.22 ( 0 .09) African American 4.59 ( 0 .09) 4.28 ( 0 .08) 4.38 ( 0 .09) 4.42 ( 0 .09) 4.47 ( 0 .12) 4.28 ( 0 .08) F (4,627) = 1.04, p = .388, partial 2 = .01

PAGE 138

126 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure V No covariates Caucasian 3.87 ( 0 .77) 3.60 ( 0 .66) 3.80 ( 0 .75) 3.79 ( 0 .69) 3.40 ( 0 .74) 3.43 ( 0 .89) Hispanic 3.87 ( 0 .79) 3.55 ( 0 .78) 3.73 ( 0 .70) 3.68 ( 0 .80) 3.55 ( 0 .56) 3.30 (1.03) African American 3.89 ( 0 .83) 3.69 ( 0 .57) 3.86 ( 0 .69) 3.83 ( 0 .73) 3.43 ( 0 .89) 3.45 (1.03) F (4,632) = 0.40, p = .808, partial 2 < .01 BMI Caucasian 3.87 ( 0 .10) 3.60 ( 0 .09) 3.80 ( 0 .11) 3.80 ( 0 .11) 3.36 ( 0 .14) 3.44 ( 0 .09) Hispanic 3.87 ( 0 .11) 3.56 ( 0 .10) 3.72 ( 0 .12) 3.68 ( 0 .11) 3.53 ( 0 .15) 3.30 ( 0 .10) African American 3.89 ( 0 .11) 3.69 ( 0 .10) 3.86 ( 0 .11) 3.83 ( 0 .11) 3.40 ( 0 .15) 3.45 ( 0 .10) F (4,628) = 0.40, p = .806, partial 2 < .01 EDI BD Caucasian 3.87 ( 0 .10) 3.60 ( 0 .09) 3.80 ( 0 .11) 3.80 ( 0 .11) 3.37 ( 0 .14) 3.44 ( 0 .09) H ispanic 3.86 ( 0 .11) 3.57 ( 0 .10) 3.72 ( 0 .11) 3.69 ( 0 .11) 3.50 ( 0 .15) 3.30 ( 0 .10) African American 3.89 ( 0 .11) 3.68 ( 0 .10) 3.86 ( 0 .11) 3.83 ( 0 .11) 3.43 ( 0 .15) 3.45 ( 0 .10) F (4,630) = 0.34, p = .850, partial 2 < .01

PAGE 139

127 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 3.92 ( 0 .11) 3.57 ( 0 .09) 3.86 ( 0 .12) 3.75 ( 0 .11) 3.42 ( 0 .14) 3.37 ( 0 .10) Hispanic 3.93 ( 0 .11) 3.52 ( 0 .10) 3.84 ( 0 .12) 3.63 ( 0 .11) 3.59 ( 0 .14) 3.17 ( 0 .10) African American 3.91 ( 0 .11) 3.67 ( 0 .10) 3.89 ( 0 .13) 3.81 ( 0 .11) 3.44 ( 0 .14) 3.39 ( 0 .11) F (4,620) = 0.43, p = .780, partial 2 < .01 PACS Caucasia n 3.88 ( 0 .10) 3.61 ( 0 .09) 3.79 ( 0 .11) 3.79 ( 0. 10) 3.37 ( 0 .14) 3.42 ( 0 .10) Hispanic 3.88 ( 0 .11) 3.57 ( 0 .10) 3.71 ( 0 .12) 3.68 ( 0 .11) 3.53 ( 0 .15) 3.28 ( 0 .10) African American 3.89 ( 0 .11) 3.68 ( 0 .10) 3.87 ( 0 .11) 3.83 ( 0 .11) 3.42 ( 0 .15) 3.45 ( 0 .10) F (4,628) = 0.40, p = .803, partial 2 < .01 SATAQ Caucasian 3.87 ( 0 .10) 3.58 ( 0 .09) 3.80 ( 0 .11) 3.79 ( 0 .11) 3.38 ( 0 .14) 3.44 ( 0 .10) Hispanic 3.90 ( 0 .11) 3.51 ( 0 .10) 3.72 ( 0 .11) 3.68 ( 0 .11) 3.57 ( 0 .15) 3.33 ( 0 .10) African American 3.89 ( 0 .11) 3.68 ( 0 .10) 3.86 ( 0 .11) 3.83 ( 0 .11) 3.41 ( 0 .15) 3.46 ( 0 .10) F (4,627) = 0.38, p = .816, partial 2 < .01

PAGE 140

128 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure W No covariates Caucasian 4.32 ( 0 .80) 4.26 ( 0 .57) 4.33 ( 0 .90) 4.29 ( 0 .67) 4.44 ( 0 .63) 4.27 ( 0 .68) Hispanic 4.74 ( 0 .56) 4.50 ( 0 .53) 4.53 ( 0 .70) 4.40 ( 0 .57) 4.63 ( 0 .71) 4.32 ( 0 .56) African American 4.56 ( 0 .60) 4.32 ( 0 .53) 4.47 ( 0 .58) 4.34 ( 0 .65) 4.37 ( 0 .64) 4.24 ( 0 .63) F (4,600) = 0.50, p = .724, partial 2 < 01 BMI Caucasian 4.28 ( 0 .10) 4.27 ( 0 .09) 4.32 ( 0 .10) 4.28 ( 0 .10) 4.45 ( 0 .13) 4.26 ( 0 .09) Hispanic 4.74 ( 0 .08) 4.49 ( 0 .07) 4.54 ( 0 .09) 4.41 ( 0 .08) 4.65 ( 0 .11) 4.30 ( 0 .08) African American 4.53 ( 0 .09) 4.35 ( 0 .07) 4.46 ( 0 .09) 4.34 ( 0 .07) 4.35 ( 0 .11) 4.24 ( 0 .08) F (4,596) = 0.50, p = .726, partial 2 < .01 EDI BD Caucasian 4.29 ( 0 .10) 4.26 ( 0 .09) 4.33 ( 0 .10) 4.27 ( 0 .10) 4.44 ( 0 .13) 4.26 ( 0 .09) Hispanic 4.73 ( 0 .08) 4.52 ( 0 .07) 4.53 ( 0 .09) 4.42 ( 0 .08) 4.62 ( 0 .11) 4.30 ( 0 .08) African American 4.54 ( 0 .08) 4.32 ( 0 .07) 4.4 7 ( 0 .09) 4.33 ( 0 .09) 4.38 ( 0 .11) 4.24 ( 0 .08) F (4,597) = 0.45, p = .764, partial 2 < .01

PAGE 141

129 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 4.24 ( 0 .10) 4.29 ( 0 .09) 4.21 ( 0 .11) 4.31 ( 0 .10) 4.41 ( 0 .12) 4.32 ( 0. 09) Hispanic 4.74 ( 0 .09) 4.49 ( 0 .07) 4.51 ( 0 .10) 4.41 ( 0 .09) 4.65 ( 0 .10) 4.30 ( 0 .08) African American 4.55 ( 0 .09) 4.32 ( 0 .07) 4.50 ( 0 .10) 4.32 ( 0 .09) 4.39 ( 0 .11) 4.24 ( 0 .08) F (4,590) = 0.58, p = .671, partial 2 < .01 PACS Caucasian 4.30 ( 0 .10) 4.29 ( 0 .08) 4.30 ( 0 .10) 4.28 ( 0 .10) 4.40 ( 0 .13) 4.25 ( 0 .09) Hispani c 4.74 ( 0 .08) 4.53 ( 0 .07) 4.51 ( 0 .09) 4.42 ( 0 .08) 4.62 ( 0 .11) 4.29 ( 0 .08) African American 4.55 ( 0 .08) 4.35 ( 0 .07) 4.45 ( 0 .09) 4.34 ( 0 .08) 4.35 ( 0 .11) 4.22 ( 0 .08) F (4,596) = 0.44, p = .773, partial 2 < .01 SATAQ Caucasian 4.29 ( 0 .10) 4.27 ( 0 .09) 4.33 ( 0 .10) 4.28 ( 0 .10) 4.44 ( 0 .13) 4.2 5 ( 0 .09) Hispanic 4.73 ( 0 .08) 4.50 ( 0 .08) 4.53 ( 0 .09) 4.41 ( 0 .08) 4.67 ( 0 .11) 4.30 ( 0 .08) African American 4.54 ( 0 .08) 4.34 ( 0 .07) 4.47 ( 0 .09) 4.34 ( 0 .09) 4.37 ( 0 .11) 4.23 ( 0 .08) F (4,596) = 0.53, p = .707, partial 2 < .01

PAGE 142

130 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure Z No covariates Caucasian 4.87 ( 0 .70) 4.53 ( 0 .95) 4.75 ( 0 .69) 4.83 ( 0 .61) 4.46 (1.00) 4.29 (1.00) Hispanic 5.00 ( 0 .54) 4.62 ( 0 .84) 5.06 ( 0 .61) 4.85 ( 0 .63) 4.69 ( 0 .85) 4.49 ( 0 .95) African American 4.84 ( 0 .79) 4.45 ( 0 .8 3) 4.61 ( 0 .80) 4.64 ( 0 .68) 4.34 ( 0 .99) 4.34 (1.03) F (4,643) = 0.53, p = .711, partial 2 < .01 BMI Caucasian 4.87 ( 0 .12) 4.52 ( 0 .10) 4.74 ( 0 .12) 4.83 ( 0 .12) 4.44 ( 0 .15) 4.28 ( 0 .11) Hispanic 5.00 ( 0 .10) 4.61 ( 0 .09) 5.06 ( 0 .11) 4.85 ( 0 .11) 4.69 ( 0 .13) 4.50 ( 0 .10) African American 4.84 ( 0 .12) 4.46 ( 0 .10) 4.61 ( 0 .12) 4.64 ( 0 .12) 4.32 ( 0 .15) 4.30 ( 0 .11) F (4,639) = 0.52, p = .722, partial 2 < .01 EDI BD Caucasian 4.88 ( 0 .12) 4.50 ( 0 .10) 4.75 ( 0 .12) 4.82 ( 0 .12) 4.50 ( 0 .15) 4.28 ( 0 .11) Hispanic 4.99 ( 0 .10) 4.64 ( 0 .11) 5.05 ( 0 .11) 4.86 ( 0 .11) 4.65 ( 0 .14) 4.50 ( 0 .10) African American 4.86 ( 0 .12) 4.41 ( 0 .10) 4.62 ( 0 .12) 4.62 ( 0 .12) 4.42 ( 0 .15) 4.29 ( 0 .11) F (4,640) = 0.64, p = .629, partial 2 < .01

PAGE 143

131 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 4.86 ( 0 .12) 4.52 ( 0 .10) 4.70 ( 0 .13) 4.84 ( 0 .12) 4.45 ( 0 .1 5) 4.29 ( 0 .11) Hispanic 4.99 ( 0 .10) 4.62 ( 0 .09) 5.03 ( 0 .12) 4.86 ( 0 .11) 4.69 ( 0 .13) 4.56 ( 0 .10) African American 4.87 ( 0 .12) 4.44 ( 0 .10) 4.66 ( 0 .13) 4.61 ( 0 .12) 4.36 ( 0 .15) 4.23 ( 0 .11) F (4,631) = 0.55, p = .697, partial 2 < .01 PACS Caucas ian 4.87 ( 0 .12) 4.51 ( 0 .10) 4.75 ( 0 .12) 4.83 ( 0 .12) 4.47 ( 0 .15) 4.28 ( 0 .11) Hispanic 5.00 ( 0 .10) 4.62 ( 0 .09) 5.06 ( 0 .11) 4.85 ( 0 .10) 4.69 ( 0 .13) 4.49 ( 0 .10) African American 4.83 ( 0 .12) 4.45 ( 0 .10) 4.61 ( 0 .1 2) 4.64 ( 0 .12) 4.35 ( 0 .15) 4.31 ( 0 .11) F (4,639) = 0.55, p = .698, partial 2 < .01 SATAQ Caucasian 4.88 ( 0 .12) 4.48 ( 0 .10) 4.74 ( 0 .12) 4.83 ( 0 .11) 4.51 ( 0 .15) 4.31 ( 0 .11) Hispanic 5.00 ( 0 .11) 4.62 ( 0 .09) 5.06 ( 0 .11) 4.85 ( 0 .11) 4.71 ( 0 .14) 4.5 0 ( 0 .10) African American 4.88 ( 0 .12) 4.42 ( 0 .10) 4.60 ( 0 .12) 4.64 ( 0 .12) 4.38 ( 0 .15) 4.32 ( 0 .11) F (4,637) = 0.74, p = .567, partial 2 = .01

PAGE 144

132 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure AA No covariates Caucasian 6.13 ( 0 .78) 5.99 ( 0 .82) 5.77 ( 0 .95) 6.11 ( 0 .78) 5.81 ( 0 .97) 5.85 ( 0 .98) Hispanic 5.94 (1.00) 5.93 ( 0 .80) 5.86 ( 0 .80) 5.98 ( 0 .78) 5.57 ( 0 .80) 5.78 ( 0 .87) African American 5.66 ( 0 .84) 5.57 ( 0 .86) 5.39 ( 0 .83) 5.60 ( 0 .79) 5.36 ( 0 .79) 5.37 ( 0 .99) F (4,646) = 0.82, p = .515, partial 2 = .01 BMI Caucasian 6.14 ( 0 .12) 5.94 ( 0 .11) 5.79 ( 0 .12) 6.10 ( 0 .12) 5. 92 ( 0 .15) 5.87 ( 0 .11) Hispanic 5.95 ( 0 .12) 5.89 ( 0 .10) 5.89 ( 0 .12) 5.96 ( 0 .12) 5.56 ( 0 .15) 5.79 ( 0 .11) African American 5.62 ( 0 .12) 5.53 ( 0 .10) 5.41 ( 0 .12) 5.59 ( 0 .12) 5.40 ( 0 .16) 5.38 ( 0 .11) F (4,642) = 0.91, p = .458, partial 2 = .01 EDI BD Caucasian 6.11 ( 0 .12) 6.00 ( 0 .10) 5.76 ( 0 .12) 6.12 ( 0 .12) 5.84 ( 0 .16) 5.86 ( 0 .11) Hispanic 5.93 ( 0 .12) 5.92 ( 0 .10) 5.87 ( 0 .12) 5.98 ( 0 .12) 5.56 ( 0 .16) 5.78 ( 0 .11) African American 5.59 ( 0 .12) 5.61 ( 0 1 0) 5.37 ( 0 .12) 5.62 ( 0 .12) 5.27 ( 0 .16) 5.38 ( 0 .11) F (4,644) = 0.73, p = .571, partial 2 = .01

PAGE 145

133 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 6.12 ( 0 .12) 5.97 ( 0 .10) 5.84 ( 0 .14) 6.11 ( 0 .12) 5.87 ( 0 .15) 5.86 ( 0 .11) Hispanic 5.91 ( 0 .12) 5.94 ( 0 .10) 5.81 ( 0 .13) 6.00 ( 0 .12) 5.53 ( 0 15) 5.82 ( 0 .11) African American 5.60 ( 0 .12) 5.58 ( 0 .10) 5.40 ( 0 .14) 5.61 ( 0 .12) 5.36 ( 0 .15) 5.39 ( 0 .11) F (4,635) = 0.55, p = .698, partial 2 < .01 PACS Caucasian 6.12 ( 0 .12) 6.00 ( 0 .10) 5.76 ( 0 .12) 6.11 ( 0 .12) 5.85 ( 0 .16) 5.84 ( 0 .11) Hisp anic 5.91 ( 0 .12) 5.89 ( 0 .10) 5.89 ( 0 .12) 5.98 ( 0 .12) 5.59 ( 0 .12) 5.79 ( 0 .11) African American 5.61 ( 0 .12) 5.57 ( 0 .10) 5.39 ( 0 .12) 5.60 ( 0 .12) 5.36 ( 0 .15) 5.36 ( 0 .11) F (4,642) = 0.87, p = .482, partial 2 = .01 SATAQ Caucasian 6.12 ( 0 .12) 5.97 ( 0 .11) 5.76 ( 0 .12) 6.11 ( 0 .12) 5.90 ( 0 .16) 5.87 ( 0 .11) Hispanic 5.94 ( 0 .12) 5.87 ( 0 .19) 5.86 ( 0 .12) 5.98 ( 0 .12) 5.62 ( 0 .15) 5.82 ( 0 .11) African American 5.60 ( 0 .12) 5.53 ( 0 .10) 5.39 ( 0 .12) 5.60 ( 0 .12) 5.41 ( 0 .16) 5.40 ( 0 .11) F (4,640) = 0.73, p = .574, partial 2 = .01

PAGE 146

134 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure DD No covariate s Caucasian 5.91 ( 0 .76) 5.62 ( 0 .82) 5.69 ( 0 .79) 5.67 ( 0 .88) 5.50 ( 0 .61) 5.33 (1.02) Hispanic 5.71 ( 0 .74) 5.67 ( 0 .76) 5.77 ( 0 .84) 5.68 ( 0 .83) 5.54 ( 0 .99) 5.46 ( 0 .89) African American 5.85 (1.16) 5.56 ( 0 .87) 5.86 ( 0 .76) 5.60 ( 0 .82) 5.29 (1.04) 5.46 ( 0 .96) F (4,618) = 1.15, p = .330, partial 2 = .01 BMI Caucasian 5.94 ( 0 .12) 5.55 ( 0 .10) 5.70 ( 0 .12) 5.64 ( 0 .12) 5.57 ( 0 .15) 5.35 ( 0 .10) Hispanic 5.72 ( 0 .11) 5.64 ( 0 .10) 5.75 ( 0 .12) 5.71 ( 0 .12) 5 .57 ( 0 .25) 5.46 ( 0 .10) African American 5.79 ( 0 .12) 5.61 ( 0 .11) 5.83 ( 0 .13) 5.63 ( 0 .13) 5.40 ( 0 .16) 5.44 ( 0 .11) F (4,619) = 1.02, p = .395, partial 2 = .01 EDI BD Caucasian 5.90 ( 0 .12) 5.62 ( 0 .10) 5.66 ( 0 .12) 5.68 ( 0 .12) 5.51 ( 0 .16) 5.33 ( 0 .10) Hispanic 5.73 ( 0 .11) 5.63 ( 0 .10) 5.75 ( 0 .12) 5.70 ( 0 .11) 5.60 ( 0 .15) 5.45 ( 0 .10) African American 5.85 ( 0 .12) 5.48 ( 0 .11) 5.88 ( 0 .13) 5.57 ( 0 .13) 5.54 ( 0 .17) 5.45 ( 0 .11) F (4,618) = 0.92, p = .451, partial 2 = 01

PAGE 147

135 Rater R ace Caucasian Hispanic African American Target R ace Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS Caucasian 5.91 ( 0 .12) 5.62 ( 0 .10) 5.66 ( 0 .13) 5.67 ( 0 .12) 5.52 ( 0 .15) 5.35 ( 0 .11) Hispanic 5.71 ( 0 .12) 5.64 ( 0 .12) 5.74 ( 0 .13) 5.72 ( 0 .12) 5.54 ( 0 .15) 5.48 ( 0 .11) African American 5.84 ( 0 .13) 5.59 ( 0 .13) 5.90 ( 0 .14) 5.59 ( 0 .13) 5.43 ( 0 .16) 5.46 ( 0 .12) F (4,606) = 1.15, p = .334, partial 2 = .01 PACS Ca ucasian 5.88 ( 0 .12) 5.55 ( 0 .10) 5.71 ( 0 .12) 5.67 ( 0 .10) 5.59 ( 0 .15) 5.36 ( 0 .10) Hispanic 5.70 ( 0 .11) 5.62 ( 0 .10) 5.77 ( 0 .12) 5.71 ( 0 .11) 5.59 ( 0 .15) 5.48 ( 0 .10) African American 5.82 ( 0 .13) 5.52 ( 0 .11) 5.88 ( 0 .13) 5.59 ( 0 .13) 5.45 ( 0 .16) 5.45 ( 0 .11) F (4,615) = 1.06, p = .374, partial 2 = .01 SATAQ Caucasian 5.93 ( 0 .12) 5.55 ( 0 .10) 5.65 ( 0 .12) 5.67 ( 0 .12) 5.56 ( 0 .15) 5.37 ( 0 .11) Hispanic 5.72 ( 0 .11) 5.58 ( 0 .10) 5.73 ( 0 .12) 5.70 ( 0 .11) 5.64 ( 0 .15) 5.51 ( 0 .10) African American 5.83 ( 0 .13) 5.51 ( 0 .11) 5.86 ( 0 .13) 5.59 ( 0 .13) 5.44 ( 0 .17) 5.48 ( 0 .11) F (4,612) = 1.26, p = .287, partial 2 = .01

PAGE 148

136 Table L.2. Repeated measures effects: Two way interactions (target race X covariate) Covariate F p and partial 2 values Figure A BMI F (2, 630) = 1.35, p = .259, partial 2 < .01 EDI BD F (2,632) = 0.65, p = .519, partial 2 < .01 DMS F (2,624) = 0.18, p = .627, partial 2 < .01 PACS F (2,630) = 0.66, p = .514, partial 2 < .01 SATAQ F (2,629) = 4.60, p = .011, partial 2 = .01 Figure B BMI F (2, 634) = 2.83, p = .060, partial 2 = .01 EDI BD F (2,634) = 0.19, p = .822, partial 2 < .01 DMS F (2,629) = 0.68, p = .505, partial 2 < .01 PACS F (2,632) = 1.80, p = .167, partial 2 = .01 SATAQ F (2,629) = 0.49, p = .613, partial 2 < .01 Figure C BMI F (2, 633) = 1.22, p = .295, partial 2 < .01 EDI BD F (2, 636) = 0.24, p = .786, partial 2 < .01 DMS F (2,627) = 0.40, p = .667, partial 2 < .01

PAGE 149

137 Covariate F p and partial 2 values PACS F (2,634) = 0.08, p = .922, partial 2 < .01 SATAQ F (2,632) = 1.35, p = .260, partial 2 < .01 Figure D BMI F (2,635) = 1.20, p = .302, partial 2 < .01 EDI BD F (2,638) = 0.10, p = .902, partial 2 < .01 DMS F (2,627) = 0.89, p = .410, partial 2 < .01 PACS F (2,636) = 0.21, p = .813, partial 2 < .01 SATAQ F (2,634) = 3.66, p = .027, partial 2 = .01 Fig ure G BMI F (2, 611) = 0.70, p = .510, partial 2 < .01 EDI BD F (2,614) = 1.61, p = .202, partial 2 = .01 DMS F (2,604) = 0.38, p = .670, partial 2 < .01 PACS F (2,613) = 2.07, p = .129, partial 2 = .01 SATAQ F (2,611) = 0.14, p = .859, partial 2 < .01 Figure H BMI F (2,614) = 0.01, p = .987, partial 2 < .01 EDI BD F (2,616) = 0.31, p = .722, partial 2 < .01 DMS F (2,609) = 0.82, p = .436, partial 2 < .01

PAGE 150

138 Covariate F p and partial 2 values PACS F (2,613) = 1.87, p = .156, partial 2 = .01 SATAQ F (2,612) = 1.66, p = .193, partial 2 = .01 Figure K BMI F (2,636) = 0.53, p = .589, partial 2 < .01 EDI BD F (2,638) = 0.31, p = .732, partial 2 < .01 DMS F (2,629) = 0.11, p = .894, partial 2 < .01 PACS F (2,636) = 0.40, p = .672, partial 2 < .01 SATAQ F (2,634) = 1.25, p = .289, partia l 2 < .01 Figure L BMI F (2,631) = 1.91, p = .150, partial 2 = .01 EDI BD F (2,632) = 0.48, p = .617, partial 2 < .01 DMS F (2,622) = 0.02, p = .976, partial 2 < .01 PACS F (2,631) = 0.18, p = .831, partial 2 < .01 SATAQ F (2,630) = 0.27, p = .764, p artial 2 < .01 Figure O BMI F (2,642) = 1.22, p = .295, partial 2 < .01 EDI BD F (2,644) = 0.61, p = .544, partial 2 < .01 DMS F (2,636) = 0.08, p = .921, partial 2 < .01

PAGE 151

139 Covariate F p and partial 2 values PACS F (2,642) = 0.09, p = .914, partial 2 < .01 SATAQ F (2,639) = 1.37, p = .0 51, partial 2 = .01 Figure P BMI F (2,608) = 2.55, p = .082, partial 2 = .01 EDI BD F (2,609) = 1.37, p = .254, partial 2 < .01 DMS F (2,600) = 0.49, p = .599, partial 2 < .01 PACS F (2,608) = 2.09, p = .128, partial 2 = .01 SATAQ F (2,607) = 0.80, p = .915, partial 2 < .01 Figure Q BMI F (2,647) = 0.14, p = .865, partial 2 < .01 EDI BD F (2,649) = 0.32, p = .725, partial 2 < .01 DMS F (2,642) = 0.26, p = .774, partial 2 < .01 PACS F (2,647) = 0.65, p = .521, partial 2 < .01 SATAQ F (2,646) = 1. 19, p = .304, partial 2 < .01 Figure R BMI F (2,615) = 0.05, p = .947, partial 2 < .01 EDI BD F (2,618) = 2.22, p = .112, partial 2 = .01 DMS F (2,613) = 0.67, p = .509, partial 2 < .01

PAGE 152

140 Covariate F p and partial 2 values PACS F (2,615) = 0.53, p = .581, partial 2 < .01 SATAQ F (2,607) = 0.10, p = .897, partial 2 < .01 Figure S BMI F (2,629) = 0.30, p = .743, partial 2 < .01 EDI BD F (2,631) = 1.20, p = .301, partial 2 < .01 DMS F (2,624) = 1.38, p = .253, partial 2 < .01 PACS F (2,629) = 0.27, p = .764, partial 2 < .01 SATAQ F (2 ,627) = 0.18, p = .835, partial 2 < .01 Figure V BMI F (2,628) = 0.02, p = .983, partial 2 < .01 EDI BD F (2,630) = 0.63, p = .531, partial 2 < .01 DMS F (2,620) = 0.64, p = .523, partial 2 < .01 PACS F (2,628) = 0.51, p = .597, partial 2 < .01 SATA Q F (2,627) = 0.64, p = .527, partial 2 < .01 Figure W BMI F (2,596) = 0.85, p = .422, partial 2 < .01 EDI BD F (2,597) = 0.50, p = .595, partial 2 < .01 DMS F (2,590) = 2.47, p = .089, partial 2 = .01

PAGE 153

141 Covariate F p and partial 2 values PACS F (2,596) = 0.03, p = .962, partial 2 < .01 SATAQ F (2,596) = 0.28, p = .743, partial 2 < .01 Figure Z BMI F (2,639) = 0.09, p = .913, partial 2 < .01 EDI BD F (2,640) = 3.06, p = .048, partial 2 = .01 DMS F (2,631) = 1.09, p = .338, partial 2 < .01 PACS F (2,639) = 0.20, p = .814, partial 2 < .01 SATAQ F (2,637) = 0.93, p = .395, partial 2 < .01 Figure AA BMI F (2,642) = 0.02, p = .980, partial 2 < .01 EDI BD F (2,644) = 2.09, p = .126, partial 2 = .01 DMS F (2,635) = 0.25, p = .779, partial 2 < .01 PACS F (2,642) = 2.30, p = .101, partia l 2 = .01 SATAQ F (2,640) = 1.07, p = .343, partial 2 < .01 Figure DD BMI F (2,619) = 10.27, p < .001 partial 2 = .03 EDI BD F (2,618) = 2.67, p = .072, partial 2 = .01 DMS F (2,606) = 0.22, p = .797, partial 2 < .01

PAGE 154

142 Covariate F p and partial 2 values PACS F (2,615) = 1.32, p = .377, partial 2 < .01 SATAQ F (2,612) = 0.97, p = .193, partial 2 = .01 Table L.3. Repeated measures effects: Two way interactions (target race X rater gender) Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure A No covariates 3.83 (1.00) 3.92 ( 0 .74) 3.65 ( 0 .98) 3.81 ( 0 .68) 3.86 (1.01) 3.95 ( 0 .67) F (2, 630) = 0.18, p = .832, partial 2 < .01 BMI 3.88 ( 0 .08) 3.92 ( 0 .06) 3.73 ( 0 .07) 3.79 ( 0 .06) 3.92 ( 0 .07) 3.92 ( 0 .06) F (2,632) = 0.22, p = .800, partial 2 < .01 EDI BD 3.83 ( 0 .08) 3.95 ( 0 .06) 3.65 ( 0 .07) 3.84 ( 0 .06) 3.83 ( 0 .07) 3.98 ( 0 .06) F (2,633) = 0.22, p = .926, partial 2 < .01

PAGE 155

143 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS 3.86 ( 0 .08) 3.94 ( 0 .07) 3.71 ( 0 .08) 3.81 ( 0 .07) 3.87 ( 0 .08) 3.95 ( 0 .07) F (2,624) = 0.01, p = .991, partial 2 < .01 PACS 3.84 ( 0 .07) 3.94 ( 0 .06) 3.68 ( 0 .07) 3.83 ( 0 .06) 3.85 ( 0 .07) 3.96 ( 0 .06) F (2,630) = 0.17, p = .843, partial 2 < .01 SATAQ 3.89 ( 0 .07) 3.93 ( 0 .06) 3.66 ( 0 .07) 3.83 ( 0 .06) 3.90 ( 0 .07) 3.95 ( 0 .06) F (2,629) = 1.09, p = .337, partial 2 < .01 Figure B No covariates 3.50 (1.01) 3.75 ( 0 .70) 3.53 ( 0 .92) 3.71 ( 0 .74) 3.44 ( 0 .93) 3.76 ( 0 .75) F (2,636) = 0.96, p = .384, partial 2 < .01 BMI 3.55 ( 0 .08) 3.73 ( 0 .06) 3.60 ( 0 .07) 3.69 ( 0 .06) 3.50 ( 0 .08) 3.72 ( 0 .06) F (2,634) = 0.97, p = .378, partial 2 < .01

PAGE 156

144 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values EDI BD 3.51 ( 0 .08) 3.76 ( 0 .06) 3.52 ( 0 .08) 3.73 ( 0 .06) 3.41 ( 0 .08) 3.77 ( 0 .06) F (2,634) = 1.14, p = .322, partial 2 < .01 DMS 3.52 ( 0 .08) 3.76 ( 0 .07) 3.53 ( 0 .08) 3.71 ( 0 .07) 3.47 ( 0 .08) 3.72 ( 0 .07) F (2,629) = 0.21, p = .808, partial 2 < .01 PACS 3.53 ( 0 .08) 3.74 ( 0 .06) 3.55 ( 0 .07) 3.71 ( 0 .06) 3.43 ( 0 .07) 3.76 ( 0 .06) F (2,632) = 1.43, p = .241, partial 2 < .01 SATAQ 3.51 ( 0 .08) 3.74 ( 0 .06) 3.54 ( 0 .07 ) 3.72 ( 0 .06) 3.46 ( 0 .08) 3.76 ( 0 .06) F (2,629) = 0.70, p = .494, partial 2 < .01 Figure C No covariates 2.21 (1.11) 2.50 (1.18) 2.50 (1.12) 2.66 (1.07) 2.01 (1.13) 2.35 (1.11) F (2,638) = 0.99, p = .372, partial 2 < .01

PAGE 157

145 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values BMI 2.20 ( 0 .11) 2.50 ( 0 .09) 2.52 ( 0 .10) 2.47 ( 0 .08) 2.06 ( 0 .10) 2.35 ( 0 .08) F (2,633) = 1.07, p = .343, partial 2 < .01 EDI BD 2.20 ( 0 .11) 2.50 ( 0 .09) 2.48 ( 0 .10) 2.67 ( 0 .08) 2.00 ( 0 .10) 2.38 ( 0 .08) F (2,636) = 0.99, p = .374, partial 2 < .01 DMS 2.18 ( 0 .11) 2.49 ( 0 .09) 2.52 ( 0 .11) 2.6 3 ( 0 .09) 2.03 ( 0 .11) 2.35 ( 0 .09) F (2,627) = 1.13, p = .323, partial 2 < .01 PACS 2.20 ( 0 .10) 2.49 ( 0 .09) 2.49 ( 0 .10) 2.66 ( 0 .08) 2.02 ( 0 .10) 2.36 ( 0 .08) F (2,633) = 1.07, p = .342, partial 2 < .01 SATAQ 2.23 ( 0 .10) 2.49 ( 0 .08) 2.51 ( 0 .10) 2.66 ( 0 .08) 2. 05 ( 0 .10) 2.35 ( 0 .08) F (2,632) = 0.74, p = .479, partial 2 < .01

PAGE 158

146 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure D No covariates 4.33 ( 0 .82) 4.15 ( 0 .56) 4.18 ( 0 .84) 4.10 ( 0 .67) 4.14 ( 0 .86) 4.09 ( 0 .53) F (2,640) = 1.41, p = .246, partial 2 < .01 BMI 4.35 ( 0 .06) 4.13 ( 0 .05) 4.16 ( 0 .07) 4.11 ( 0 05) 4.14 ( 0 .06) 4.08 ( 0 .05) F (2,635) = 2.04, p = .132, partial 2 = .01 EDI BD 4.30 ( 0 .06) 4.16 ( 0 .05) 4.14 ( 0 .07) 4.12 ( 0 .05) 4.13 ( 0 .06) 4.09 ( 0 .05) F (2,638) = 1.07, p = .343, partial 2 < .01 DMS 4.32 ( 0 .07) 4.15 ( 0 .06) 4.12 ( 0 .07) 4.14 ( 0 .06) 4.15 ( 0 .07) 4.07 ( 0 .06) F (2,627) = 1.61, p = .201, partial 2 = .01 PACS 4.32 ( 0 .06) 4.14 ( 0 .05) 4.16 ( 0 .06) 4.11 ( 0 .05) 4.13 ( 0 .06) 4.09 ( 0 .05) F (2,636) = 1.47, p = .230, partial 2 = .01

PAGE 159

147 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values SATAQ 4.33 ( 0 .06) 4.14 ( 0 .05) 4.14 ( 0 .06) 4.12 ( 0 .05) 4.14 ( 0 .06) 4.08 ( 0 .05) F (2,634) = 2.12, p = .121, partial 2 = .01 Figure G No covariates 3.85 ( 0 .57) 3.69 ( 0 .65) 3.93 ( 0 .50) 3.62 ( 0 .74) 3.85 ( 0 .70) 3.72 ( 0 .67) F (2,617) = 2.71, p = .070, partial 2 = .01 BMI 3.81 ( 0 .06) 3.69 ( 0 .05) 3.91 ( 0 .06) 3.63 ( 0 .06) 3.84 ( 0 .06) 3.72 ( 0 .05) F (2,617) = 2.25, p = .110, partial 2 = .01 EDI BD 3.83 ( 0 .06) 3.68 ( 0 .05) 3.93 ( 0 .06) 3.61 ( 0 .05) 3.82 ( 0 .06) 3.74 ( 0 .05) F (2,614) = 3.69, p = .028, partial 2 = .01 DMS 3.81 ( 0 .06) 3.69 ( 0 .05) 3.90 ( 0 .06) 3.65 ( 0 .05) 3.85 ( 0 .07) 3.72 ( 0 .06) F (2,604) = 1.04, p = .350, partial 2 < .01

PAGE 160

148 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values PACS 3.83 ( 0 .05) 3.68 ( 0 .05) 3.93 ( 0 .06) 3.61 ( 0 .05) 3.83 ( 0 .06) 3.73 ( 0 .05) F (2,613) = 3.19, p = .044, partial 2 = .01 SATAQ 3.83 ( 0 .06) 3.68 ( 0 .05) 3.92 ( 0 .06) 3.62 ( 0 .05) 3.85 ( 0 .06) 3.72 ( 0 .05) F (2,611) = 2.55, p = .082, partial 2 = .01 Figure H No covariates 4.61 ( 0 .83) 4.40 ( 0 .67) 4.46 ( 0 .63) 4.35 ( 0 .63) 4.36 ( 0 .71) 4.20 ( 0 .55) F (2,618) = 0.56, p = .564, partial 2 < .01 BMI 4.61 ( 0 .07) 4.39 ( 0 .06) 4.45 ( 0 .06) 4.33 ( 0 .05) 4.38 ( 0 .06) 4.19 ( 0 .05) F (2, 614) = 0.56, p = .565, partial 2 < .01 EDI BD 4.59 ( 0 .07) 4.40 ( 0 .06) 4.43 ( 0 .06) 4.35 ( 0 .06) 4.37 ( 0 .06) 4.19 ( 0 .05) F (2,616) = 0.71, p = .485, partial 2 < .01

PAGE 161

149 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS 4.55 ( 0 .07) 4.43 ( 0 .06) 4.44 ( 0 .06) 4.34 ( 0 .05) 4.34 ( 0 .06) 4.22 ( 0 .05) F (2,609) = 0.03 p = .967, partial 2 < .01 PACS 4.59 ( 0 .07) 4.40 ( 0 .06) 4.44 ( 0 .06) 4.34 ( 0 .05) 4.39 ( 0 .06) 4.19 ( 0 .05) F (2,613) = 0.69, p = .495, partial 2 < .01 SATAQ 4.61 ( 0 .07) 4.39 ( 0 .06) 4.43 ( 0 .06) 4.34 ( 0 .05) 4.38 ( 0 .06) 4.19 ( 0 .05) F (2,612) = 0.91, p = .401, partial 2 < .01 Figure K No covariates 4.41 ( 0 .65) 4.39 ( 0 .63) 4.49 ( 0 .59) 4.42 ( 0 .56) 4.57 ( 0 .64) 4.44 ( 0 .58) F (2,640) = 0.95, p = .387, partial 2 < .01 BMI 4.39 ( 0 .06) 4.37 ( 0 .05) 4.49 ( 0 .05) 4.41 ( 0 .04) 4.57 ( 0 .05) 4.42 ( 0 .04) F (2,636) = 1.01, p = .365, partial 2 < .01

PAGE 162

150 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values EDI BD 4.38 ( 0 .06) 4.38 ( 0 .05) 4.50 ( 0 .05) 4.40 ( 0 .04) 4.55 ( 0 .06) 4.43 ( 0 .05) F (2,638) = 0.85, p = .427, partial 2 < .01 DMS 4.41 ( 0 .06) 4.36 ( 0 .05) 4.52 ( 0 .06) 4.39 ( 0 .05) 4.55 ( 0 .06) 4.42 ( 0 .05) F (2,629) = 0.47, p = .626, parti al 2 < .01 PACS 4.38 ( 0 .06) 4.37 ( 0 .05) 4.49 ( 0 .05) 4.41 ( 0 .04) 4.55 ( 0 .05) 4.43 ( 0 .04) F (2,636) = 0.98, p = .375, partial 2 < .01 SATAQ 4.41 ( 0 .06) 4.38 ( 0 .05) 4.50 ( 0 .05) 4.41 ( 0 .04) 4.56 ( 0 .05) 4.43 ( 0 .04) F (2,634) = 0.78, p = .459, partial 2 < .01 Figure L No covariates 5.16 ( 0 .77) 5.07 ( 0 .76) 4.89 ( 0 .74) 4.75 ( 0 .75) 5.10 ( 0 .66) 4.99 ( 0 .73) F (2,634) = .32, p = .724, partial 2 < .01

PAGE 163

151 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values BMI 5.16 ( 0 .07) 5.05 ( 0 .06) 4.88 ( 0 .07) 4.76 ( 0 .06) 5.06 ( 0 .06) 5.00 ( 0 .05) F (2,631) = 0.20, p = .819, partial 2 < .01 EDI BD 5.11 ( 0 .07) 5.07 ( 0 .06) 4.88 ( 0 .07) 4.77 ( 0 .06) 5.06 ( 0 .06) 5.01 ( 0 .05) F (2,632) = 0.28, p = .751, partial 2 < .01 DMS 5.14 ( 0 .08) 5.08 ( 0 .06) 4.88 ( 0 .07) 4.75 ( 0 .06) 5.05 ( 0 .07) 5.03 ( 0 .05) F (2,622) = 0.46, p = .630, partial 2 < .01 PACS 5.14 ( 0 .07) 5.06 ( 0 .06) 4.89 ( 0 .07) 4.75 ( 0 .05) 5.07 ( 0 .06) 5.01 ( 0 .05) F (2,631) = 0.34, p = .707, partial 2 < .01 SATAQ 5.14 ( 0 .07) 5.06 ( 0 .06) 4.89 ( 0 .07) 4.76 ( 0 .05) 5.07 ( 0 .06) 5.00 ( 0 .05) F (2,630) = 0.24, p = .786, partial 2 < .01

PAGE 164

152 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure O No co variates 5.50 ( 0 .67) 5.32 ( 0 .86) 5.08 (1.02) 5.11 ( 0 .78) 5.36 ( 0 .79) 5.24 ( 0 .83) F (2,646) = 2.08, p = .126, partial 2 = .01 BMI 5.50 ( 0 .07) 5.20 ( 0 .06) 5.06 ( 0 .08) 5.11 ( 0 .07) 5.37 ( 0 .07) 5.24 ( 0 .06) F (2,642) = 2.63, p = .073, partial 2 = .01 EDI BD 5. 47 ( 0 .07) 5.32 ( 0 .06) 5.07 ( 0 .08) 5.10 ( 0 .07) 5.37 ( 0 .07) 5.23 ( 0 .06) F (2,644) = 1.55, p = .213, partial 2 = .01 DMS 5.47 ( 0 .08) 5.33 ( 0 .06) 5.08 ( 0 .09) 5.11 ( 0 .07) 5.36 ( 0 .08) 5.24 ( 0 .07) F (2,636) = 0.93, p = .394, partial 2 < .01 PACS 5.49 ( 0 .07) 5.3 1 ( 0 .06) 5.08 ( 0 .08) 5.09 ( 0 .07) 5.37 ( 0 .07) 5.23 ( 0 .06) F (2,642) = 1.74, p = .177, partial 2 = .01

PAGE 165

153 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values SATAQ 5.49 ( 0 .07) 5.31 ( 0 .06) 5.06 ( 0 .08) 5.11 ( 0 .07) 5.38 ( 0 .07) 5.23 ( 0 .06) F (2,639) = 2.56, p = .079, partial 2 = .01 Figure P No covariates 4.32 ( 0 .65) 4.16 ( 0 .42) 4.18 ( 0 .69) 4.05 ( 0 .49) 4.19 ( 0 .69) 4.16 ( 0 .55) F (2,611) = 1.78, p = .171, partial 2 = .01 BMI 4.33 ( 0 .05) 4.15 ( 0 .04) 4.18 ( 0 .05) 4.03 ( 0 .04) 4.17 ( 0 .05) 4.17 ( 0 .05) F (2,608) = 2.79, p = .065, partial 2 = .01 EDI BD 4.32 ( 0 .05) 4.16 ( 0 .04) 4.17 ( 0 .05) 4.04 ( 0 .04) 4.21 ( 0 .05) 4.14 ( 0 .05) F (2,609) = 0.78, p = .452, partial 2 < .01 DMS 4.34 ( 0 .05) 4.14 ( 0 .04) 4.20 ( 0 .06) 4.01 ( 0 .05) 4.19 ( 0 .06) 4.15 ( 0 .05) F (2,600) = 1.98, p = .142, partial 2 = .01 PACS 4.32 ( 0 .05) 4.16 ( 0 .04) 4.16 ( 0 .05) 4.04 ( 0 .04) 4.19 ( 0 .05) 4.15 ( 0 .04) F (2,608) = 1.32, p = .266, partial 2 < .01 SATAQ 4.32 ( 0 .05) 4.16 ( 0 .04) 4.16 ( 0 .05) 4.04 ( 0 .04) 4.19 ( 0 .05) 4.15 ( 0 .04) F (2,607) = 1.61, p = .202, partial 2 = .01

PAGE 166

154 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure Q No covariates 4.08 ( 0 .40) 3.99 ( 0 .47) 3.98 ( 0 .56) 3.99 ( 0 .32) 4.01 ( 0 .53) 3.99 ( 0 .32) F (2,651) = 1.07, p = .342, partial 2 < .01 BMI 4.08 ( 0 .04) 4.00 ( 0 .03) 3.99 ( 0 .04) 3.99 ( 0 .03) 4.02 ( 0 .04) 3.99 ( 0 .03) F (2,647) = 0.74, p = .476, partial 2 < .01 EDI BD 4.08 ( 0 .04) 4.01 ( 0 .03) 3.97 ( 0 .04 ) 4.00 ( 0 .03) 4.02 ( 0 .04) 3.99 ( 0 .03) F (2,649) = 1.09, p = .336, partial 2 < .01 DMS 4.09 ( 0 .04) 4.01 ( 0 .04) 3.98 ( 0 .04) 4.01 ( 0 .03) 4.00 ( 0 .04) 4.00 ( 0 .03) F (2,642) = 1.14, p = .322, partial 2 < .01 PACS 4.08 ( 0 .04) 4.01 ( 0 .03) 3.97 ( 0 .04) 4.00 ( 0 .03) 4.02 ( 0 .04) 3.99 ( 0 .03) F (2,647) = 1.04, p = .355, partial 2 < .01

PAGE 167

155 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values SATAQ 4.09 ( 0 .04) 4.01 ( 0 .03) 4.00 ( 0 .04) 3.99 ( 0 .03) 4.03 ( 0 .04) 3.99 ( 0 .03) F (2,646) = 0.64, p = .526, partial 2 < .01 Figure R No covariates 3.56 ( 0 .78) 3.35 ( 0 .77) 3.50 ( 0 .75) 3.3 7 ( 0 .82) 3.34 ( 0 .74) 3.22 ( 0 .83) F (2,619) = 0.47, p = .621, partial 2 < .01 BMI 3.56 ( 0 .07) 3.36 ( 0 .06) 3.47 ( 0 .07) 3.37 ( 0 .06) 3.33 ( 0 .07) 3.22 ( 0 .06) F (2,615) = 0.48, p = .610, partial 2 < .01 EDI BD 3.59 ( 0 .07) 3.35 ( 0 .06) 3.46 ( 0 .07) 3.38 ( 0 .06) 3. 32 ( 0 .07) 3.22 ( 0 .06) F (2,618) = 1.26, p = .285, partial 2 < .01 DMS 3.61 ( 0 .08) 3.32 ( 0 .06) 3.51 ( 0 .07) 3.32 ( 0 .06) 3.41 ( 0 .08) 3.15 ( 0 .06) F (2,613) = 0.29, p = .743, partial 2 < .01

PAGE 168

156 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values PACS 3.56 ( 0 .07) 3.36 ( 0 .06) 3.47 ( 0 .07) 3.37 ( 0 .06) 3.32 ( 0 .07) 3.2 2 ( 0 .06) F (2,615) = 0.65, p = .617, partial 2 < .01 SATAQ 3.59 ( 0 .07) 3.36 ( 0 .06) 3.52 ( 0 .07) 3.36 ( 0 .06) 3.32 ( 0 .07) 3.22 ( 0 .06) F (2,607) = 0.75, p = .469, partial 2 < .01 Figure S No covariates 4.49 ( 0 .75) 4.27 ( 0 .57) 4.48 ( 0 .76) 4.31 ( 0 .63) 4.49 ( 0 .77) 4.33 ( 0 .57) F (2,633) = 0.10, p = .903, partial 2 < .01 BMI 4.45 ( 0 .06) 4.27 ( 0 .05) 4.46 ( 0 .06) 4.31 ( 0 .05) 4.50 ( 0 .06) 4.32 ( 0 .05) F (2,629) = 0.07, p = .936, partial 2 < .01 EDI BD 4.45 ( 0 .06) 4.28 ( 0 .05) 4.46 ( 0 .06) 4.30 ( 0 .05) 4.47 ( 0 .06) 4.34 ( 0 .05) F (2,631) = 0.13, p = .877, partial 2 < .01

PAGE 169

157 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS 4.51 ( 0 .07) 4.24 ( 0 .05) 4.48 ( 0 .07) 4.31 ( 0 .05) 4.47 ( 0 .07) 4.34 ( 0 .05) F (2,624) = 1.07, p = .345, partial 2 < .01 PACS 4.44 ( 0 .06) 4.28 ( 0 .05) 4.45 ( 0 .06) 4.31 ( 0 .05) 4.47 ( 0 .06) 4.34 ( 0 .05) F (2,629 ) = 0.12, p = .884, partial 2 < .01 SATAQ 4.45 ( 0 .06) 4.27 ( 0 .05) 4.45 ( 0 .06) 4.31 ( 0 .05) 4.48 ( 0 .06) 4.33 ( 0 .05) F (2,627) = 0.14, p = .871, partial 2 < .01 Figure V No covariates 3.73 ( 0 .77) 3.59 ( 0 .76) 3.74 ( 0 .72) 3.50 ( 0 .89) 3.77 ( 0 .82) 3.65 ( 0 .82) F (2,632) = 1.03, p = .358, partial 2 < .01 BMI 3.68 ( 0 .07) 3.61 ( 0 .06) 3.70 ( 0 .07) 3.51 ( 0 .06) 3.72 ( 0 .07) 3.66 ( 0 .06) F (2,628) = 0.93, p = .395 partial 2 < .01

PAGE 170

158 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values EDI BD 3.68 ( 0 .07) 3.61 ( 0 .06) 3.69 ( 0 .07) 3.52 ( 0 .06) 3.73 ( 0 .07) 3.66 ( 0 .06) F (2,630) = 0.53, p = .585, partial 2 < .01 DMS 3.74 ( 0 .07) 3.56 ( 0 .06) 3.79 ( 0 .08) 3.44 ( 0 .06) 3.75 ( 0 .08) 3.62 ( 0 .07) F (2,620) = 1.70, p = .185, partial 2 = .01 PACS 3.68 ( 0 .07) 3.61 ( 0 .06) 3.71 ( 0 .07) 3.51 ( 0 .06) 3.73 ( 0 .07) 3.65 ( 0 .06) F (2,628) = 0.88, p = .41 3, partial 2 < .01 SATAQ 3.68 ( 0 .07) 3.61 ( 0 .06) 3.73 ( 0 .07) 3.50 ( 0 .06) 3.72 ( 0 .07) 3.66 ( 0 .06) F (2,627) = 1.35, p = .261, partial 2 < .01 Figure W No covariates 4.35 ( 0 .80) 4.27 ( 0 .63) 4.63 ( 0 .66) 4.41 ( 0 .55) 4.48 ( 0 .60) 4.30 ( 0 .60) F (2,600) = 1.47, p = .232, partial 2 = .01

PAGE 171

159 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values BMI 4.35 ( 0 .07) 4.27 ( 0 .05) 4.64 ( 0 .05) 4.40 ( 0 .05) 4.44 ( 0 .06) 4.31 ( 0 .05) F (2,596) = 1.62, p = .201, partial 2 = .01 EDI BD 4.35 ( 0 .07) 4.26 ( 0 .05) 4.62 ( 0 .06) 4.41 ( 0 .05) 4.46 ( 0 .06) 4.30 ( 0 .05) F (2,597) = 0.90, p = .402, partial 2 < .01 DMS 4.29 ( 0 .07) 4.31 ( 0 .06) 4.63 ( 0 .06) 4.40 ( 0 .05) 4.48 ( 0 .06) 4.29 ( 0 .05) F (2,590) = 3.45, p = .035, partial 2 = .01 PACS 4.33 ( 0 .06) 4.27 ( 0 .05) 4.62 ( 0 .05) 4.41 ( 0 .04) 4.45 ( 0 .05) 4.30 ( 0 .05) F (2,596) = 1.46, p = .233, partial 2 = .01 SATAQ 4.35 ( 0 .06) 4.27 ( 0 .05) 4.64 ( 0 .05) 4.40 ( 0 .04) 4.46 ( 0 .05) 4.30 ( 0 .05) F (2,596) = 1.62, p = .200, partial 2 = .01

PAGE 172

160 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure Z No covariates 4.73 ( 0 .79) 4.53 ( 0 .91) 4.95 ( 0 .67) 4.64 ( 0 .84) 4.63 ( 0 .86) 4.47 ( 0 .88) F (2,643) = 0.98, p = .376, parti al 2 .01 BMI 4.68 ( 0 .08) 4.55 ( 0 .06) 4.92 ( 0 .07) 4.65 ( 0 .06) 4.59 ( 0 .08) 4.47 ( 0 .06) F (2,639) = 1.10, p = .334, partial 2 < .01 EDI BD 4.71 ( 0 .08) 4.53 ( 0 .06) 4.90 ( 0 .07) 4.67 ( 0 .06) 4.63 ( 0 .08) 4.44 ( 0 .06) F (2,640) = 0.13, p = .872, partial 2 < .01 DMS 4.67 ( 0 .08) 4.55 ( 0 .07) 4.90 ( 0 .07) 4.68 ( 0 .08) 4.63 ( 0 .08) 4.42 ( 0 .07) F (2,631) = 0.40, p = .668, partial 2 < .01 PACS 4.70 ( 0 .07) 4.54 ( 0 .06) 4.91 ( 0 .07) 4.65 ( 0 .06) 4.60 ( 0 .08) 4.46 ( 0 .06) F (2,639) = 0.83, p = .437, partial 2 < .01

PAGE 173

161 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values SATAQ 4.71 ( 0 .08) 4.54 ( 0 .06) 4.92 ( 0 .07) 4.66 ( 0 .06) 4.62 ( 0 .07) 4.46 ( 0 .06) F (2,637) = 0.66, p = .516, partial 2 < .01 Figure AA No covariates 5.92 ( 0 .90) 5.98 ( 0 .87) 5.82 ( 0 .89) 5.89 ( 0 .82) 5.49 ( 0 .83) 5.51 ( 0 .89) F (2,646) = 0.14, p = .866, partial 2 < .01 BM I 5.95 ( 0 .08) 5.97 ( 0 .06) 5.80 ( 0 .08) 5.88 ( 0 .06) 5.48 ( 0 .08) 5.50 ( 0 .06) F (2,642) = 0.19, p = .828, partial 2 < .01 EDI BD 5.90 ( 0 .08) 5.99 ( 0 .07) 5.79 ( 0 .08) 5.89 ( 0 .06) 5.41 ( 0 .08) 5.54 ( 0 .06) F (2,644) = 0.06, p = .944, partial 2 < .01 DMS 5.94 ( 0 .0 9) 5.98 ( 0 .07) 5.75 ( 0 .08) 5.92 ( 0 .07) 5.45 ( 0 .09) 5.52 ( 0 .07) F (2,635) = 0.53, p = .591, partial 2 < .01

PAGE 174

162 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values PACS 5.91 ( 0 .08) 5.98 ( 0 .06) 5.80 ( 0 .08) 5.89 ( 0 .06) 5.45 ( 0 .08) 5.51 ( 0 .06) F (2,642) = 0.03, p = .975, partial 2 < .01 SATAQ 5.93 ( 0 .08) 5.98 ( 0 06) 5.80 ( 0 .08) 5.89 ( 0 .06) 5.47 ( 0 .08) 5.51 ( 0 .06) F (2,640) = 0.07, p = .931, partial 2 < .01 Figure DD No covariates 5.73 ( 0 .75) 5.54 ( 0 .92) 5.69 ( 0 .84) 5.60 ( 0 .83) 5.71 (1.02) 5.54 ( 0 .89) F (2,618) = 0.64, p = .523, partial 2 < .01 BMI 5.74 ( 0 .08) 5 .51 ( 0 .06) 5.68 ( 0 .08) 5.60 ( 0 .06) 5.67 ( 0 .08) 5.56 ( 0 .07) F (2,619) = 1.01, p = .365, partial 2 < .01 EDI BD 5.69 ( 0 .08) 5.54 ( 0 .06) 5.69 ( 0 .08) 5.59 ( 0 .06) 5.76 ( 0 .08) 5.50 ( 0 .07) F (2,618) = 1.02, p = .361, partial 2 < .01

PAGE 175

163 Target R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS 5.70 ( 0 .08) 5.55 ( 0 .07) 5.66 ( 0 .08) 5.62 ( 0 .07) 5.72 ( 0 .09) 5.53 ( 0 .07) F (2,606) = 0.72, p = .484, partial 2 < .01 PACS 5.73 ( 0 .07) 5.53 ( 0 .06) 5.69 ( 0 .07) 5.60 ( 0 .06) 5.72 ( 0 .08) 5.52 ( 0 .07) F (2,615) = 0.68, p = .503, partial 2 < .01 SATAQ 5.72 ( 0 .08) 5.53 ( 0 .06) 5.70 ( 0 .07) 5.60 ( 0 .06) 5.71 ( 0 .08) 5.53 ( 0 .07) F (2,612) = 0.37, p = .682, partial 2 < .01

PAGE 176

164 Table L.4. Repeated measures effects: Two way interactions (target race X rater race) Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values Figure A No covariates 3.88 ( 0 .92) 3.83 ( 0 .85) 3.93 ( 0 .79) 3.72 ( 0 .88) 3.67 ( 0 .86) 3.85 ( 0 .68) 3.96 ( 0 .83) 3.76 ( 0 .83) 4.01 ( 0 .81) F (4, 634) = 0.73, p = .569, partial 2 = .01 BMI 3.90 ( 0 .08) 3.84 ( 0 .08) 3.96 ( 0 .09) 3.70 ( 0 .07) 3.67 ( 0 .08) 3.91 ( 0 .09) 3.94 ( 0 .07) 3.76 ( 0 .08) 4.06 ( 0 .09) F ( 4,630) = 0.86, p = .485, partial 2 = .01 EDI BD 3.92 ( 0 .08) 3.84 ( 0 .08) 3.92 ( 0 .09) 3.73 ( 0 .07) 3.67 ( 0 .08) 3.83 ( 0 .09) 3.97 ( 0 .07) 3.76 ( 0 .08) 3.97 ( 0 .09) F (4,632) = 0.66, p = .619, partial 2 < .01 DMS 3.91 ( 0 .08) 3.83 ( 0 .08) 3.95 ( 0 .09) 3.73 ( 0 .07) 3 .67 ( 0 .08) 3.88 ( 0 .09) 3.96 ( 0 .08) 3.75 ( 0 .08) 4.02 ( 0 .09) F (4,624) = 0.77, p = .541, partial 2 = .01 PACS 3.93 ( 0 .08) 3.83 ( 0 .08) 3.92 ( 0 .09) 3.74 ( 0 .08) 3.66 ( 0 .08) 3.86 ( 0 .09) 3.99 ( 0 .08) 3.74 ( 0 .08) 3.98 ( 0 .09) F (4,630) = 0.94, p = .437, partial 2 = .01

PAGE 177

165 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values SATAQ 3.91 ( 0 .08) 3.83 ( 0 .08) 4.00 ( 0 .09) 3.75 ( 0 .08) 3.68 ( 0 .08) 3.80 ( 0 .09) 3.96 ( 0 .07) 3.76 ( 0 .08) 4.06 ( 0 .09) F (4,629) = 0.66, p = .619, partial 2 < .01 Figure B No covariates 3.68 ( 0 .80) 3.56 ( 0 .93) 3.69 ( 0 .83) 3.58 ( 0 .89) 3.64 ( 0 .85) 3.69 ( 0 .70) 3.63 ( 0 .73) 3.57 ( 0 .89) 3.67 ( 0 .93) F (4,636) = 0.47, p = .752, partial 2 < .01 BMI 3.66 ( 0 .08) 3.58 ( 0 .08) 3.69 ( 0 .09) 3.55 ( 0 .07) 3.64 ( 0 .08) 3.73 ( 0 .09) 3.58 ( 0 .07) 3.57 ( 0 .08) 3.68 ( 0 .09) F (4,634) = 0.65, p = .625, partial 2 < .01 EDI BD 3.67 ( 0 .08) 3.58 ( 0 .08) 3.64 ( 0 .09) 3.58 ( 0 .07) 3.64 ( 0 .08) 3.66 ( 0 .09) 3.61 ( 0 .07) 3.57 ( 0 .08) 3.60 ( 0 .09) F (4,634) = 0.49, p = .742, partial 2 < .01 DMS 3.66 ( 0 .08) 3.60 ( 0 .08) 3.66 ( 0 .09) 3.57 ( 0 .07) 3.63 ( 0 .08) 3.66 ( 0 .09) 3.60 ( 0 .07) 3.58 ( 0 .08) 3.61 ( 0 .0 9) F (4,629) = 0.38, p = .825, partial 2 < .01

PAGE 178

166 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values PACS 3.66 ( 0 .08) 3.58 ( 0 .08) 3.66 ( 0 .09) 3.57 ( 0 .08) 3.64 ( 0 .08) 3.68 ( 0 .09) 3.62 ( 0 .08) 3.56 ( 0 .08) 3.61 ( 0 .09) F (4,632) = 0.64, p = .631, partial 2 < .01 SATAQ 3.66 ( 0 .08) 3.58 ( 0 .08) 3.65 ( 0 .10) 3.57 ( 0 08) 3.64 ( 0 .08) 3.68 ( 0 .09) 3.61 ( 0 .08) 3.57 ( 0 .08) 3.64 ( 0 .09) F (4,629) = 0.50, p = .731, partial 2 < .01 Figure C No covariates 2.48 (1.16) 2.25 (1.09) 2.36 (1.23) 2.69 (1.10) 2.57 (1.05) 2.49 (1.13) 2.23 (1.19) 2.18 (1.12) 2.22 (1.08) F (4,638) = 0.70 p = .595, partial 2 < .01 BMI 2.48 ( 0 .11) 2.24 ( 0 .12) 2.33 ( 0 .13) 2.68 ( 0 .10) 2.56 ( 0 .11) 2.51 ( 0 .12) 2.21 ( 0 .10) 2.17 ( 0 .11) 2.24 ( 0 .12) F (4,633) = 0.78, p = .535, partial 2 < .01 EDI BD 2.48 ( 0 .11) 2.24 ( 0 .12) 2.32 ( 0 .13) 2.70 ( 0 .10) 2.56 ( 0 .11) 2. 47 ( 0 .12) 2.22 ( 0 .10) 2.17 ( 0 .11) 2.18 ( 0 .12) F (4,636) = 0.64, p = .634, partial 2 < .01

PAGE 179

167 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values DMS 2.47 ( 0 .10) 2.24 ( 0 .12) 2.29 ( 0 .12) 2.70 ( 0 .10) 2.58 ( 0 .11) 2.45 ( 0 .12) 2.21 ( 0 .10) 2.18 ( 0 .12) 2.17 ( 0 .12) F (4,627) = 0.77, p = .543, partial 2 < .01 PACS 2.4 8 ( 0 .11) 2.24 ( 0 .12) 2.31 ( 0 .13) 2.70 ( 0 .10) 2.56 ( 0 .11) 2.47 ( 0 .12) 2.23 ( 0 .10) 2.16 ( 0 .11) 2.18 ( 0 .12) F (4,634) = 0.61, p = .657, partial 2 < .01 SATAQ 2.44 ( 0 .11) 2.23 ( 0 .12) 2.40 ( 0 .13) 2.68 ( 0 .10) 2.55 ( 0 .11) 2.52 ( 0 .12) 2.18 ( 0 .10) 2.16 ( 0 .11) 2.27 ( 0 .13) F (4,632) = 0.94, p = .438, partial 2 = .01 Figure D No covariates 4.30 ( 0 .77) 4.26 ( 0 .65) 4.10 ( 0 .58) 4.10 ( 0 .73) 4.15 ( 0 .79) 4.15 ( 0 .72) 4.14 ( 0 .73) 4.04 ( 0 .71) 4.15 ( 0 .61) F (4,640) = 1.09, p = .360, partial 2 = .01 BMI 4.28 ( 0 .06) 4.25 ( 0 .07 ) 4.18 ( 0 .07) 4.13 ( 0 .07) 4.15 ( 0 .07) 4.12 ( 0 .08) 4.15 ( 0 .06) 4.04 ( 0 .07) 4.15 ( 0 .08) F (4,635) = 0.90, p = .464, partial 2 = .01

PAGE 180

168 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values EDI BD 4.30 ( 0 .06) 4.25 ( 0 .07) 4.14 ( 0 .07) 4.14 ( 0 .07) 4.15 ( 0 .07) 4.10 ( 0 .08) 4.15 ( 0 .06) 4.04 ( 0 .07) 4.13 ( 0 .08) F (4,638) = 1.13, p = .341, partial 2 = .01 DMS 4.29 ( 0 .06) 4.24 ( 0 .07) 4.17 ( 0 .07) 4.13 ( 0 .07) 4.13 ( 0 .07) 4.13 ( 0 .08) 4.15 ( 0 .06) 4.04 ( 0 .07) 4.14 ( 0 .07) F (4,627) = 0.75, p = .555, partial 2 = .01 PACS 4.30 ( 0 .06) 4.25 ( 0 .07) 4.15 ( 0 .07) 4.14 ( 0 .07) 4.15 ( 0 .07) 4.12 ( 0 .08) 4.16 ( 0 .06) 4.04 ( 0 .07) 4.13 ( 0 .07) F (4,636) = 1.05, p = .382, partial 2 = .01 SATAQ 4.28 ( 0 .06) 4.25 ( 0 .07) 4.19 ( 0 .08) 4.14 ( 0 .07) 4.16 ( 0 .07) 4.08 ( 0 .08) 4.14 ( 0 .06) 4.04 ( 0 .07) 4.14 ( 0 .08) F (4,634) = 0.81, p = .516, partial 2 = .01 Fig ure G No covariates 3.84 ( 0 .57) 3.83 ( 0 .57) 3.58 ( 0 .71) 3.79 ( 0 .64) 3.86 ( 0 .60) 3.60 ( 0 .74) 3.77 ( 0 .64) 3.76 ( 0 .73) 3.80 ( 0 .69) F (4,617) = 2.58, p = .039, partial 2 = .02

PAGE 181

169 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values BMI 3.85 ( 0 .06) 3.83 ( 0 .06) 3.57 ( 0 .07) 3.81 ( 0 .06) 3.86 ( 0 .06) 3.65 ( 0 .07) 3.79 ( 0 .06) 3.76 ( 0 .07) 3.80 ( 0 .08) F (4,611) = 2.62, p = .037, partial 2 = .02 EDI BD 3.85 ( 0 .06) 3.83 ( 0 .06) 3.59 ( 0 .07) 3.80 ( 0 .06) 3.86 ( 0 .06) 3.66 ( 0 .07) 3.79 ( 0 .06) 3.76 ( 0 .07) 3.78 ( 0 .08) F (4,614) = 2.00, p = .096, partial 2 = .01 DMS 3.85 ( 0 .06) 3.82 ( 0 .06) 3.58 ( 0 .07) 3.80 ( 0 .06) 3.85 ( 0 .06) 3.66 ( 0 .07) 3.79 ( 0 .06) 3.76 ( 0 .07) 3.80 ( 0 .07) F (4,604) = 2.45, p = .049, partial 2 = .02 PACS 3.83 ( 0 .06) 3.83 ( 0 .06) 3.60 ( 0 .07) 3.79 ( 0 .06) 3.86 ( 0 .06) 3.66 ( 0 .07) 3.80 ( 0 .06) 3.75 ( 0 .07) 3.79 ( 0 .08) F (4,613 ) = 2.05, p = .090, partial 2 = .01 SATAQ 3.84 ( 0 .06) 3.82 ( 0 .06) 3.61 ( 0 .07) 3.80 ( 0 .06) 3.85 ( 0 .06) 3.66 ( 0 .07) 3.79 ( 0 .06) 3.76 ( 0 .07) 3.81 ( 0 .08) F (4,611) = 1.98, p = .100, partial 2 = .01

PAGE 182

170 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values Figure H No covariates 4.59 ( 0 .72) 4.48 ( 0 .75) 4.37 ( 0 .77) 4.46 ( 0 .63) 4.38 ( 0 .63) 4.32 ( 0 .63) 4.34 ( 0 .65) 4.25 ( 0 .69) 4.20 ( 0 .53) F (4,618) = 0.35, p = .838, partial 2 < .01 BMI 4.60 ( 0 .07) 4.49 ( 0 .07) 4.40 ( 0 .08) 4.46 ( 0 .06) 4.38 ( 0 .06) 4.33 ( 0 .07) 4.34 ( 0 .06) 4.26 ( 0 .06) 4.27 ( 0 .07) F (4,614) = 0.34, p = .846, partial 2 < .01 EDI BD 4.61 ( 0 .07) 4.49 ( 0 .07) 4.38 ( 0 .08) 4.47 ( 0 .06) 4.38 ( 0 .06) 4.31 ( 0 .07) 4.34 ( 0 .06) 4.26 ( 0 .06) 4.25 ( 0 .07) F (4,616) = 0.39, p = .810, partial 2 < .01 DMS 4.60 ( 0 .07) 4.48 ( 0 .08) 4.39 ( 0 .08) 4.47 ( 0 .06) 4.39 ( 0 .06) 4.32 ( 0 .07) 4 .34 ( 0 .06) 4.25 ( 0 .06) 4.26 ( 0 .07) F (4,609) = 0.32, p = .858, partial 2 < .01 PACS 4.60 ( 0 .07) 4.49 ( 0 .07) 4.39 ( 0 .08) 4.47 ( 0 .06) 4.38 ( 0 .06) 4.31 ( 0 .07) 4.32 ( 0 .06) 4.27 ( 0 .06) 4.28 ( 0 .07) F (4,613) = 0.57, p = .677, partial 2 < .01

PAGE 183

171 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values SATAQ 4.59 ( 0 .07) 4.48 ( 0 .07) 4.42 ( 0 .08) 4.47 ( 0 .06) 4.38 ( 0 .06) 4.31 ( 0 .07) 4.34 ( 0 .06) 4.25 ( 0 .06) 4.27 ( 0 .07) F (4,612) = 0.28, p = .885, partial 2 < .01 Figure K No covariates 4.52 ( 0 .60) 4.39 ( 0 .63) 4.26 ( 0 .66) 4.50 ( 0 .59) 4.43 ( 0 .60) 4.41 ( 0 .53) 4.63 ( 0 .67) 4.50 ( 0 .59) 4.31 ( 0 .50) F (4,640) = 1.56, p = .185, partial 2 = .01 BMI 4.49 ( 0 .06) 4.39 ( 0 .06) 4.26 ( 0 .07) 4.50 ( 0 .05) 4.43 ( 0 .06) 4.42 ( 0 .06) 4.64 ( 0 .05) 4.50 ( 0 .06) 4.34 ( 0 .07) F (4,636) = 1.16, p = .326, partial 2 = .01 EDI BD 4.50 ( 0 .06) 4.39 ( 0 .06) 4.25 ( 0 .07) 4.49 ( 0 .05) 4.43 ( 0 .06) 4.44 ( 0 .06) 4.65 ( 0 .05) 4.50 ( 0 .06) 4.32 ( 0 .07) F (4,638) = 1.70, p = .149, partial 2 = .01 DMS 4.51 ( 0 .06) 4.41 ( 0 .06) 4.24 ( 0 .07) 4.50 ( 0 .05) 4.46 ( 0 .06) 4.42 ( 0 .06) 4.65 ( 0 .05) 4.49 ( 0 .06) 4.32 ( 0 .06) F (4,629) = 1.45, p = .218, partial 2 = .01

PAGE 184

172 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values PACS 4.50 ( 0 .06) 4.40 ( 0 .06) 4.25 ( 0 .07) 4.50 ( 0 .05) 4.43 ( 0 .06) 4.41 ( 0 .06) 4.64 ( 0 .06) 4.50 ( 0 .06) 4.33 ( 0 .07) F (4,636) = 1.22, p = .301, partial 2 = .01 SATAQ 4.50 ( 0 .06) 4.39 ( 0 .06) 4.29 ( 0 .07) 4.48 ( 0 .05) 4.43 ( 0 .06) 4.45 ( 0 07) 4.62 ( 0 .06) 4.49 ( 0 .06) 4.37 ( 0 .07) F (4,634) = 1.07, p = .371 partial 2 = .01 Figure L No covariates 5.21 ( 0 .76) 5.12 ( 0 .74) 4.95 ( 0 .79) 4.87 ( 0 .72) 4.86 ( 0 .74) 4.68 ( 0 .79) 5.12 ( 0 .67) 5.18 ( 0 .64) 4.78 ( 0 .74) F (4,631) = 1.50, p = .200, partial 2 = .01 BMI 5.19 ( 0 .07) 5.12 ( 0 .08) 5.00 ( 0 .08) 4.88 ( 0 .07) 4.86 ( 0 .07) 4.73 ( 0 .08) 5.11 ( 0 .06) 5.19 ( 0 .07) 4.80 ( 0 .08) F (4,631) = 1.50, p = .200, partial 2 = .01 EDI BD 5.21 ( 0 .07) 5.12 ( 0 .08) 4.95 ( 0 .08) 4.88 ( 0 .07) 4.86 ( 0 .07) 4.73 ( 0 .08) 5.11 ( 0 .06) 5.1 8 ( 0 .07) 4.80 ( 0 .08) F (4,632) = 1.42, p = .226, partial 2 = .01

PAGE 185

173 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values DMS 5.21 ( 0 .07) 5.12 ( 0 .08) 4.99 ( 0 .08) 4.88 ( 0 .07) 4.86 ( 0 .08) 4.72 ( 0 .08) 5.12 ( 0 .06) 5.18 ( 0 .06) 4.82 ( 0 .07) F (4,622) = 1.26, p = .285, partial 2 = .01 PACS 5.19 ( 0 .07) 5.12 ( 0 .08) 5.00 ( 0 .08) 4.87 ( 0 .07) 4.86 ( 0 .07) 4.73 ( 0 .08) 5.10 ( 0 .06) 5.19 ( 0 .07) 4.82 ( 0 .08) F (4,631) = 1.40, p = .235, partial 2 = .01 SATAQ 5.19 ( 0 .07) 5.11 ( 0 .08) 5.00 ( 0 .09) 4.87 ( 0 .07) 4.85 ( 0 .07) 4.83 ( 0 .08) 5.10 ( 0 .06) 5.18 ( 0 .07) 4.83 ( 0 .08) F (4,630) = 1.35, p = .252, partial 2 = .01 Figure O No covariates 5.51 ( 0 .88) 5.40 ( 0 .65) 5.23 ( 0 .77) 5.13 ( 0 .90) 5.25 ( 0 .83) 4.89 ( 0 .92) 5.39 ( 0 .78) 5.36 ( 0 .80) 5.11 ( 0 .84) F (4,646) = 1.28, p = .277, partial 2 = .01 BMI 5.52 ( 0 .07) 5.40 ( 0 .08) 5.28 ( 0 .09) 5.12 ( 0 .08) 5.25 ( 0 .09) 4.88 ( 0 .10) 5.40 ( 0 .07) 5.36 ( 0 .08) 5.15 ( 0 .09) F (4,642) = 1.24, p = .294, partial 2 = .01

PAGE 186

174 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values EDI BD 5.53 ( 0 .07) 5.40 ( 0 .08) 5.25 ( 0 .09) 5.12 ( 0 .08) 5.25 (. 0 09) 4.89 ( 0 .10) 5.40 ( 0 .07) 5.36 ( 0 .08) 5.14 ( 0 .09) F (4,644) = 1.25, p = .290, partial 2 = .01 DMS 5.53 ( 0 .07) 5.39 ( 0 .08) 5.27 ( 0 .08) 5.12 ( 0 .08) 5.27 ( 0 .09) 4.88 ( 0 .10) 5.41 ( 0 .07) 5.36 ( 0 .08) 5.13 ( 0 .09) F (4,636) = 1.52, p = .195, partial 2 = .01 PACS 5.51 ( 0 .07) 5.41 ( 0 .08) 5.29 ( 0 .09) 5.09 ( 0 .08) 5.27 ( 0 .09) 4.90 ( 0 .10) 5.38 ( 0 .07) 5.37 ( 0 .08) 5.15 ( 0 .09) F (4,642) = 1.34, p = .253, partial 2 = .01 SATAQ 5.51 ( 0 .07) 5.40 ( 0 .08) 5.30 ( 0 .09) 5.12 ( 0 .08) 5.25 ( 0 .09) 4.88 ( 0 .10) 5.36 ( 0 .07) 5.35 ( 0 .08) 5.20 ( 0 .09) F (4,639) = 1.41, p = .228, partial 2 = .01 Figure P No covariates 4.30 ( 0 .59) 4.20 ( 0 .49) 4.16 ( 0 .50) 4.18 ( 0 .51) 4.10 ( 0 .63) 4.01 ( 0 .62) 4.26 ( 0 .58) 4.09 ( 0 .63) 4.14 ( 0 .62) F (2,611) = 0.73, p = .565, partial 2 < .01

PAGE 187

175 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values BMI 4.30 ( 0 .05) 4.21 ( 0 .05) 4.22 ( 0 .06) 4.18 ( 0 .05) 4.10 ( 0 .06) 4.03 ( 0 .06) 4.27 ( 0 .05) 4.09 ( 0 .06) 4.15 ( 0 .06) F (2,608) = 0.71, p = .579, partial 2 < .01 EDI BD 4.30 ( 0 .05) 4.21 ( 0 .05) 4.21 ( 0 .06) 4.19 ( 0 .05) 4.10 ( 0 .06) 4.02 ( 0 .06) 4.26 ( 0 .05) 4.09 ( 0 .06) 4.18 ( 0 .07) F (2,609) = 0.86, p = .481, partial 2 = .01 DMS 4.30 ( 0 .05) 4.21 ( 0 .05) 4.21 ( 0 .06) 4.19 ( 0 .05) 4.11 ( 0 .06) 4.02 ( 0 .06) 4.27 ( 0 .05) 4.09 ( 0 .06) 4.15 ( 0 .06) F (2,600) = 0.80, p = .521, partial 2 = .01 PACS 4.30 ( 0 .05) 4.21 ( 0 .05) 4.20 ( 0 .06) 4.20 ( 0 .05) 4.09 ( 0 .06) 4.00 ( 0 .06) 4.26 ( 0 .06) 4.09 ( 0 .06) 4.17 ( 0 .07) F (2,608) = 0.88, p = .473, parti al 2 = .01 SATAQ 4.30 ( 0 .05) 4.21 ( 0 .05) 4.21 ( 0 .06) 4.17 ( 0 .05) 4.10 ( 0 .06) 4.03 ( 0 .06) 4.26 ( 0 .06) 4.09 ( 0 .06) 4.16 ( 0 .07) F (2,607) = 0.70, p = .594, partial 2 < .01

PAGE 188

176 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values Figure Q No covariates 4.00 ( 0 .41) 4.10 ( 0 .45) 4.00 ( 0 .48) 4.02 ( 0 .48) 3.96 ( 0 .44) 3.97 ( 0 .37) 4.00 ( 0 .47) 4.02 ( 0 .28) 4.01 ( 0 .49) F (2,651) = 1.28, p = .275, partial 2 = .01 BMI 4.01 ( 0 .04) 4.10 ( 0 .04) 4.02 ( 0 .05) 4.03 ( 0 .04) 3.96 ( 0 .04) 3.99 ( 0 .05) 3.97 ( 0 .04) 4.02 ( 0 .04) 4.04 ( 0 .05) F (2,647) = 1.27, p = .279, partial 2 = .01 EDI BD 4.01 ( 0 .04) 4.10 ( 0 .04) 4.03 ( 0 .05) 4.03 ( 0 .04) 3.96 ( 0 .04) 3.97 ( 0 .05) 3.97 ( 0 .04) 4.02 ( 0 .04) 4.03 ( 0 .05) F (2,649) = 1.37, p = .243, partial 2 = .01 DMS 4.01 ( 0 .04) 4.10 ( 0 .05) 4.03 ( 0 .05) 4.03 ( 0 .04) 3.97 ( 0 .04) 3.98 ( 0 .05) 3.97 ( 0 .04) 4.00 ( 0 .04) 4. 04 ( 0 .05) F (2,642) = 1.14, p = .335, partial 2 = .01 PACS 4.02 ( 0 .04) 4.10 ( 0 .04) 4.02 ( 0 .05) 4.04 ( 0 .04) 3.95 ( 0 .04) 3.96 ( 0 .05) 3.97 ( 0 .04) 4.02 ( 0 .04) 4.03 ( 0 .05) F (2,647) = 1.41, p = .228, partial 2 = .01

PAGE 189

177 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values SATAQ 4.01 ( 0 .04) 4.10 ( 0 .04) 4.03 ( 0 .05) 4 .04 ( 0 .04) 3.96 ( 0 .04) 3.99 ( 0 .05) 3.95 ( 0 .04) 4.02 ( 0 .04) 4.05 ( 0 .05) F (2,646) = 1.86, p = .116, partial 2 = .01 Figure R No covariates 3.43 ( 0 .72) 3.53 ( 0 .71) 3.35 ( 0 .92) 3.46 ( 0 .75) 3.44 ( 0 .64) 3.37 ( 0 .98) 3.30 ( 0 .65) 3.29 ( 0 .70) 3.21 ( 0 .92) F (2,619) = 0.25, p = .904, partial 2 < .01 BMI 3.45 ( 0 .07) 3.53 ( 0 .08) 3.40 ( 0 .08) 3.48 ( 0 .07) 3.45 ( 0 .08) 3.33 ( 0 .09) 3.31 ( 0 .07) 3.30 ( 0 .07) 3.22 ( 0 .09) F (2,615) = 0.28, p = .888, partial 2 < .01 EDI BD 3.44 ( 0 .07) 3.53 ( 0 .08) 3.43 ( 0 .09) 3.49 ( 0 .07) 3.46 ( 0 .08) 3.32 ( 0 .09) 3.31 ( 0 .07) 3.30 ( 0 .08) 3.21 ( 0 .09) F (2,618) = 0.43, p = .782, partial 2 < .01 DMS 3.46 ( 0 .07) 3.55 ( 0 .08) 3.40 ( 0 .08) 3.48 ( 0 .07) 3.46 ( 0 .08) 3.30 ( 0 .08) 3.31 ( 0 .07) 3.33 ( 0 .08) 3.19 ( 0 .08) F (2,613) = 0.25, p = .905, partial 2 < .01

PAGE 190

178 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values P ACS 3.46 ( 0 .07) 3.53 ( 0 .08) 3.39 ( 0 .08) 3.51 ( 0 .07) 3.44 ( 0 .08) 3.31 ( 0 .09) 3.33 ( 0 .07) 3.29 ( 0 .08) 3.20 ( 0 .09) F (2,615) = 0.38, p = .819, partial 2 < .01 SATAQ 3.46 ( 0 .07) 3.53 ( 0 .08) 3.44 ( 0 .09) 3.50 ( 0 .07) 3.45 ( 0 .08) 3.36 ( 0 .09) 3.30 ( 0 .07) 3.30 ( 0 .0 8) 3.21 ( 0 .09) F (2,607) = 0.31, p = .865, partial 2 < .01 Figure S No covariates 4.38 ( 0 .65) 4.45 ( 0 .64) 4.24 ( 0 .68) 4.46 ( 0 .65) 4.41 ( 0 .71) 4.25 ( 0 .70) 4.43 ( 0 .68) 4.40 ( 0 .66) 4.36 ( 0 .65) F (2,633) = 1.22, p = .302, partial 2 = .01 BMI 4.41 ( 0 .06) 4.4 3 ( 0 .07) 4.25 ( 0 .07) 4.48 ( 0 .06) 4.41 ( 0 .07) 4.26 ( 0 .08) 4.44 ( 0 .06) 4.40 ( 0 .07) 4.40 ( 0 .07) F (2,629) = 1.40, p = .232, partial 2 = .01 EDI BD 4.41 ( 0 .06) 4.43 ( 0 .06) 4.25 ( 0 .07) 4.48 ( 0 .06) 4.41 ( 0 .07) 4.26 ( 0 .08) 4.44 ( 0 .06) 4.40 ( 0 .07) 4.36 ( 0 .07) F (2 ,631) = 0.91, p = .457, partial 2 = .01

PAGE 191

179 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values DMS 4.42 ( 0 .06) 4.45 ( 0 .07) 4.26 ( 0 .07) 4.49 ( 0 .06) 4.41 ( 0 .07) 4.28 ( 0 .07) 4.44 ( 0 .06) 4.38 ( 0 .07) 4.40 ( 0 .07) F (2,624) = 1.67, p = .156, partial 2 = .01 PACS 4.43 ( 0 .06) 4.42 ( 0 .07) 4.24 ( 0 .07) 4.50 ( 0 .06) 4.4 0 ( 0 .07) 4.24 ( 0 .08) 4.47 ( 0 .06) 4.38 ( 0 .07) 4.36 ( 0 .07) F (2,629) = 1.05, p = .380, partial 2 = .01 SATAQ 4.40 ( 0 .06) 4.43 ( 0 .06) 4.26 ( 0 .07) 4.48 ( 0 .06) 4.41 ( 0 .07) 4.25 ( 0 .08) 4.34 ( 0 .06) 4.40 ( 0 .07) 4.37 ( 0 .07) F (2,627) = 1.11, p = .352, partial 2 = .01 Figure V No covariates 3.71 ( 0 .72) 3.80 ( 0 .72) 3.42 ( 0 .84) 3.69 ( 0 .80) 3.70 ( 0 .75) 3.38 ( 0 .91) 3.77 ( 0 .72) 3.85 ( 0 .70) 3.44 ( 0 .98) F (2,632) = 0.28, p = .889, partial 2 < .01 BMI 3.73 ( 0 .07) 3.80 ( 0 .08) 3.40 ( 0 .08) 3.71 ( 0 .07) 3.70 ( 0 .08) 3.41 ( 0 .09 ) 3.79 ( 0 .07) 3.85 ( 0 .08) 3.43 ( 0 .09) F (2,628) = 0.28, p = .890, partial 2 < .01

PAGE 192

180 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values EDI BD 3.73 ( 0 .07) 3.80 ( 0 .07) 3.41 ( 0 .08) 3.72 ( 0 .07) 3.70 ( 0 .08) 3.40 ( 0 .09) 3.79 ( 0 .07) 3.85 ( 0 .08) 3.44 ( 0 .09) F (2,630) = 0.21, p = .932, partial 2 < .01 DMS 3.74 ( 0 07) 3.81 ( 0 .08) 3.40 ( 0 .08) 3.72 ( 0 .07) 3.73 ( 0 .08) 3.38 ( 0 .09) 3.79 ( 0 .07) 3.85 ( 0 .08) 3.41 ( 0 .09) F (2,620) = 0.12 p = .972, partial 2 < .01 PACS 3.74 ( 0 .07) 3.79 ( 0 .08) 3.40 ( 0 .08) 3.72 ( 0 .07) 3.70 ( 0 .08) 3.41 ( 0 .09) 3.79 ( 0 .07) 3.85 ( 0 .08) 3.44 ( 0 .09) F (2,628) = 0.26, p = .900, partial 2 < .01 SATAQ 3.73 ( 0 .07) 3.80 ( 0 .08) 3.41 ( 0 .09) 3.70 ( 0 .07) 3.70 ( 0 .08) 3.45 ( 0 .09) 3.78 ( 0 .07) 3.85 ( 0 .08) 3.44 ( 0 .09) F (2,627) = 0.39, p = .815, partial 2 < .01 Figure W No covariates 4.28 ( 0 .68) 4.31 ( 0 .64) 4.3 3 ( 0 .67) 4.60 ( 0 .55) 4.47 ( 0 .64) 4.43 ( 0 .63) 4.43 ( 0 .60) 4.40 ( 0 .62) 4.28 ( 0 .64) F (2,600) = 1.70, p = .152, partial 2 = .01

PAGE 193

181 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values BMI 4.27 ( 0 .07) 4.30 ( 0 .07) 4.35 ( 0 .08) 4.61 ( 0 .05) 4.47 ( 0 .06) 4.47 ( 0 .07) 4.44 ( 0 .06) 4.40 ( 0 .06) 4.29 ( 0 .07) F (2,596) = 1.81, p = .130, partial 2 = .01 EDI BD 4.27 ( 0 .07) 4.30 ( 0 .07) 4.35 ( 0 .08) 4.62 ( 0 .05) 4.47 ( 0 .06) 4.46 ( 0 .07) 4.43 ( 0 .06) 4.40 ( 0 .06) 4.31 ( 0 .07) F (2,597) = 1.77, p = .138, partial 2 = .01 DMS 4.27 ( 0 .06) 4.26 ( 0 .07) 4.36 ( 0 .06) 4.61 ( 0 .05) 4.46 ( 0 .06) 4.48 ( 0 .06) 4.44 ( 0 .06) 4.41 ( 0 .06) 4.31 ( 0 .07) F (2,590) = 2.07, p = .088 partial 2 = .01 PACS 4.30 ( 0 .07) 4.29 ( 0 .07) 4.33 ( 0 .08) 4.64 ( 0 .06) 4.46 ( 0 .06) 4.45 ( 0 .07) 4.45 ( 0 .06) 4.39 ( 0 .06) 4.28 ( 0 .07) F (2,596) = 1.57, p = .184, partial 2 = .01 SATAQ 4.28 ( 0 .07) 4.30 ( 0 .07) 4.34 ( 0 .08) 4.62 ( 0 .06) 4.47 ( 0 .06) 4.48 ( 0 .07) 4.44 ( 0 .06) 4.40 ( 0 .06) 4.30 ( 0 .07) F (2,596) = 1.50, p = .204 partial 2 = .01

PAGE 194

182 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values Figure Z No covariates 4.67 ( 0 .87) 4.79 ( 0 .65) 4.34 (1.00) 4.78 ( 0 .75) 4.95 ( 0 .63) 4.56 ( 0 .92) 4.62 ( 0 .83) 4 .63 ( 0 .74) 4.34 (1.01) F (2,643) = 0.64, p = .630, partial 2 < .01 BMI 4.70 ( 0 .08) 4.79 ( 0 .08) 4.36 ( 0 .09) 4.81 ( 0 .07) 4.95 ( 0 .08) 4.59 ( 0 .08) 4.65 ( 0 .08) 4.63 ( 0 .08) 4.31 ( 0 .09) F (2,639) = 0.68, p = .606 partial 2 < .01 EDI BD 4.69 ( 0 .08) 4.79 ( 0 .08) 4 .39 ( 0 .09) 4.81 ( 0 .07) 4.96 ( 0 .08) 4.58 ( 0 .08) 4.63 ( 0 .08) 4.62 ( 0 .08) 4.36 ( 0 .09) F (2,640) = 0.52, p = .722 partial 2 < .01 DMS 4.69 ( 0 .08) 4.77 ( 0 .09) 4.37 ( 0 .09) 4.80 ( 0 .07) 4.94 ( 0 .07) 4.62 ( 0 .08) 4.65 ( 0 .08) 4.64 ( 0 .08) 4.29 ( 0 .09) F (2,631) = 0.77, p = .542 partial 2 = .01 PACS 4.69 ( 0 .08) 4.79 ( 0 .08) 4.37 ( 0 .09) 4.81 ( 0 .07) 4.95 ( 0 .08) 4.59 ( 0 .08) 4.64 ( 0 .08) 4.63 ( 0 .08) 4.64 ( 0 .08) F (2,639) = 0.54, p = .705 partial 2 < .01

PAGE 195

183 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values SATAQ 4.68 ( 0 .08) 4.78 ( 0 .08) 4.41 ( 0 .09) 4.81 ( 0 .07) 4.95 ( 0 .08) 4.60 ( 0 .09) 4.65 ( 0 .08) 4.62 ( 0 .08) 4.35 ( 0 .09) F (2,637) = 0.60, p = .662 partial 2 < .01 Figure AA No covariates 6.05 ( 0 .80) 5.94 ( 0 .88) 5.84 ( 0 .97) 5.94 ( 0 .89) 5.92 ( 0 .80) 5.71 ( 0 .85) 5.61 ( 0 .85) 5.50 ( 0 .81) 5.37 ( 0 .93) F (4,646) = 0.40, p = .806, partial 2 < .01 BMI 6.04 ( 0 .08) 5.94 ( 0 .09) 5.89 ( 0 .10) 5.92 ( 0 .08) 5.93 ( 0 .08) 5.68 ( 0 .09) 5.58 ( 0 .08) 5.50 ( 0 .08) 5.39 ( 0 .10) F (4,642) = 0.49, p = .745, partial 2 < .01 EDI BD 6.05 ( 0 .08) 5.94 ( 0 .09) 5.85 ( 0 .10) 5.93 ( 0 .08) 5.92 ( 0 .08) 5.67 ( 0 .09) 5.60 ( 0 .08) 5.50 ( 0 .08) 5.33 ( 0 .09) F (4,644) = 0.37, p = .830, partial 2 < .01 DMS 6.04 ( 0 .08) 5.97 ( 0 .09) 5.86 ( 0 .09) 5.93 ( 0 .08) 5.90 ( 0 .09) 5.68 ( 0 .09) 5.59 ( 0 .08) 5.50 ( 0 .09) 5.37 ( 0 .09) F (4,635) = 0.19, p = .944, partial 2 < .01

PAGE 196

184 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values PACS 6.06 ( 0 .08) 5.94 ( 0 .09) 5 .84 ( 0 .10) 5.90 ( 0 .08) 5.94 ( 0 .08) 5.69 ( 0 .09) 5.59 ( 0 .08) 5.50 ( 0 .09) 5.36 ( 0 .09) F (4,642) = 0.45, p = .775, partial 2 < .01 SATAQ 6.05 ( 0 .08) 5.94 ( 0 .09) 5.88 ( 0 .10) 5.90 ( 0 .08) 5.92 ( 0 .08) 5.72 ( 0 .10) 5.57 ( 0 .08) 5.49 ( 0 .09) 5.40 ( 0 .10) F (4,640) = 0.3 0, p = .877, partial 2 < .01 Figure DD No covariates 5.74 ( 0 .81) 5.68 ( 0 .83) 5.39 ( 0 .90) 5.69 ( 0 .75) 5.72 ( 0 .83) 5.49 (. 0 92) 5.69 (1.01) 5.73 ( 0 .80) 5.40 ( 0 .99) F (2,618) = 0.55, p = .698, partial 2 < .01 BMI 5.74 ( 0 .08) 5.67 ( 0 .08) 5.46 ( 0 .09) 5.68 ( 0 .08) 5.73 ( 0 .08) 5.51 ( 0 .09) 5.70 ( 0 .08) 5.73 ( 0 .09) 5.42 ( 0 .10) F (2,619) = 0.47, p = .751, partial 2 < .01 EDI BD 5.76 ( 0 .08) 5.67 ( 0 .08) 5.42 ( 0 .09) 5.68 ( 0 .08) 5.73 ( 0 .08) 5.52 ( 0 .09) 5.67 ( 0 .08) 5.73 ( 0 .09) 5.49 ( 0 .10) F (2,618) = 0.75, p = .553, pa rtial 2 = .01

PAGE 197

185 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucas ian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values DMS 5.76 ( 0 .08) 5.67 ( 0 .09) 5.44 ( 0 .09) 5.67 ( 0 .08) 5.73 ( 0 .08) 5.51 ( 0 .09) 5.69 ( 0 .08) 5.74 ( 0 .09) 5.44 ( 0 .10) F (2,606) = 0.68, p = .599, partial 2 < .01 PACS 5.71 ( 0 .08) 5.69 ( 0 .08) 5.47 (. 0 09) 5.66 ( 0 .08) 5.74 ( 0 .08) 5.53 ( 0 .09) 5.67 ( 0 .08) 5.74 ( 0 .09) 5.45 ( 0 .10) F (2,615) = 0.35, p = .836, partial 2 < .01 SATAQ 5.74 ( 0 .08) 5.66 ( 0 .08) 5.47 ( 0 .10) 5.65 ( 0 .08) 5.72 ( 0 .08) 5.58 ( 0 .09) 5.67 ( 0 .08) 5.72 ( 0 .09) 5.46 ( 0 .10) F (2,612) = 0.81, p = .514, partial 2 = .01

PAGE 198

186 Table L.5. Repeat ed measure s effects: Main effects of target race Target Race Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure A No covariates 3.88 ( 0 .86) a 3.74 ( 0 .82) b 3.91 ( 0 .83) a F (2, 634) = 6 .56, p = .002, partial 2 = .02 BMI 3.90 ( 0 .05) 3.76 ( 0 .05) 3.92 ( 0 .05) F (2,630) = 1.77, p = .172, partial 2 = .01 EDI BD 3.89 ( 0 .05) 3.74 ( 0 .05) 3.90 ( 0 .05) F (2,632) = 0.96, p = .381, partial 2 < .01 DMS 3.90 ( 0 .05) 3.76 ( 0 .05) 3.91 ( 0 .05) F (2,624) = 0.09, p = .910, partial 2 < .01 PACS 3.89 ( 0 .05) 3.75 ( 0 .05) 3.91 ( 0 .05) F (2,630) = 1.26, p = .285, partial 2 < .01 SATAQ 3.91 ( 0 .05) 3.74 ( 0 .05) 3.92 ( 0 .05) F (2,629) = 1.46, p = .233, partial 2 = .01 Figure B No covariates 3.65 ( 0 .85) 3.63 ( 0 .82) 3.62 ( 0 .84) F (2,636) = 0.30, p = .735, partial 2 < .01 BMI 3.64 ( 0 .05) 3.64 ( 0 .05) 3.61 ( 0 .05) F (2,634) = 2.92, p = .055, partial 2 = .01 EDI BD 3.63 ( 0 .05) 3.62 ( 0 .05) 3.59 ( 0 .05) F (2,634) = 0.11, p = .890, partial 2 < .01 DMS 3.64 ( 0 .05) 3.62 ( 0 .05 ) 3.60 ( 0 .05) F (2,629) = 0.51, p = .602, partial 2 < .01 PACS 3.63 ( 0 .05) 3.63 ( 0 .05) 3.60 ( 0 .05) F (2,632) = 1.59, p = .205, partial 2 = .01 SATAQ 3.63 ( 0 .05) 3.63 ( 0 .05) 3.61 ( 0 .05) F (2, 629) = 0.38, p = .684, partial 2 < .01

PAGE 199

187 Target Race Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure C No covariates 2.37 (1.16) a 2.59 (1.09) b 2.21 (1.13) c F (2,638) = 18.36, p < .001 partial 2 = .05 BMI 2.35 ( 0 .07) 2.58 ( 0 .06) 2.20 ( 0 .07) F (2,633) = 1.66, p = .191, partial 2 = .01 EDI BD 2.35 ( 0 .07) 2.58 ( 0 .06) 2.19 ( 0 .07) F (2,636) = 0.53, p = .588, partial 2 < .0 1 DMS 2.34 ( 0 .07) 2.58 ( 0 .06) 2.19 ( 0 .07) F (2,627) = 1.19, p = .304, partial 2 < .01 PACS 2.34 ( 0 .07) 2.57 ( 0 .06) 2.19 ( 0 .07) F (2,634) = 0.13, p = .874, partial 2 < .01 SATAQ 2.36 ( 0 .07) a 2.59 ( 0 .06) b 2.20 ( 0 .07) c F (2,632) = 5.94, p = .003, partial 2 = .02 Figure D No covariates 4.23 ( 0 .69) a 4.13 ( 0 .74) b 4.11 ( 0 .69) b F (2,640) = 4.31, p = .014, partial 2 = .01 BMI 4.24 ( 0 .04) 4.14 ( 0 .04) 4.11 ( 0 .04) F (2,635) = 0.51, p = .597, partial 2 < .01 EDI BD 4.23 ( 0 .04) 4.13 ( 0 .04) 4.11 ( 0 .04) F (2,638) = 0 .44, p = .641, partial 2 < .01 DMS 4.23 ( 0 .04) 4.13 ( 0 .04) 4.11 ( 0 .04) F (2,627) = 1.11, p = .331, partial 2 < .01 PACS 4.23 ( 0 .04) 4.13 ( 0 .04) 4.11 ( 0 .04) F (2,636) = 0.11, p = .893, partial 2 < .01 SATAQ 4.24 ( 0 .04) 4.13 ( 0 .04) 4.11 ( 0 .04) F (2,634) = 1.99, p = .138, partial 2 = .01 Figure G No covariates 3.76 ( 0 .62) 3.75 ( 0 .66) 3.77 ( 0 .69) F (2,617) = 0.23, p = .785, partial 2 < .01

PAGE 200

188 Target Race Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values BMI 3.75 ( 0 .04) 3.77 ( 0 .04) 3.78 ( 0 .04) F (2,611) = 0.66, p = .510, partial 2 < .01 EDI BD 3.75 ( 0 .04) 3.77 ( 0 .04) 3.78 ( 0 .04) F (2,614) = 1.69, p = .186, partial 2 = .01 DMS 3.75 ( 0 .04) 3.77 ( 0 .04) 3.78 ( 0 .04) F (2,604) = 0.43, p = .644, partial 2 < .01 PACS 3.76 ( 0 .04) 3.77 ( 0 .04) 3.78 ( 0 .04) F (2,613) = 2.20, p = .114, partial 2 = .01 SATAQ 3.76 ( 0 .04) 3.77 ( 0 .04 ) 3.78 ( 0 .04) F (2,611) = 0.23, p = .788, partial 2 < .01 Figure H No covariates 4.49 ( 0 .75) a 4.39 ( 0 .63) b 4.27 ( 0 .63) c F (2,618) = 10.34, p < .001 partial 2 = .03 BMI 4.50 ( 0 .04) 4.39 ( 0 .04) 4.29 ( 0 .04) F (2,614) = 0.38, p = .672, partial 2 < .01 EDI BD 4.49 ( 0 .04) 4.39 ( 0 .06) 4.28 ( 0 .04) F (2,616) = 0.98, p = .371, partial 2 < .01 DMS 4.49 ( 0 .04) 4.39 ( 0 .04) 4.28 ( 0 .04) F (2,609) = 0.80, p = .443, partial 2 < .01 PACS 4.49 ( 0 .04) 4.39 ( 0 .04) 4.29 ( 0 .04) F (2,613) = 2.97, p = .055, partial 2 = .01 SATAQ 4.50 ( 0 .04) 4.39 ( 0 .04) 4.29 ( 0 .04) F (2,612) = 1.93, p = .149, partial 2 = .01 Figure K No covariates 4.40 ( 0 .63) 4.45 ( 0 .58) 4.49 ( 0 .61) F (2,640) = 3.47, p = .032, partial 2 = .01 BMI 4.38 ( 0 .04) 4.45 ( 0 .03) 4.49 ( 0 .03) F (2,636) = 0.48, p = .62 0, partial 2 < .01 EDI BD 4.38 ( 0 .04) 4.45 ( 0 .03) 4.50 ( 0 .03) F (2,638) = 0.47, p = .626, partial 2 < .01

PAGE 201

189 Target Race Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values DMS 4.38 ( 0 .04) 4.46 ( 0 .03) 4.48 ( 0 .03) F (2,629) = 0.09, p = .915, partial 2 < .01 PACS 4.38 ( 0 .04) 4.45 ( 0 .03) 4.49 ( 0 .03) F (2,636) = 0.35, p = .702, partial 2 < .01 SATAQ 4.39 ( 0 .04) 4.45 ( 0 .03) 4.50 ( 0 .03) F (2,634) = 0.42, p = .655, partial 2 < .01 Figure L No covariates 5.10 ( 0 .77) a 4.81 ( 0 .75) b 5.04 ( 0 .70) a F (2,634) = 17.31, p < .001 partial 2 = .05 BMI 5.10 ( 0 .04) 4.82 ( 0 .04) 5.03 ( 0 04) F (2,631) = 0.58, p = .559, partial 2 < .01 EDI BD 5.09 ( 0 .04) 4.82 ( 0 .04) 5.03 ( 0 .04) F (2,632) = 1.29, p = .277, partial 2 < .01 DMS 5.11 ( 0 .04) 4.82 ( 0 .04) 5.04 ( 0 .04) F (2,622) = 0.51, p = .598, partial 2 < .01 PACS 5.10 ( 0 .04) 4.82 ( 0 .04) 5.04 ( 0 .04) F (2,631) = 0.03, p = .970, partial 2 < .01 SATAQ 5.10 ( 0 .04) 4.82 ( 0 .04) 5.03 ( 0 .04) F (2,630) = 0.99, p = .370, partial 2 < .01 Figure O No covariates 5.39 ( 0 .79) a 5.10 ( 0 .90) b 5.29 ( 0 .81) a F (2,646) = 17.45, p < .001 partial 2 = .05 BMI 5.40 ( 0 .05) 5.09 ( 0 .05) 5.30 ( 0 .05) F (2,642) = 0.25, p = .774, partial 2 < .01 EDI BD 5.40 ( 0 .05) 5.09 ( 0 .05) 5.30 ( 0 .05) F (2,644) = 1.87, p = .156, partial 2 = .01 DMS 5.40 ( 0 .05) 5.09 ( 0 .05) 5.30 ( 0 .05) F (2,636) = 0.12, p = .887, partial 2 < .01 PACS 5 .40 ( 0 .05) 5.09 ( 0 .05) 5.30 ( 0 .05) F (2,642) = 0.77, p = .459, partial 2 < .01

PAGE 202

190 Target Race Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values SATAQ 5.40 ( 0 .05) 5.08 ( 0 .05) 5.31 ( 0 .05) F (2,639) = 2.25, p = .106, partial 2 = .01 Figure P No covariates 4.23 ( 0 .54) a 4.10 ( 0 .59) b 4.18 ( 0 .61) b F (2,611) = 6.74, p = .002, partial 2 = .02 BMI 4.24 ( 0 .03) 4.11 ( 0 .03) 4.17 ( 0 .03) F (2,608) = 3.02, p = .052, partial 2 = .01 EDI BD 4.24 ( 0 .03) 4.10 ( 0 .03) 4.17 ( 0 .03) F (2,609) = 1.59, p = .206, partial 2 = .01 DMS 4.24 ( 0 .03) 4.11 ( 0 .03) 4.17 ( 0 .03) F (2,600) = 0.57, p = .55 6, partial 2 < .01 PACS 4.24 ( 0 .03) 4.10 ( 0 .03) 4.17 ( 0 .03) F (2,608) = 1.64, p = .197, partial 2 = .01 SATAQ 4.24 ( 0 .03) 4.10 ( 0 .03) 4.17 ( 0 .03) F (2,607) = 1.19, p = .302, partial 2 < .01 Figure Q No covariates 4.03 ( 0 .45) 3.99 ( 0 .43) 4.00 ( 0 .42) F ( 2,651) = 1.54, p = .215, partial 2 = .01 BMI 4.04 ( 0 .03) 3.99 ( 0 .03) 4.01 ( 0 .02) F (2,647) = 0.22, p = .802, partial 2 < .01 EDI BD 4.05 ( 0 .03) 3.99 ( 0 .03) 4.01 ( 0 .02) F (2,649) = 0.33, p = .721, partial 2 < .01 DMS 4.05 ( 0 .03) 3.99 ( 0 .02) 4.00 ( 0 .02) F (2,642) = 0.46, p = .632, partial 2 < .01 PACS 4.04 ( 0 .03) 3.99 ( 0 .03) 4.01 ( 0 .02) F (2,647) = 0.73, p = .484, partial 2 < .01 SATAQ 4.05 ( 0 .03) 4.00 ( 0 .02) 4.01 ( 0 .02) F (2,646) = 1.97, p = .140, partial 2 = .01

PAGE 203

191 Target Race Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure R No covariates 3.44 ( 0 .78) a 3 .43 ( 0 .78) a 3.27 ( 0 .80) b F (2,619) = 7.12, p = .001, partial 2 = .02 BMI 3.46 ( 0 .04) 3.42 ( 0 .05) 3.27 ( 0 .05) F (2,615) = 0.07, p = .932, partial 2 < .01 EDI BD 3.47 ( 0 .04) 3.42 ( 0 .06) 3.27 ( 0 .05) F (2,618) = 1.72, p = .182, partial 2 = .01 DMS 3.47 ( 0 .0 5) 3.41 ( 0 .04) 3.28 ( 0 .05) F (2,613) = 0.31, p = .731, partial 2 < .01 PACS 3.46 ( 0 .04) 3.42 ( 0 .05) 3.27 ( 0 .05) F (2,615) = 0.56, p = .569, partial 2 < .01 SATAQ 3.48 ( 0 .04) 3.44 ( 0 .04) 3.27 ( 0 .05) F (2,607) = 1.28, p = .277, partial 2 < .01 Figure S N o covariates 4.36 ( 0 .66) 4.38 ( 0 .69) 4.40 ( 0 .66) F (2,633) = 0.58, p = .559, partial 2 < .01 BMI 4.36 ( 0 .04) 4.38 ( 0 .04) 4.41 ( 0 .04) F (2,629) = 0.17, p = .844, partial 2 < .01 EDI BD 4.36 ( 0 .04) 4.38 ( 0 .04) 4.40 ( 0 .04) F (2,631) = 1.32, p = .268, partial 2 < .01 DMS 4.37 ( 0 .04) 4.39 ( 0 .04) 4.40 ( 0 .04) F (2,624) = 1.12, p = .329, partial 2 < .01 PACS 4.36 ( 0 .04) 4.38 ( 0 .04) 4.40 ( 0 .04) F (2,629) = 0.37, p = .688, partial 2 < .01 SATAQ 4.36 ( 0 .04) 4.38 ( 0 .04) 4.40 ( 0 .04) F (2,627) = 0.40, p = .673, parti al 2 < .01 Figure V No covariates 3.65 ( 0 .77) 3.60 ( 0 .83) 3.70 ( 0 .82) F (2,632) = 1.12, p = .328, partial 2 < .01

PAGE 204

192 Target Race Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values BMI 3.64 ( 0 .04) 3.61 ( 0 .05) 3.69 ( 0 .05) F (2,628) = 0.01, p = .992, partial 2 < .01 EDI BD 3.65 ( 0 .04) 3.61 ( 0 .05) 3.69 ( 0 .05) F (2,630) = 0.35, p = .702, partial 2 < .01 DMS 3.65 ( 0 .04) 3.61 ( 0 .05) 3.68 ( 0 .05) F (2,620) = 0.41, p = .659, partial 2 < .01 PACS 3.64 ( 0 .04) 3.61 ( 0 .05) 3.69 ( 0 .05) F (2,628) = 0.34, p = .708, partial 2 < .01 SATAQ 3.65 ( 0 .04) 3.62 ( 0 .05) 3.69 ( 0 .05) F (2,627 ) = 1.05, p = .351, partial 2 < .01 Figure W No covariates 4.30 ( 0 .71) a 4.51 ( 0 .61) b 4.38 ( 0 .61) a F (2,600) = 12.83, p < .001 partial 2 = .04 BMI 4.31 ( 0 .04) 4.52 ( 0 .04) 4.38 ( 0 .04) F (2,596) = 0.38, p = .671, partial 2 < .01 EDI BD 4.31 ( 0 .04) 4.52 ( 0 .03) 4.38 ( 0 .04) F (2,597) = 1.75, p = .176, partial 2 = .01 DMS 4.30 ( 0 .04) 4.52 ( 0 .03) 4.39 ( 0 .04) F (2,590) = 3.90, p = .023 partial 2 = .01 PACS 4.30 ( 0 .04) 4.52 ( 0 .03) 4.38 ( 0 .04) F (2,596) = 0.21, p = .802, partial 2 < .01 SATAQ 4.31 ( 0 .04) 4.52 ( 0 .04) 4.38 ( 0 .04) F (2,596) = 0.52, p = .583 partial 2 < .01 Figure Z No covariates 4.61 ( 0 .87) a 4.77 ( 0 .79) b 4.54 ( 0 .87) a F (2,642) = 12.84, p < .001 partial 2 = .04 BMI 4.62 ( 0 .05) a 4.79 ( 0 .04) b 4.53 ( 0 .05) a F (2,639) = 0.13, p = .875 partial 2 < 01 EDI BD 4.62 ( 0 .05) a 4.78 ( 0 .04) b 4.54 ( 0 .04) a F (2,640) = 5.58, p = .004 partial 2 = .02

PAGE 205

193 Target Race Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values DMS 4.61 ( 0 .05) 4.79 ( 0 .04) 4.53 ( 0 .05) F (2,631) = 0.51, p = .599 partial 2 < .01 PACS 4.62 ( 0 .05) 4.78 ( 0 .04) 4.53 ( 0 .05) F (2,639) = 0.82, p = .441 partial 2 < .01 SATAQ 4.62 ( 0 .05) 4.79 ( 0 .04) 4.54 ( 0 .05) F (2,637) = 3.84, p = .022 partial 2 = .01 Figure AA No covariates 5.95 ( 0 .88) a 5.86 ( 0 .85) a 5.50 ( 0 .87) b F (2,646) = 36.20, p < .001 partial 2 = .10 BMI 5.96 ( 0 .05) 5.84 ( 0 .05) 5.49 ( 0 .05) F (2,642) = 0 .83, p = .435, partial 2 < .01 EDI BD 5.95 ( 0 .05) 5.84 ( 0 .05) 5.47 ( 0 .05) F (2,644) = 0.58, p = .559, partial 2 < .01 DMS 5.96 ( 0 .05) 5.84 ( 0 .05) 5.49 ( 0 .05) F (2,635) = 1.23, p = .294, partial 2 < .01 PACS 5.95 ( 0 .05) 5.84 ( 0 .05) 5.48 ( 0 .05) F (2,642) = 2.54, p = .080, partial 2 = .01 SATAQ 5.96 ( 0 .05) a 5.85 ( 0 .05) a 5.49 ( 0 .05) b F (2,640) = 5.68, p = .004, partial 2 = .02 Figure DD No covariates 5.61 ( 0 .87) 5.64 ( 0 .83) 5.61 ( 0 .95) F (2,618) = 0.01, p = .901, partial 2 < .01 BMI 5.63 ( 0 .05) 5.64 ( 0 05) 5.61 ( 0 .05) F (2,619) = 9.95, p < .001 partial 2 = .03 EDI BD 5.62 ( 0 .05) 5.64 ( 0 .05) 5.63 ( 0 .05) F (2,618) = 2.58, p = .078, partial 2 = .01 DMS 5.62 ( 0 .05) 5.64 ( 0 .05) 5.63 ( 0 .05) F (2,606) = 0.18, p = .828, partial 2 < .01 PACS 5.63 ( 0 .05) 5.64 ( 0 .05) 5.62 ( 0 .05) F (2,615) = 1.31, p = .271, partial 2 < .01

PAGE 206

194 Target Race Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values SATAQ 5.62 ( 0 .05) 5.65 ( 0 .05) 5.62 ( 0 .05) F (2,612) = 0.70, p = .492, partial 2 < .01 Table L.6. Between subjects effects: Rater gender X rater race interaction Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure A No covariates 3.88 (0.77) 3.85 (0.56) 3.57 ( 0 .95) 3.94 ( 0 .39) 3.98 ( 0 72) 3.91 ( 0 .50) F (2, 322) = 3.41, p = .034, partial 2 = .02 BMI 3.90 (0.09) 3.80 (0.08) 3.60 ( 0 .09) 3.91 ( 0 .09) 4.02 ( 0 .11) 3.93 ( 0 .08) F (2,320) = 3.43, p = .033, partial 2 = .02 EDI BD 3.85 (0.09) 3.89 (0.08) 3.56 ( 0 .09) 3.96 ( 0 .09) 3.90 ( 0 .12) 3.92 ( 0 .08) F (2,321) = 2.81, p = .062, partial 2 = .02

PAGE 207

195 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS 3.89 (0.10) 3.85 (0.08) 3.57 ( 0 .10) 3.93 ( 0 .09) 3.98 ( 0 .11) 3.92 ( 0 .09) F (2,317) = 3.19, p = .042, partial 2 = .02 PACS 3.89 (0.09) 3.89 (0.08) 3.55 ( 0 .09) 3.94 ( 0 .09) 3.94 ( 0 .11) 3.91 ( 0 .08) F (2,320) = 3.42, p = .034, partial 2 = .02 SATAQ 3.86 (0.09) 3.88 (0.08) 3.57 ( 0 .09) 3.94 ( 0 .09) 4.01 ( 0 .12) 3.89 ( 0 .08) F (2,319) = 3.72, p = .025, partial 2 = .02 Figure B No covariates 3.50 (0.71) 3.73 (0.56) 3.45 ( 0 .88) 3.74 ( 0 .52) 3.57 ( 0 .76) 3.74 ( 0 .64) F (2,322) = 0.19, p = .828, partial 2 < .01 BMI 3.53 (0.09) 3.66 (0.08) 3.49 ( 0 .10) 3.71 ( 0 .09) 3.63 ( 0 .12) 3.76 ( 0 .08) F (2,320) = .19, p = .828, partial 2 < .01

PAGE 208

196 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values EDI BD 3.48 (0.09) 3.76 (0.08) 3.44 ( 0 .10) 3.76 ( 0 .09) 3.52 ( 0 .12) 3.75 ( 0 .08) F (2,321) = 0 .10, p = .908, partial 2 < .01 DMS 3.50 (0.09) 3.73 (0.08) 3.46 ( 0 .11) 3.74 ( 0 .10) 3.57 ( 0 .12) 3.71 ( 0 .09) F (2,317) = 0.21, p = .808, partial 2 < .01 PACS 3.50 (0.09) 3.73 (0.08) 3.44 ( 0 .10) 3.74 ( 0 .09) 3.56 ( 0 .12) 3.74 ( 0 .09) F (2,320) = 0.20, p = .820 partial 2 < .01 SATAQ 3.49 (0.09) 3.74 (0.08) 3.45 ( 0 .10) 3.74 ( 0 .09) 3.57 ( 0 .12) 3.74 ( 0 .09) F (2,319) = 0.20, p = .815, partial 2 < .01 Figure C No covariates 2.32 (0.93) 2.61 (0.87) 2.13 ( 0 .97) 2.51 ( 0 .85) 2.26 ( 0 .91) 2.40 (1.00) F (2,320) = 0.36, p = .697, partial 2 < .01

PAGE 209

197 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values BMI 2.34 (0.13) 2.56 (0.11) 2.15 ( 0 .13) 2.50 ( 0 .13) 2.30 ( 0 .17) 2.42 ( 0 .11) F (2,318) = 0.35, p = .704, partial 2 < .01 EDI BD 2.32 (0.13) 2.62 (0.11) 2.13 ( 0 .13) 2.51 ( 0 .13) 2.24 ( 0 .17) 2.41 ( 0 .11) F (2,319) = 0.32, p = .728, p artial 2 < .01 DMS 2.32 (0.12) 2.60 (0.11) 2.16 ( 0 .15) 2.51 ( 0 .13) 2.26 ( 0 .16) 2.36 ( 0 .12) F (2,315) = 0.46, p = .632, partial 2 < .01 PACS 2.32 (0.13) 2.61 (0.11) 2.13 ( 0 .13) 2.51 ( 0 .13) 2.25 ( 0 .16) 2.39 ( 0 .12) F (2,318) = 0.43, p = .654, partial 2 < 01 SATAQ 2.32 (0.13) 2.55 (0.11) 2.12 ( 0 .13) 2.50 ( 0 .13) 2.35 ( 0 .17) 2.45 ( 0 .12) F (2,317) = 0.58, p = .558, partial 2 < .01

PAGE 210

198 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure D No covariates 4.23 (0.71) 4.15 (0.44) 4.22 ( 0 .68) 4.07 ( 0 .50) 4.17 ( 0 .52) 4.11 ( 0 .37) F (2,322) = 0.21, p = .813, partia l 2 < .01 BMI 4.24 (0.07) 4.14 (0.07) 4.23 ( 0 .08) 4.07 ( 0 .08) 4.18 ( 0 .09) 4.11 ( 0 .07) F (2, 320) = 0.20, p = .817, part i al 2 < .01 EDI BD 4.22 (0.07) 4.17 (0.06) 4.22 ( 0 .08) 4.08 ( 0 .08) 4.13 ( 0 .10) 4.12 ( 0 .07) F (2,321) = 0.31, p = .732, partial 2 < .01 DMS 4.22 (0.08) 4.15 (0.07) 4.20 ( 0 .09) 4.08 ( 0 .08) 4.17 ( 0 .09) 4.13 ( 0 .07) F (2,317) = 0.16, p = .856, partial 2 < .01 PACS 4.23 (0.07) 4.16 (0.07) 4.22 ( 0 .08) 4.07 ( 0 .08) 4.16 ( 0 .09) 4.11 ( 0 .07) F (2,320) = 0.21, p = .811, partial 2 < .01

PAGE 211

199 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values SATAQ 4.21 ( 0.07) 4.16 (0.07) 4.23 ( 0 .08) 4.07 ( 0 .08) 4.17 ( 0 .10) 4.11 ( 0 .07) F (2,319) = 0.27, p = .764, partial 2 < .01 Figure G No covariates 3.93 (0.44) 3.69 (0.45) 3.90 ( 0 .50) 3.72 ( 0 .49) 3.73 ( 0 .31) 3.62 ( 0 .61) F (2,323) = 0.51, p = .601, partial 2 < .01 BMI 3.94 (0.07) 3.70 (0.06) 3.90 ( 0 .07) 3.73 ( 0 .07) 3.73 ( 0 .09) 3.62 ( 0 .06) F (2,321) = 0.42, p = .657, partial 2 < .01 EDI BD 3.94 (0.07) 3.69 (0.06) 3.90 ( 0 .07) 3.72 ( 0 .07) 3.74 ( 0 .09) 3.62 ( 0 .06) F (2,322) = 0.50, p = .605, partial 2 < .01 DMS 3.93 (0.07) 3.69 (0.06) 3.89 ( 0 .08) 3.73 ( 0 .09) 3.73 ( 0 .09) 3.63 ( 0 .07) F (2,318) = 0.58, p = .562, partial 2 < .01

PAGE 212

200 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values PACS 3.94 (0.07) 3.68 (0.06) 3.91 ( 0 .07) 3.72 ( 0 .07) 3.75 ( 0 .07) 3.62 ( 0 .06) F (2,321) = 0.44, p = .644, partial 2 < .01 SATAQ 3.95 (0.07) 3.67 (0.06 ) 3.90 ( 0 .07) 3.72 ( 0 .07) 3.76 ( 0 .09) 3.63 ( 0 .06) F (2,320) = 0.70, p = .497, partial 2 < .01 Figure H No covariates 4.56 (0.46) 4.39 (0.47) 4.46 ( 0 .57) 4.29 ( 0 .41) 4.38 ( 0 .52) 4.26 ( 0 .41) F (2,323) = 0.01, p = .916, partial 2 < .01 BMI 4.57 (0.06) 4.36 (0.06) 4.47 ( 0 .07) 4.28 ( 0 .07) 4.40 ( 0 .08) 4.26 ( 0 .06) F (2,321) = 0.13, p = .881, partial 2 < .01 EDI BD 4.56 (0.06) 4.39 (0.06) 4.46 ( 0 .07) 4.39 ( 0 .07) 4.37 ( 0 .09) 4.26 ( 0 .06) F (2,322) = 0.10, p = .903, partial 2 < .01

PAGE 213

201 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS 4.54 (0.07) 4.40 (0.06) 4.4 3 ( 0 .07) 4.31 ( 0 .07) 4.36 ( 0 .08) 4.28 ( 0 .06) F (2,318) = 0.01 p = .915, partial 2 < .01 PACS 4.55 (0.06) 4.37 (0.06) 4.47 ( 0 .07) 4.29 ( 0 .07) 4.39 ( 0 .08) 4.26 ( 0 .06) F (2,321) = 0.08, p = .928, partial 2 < .01 SATAQ 4.56 (0.06) 4.37 (0.06) 4.46 ( 0 .07) 4.2 9 ( 0 .07) 4.40 ( 0 .08) 4.27 ( 0 .06) F (2,320) = 0.07, p = .929, partial 2 < .01 Figure K No covariates 4.56 (0.43) 4.53 (0.45) 4.51 ( 0 .44) 4.36 ( 0 .42) 4.34 ( 0 .43) 4.33 ( 0 .37) F (2,320) = 0.74, p = .476, partial 2 = .01 BMI 4.58 (0.06) 4.51 (0.05) 4.53 ( 0 .0 6) 4.36 ( 0 .06) 4.35 ( 0 .08) 4.33 ( 0 .05) F (2,318) = 0.79, p = .456, partial 2 = .01

PAGE 214

202 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values EDI BD 4.57 (0.06) 4.52 (0.05) 4.52 ( 0 .06) 4.36 ( 0 .06) 4.35 ( 0 .08) 4.33 ( 0 .05) F (2,319) = 0.67, p = .513, partial 2 < .01 DMS 4.58 (0.06) 4.52 (0.05) 4.54 ( 0 .07) 4.35 ( 0 06) 4.35 ( 0 .08) 4.30 ( 0 .06) F (2,315) = 0.87, p = .420, partial 2 = .01 PACS 4.56 (0.06) 4.53 (0.05) 4.52 ( 0 .06) 4.37 ( 0 .06) 4.34 ( 0 .08) 4.32 ( 0 .05) F (2,318) = 0.68, p = .507, partial 2 < .01 SATAQ 4.57 (0.06) 4.50 (0.05) 4.51 ( 0 .06) 4.36 ( 0 .06) 4.40 ( 0 .08) 4.34 ( 0 .05) F (2,317) = 0.38, p = .682, partial 2 < .01 Figure L No covariates 5.10 (0.50) 5.03 (0.59) 5.08 ( 0 .59) 5.03 ( 0 .42) 4.89 ( 0 .53) 4.78 ( 0 .56) F (2,321) = 0.07, p = .928, partial 2 < .01

PAGE 215

203 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values BMI 5.11 (0.08) 5.01 (0.07) 5.09 ( 0 .08) 5.02 ( 0 .08) 4 .90 ( 0 .10) 4.78 ( 0 .07) F (2,319) = 0.05, p = .956, partial 2 < .01 EDI BD 5.10 (0.08) 5.04 (0.06) 5.07 ( 0 .08) 5.03 ( 0 .07) 4.87 ( 0 .10) 4.78 ( 0 .07) F (2,320) = 0.05, p = .953, partial 2 < .01 DMS 5.10 (0.08) 5.03 (0.07) 5.08 ( 0 .09) 5.03 ( 0 .08) 4.89 ( 0 .10) 4.80 ( 0 .07) F (2,316) = 0.03, p = .969, partial 2 < .01 PACS 5.10 (0.08) 5.01 (0.06) 5.09 ( 0 .08) 5.02 ( 0 .07) 4.91 ( 0 .10) 4.79 ( 0 .07) F (2,319) = 0.06, p = .938, partial 2 < .01 SATAQ 5.11 (0.08) 4.99 (0.07) 5.07 ( 0 .08) 5.02 ( 0 .07) 4.92 ( 0 .10) 4.80 ( 0 .07) F (2,319) = 0.12, p = .886, partial 2 < .01

PAGE 216

204 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure O No covariates 5.39 (0.58) 5.31 (0.56) 5.37 ( 0 .69) 5.30 ( 0 .51) 5.12 ( 0 .61) 5.06 ( 0 .72) F (2,325) = 0.02, p = .985, partial 2 < .01 BMI 5.40 (0.08) 5.30 (0.07) 5.38 ( 0 .09) 5.30 ( 0 .09) 5.15 ( 0 .11) 5.06 ( 0 .08) F (2,323) = 0.01, p = .990, partial 2 < .01 EDI BD 5.40 (0.08) 5.30 (0.07) 5.38 ( 0 .09) 5.30 ( 0 .09) 5.13 ( 0 .11) 5.06 ( 0 .08) F (2,324) = 0.01, p = .988, partial 2 < .01 DMS 5.40 (0.09) 5.31 (0.07) 5.38 ( 0 .10) 5.30 ( 0 .09) 5.12 ( 0 .11) 5.07 ( 0 .08) F (2, 320) = 0.02, p = .973, partial 2 < .01 PACS 5.38 (0.08) 5.27 (0.07) 5.40 ( 0 .09) 5.30 ( 0 .08) 5.16 ( 0 .11) 5.07 ( 0 .08) F (2,323) = 0.004, p = .996, partial 2 < .01

PAGE 217

205 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values SATAQ 5.40 (0.08) 5.26 (0.08) 5.37 ( 0 .09) 5.30 ( 0 .09) 5.17 ( 0 .11) 5.09 ( 0 .08) F (2,322) = 0.1 1, p = .894, partial 2 < .01 Figure P No covariates 4.32 (.53) 4.18 (0.40) 4.15 ( 0 .43) 4.12 ( 0 .31) 4.22 ( 0 .57) 4.04 ( 0 .36) F (2,323) = 0.77, p = .466, partial 2 = .01 BMI 4.31 (0.06) 4.19 (0.05) 4.14 ( 0 .06) 4.12 ( 0 .06) 4.22 ( 0 .08) 4.04 ( 0 .05) F (2,321) = 0.85, p = .430, partial 2 = .01 EDI BD 4.32 (0.06) 4.18 (0.05) 4.15 ( 0 .06) 4.12 ( 0 .06) 4.23 ( 0 .08) 4.04 ( 0 .05) F (2,322) = 0.79, p = .454, partial 2 = .01 DMS 4.33 (0.06) 4.17 (0.05) 4.17 ( 0 .07) 4.10 ( 0 .06) 4.23 ( 0 .08) 4.02 ( 0 .06) F (2,318) = 0.64, p = .530, partial 2 < .01

PAGE 218

206 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values PACS 4.32 (0.06) 4.19 (0.05) 4.14 ( 0 .06) 4.12 ( 0 .06) 4.21 ( 0 .08) 4.04 ( 0 .05) F (2,321) = 0.76, p = .471, partial 2 = .01 SATAQ 4.30 (0.06) 4.19 (0.05) 4.15 ( 0 .06) 4.12 ( 0 .06) 4.22 ( 0 .08) 4.04 ( 0 .05) F (2,320) = 0.77, p = .466, part ial 2 < .01 Figure Q No covariates 4.03 (0.35) 3.97 (0.14) 3.98 ( 0 .18) 4.07 ( 0 .27) 4.09 ( 0 .33) a 3.95 ( 0 .33) b F (2,326) = 4.38, p = .013, partial 2 = .03 BMI 4.03 (0.04) 3.97 (0.03) 3.98 ( 0 .04) 4.07 ( 0 .04) 4.08 ( 0 .05) 3.95 ( 0 .03) F (2,324) = 4.05, p = .0 18, partial 2 = .02 EDI BD 4.03 (0.04) 3.98 (0.03) 3.97 ( 0.0 4) 4.07 ( 0 .04) 4.07 ( 0 .05) 3.95 ( 0 .03) F (2,325) = 3.97, p = .020, partial 2 = .02

PAGE 219

207 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS 4.02 (0.04) 3.98 (0.03) 3.97 (. 0 04) 4.08 ( 0 .04) 4.08 ( 0 .05) 3.96 ( 0 .04) F (2,321) = 4.50, p = .012, partial 2 = .03 PACS 4.04 (0.04) 3.99 (0.03) 3.97 ( 0 .04) 4.07 ( 0 .04) 4.07 ( 0 .05) a 3.94 ( 0 .03) b F (2,324) = 4.40, p = .013, partial 2 = .03 SATAQ 4.04 ( 0 .04) 3.96 ( 0 .03) 3.98 ( 0 .04) 4.07 ( 0 .04) 4.10 ( 0 .05) a 3.95 ( 0 .03) b F(2, 323) = 4.72, p = .010, partial 2 = 03 Figure R No covariates 3.50 (0.47) 3.32 (0.62) 3.53 ( 0 .44) 3.32 ( 0 .49) 3.36 ( 0 .63) 3.29 ( 0 .69) F (2 320) = 0.35, p = .706, partial 2 < .01 BMI 3.50 (0.08) 3.32 (0.07) 3.53 ( 0 .08) 3.33 ( 0 .08) 3.36 ( 0 .10) 3.29 ( 0 .07) F (2 318) = 0.35, p = .708, partial 2 < .01

PAGE 220

208 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values EDI BD 3.50 (0.08) 3.33 (0.07) 3.53 ( 0 .08) 3.33 ( 0 .08) 3.35 ( 0 .10) 3.29 ( 0 .07) F (2 319) = 0.38, p = .687, partial 2 < .01 DMS 3.54 (0.08) 3.29 (0.07) 3.60 ( 0 .09) 3.29 ( 0 .08) 3.39 ( 0 .10) 3.21 ( 0 .08) F (2 315) = 0.34, p = .713, partial 2 < .01 P ACS 3.51 (0.08) 3.35 (0.07) 3.51 ( 0 .08) 3.33 ( 0 .08) 3.32 ( 0 .10) 3.27 ( 0 .07) F (2 318) = 0.34, p = .709, partial 2 < .01 SATAQ 3.52 (0.08) 3.33 (0.07) 3.53 ( 0 .08) 3.33 ( 0 .08) 3.39 ( 0 .10) 3.29 ( 0 .07) F (2 317) = 0.22, p = .805, partial 2 < .01 Figure S No covariates 4.59 (0.68) 4.20 (0.42) 4.44 ( 0 .62) 4.38 ( 0 .50) 4.38 ( 0 .61) 4.22 ( 0 .48) F (2,317) = 1.37, p = .256, partial 2 = .01

PAGE 221

209 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values BMI 4.59 (0.08) 4.20 (0.07) 4.44 ( 0 .08) 4.38 ( 0 .08) 4.38 ( 0 .10) 4.23 ( 0 .07) F (2,315) = 1.36, p = .258, partial 2 = .01 EDI BD 4.59 (0.08) 4.31 (0.07) 4.44 ( 0 .08) 4.39 ( 0 .08) 4.36 ( 0 .10) 4.23 ( 0 .07) F (2,316) = 1.31, p = .272, partial 2 = .01 DMS 4.61 (0.08) 4.29 (0.07) 4.46 ( 0 .09) 4.37 ( 0 .08) 4.39 ( 0 .10) 4.23 ( 0 .07) F (2,312) = 1.30, p = .273, partial 2 = .01 PACS 4.60 (0.08) 4.33 (0.07) 4.42 ( 0 .08) 4.39 ( 0 .07) 4.34 ( 0 .10) 4.22 ( 0 .07) F (2,315) = 1.51, p = .222, partial 2 = .01 SATAQ 4.56 (0.08) 4.31 (0.07) 4.44 ( 0 .08) 4.39 ( 0 .07) 4.37 ( 0 .10) 4.22 ( 0 .07) F (2,314) = 0.88, p = .418, partial 2 = .01

PAGE 222

210 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure V No covariates 3.88 (0.55) 3.61 (0.49) 3.80 ( 0 .51) 3.77 ( 0 .58) 3.45 ( 0 .52) 3.40 ( 0 .78) F (2,323) = 1.46, p = .234, partial 2 = .01 BMI 3.87 (0.08) 3.62 (0.07) 3.79 ( 0 .08) 3.77 ( 0 .08) 3.43 ( 0 .11) 3.40 ( 0 .07) F (2,321) = 1.42, p = .243 partial 2 = .01 EDI BD 3.87 (0.08) 3.62 (0.07) 3.79 ( 0 .08) 3.77 ( 0 .08) 3.43 ( 0 .11) 3.40 ( 0 .07) F (2,322) = 1.44, p = .238, partial 2 = .01 DMS 3.92 (0.08) 3.58 (0.07) 3.86 ( 0 .09) 3.73 ( 0 .08) 3.48 ( 0 .10) 3.31 ( 0 .08) F (2,318) = 0.98, p = .376, partial 2 = .01 PACS 3.88 (0.08) 3.62 (0.07) 3.79 ( 0 .08) 3.77 ( 0 .08) 3.44 ( 0 .11) 3.38 ( 0 .07) F (2,321) = 1.44, p = .240, partial 2 = .01

PAGE 223

211 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values SATAQ 3.89 (0.08) 3.59 (0.07) 3.80 ( 0 .08) 3.77 ( 0 .08) 3.45 ( 0 .11) 3.41 ( 0 .07) F (2,320) = 1.80, p = .167, partial 2 = .01 Figure W No covariates 4.52 (0.49) 4.36 (0.3 9) 4.44 ( 0 .55) 4.34 ( 0 .43) 4.49 ( 0 .57) 4.27 ( 0 .46) F (2,317) = 0.39, p = .680, partial 2 < .01 BMI 4.52 (0.07) 4.37 (0.06) 4.44 ( 0 .07) 4.34 ( 0 .07) 4.48 ( 0 .08) 4.25 ( 0 .06) F (2,315) = 0.37, p = .690, partial 2 < .01 EDI BD 4.52 (0.07) 4.36 (0.06) 4.44 ( 0 07) 4.34 ( 0 .07) 4.48 ( 0 .09) 4.27 ( 0 .06) F (2,316) = 0.35, p = .706, partial 2 < .01 DMS 4.51 (0.07) 4.37 (0.06) 4.41 ( 0 .08) 4.35 ( 0 .07) 4.48 ( 0 .08) 4.29 ( 0 .06) F (2,312) = 0.52, p = .595, partial 2 < .01

PAGE 224

212 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values PACS 4.53 (0.07) 4.39 (0.06) 4.42 ( 0 .07) 4.34 ( 0 .0 7) 4.45 ( 0 .08) 4.26 ( 0 .06) F (2,315) = 0.40, p = .673, partial 2 < .01 SATAQ 4.52 (0.07) 4.37 (0.06) 4.44 ( 0 .07) 4.34 ( 0 .07) 4.49 ( 0 .09) 4.26 ( 0 .06) F (2,315) = 0.47, p = .626, partial 2 < .01 Figure Z No covariates 4.90 (0.53) 4.53 (0.68) 4.80 ( 0 .54) 4 .77 ( 0 .50) 4.50 ( 0 .75) 4.36 ( 0 .74) F (2,325) = 2.24, p = .108, partial 2 = .01 BMI 4.90 (0.09) 4.53 (0.08) 4.80 ( 0 .09) 4.77 ( 0 .09) 4.48 ( 0 .11) 4.36 ( 0 .08) F (2,323) = 2.27, p = .043, partial 2 = .02 EDI BD 4.91 (0.09) 4.52 (0.08) 4.81 ( 0 .09) 4.77 ( 0 .09) 4.52 ( 0 .11) 4.36 ( 0 .08) F (2,324) = 2.31, p = .101, partial 2 = .01

PAGE 225

213 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values DMS 4.91 (0.09) 4.52 (0.08) 4.80 ( 0 .10) 4.77 ( 0 .09) 4.50 ( 0 .11) 4.36 ( 0 .08) F (2,320) = 2.80, p = .100, partial 2 = .01 PACS 4.90 (0.09) 4.53 (0.08) 4.81 ( 0 .09) 4.77 ( 0 .09) 4.50 ( 0 .11) 4 .36 ( 0 .08) F (2,323) = 2.20, p = .112, partial 2 = .01 SATAQ 4.92 (0.09) 4.51 (0.08) 4.80 ( 0 .09) 4.77 ( 0 .09) 4.53 ( 0 .11) 4.38 ( 0 .08) F (2,322) = 2.61, p = .075, partial 2 = .02 Figure AA No covariates 5.88 (0.60) 5.83 (0.61) 5.67 ( 0 .63) 5.90 ( 0 .57) 5.59 ( 0 .69) 5.67 ( 0 .73) F (2,323) = 1.33, p = .265, partial 2 = .01 BMI 5.90 (0.09) 5.79 (0.08) 5.70 ( 0 .09) 5.88 ( 0 .09) 5.62 ( 0 .12) 5.68 ( 0 .08) F (2,321) = 1.56, p = .211, partial 2 = .01

PAGE 226

214 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values EDI BD 5.87 (0.09) 5.84 (0.08) 5.67 ( 0 .09) 5.91 ( 0 .09) 5.56 ( 0 .12) 5.6 7 ( 0 .08) F (2,322) = 1.26, p = .283, partial 2 = .01 DMS 5.88 (0.09) 5.83 (0.08) 5.68 ( 0 .10) 5.90 ( 0 .09) 5.59 ( 0 .11) 5.69 ( 0 .08) F (2,318) = 1.26, p = .284, partial 2 = .01 PACS 5.91 (0.08) 5.98 (0.06) 5.80 ( 0 .08) 5.89 ( 0 .06) 5.45 ( 0 .08) 5.51 ( 0 .06) F (2, 321) = 1.32, p = .270, partial 2 = .01 SATAQ 5.89 (0.09) 5.79 (0.08) 5.67 ( 0 .09) 5.90 ( 0 .09) 5.64 ( 0 .12) 5.70 ( 0 .08) F (2,320) = 1.76, p = .175, partial 2 = .01 Figure DD No covariates 5.82 (0.71) 5.60 (0.65) 5.75 ( 0 .61) 5.66 ( 0 .57) 5.49 ( 0 .57) 5.41 ( 0 .80) F (2,318) = 0.34, p = .713, partial 2 < .01 BMI 5.82 (0.09) 5.60 (0.08) 5.76 ( 0 .10) 5.66 ( 0 .09) 5.51 ( 0 .12) 5.41 ( 0 .08) F (2,316) = 0.31, p = .734, partial 2 < .01

PAGE 227

215 Rater R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values EDI BD 5.83 (0.09) 5.58 (0.08) 5.77 ( 0 .10) 5.65 ( 0 .09) 5.55 ( 0 .12) 5.41 ( 0 .08) F (2,31 7) = 0.34, p = .714, partial 2 < .01 DMS 5.82 (0.10) 5.60 (0.08) 5.77 ( 0 .11) 5.66 ( 0 .10) 5.50 ( 0 .12) 5.43 ( 0 .09) F (2,313) = 0.37, p = .690, partial 2 < .01 PACS 5.80 (0.09) 5.56 (0.08) 5.79 ( 0 .10) 5.66 ( 0 .09) 5.54 ( 0 .12) 5.43 ( 0 .08) F (2,316) = 0.29, p = .751, partial 2 < .01 SATAQ 5.83 (0.09) 5.55 (0.08) 5.75 ( 0 .09) 5.65 ( 0 .09) 5.55 ( 0 .12) 5.45 ( 0 .08) F (2,315) = 0.72, p = .489, partial 2 = .01 Table L.7. Between subjects effects: Main effects of rater gender Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure A No covariates 3.79 ( 0 .84) 3.89 ( 0 .50) F (1,322) = 1.50, p = 222, partial 2 = .01 BMI 3.84 ( 0 .06) 3.88 ( 0 .05) F (1,320) = 0.23, p = .634, partial 2 < .01

PAGE 228

216 Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values EDI BD 3.77 ( 0 .06) 3.92 ( 0 .05) F (1,3 21) = 3.63, p = .058, partial 2 = .01 DMS 3.81 ( 0 .06) 3.90 ( 0 .05) F (1,317) = 0.90, p = .344, partial 2 < .01 PACS 3.79 ( 0 .06) 3.91 ( 0 .05) F (1,320) = 2.48, p = .117, partial 2 = .01 SATAQ 3.82 ( 0 .06) 3.90 ( 0 .05) F (1,319) = 1.32, p = .251, partial 2 < .01 Figure B No covariates 3.50 ( 0 .78) a 3.74 ( 0 .58) b F (1,322) = 9.25, p = .003, partial 2 = .03 BMI 3.55 ( 0 .06) 3.71 ( 0 .05) F (1,320) = 4.26, p = .040, partial 2 = .01 EDI BD 3.48 ( 0 .06) a 3.75 ( 0 .05) b F (1,321) = 11.48, p = .001, partial 2 = .04 DMS 3.51 ( 0 .07) 3.73 ( 0 .06) F (1,317) = 5.49, p = .020, partial 2 = .02 PACS 3.50 ( 0 .06) a 3.74 ( 0 .05) b F (1,320) = 9.05, p = .003, partial 2 = .03 SATAQ 3.50 ( 0. 06) a 3.74 0 (05) b F (1,319) = 9.17, p = .003, partial 2 = .03 Figure C No covariates 2.24 ( 0 .94 ) a 2.51 ( 0 .91) b F (1,320) = 6.57, p = .011, partial 2 = .02 BMI 2.26 ( 0 .08) 2.50 ( 0 .07) F (1,318) = 4.60, p = .033, partial 2 = .01 EDI BD 2.23 ( 0 .09) a 2.51 ( 0 .07) b F (1,319) = 6.43, p = .012, partial 2 = .02 DMS 2.25 ( 0 .09) 2.49 ( 0 .07) F (1,315) = 3.68, p = .056, partial 2 = .01

PAGE 229

217 Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values PACS 2.24 ( 0 .08) a 2.50 ( 0 .07) b F (1,318) = 6.18, p = .013, partial 2 = .02 SATAQ 2.27 ( 0 .08) 2.50 ( 0 .07) F (1,317) = 4.82, p = .029, partial 2 = .02 Figure D No covariates 4.21 ( 0 .65) 4.11 ( 0 .43) F (1,322) = 2.59, p = .109, p artial 2 = .01 BMI 4.22 ( 0 .05) 4.11 ( 0 .04) F (1,320) = 2.98, p = .086, partial 2 = .01 EDI BD 4.19 ( 0 .05) 4.13 ( 0 .04) F (1,321) = 1.07, p = .303, partial 2 < .01 DMS 4.20 ( 0 .05) 4.12 ( 0 .04) F (1,317) = 1.17, p = .280, partial 2 < .01 PACS 4.20 ( 0 .05) 4.11 ( 0 .04) F (1,320) = 2.06, p = .152, partial 2 = .01 SATAQ 4.20 ( 0 .05) 4.11 ( 0 .04) F (1,319) = 2.14, p = .145, partial 2 = .01 Figure G No covariates 3.88 ( 0 .44) a 3.67 ( 0 .52) b F (1,323) = 10.65, p = .001, partial 2 = .03 BMI 3.85 ( 0 .04) a 3.68 ( 0 .04) b F (1,321) = 9.09, p = .003, partial 2 = .03 EDI BD 3.86 ( 0 .04) a 3.68 ( 0 .04) b F (1,322) = 9.51, p = .002, partial 2 = .03 DMS 3.85 ( 0 .05) a 3.69 ( 0 .04) b F (1,318) = 6.12, p = .014, partial 2 = .02 PACS 3.86 ( 0 .04) a 3.67 ( 0 .04) b F (1,321) = 11.56, p = .00 1, partial 2 = .04 SATAQ 3.87 ( 0 .04) a 3.67 ( 0 .04) b F (1,320) = 11.87, p = .001, partial 2 = .04

PAGE 230

218 Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure H No covariates 4.48 ( 0 .52) a 4.31 ( 0 .43) b F (1,323) = 8.44, p = .004, partial 2 = .03 BMI 4.48 ( 0 .04) a 4.30 ( 0 .04) b F (1,321) = 10.44, p = .001, part ial 2 = .03 EDI BD 4.46 ( 0 .04) a 4.31 ( 0 .04) b F(1,322) = 6.93, p = .009, partial 2 = .02 DMS 4.44 ( 0 .05) 4.33 ( 0 .04) F (1,318) = 3.02, p = .083, partial 2 = .01 PACS 4.47 ( 0 .04) a 4.31 ( 0 .04) b F (1,321) = 9.23, p = .003, partial 2 = .03 SATAQ 4.47 ( 0 .0 4) a 4.31 ( 0. 04) b F (1,320) = 9.33, p = .002, partial 2 = .03 Figure K No covariates 4.49 ( 0 .44) 4.41 ( 0 .43) F (1,320) = 1.81, p = .179, partial 2 = .01 BMI 4.48 ( 0 .04) 4.40 ( 0 .03) F (1,318) = 2.86, p = .092, partial 2 = .01 EDI BD 4.48 ( 0 .04) 4.40 ( 0 .0 3) F (1,319) = 2.07, p = .151, partial 2 = .01 DMS 4.49 ( 0 .04) 4.39 ( 0 .03) F (1,315) = 2.88, p = .090, partial 2 = .01 PACS 4.47 ( 0 .04) 4.40 ( 0 .03) F (1,318) = 2.09, p = .150, partial 2 = .01 SATAQ 4.49 ( 0 .04) 4.40 ( 0 .03) F (1,317) = 3.18, p = .075, part ial 2 = .01 Figure L No covariates 5.04 ( 0 .54) 4.94 ( 0 .55) F (1,321) = 1.71, p = .191, partial 2 = .01

PAGE 231

2 19 Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values BMI 5.03 ( 0 .05) 4.94 ( 0 .04) F (1,319) = 2.18, p = .141, partial 2 = .01 EDI BD 5.01 ( 0 .05) 4.95 ( 0 .04) F (1,320) = 1.02, p = .313, partial 2 < .01 DMS 5.02 ( 0 .05) 4.95 ( 0 .04) F (1,316) = 0.91, p = .340, partial 2 < .01 PACS 5.03 ( 0 .05) 4.94 ( 0 .04) F (1,319) = 2.21, p = .138, partial 2 = .01 SATAQ 5.03 ( 0 .05) 4.94 ( 0 .04) F (1,319) = 2.29, p = .131, partial 2 = .01 Figure O No covariates 5.32 ( 0 .64 ) 5.22 ( 0 .62) F (1,325) = 1.07, p = .303, partial 2 < .01 BMI 5.31 ( 0 .06) 5.22 ( 0 .05) F (1,323) = 1.54, p = .215, partial 2 = .01 EDI BD 5.30 ( 0 .06) 5.22 ( 0 .05) F (1,324) = 1.29, p = .257, partial 2 < .01 DMS 5.30 ( 0 .06) 5.23 ( 0 .05) F (1,320) = 0.74, p = .390, partial 2 < .01 PACS 5.31 ( 0 .05) 5.21 ( 0 .05) F (1,323) = 2.04, p = .154, partial 2 = .01 SATAQ 5.31 ( 0 .06) 5.22 ( 0 .05) F (1,322) = 1.81, p = .180, partial 2 = .01 Figure P No covariates 4.23 ( 0 .51) 4.11 ( 0 .37) F (1,323) = 5.34, p = .021, partial 2 = .02 BMI 4.23 ( 0 .04) 4.12 ( 0 .03) F (1,321) = 3.50, p = .035, partial 2 = .01 EDI BD 4.23 ( 0 .04) 4.11 ( 0 .03) F (1,322) = 5.15, p = .024, partial 2 = .02

PAGE 232

220 Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values DMS 4.25 ( 0 .04) a 4.10 ( 0 .04) b F (1,318) = 6.04, p = .014, partial 2 = .02 PACS 4.22 ( 0 .04) 4.12 ( 0 .03) F (1,321) = 4.53, p = .034, partial 2 = .01 SATAQ 4.22 ( 0 .04) 4.12 ( 0 .03) F (1,320) = 4.66, p = .032, partial 2 = .01 Figure Q No covariates 4.02 ( 0 .30) 3.99 ( 0 .26) F (1,326) = 1.34, p = .248, partial 2 < .01 BMI 4.03 ( 0 .02) 4.00 ( 0 .02) F (1,324 ) = 1.24, p = .266, partial 2 < .01 EDI BD 4.03 ( 0 .03) 4.00 ( 0 .02) F (1,325) = 0.57, p = .451, partial 2 < .01 DMS 4.02 ( 0 .03) 4.01 ( 0 .02) F (1,321) = 0.14, p = .710, partial 2 < .01 PACS 4.02 ( 0 .02) 4.00 ( 0 .02) F (1,324) = 0.58, p = .448, partial 2 < .01 SATAQ 4.04 ( 0 .02) 4.00 ( 0 .02) F (1,323) = 1.91, p = .167, partial 2 = .01 Figure R No covariates 3.47 ( 0 .51) 3.31 ( 0 .61) F (1,320) = 5.24, p = .023, partial 2 = .02 BMI 3.45 ( 0 .05) 3.32 ( 0 .04) F (1,318) = 4.19, p = .042, partial 2 = .01 EDI BD 3.4 6 ( 0 .05) 3.32 ( 0 .04) F (1,319) = 4.22, p = .041, partial 2 = .01 DMS 3.51 ( 0 .06) a 3.26 ( 0 .05) b F (1,315) = 10.56, p = .001, partial 2 = .03 PACS 3.45 ( 0 .05) 3.32 ( 0 .04) F (1,318) = 3.85, p = .051, partial 2 = .01

PAGE 233

221 Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values SATAQ 3.48 ( 0 .05) a 3.31 ( 0 .04) b F (1,317) = 6.13, p = .014, partial 2 = .02 Figure S No covariates 4.48 ( 0 .64) a 4.30 ( 0 .47) b F(1, 317) = 7.04 p = .008, partial 2 = .02 BMI 4.47 ( 0 .05) a 4.30 ( 0 .04) b F (1 315) = 6.60, p = .011, partial 2 = .02 EDI BD 4.46 ( 0 .05) 4.31 ( 0 .04) F (1 316) = 5.36, p = .021, partial 2 = .02 DMS 4.49 ( 0 .05) a 4.30 ( 0 .04) b F (1 312) = 5.48, p = .012, partial 2 = .02 PACS 4.45 ( 0 .05) 4.31 ( 0 .04) F (1 315) = 5.09, p = .025, partial 2 = .02 SATAQ 4.46 ( 0 .05) 4.30 ( 0 .04) F (1 314) = 5.97, p = .015, partial 2 = .02 Figure V No covariates 3.75 ( 0 .55) 3.58 ( 0 .64) F (1 323) = 3.17, p = .076, partial 2 = .01 BMI 3.70 ( 0 .05) 3.59 ( 0 .04) F (1 321) = 2.21, p = .138, partial 2 = .01 EDI BD 3.70 ( 0 .05) 3.60 ( 0 .04) F (1 322) = 2.10, p = .148, partial 2 = .01 DMS 3.76 ( 0 .06) a 3.5 4 ( 0 .05) b F(1, 318) = 7.24, p = .008, partial 2 = .02 PACS 3.70 ( 0 .05) 3.59 ( 0 .04) F (1 321) = 2.80, p = .095, partial 2 = .01 SATAQ 3.71 ( 0 .05) 3.59 ( 0 .04) F (1 320) = 3.23, p = .073, partial 2 = .01

PAGE 234

222 Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure W No covariates 4.48 ( 0 .53) a 4.32 ( 0 .43) b F ( 1 317) = 8.87, p = .003, partial 2 = .03 BMI 4.47 ( 0 .04) a 4.33 ( 0 .04) b F(1 315) = 7.22, p = .008, partial 2 = .02 EDI BD 4.48 ( 0 .04) a 4.32 ( 0 .04) b F(1, 316) = 7.38, p = .007, partial 2 = .02 DMS 4.47 ( 0 .05) 433 ( 0 .04) F (1,312) = 4.02, p = .046, partia l 2 = .01 PAC 4.47 ( 0 .04) a 4.33 ( 0 .04) b F (1,315) = 6.39, p = .012, partial 2 = .02 SATAQ 4.84 ( 0 .04) a 4.23 ( 0 .04) b F (1,315) = 8.39, p = .004, partial 2 = .03 Figure Z No covariates 4.77 ( 0 .61) a 4.54 ( 0 .67) b F (1,325) = 6.29, p = .013, partial 2 = .0 2 BMI 4.73 ( 0 .06) 4.56 ( 0 .05) F (1,323) = 5.53, p = .019, partial 2 = .02 EDI BD 4.75 ( 0 .06) a 4.55 ( 0 .05) b F(1, 324) = 6.91, p = .009, partial 2 = .02 DMS 4.74 ( 0 .06) 4.55 ( 0 .05) F (1,320) = 4.53, p = .034, partial 2 = .01 PACS 4.74 ( 0 .06) a 4.55 ( 0 .05) b F (1,323) = 6.34, p = .012, partial 2 = .02 SATAQ 4.75 ( 0 .06) a 4.55 ( 0 .05) b F(1,322) = 7.43, p = .007, partial 2 = .02 Figure AA No covariates 5.73 ( 0 .64) 5.79 ( 0 .65) F (1,323) = 1.36, p = .245, partial 2 < .01

PAGE 235

223 Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values BMI 5.74 ( 0 .06) 5.78 ( 0 .05) F (1,321) = 0.34, p = .562, partial 2 < .01 EDI BD 5.70 ( 0 .06) 5.81 ( 0 .05) F (1,322) = 1.92, p = .166, partial 2 = .01 DMS 5.71 ( 0 .06) 5.81 ( 0 .05) F (1,318) = 1.12, p = .290, partial 2 < .01 PACS 5.72 ( 0 .06) 5.79 ( 0 .05) F (1,321) = 1.02, p = .313, partial 2 < .01 SATAQ 5.73 ( 0 .06) 5.79 ( 0 .05) F (1,320) = 0.72, p = .398, partial 2 < .01 Figure DD No covariates 5.72 ( 0 .65) 5.55 ( 0 .69) F (1,318) = 2.88, p = .091, partial 2 = .01 BMI 5.70 ( 0 .06) 5.56 ( 0 .05) F (1,316) = 3.03, p = .083, partial 2 = .01 EDI BD 5.71 ( 0 .06) 5.54 ( 0 .05) F (1,317) = 4.27, p = .040, partial 2 = .01 DMS 5.69 ( 0 .07) 5.56 ( 0 .06) F (1,313) = 1.95, p = .163, partial 2 = .01 PACS 5.71 ( 0 .06) 5.55 ( 0 .05) F (1,316) = 4.27, p = .040, partial 2 = .01 SATAQ 5.71 ( 0 .06) 5.55 ( 0 .05) F (1,315) = 4.05 p = .045, partial 2 = .01

PAGE 236

224 Table L.8. Between subjects effects: Main effects of rater race Rater R ace Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure A No covariates 3.86 ( 0 .65) 3.75 ( 0 .74) 3.93 ( 0 .58) F (2,322) = 2.04, p = .132, partial 2 = .01 BMI 3.85 ( 0 .06) 3.76 ( 0 .06) 3.97 ( 0 .07) F (2,320) = 2.58, p = .077, partial 2 = .02 EDI BD 3.87 ( 0 .06) 3.76 ( 0 .06) 3.91 ( 0 .07) F (2,321) = 1.46, p = .234, partial 2 = .01 DMS 3.8 7 ( 0 .06) 3.75 ( 0 .07) 3.95 ( 0 .07) F (2,317) = 2.12, p = .122, partial 2 = .01 PACS 3.89 ( 0 .06) 3.74 ( 0 .07) 3.92 ( 0 .07) F (2,320) = 2.08, p = .126, partial 2 = .01 SATAQ 3.87 ( 0 .06) 3.76 ( 0 .06) 3.95 ( 0 .07) F (2,319) = 2.15, p = .118, partial 2 = .01 Figur e B No covariates 3.63 ( 0 .64) 3.60 ( 0 .73) 3.68 ( 0 .68) F (2,322) = 0.19, p = .824, partial 2 < .01 BMI 3.60 ( 0 .06) 3.60 ( 0 .07) 3.70 ( 0 .07) F (2,320) = 0.66, p = .520, partial 2 < .01 EDI BD 3.62 ( 0 .06) 3.60 ( 0 .07) 3.63 ( 0 .07) F (2,321) = 0.07, p = .933, p artial 2 < .01 DMS 3.61 ( 0 .06) 3.60 ( 0 .07) 3.64 ( 0 .07) F (2,317) = 0.09, p = .914, partial 2 < .01 PACS 3.62 ( 0 .06) 3.59 ( 0 .07) 3.65 ( 0 .07) F (2,320) = 0.16, p = .849, partial 2 < .01 SATAQ 3.61 ( 0 .06) 3.60 ( 0 .07) 3.65 ( 0 .08) F (2,319) = 0.24, p = .840, partial 2 < .01

PAGE 237

225 Rater R ace Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure C No covariates 2.48 ( 0 .90) 2.33 ( 0 .92) 2.36 ( 0 .97) F (2,320) = 0.83, p = .436, partial 2 = .01 BMI 2.46 ( 0 .08) 2.32 ( 0 .09) 2.36 ( 0 .10) F (2,318) = 0.62, p = .537, partial 2 < .01 EDI BD 2.47 ( 0 .08) 2.32 ( 0 .09) 2.32 ( 0 .10) F (2, 319) = 0.88, p = .415, partial 2 = .01 DMS 2.46 ( 0 .08) 2.33 ( 0 .09) 2.31 ( 0. 10) F (2,315) = 0.88, p = .417, partial 2 = .01 PACS 2.47 ( 0 .09) 2.32 ( 0 .09) 2.32 ( 0 .10) F (2,318) = 0.88, p = .416, partial 2 = .01 SATAQ 2.44 ( 0 .08) 2.31 ( 0 .09) 2.40 ( 0 .10) F ( 2,317) = 0.51, p = .604, partial 2 < .01 Figure D No covariates 4.18 ( 0 .57) 4.15 ( 0 .59) 4.13 ( 0 .42) F (2,322) = 0.28, p = .755, partial 2 < .01 BMI 4.19 ( 0 .05) 4.15 ( 0 .05) 4.15 ( 0 .16) F (2,320) = 0.20, p = .817, partial 2 < .01 EDI BD 4.20 ( 0 .05) 4.15 ( 0 .05) 4.12 ( 0 .06) F (2,321) = 0.50, p = .609, partial 2 < .01 DMS 4.19 ( 0 .05) 4.14 ( 0 .06) 4.15 ( 0 .06) F (2,318) = 2.32, p = .109, partial 2 = .01 PACS 4.20 ( 0 .05) 4.15 ( 0 .05) 4.13 ( 0 .06) F (2,321) = 2.17, p = .116, partial 2 = .01 SATAQ 4.19 ( 0 .05) 4. 15 ( 0 .05) 4.14 ( 0 .06) F (2,319) = 0.24, p = .783, partial 2 < .01 Figure G No covariates 3.80 ( 0 .46) 3.81 ( 0 .50) 3.66 ( 0 .53) F (2,323) = 2.57, p = .078, partial 2 = .02

PAGE 238

226 Rater R ace Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values BMI 3.82 ( 0 .04) 3.81 ( 0 .05) 3.67 ( 0 .05) F (2,321) = 2.60, p = .076, partial 2 = .02 EDI BD 3.81 ( 0 .04) 3.81 ( 0 .05) 3.68 ( 0 .05) F (2,322) = 2.39, p = .094, partial 2 = .02 DMS 3.81 ( 0 .04) 3.81 ( 0 .05) 3.68 ( 0 .05) F (2,315) = 2.02, p = .134, partial 2 = .01 PACS 3.81 ( 0 .04) 3.82 ( 0 .05) 3.68 ( 0 .05) F (2,318) = 2.01, p = .136, partial 2 = 01 SATAQ 3.69 ( 0 .06) 3.81 ( 0 .05) 3.69 ( 0 .06) F (2,320) = 1.59, p = .206, partial 2 = .01 Figure H No covariates 4.46 ( 0 .47) 4.27 ( 0 .50) 4.30 ( 0 .45) F (2,323) = 2.97, p = .053, partial 2 = .02 BMI 4.47 ( 0 .04) 4.38 ( 0 .05) 4.33 ( 0 .05) F (2,321) = 2.20, p = .112, partial 2 = .01 EDI BD 4.47 ( 0 .04) 4.38 ( 0 .05) 4.31 ( 0 .05) F (2,322) = 3.01, p = .051, partial 2 = .02 DMS 4.47 ( 0 .04) 4.37 ( 0 .05) 4.32 ( 0 .05) F (2,318) = 2.66, p = .072, partial 2 = .02 PACS 4.46 ( 0 .04) 4.38 ( 0 .05) 4.33 ( 0 .05) F (2,321) = 2.15, p = .118, partial 2 = .01 SATAQ 4.46 ( 0 .04) 4.37 ( 0 .05) 4.34 ( 0 .05) F (2,320) =1.99, p = .139, partial 2 = .01 Figure K No covariates 4.54 ( 0 .44) a 4.44 ( 0 .44) 4.33 ( 0 .39) b F (2,320) = 6.59, p = .002, partial 2 = .04 BMI 4.54 ( 0 .04) a 4.44 ( 0 .04) 4.34 ( 0 .05) b F (2,318) = 5.61, p = .004, partial 2 = .03 EDI BD 4.55 ( 0 .04) a 4.44 ( 0 .04) 4.34 ( 0 .05) b F (2,319) = 5.90, p = .003, partial 2 = .04

PAGE 239

227 Rater R ace Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values DMS 4.55 ( 0. 04) a 4.45 ( 0 .04) 4.33 ( 0 .05) b F (2,315) = 7.06, p = .001, partial 2 = .04 PACS 4.55 ( 0 .04) a 4.44 ( 0 .0 4) 4.33 ( 0 .05) b F (2,318) = 6.17, p = .002, partial 2 = .04 SATAQ 4.54 ( 0 .04) 4.44 ( 0 .04) 4.37 ( 0 .05) F (2,317) = 3.76, p = .024, partial 2 = .02 Figure L No covariates 5.06 ( 0 .56) a 5.05 ( 0 .51) a 4.81 ( 0 .55) b F (2,321) = 5.42, p = .005, partial 2 = .04 BMI 5.06 ( 0 .05) a 5.05 ( 0 .05) a 4.84 ( 0 .06) b F(2,319) = 4.87, p = .008, partial 2 = .03 EDI BD 5.07 ( 0 .05) a 5.05 ( 0 .05) a 4.83 ( 0 .06) b F (2,320) = 5.66, p = .004, partial 2 = .03 DMS 5.07 ( 0 .05) a 5.05 ( 0 .05) a 4.84 ( 0 .06) b F(2,316) = 4.90, p = .008, partial 2 = .03 PACS 5.05 ( 0 .05) a 5.06 ( 0 .05) a 4.85 ( 0 .06) b F (2,319) = 4.44, p = .013, partial 2 = .03 SATAQ 5.05 ( 0 .05) 5.05 ( 0 .05) 4.86 ( 0. 06) F (2,319) = 3.57, p = .029, partial 2 = .02 Figure O No covariates 5.34 ( 0 .57) a 5.34 ( 0 .60) a 5.08 ( 0 .68) b F (2,32 5) = 5.53, p = .004, partial 2 = .03 BMI 5.35 ( 0 .06) a 5.34 ( 0 .06) a 5.10 ( 0 .07) b F (2,323) = 4.58, p = .011, partial 2 = .03 EDI BD 5.35 ( 0 .06) a 5.34 ( 0 .06) a 5.09 ( 0 .07) b F(2 324) = 5.63, p = .008, partial 2 = .03 DMS 5.35 ( 0 .06) a 5.34 ( 0 .06) a 5.09 ( 0 09) b F(2 320) = 5.24, p = .006, partial 2 = .03 PACS 5.33 ( 0 .06) 5.35 ( 0 .06) 5.11 ( 0 .07) F (2 323) = 4.63, p = .018, partial 2 = .03

PAGE 240

228 Rater R ace Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values SATAQ 5.33 ( 0 .06) 5.33 ( 0 .06) 5.13 ( 0 .07) F (2 322) = 3.51, p = .047, partial 2 = .02 Figure P No covariates 4.24 ( 0 .4 7) 4.13 ( 0 .37) 4.10 ( 0 .45) F (2,323) = 2.79, p = .063, partial 2 = .02 BMI 4.25 ( 0 .04) 4.13 ( 0 .04) 4.13 ( 0 .05) F (2,321) = 2.81, p = .062, partial 2 = .02 EDI BD 4.25 ( 0 .04) 4.13 ( 0 .04) 4.14 ( 0 .05) F (2,322) = 2.64, p = .024, partial 2 = .02 DMS 4.25 ( 0 .04) 4.14 ( 0 .04) 4.13 ( 0 .05) F (2,318) = 2.89, p = .057, partial 2 = .02 PACS 4.26 ( 0 .04) 4.13 ( 0 .04) 4.13 ( 0 .05) F (2,321) = 3.04, p = .049, partial 2 = .02 SATAQ 4.24 ( 0 .04) 4.13 ( 0 .04) 4.13 ( 0 .05) F (2,320) = 2.30, p = .102, partial 2 = .01 Figure Q No covariates 4.00 ( 0 .26) 4.03 ( 0 .24) 3.99 ( 0 .33) F (2,326) = 0.21, p = .808, partial 2 < .01 BMI 4.00 ( 0 .02) 4.03 ( 0 .03) 4.01 ( 0 .03) F (2,324) = 0.22, p = .802, partial 2 < .01 EDI BD 4.00 ( 0 .02) 4.03 ( 0 .03) 4.01 ( 0 .03) F (2,325) = 0.19, p = .831, parti al 2 < .01 DMS 4.00 ( 0 .02) 4.02 ( 0 .03) 4.02 ( 0 .03) F (2,321) = 0.23, p = .798, partial 2 < .01 PACS 4.01 ( 0 .03) 4.02 ( 0 .03) 4.00 ( 0 .03) F (2,324) = 0.09, p = .919, partial 2 < .01 SATAQ 4.00 ( 0 .03) 4.02 ( 0 .03) 4.03 ( 0 .03) F (2,323) = 0.26, p = .772, par tial 2 < .01

PAGE 241

229 Rater R ace Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure R No covariates 3.40 ( 0 .57) 3.42 ( 0 .47) 3.31 ( 0 .67) F (2,320) = 0.92, p = .401, partial 2 = .01 BMI 3.41 ( 0 .05) 3.43 ( 0 .06) 3.32 ( 0 .06) F (2,318) = 1.06, p = .348, partial 2 = .01 EDI BD 3.41 ( 0 .05) 3.43 ( 0 .06) 3.32 ( 0 .06) F (2,319) = 0.97, p = .381, partial 2 = .01 DMS 3.42 ( 0 .05) 3.45 ( 0 .06) 3.30 ( 0 .06) F (2,315) = 1.77, p = .172, partial 2 = .01 PACS 3.43 ( 0 .05) 3.42 ( 0 .06) 3.30 ( 0 .06) F (2,318) = 1.50, p = .225, partial 2 = .01 SATAQ 3.42 ( 0 .05) 3.43 ( 0 .06) 3.34 ( 0 .06) F (2,31 7) = 0.70, p = .498, partial 2 < .01 Figure S No covariates 4.42 ( 0 .56) 4.41 ( 0 .55) 4.27 ( 0 .53) F (2,317) = 1.83, p = .162, partial 2 = .01 BMI 4.44 ( 0 .05) 4.41 ( 0 .05) 4.30 ( 0 .06) F (2,315) = 1.70, p = .184, partial 2 = .01 EDI BD 4.45 ( 0 .05) 4.41 ( 0 05) 4.29 ( 0 .06) F (2,316) = 2.00, p = .137, partial 2 = .01 DMS 4.45 ( 0 .05) 4.41 ( 0 .06) 4.31 ( 0 .06) F (2,312) = 1.65, p = .193, partial 2 = .01 PACS 4.46 ( 0 .05) 4.40 ( 0 .05) 4.28 ( 0 .06) F (2,315) = 2.78, p = .064, partial 2 = .02 SATAQ 4.44 ( 0 .05) 4.41 ( 0 .05) 4.29 ( 0 .06) F (2,314) = 1.66, p = .193, partial 2 = .01 Figure V No covariates 3.72 ( 0 .53) a 3.78 ( 0 .54) a 3.42 ( 0 .71) b F (2,323) = 10.32, p < .001 partial 2 = .06

PAGE 242

230 Rater R ace Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values BMI 3.75 ( 0 .05) a 3.78 ( 0 .06) a 3.41 ( 0 .07) b F (2,321) = 10.81, p < .001 partial 2 = .06 EDI BD 3.75 ( 0 .05) a 3.78 ( 0 .06) a 3.42 ( 0 .07) b F (2,322) = 10.56, p < .001 partial 2 = .06 DMS 3.75 ( 0 .05) a 3.80 ( 0 .06) a 3.40 ( 0 .06) b F (2,318) = 13.26, p < .001 partial 2 = .07 PACS 3.75 ( 0 .05) a 3.78 ( 0 .06) a 3.41 ( 0 .07) b F (2,321) = 10.83, p < .001 partial 2 = .06 SATAQ 3.74 ( 0 .05) a 3.78 ( 0 .06) a 3.43 ( 0 .07) b F (2,320) = 9.07, p < .001 partial 2 = .05 Figure W No covariates 4.43 ( 0 .44) 4.39 ( 0 .49) 4.34 ( 0 .51) F (2,317) = 0.52, p = .597, partial 2 < .01 BMI 4.44 ( 0 .04) 4.39 ( 0 .05) 4.37 ( 0 .05) F (2,315) = 0.60, p = .549, partial 2 < .01 EDI BD 4.44 ( 0 .04) 4.39 ( 0 .05) 4.37 ( 0 .05) F (2,316) = 0.56, p = .574, partial 2 < .01 DMS 4.44 ( 0 .04) 4.38 ( 0 .05) 4.38 ( 0 .05) F (2,312) = 0.55, p = .577, partial 2 < .01 PACS 4.46 ( 0 .04) 4.38 ( 0 .05) 4.35 ( 0 .05 ) F (2,315) = 1.34, p = .264, partial 2 = .01 SATAQ 4.45 ( 0 .04) 4.39 ( 0 .05) 4.37 ( 0 .05) F (2,315) = 0.61, p = .542, partial 2 < .01 Figure Z No covariates 4.69 ( 0 .64) a 4.79 ( 0 .52) a 4.41 ( 0 .74) b F (2,325) = 8.67, p < .001 partial 2 = .05 BMI 4.72 ( 0 .06 ) a 4.79 ( 0 .06) a 4.42 ( 0 .07) b F (2,323) = 10.52, p < .00 1 partial 2 = .05 EDI BD 4.71 ( 0 .06) a 4.79 ( 0 .06) a 4.44 ( 0 .07) b F (2,324) = 9.09, p = .001, partial 2 = .05

PAGE 243

231 Rater R ace Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values DMS 4.72 ( 0 .06) a 4.78 ( 0 .06) a 4.43 ( 0 .07) b F (2,320) = 8.15 p < .001 partial 2 = .05 PAC S 4.71 ( 0 .06) a 4.79 ( 0 .06) a 4.43 ( 0 .07) b F (2,323) = 8.37, p < .00 1 partial 2 = .05 SATAQ 4.71 ( 0 .06) a 4.79 ( 0 .06) a 4.45 ( 0 .07) b F (2,322) = 6.72, p = .001, partial 2 = .04 Figure AA No covariates 5.85 ( 0 .61) 5.79 ( 0 .61) 5.64 ( 0 .71) F (2,323) = 3.22, p = .041, partial 2 = .02 BMI 5.84 ( 0 .06) 5.79 ( 0 .06) 5.65 ( 0 .07) F (2,321) = 2.75, p = .106, partial 2 = .01 EDI BD 5.86 ( 0 .06) 5.79 ( 0 .06) 5.62 ( 0 .07) F (2,322) = 3.56, p = .029, partial 2 = .02 DMS 5.85 ( 0 .06) 5.79 ( 0 .06) 5.64 ( 0 .07) F (2,318) = 3.56, p = .054, partial 2 = .02 PACS 5.85 ( 0 .06) 5.79 ( 0 .06) 5.63 ( 0 .07) F (2,321) = 2.87 p = .058, partial 2 = .02 SATAQ 5.84 ( 0 .06) 5.78 ( 0 .06) 5.67 ( 0 .07) F (2,320) = 1.66, p = .192, partial 2 = .01 Figure DD No covariates 5.69 ( 0 .68) a 5.71 ( 0 .59) a 5.4 4 ( 0 .73) b F (2,318) = 4.49, p = .012, partial 2 = .03 BMI 5.71 ( 0 .06) 5.71 ( 0 .07) 5.46 ( 0 .07) F (2,316) = 3.99, p = .020, partial 2 = .03 EDI BD 5.70 ( 0 .06) 5.71 ( 0 .07) 5.48 ( 0 .07) F (2,317) = 3.31, p = .038, partial 2 = .02 DMS 5.71 ( 0 .06) 5.71 ( 0 .07) 5.46 ( 0 .07) F (2,313) = 4.10, p = .018, partial 2 = .03 PACS 5.68 ( 0 .06) 5.72 ( 0 .07) 5.49 ( 0 .07) F (2,316) = 3.15, p = .044, partial 2 = .02

PAGE 244

232 Rater R ace Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values SATAQ 5.69 ( 0 .06) 5.70 ( 0 .07) 5.50 ( 0 .08) F (2,315) = 2.37, p = .095, partial 2 = .02 Table L.9. Between subject s effects: Covariate effects Covariate F p and partial 2 values Figure A BMI F(1, 320) = 7.22, p = .008, partial 2 = .02 EDI BD F (1, 321) = 4.62, p = .032, partial 2 = .01 DMS F (1,317) = 0.07, p = .793, partial 2 < .01 PACS F (1,320) = 3.26, p = 076, partial 2 = .01 SATAQ F (1,319) = 1.00, p = .318, partial 2 < .01 Figure B BMI F (1,320) = 10.73, p = .001, partial 2 = .03 EDI BD F (1,321) = 2.43, p = .120, partial 2 = .01 DMS F (1,317) = 0.001, p = .976, partial 2 < .01 PACS F (1,320) = 0.06 p = .808, partial 2 < .01 SATAQ F (1,319) = 0.07, p = .796, partial 2 < .01

PAGE 245

233 Covariate F p and partial 2 values Figure C BMI F (1,320) = 0.36, p = .551, partial 2 < .01 EDI BD F (1,319) = 0.14, p = .706, partial 2 < .01 DMS F (1,315) = 0.00, p = 1.00, partial 2 < .01 PACS F (1,318) = 0.03, p = .865, partial 2 < .01 SATAQ F (1,317) = 3.20, p = .075, partial 2 = .01 Figure D BMI F (1,320) = 0.36, p = .551, partial 2 < .01 EDI BD F (1,321) = 1.98, p = .161, partial 2 = .01 DMS F (1,317) = 0.09, p = .770, partial 2 < .01 PACS F (1,3 20) = 0.47, p = .492, partial 2 < .01 SATAQ F (1,319) = 0.22, p = .643, partial 2 < .01 Figure G BMI F (1,321) = 0.27, p = .607, partial 2 < .01 EDI BD F (1,322) = 0.01, p = .934, partial 2 < .01 DMS F (1,318) = 0.09, p = .771, partial 2 < .01 PACS F (1,321) = 0.78, p = .378, partial 2 < .01 SATAQ F (1,320) = 1.64, p = .202, partial 2 = .01

PAGE 246

234 Covariate F p and partial 2 values Figure H BMI F (1,321) = 2.03, p = .155, partial 2 = .01 EDI BD F (1,322) = 0.10, p = .747, partial 2 < .01 DMS F (1,318) = 1.51, p = .221, partial 2 = .01 PACS F (1,321) = 1.03, p = .312, partial 2 < .01 SATAQ F (1,320) = 0.97, p = .326, partial 2 < .01 Figure K BMI F (1,318) = 2.29, p = .131, partial 2 = .01 EDI BD F (1,319) = 0.27, p = .606, partial 2 < .01 DMS F (1,315) = 1.11, p = .293, partial 2 < .01 PACS F (1,318) = 0.01, p = .785, partial 2 < .01 SATAQ F (1,317) = 3.36, p = .068, partial 2 = .01 Figure L BMI F (1,319) = 0.87, p = .351, partial 2 < .01 EDI BD F (1,320) = 0.49, p = .484, partial 2 < .01 DMS F (1,316) = 0.001, p = .973, partial 2 < .01 PACS F (1,319) = 1.49, p = .223, partial 2 = .01 SATAQ F (1,319) = 3.34, p = .068, partial 2 = .01

PAGE 247

235 Covariate F p and partial 2 values Figure O BMI F (1,323) = 0.28, p = .594, partial 2 < .01 EDI BD F (1,324) = 0.24, p = .625, partial 2 < .01 DMS F (1,320) = 0.01, p = .937, pa rtial 2 < .01 PACS F (1,323) = 4.61, p = .032, partial 2 = .01 SATAQ F (1,322) = 5.06, p = .025, partial 2 = .02 Figure P BMI F (1,321) = 0.27, p = .601, partial 2 < .01 EDI BD F (1,322) = 0.08, p = .783, partial 2 < .01 DMS F (1,318) = 1.06, p = .30 5, partial 2 < .01 PACS F (1,321) = 0.55, p = .458, partial 2 < .01 SATAQ F (1,320) = 0.06, p = .809, partial 2 < .01 Figure Q BMI F (1,324) = 0.10, p = .749, partial 2 < .01 EDI BD F (1,325) = 1.01, p = .317, partial 2 < .01 DMS F (1,321) = 1.36, p = .244, partial 2 < .01 PACS F (1,324) = 4.55, p = .034, partial 2 = .01 SATAQ F (1,323) = 0.93 p = .336, partial 2 < .01

PAGE 248

236 Covariate F p and partial 2 values Figure R BMI F (1,318) = 0.19, p = .661, partial 2 < .01 EDI BD F (1,319) = 0.11, p = .741, partial 2 < .01 DMS F (1,315) = 4.42 p = .036, partial 2 = .01 PACS F (1,318) = 3.30, p = .070, partial 2 = .01 SATAQ F (1,317) = 0.03, p = .860, partial 2 < .01 Figure S BMI F (1,315) = 0.01, p = .936, partial 2 < .01 EDI BD F (1,316) = 0.37, p = .543, partial 2 < .01 DMS F (1,312) = 0.80, p = .371, partial 2 < .01 PACS F (1,315) = 4.07, p = .044, partial 2 = .01 SATAQ F (1,314) = 0.51, p = .478, partial 2 < .01 Figure V BMI F (1,321) = 0.27, p = .603, partial 2 < .01 EDI BD F (1,322) = 0.53, p = .468, partial 2 < .01 DMS F (1,3 18) = 3.36, p = .068, partial 2 = .01 PACS F (1,321) = 0.53, p = .468, partial 2 < .01 SATAQ F (1,320) = 0.70, p = .403, partial 2 < .01

PAGE 249

237 Covariate F p and partial 2 values Figure W BMI F (1,315) = 0.22, p = .643, partial 2 < .01 EDI BD F (1,316) = 0.09, p = .764, partial 2 < .01 DMS F (1,312) = 0.30, p = .583, partial 2 < .01 PACS F (1,315) = 5.14, p = .024, partial 2 = .02 SATAQ F (1,315) = 0.68, p = .409, partial 2 < .01 Figure Z BMI F (1,323) = 0.00, p = .984, partial 2 < .01 EDI BD F (1,324) = 0.64, p = .425, partial 2 < .01 DMS F (1,320) = 0.06, p = .812, partial 2 < .01 PACS F (1,323) = 0.05, p = .830, partial 2 < .01 SATAQ F (1,322) = 0.93, p = .335, partial 2 < .01 Figure AA BMI F (1,321) = 4.17, p = .042, partial 2 = .01 EDI BD F (1,322) = 0.79, p = .376, partial 2 < .01 DMS F (1,318) = 0.03, p = .870, partial 2 < .01 PACS F (1,321) = 0.29, p = .592, partial 2 < .01 SATAQ F (1,320) = 3.01, p = .084, partial 2 = .01

PAGE 250

238 Covariate F p and partial 2 values Figure DD BMI F (1,316) = 0.03, p = .859, partial 2 < .01 EDI BD F (1,317) = 1.93, p = .166, parti al 2 = .01 DMS F (1,313) = 0.004, p = .948, partial 2 < .01 PACS F (1,316) = 5.39, p = .021, partial 2 = .02 SATAQ F (1,315) = 5.90, p = .016, partial 2 = .02

PAGE 251

239 Appendix M: Significant F p and 2 values with Means and Standard Deviations for Health A nalyses General Notes: 1. Findings in bold are significant at the p < .006 level. Italicized findings represent a trend towards significance (.006 < p < .01). Significant pairwise differences are indicated for values up to and including p = .014. 2. For analyse s without covariates, raw means are presented with standard deviations. Adjusted means are presented for all ANCOVAs with adjusted standard errors. 3. Superscripts denote means that differ significantly from each other. Subscripts denote means that differ si gnificantly from each other. Table M.1. Repeated measures effects: Significant two way interactions (target race X covariate) Covariate F p and partial 2 values Figure D EDI BD F(2 632) = 4.70, p = .010, partial 2 = .01 Figure R SATAQ F(2, 634) = 4. 81, p = .008, partial 2 = .02 Figure AA SATAQ F (2 633) = 4.35, p = .014, partial 2 = .01

PAGE 252

240 Covariate F p and partial 2 values Figure DD BMI F (2,628) = 4.39, p = .013, partial 2 = .01 Table M.2. Repeated measures effects: Significant two way interactions (target race X rater gender) T arget R ace with Rater Gender Caucasian Hispanic African American Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure D EDI BD 5.42 ( 0 .11) a 5.11 ( 0 .09) b 5.25 ( 0 10) 5.36 ( 0 .09) 5.57 ( 0 .11) 5.36 ( 0 .09) F (2, 632) = 5.33, p = .005, partial 2 = .02 PACS 5.41 (0.11) a 5.12 (0.09) b 5.29 ( 0 .10) 5.35 ( 0 .08) 5.62 ( 0 .11) 5.34 ( 0 .09) F (2,633) = 4.51, p = .012, partial 2 = .01

PAGE 253

241 Table M.3. Repeated measures effects: Signifi cant two way interactions (target race X rater race) Target Race with Rater Race Caucasian Hispanic African American Covariate Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values Figure Q No covariates 5.91 ( 0 .98) a 5.62 (1.13) b 5.29 (1.13) b 5.86 (1.07) c 6.05 (1.06) c 5.27 (1.03) d 5.90 ( 0 .99) e 5.94 ( 0 .96) e 5.49 (1.09) f F(2,629) = 3.56, p = .008, partial 2 = .02 BMI 5.92 ( 0 .10) a 5.62 ( 0 .11) a 5.35 ( 0 .12) b 5.85 ( 0 .10) c 6.05 ( 0 .10) c 5.33 ( 0 .12) d 5.88 ( 0 .10) e 5.95 ( 0 .10) e 5.54 ( 0. 11) f F(2,626) = 3.59, p = .007, partial 2 = .02 EDI BD 5.91 ( 0 .10) a 5.61 ( 0 .11) b 5.39 ( 0 .12) b 5.86 ( 0 .10 ) c 6.05 ( 0 .10) c 5.33 ( 0 .10) d 5.88 ( 0 .09) e 5.94 ( 0 .10) e 5.58 ( 0 .11) f F (2,627) = 3.70, p = .006, partial 2 = .02 DMS 5.90 ( 0 .10) a 5.61 ( 0 .11) b 5.36 ( 0 .12) b 5.84 ( 0. 09) c 6.08 ( 0 .10) c 5.35 ( 0 .10) d 5.86 ( 0 .09) e 5.92 ( 0 .10) f 5.58 ( 0 .11) f F (2,618) = 3.98, p = 004, partial 2 = .02 PACS 5.88 ( 0 .10) a 5.64 ( 0 .11) 5.41 ( 0 .12) b 5.83 ( 0 .10) c 6.06 ( 0 .10) c 5.38 ( 0 .12) d 5.84 ( 0 .09) 5.97 ( 0 .10) e 5.62 ( 0 .11) f F(2,626) = 3.56, p = .008, partial 2 = .02

PAGE 254

242 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) F p and partial 2 values SATAQ 5.89 ( 0 .10) a 5.61 ( 0 .11) 5.42 ( 0 .12) b 5.88 ( 0 .10) c 6.05 ( 0 .10) c 5.33 ( 0 .12) d 5.86 ( 0 .09) e 5.94 ( 0 .10) f 5.57 ( 0 .11) f F (2,620) = 3.69, p = .006, partial 2 = .02 Figure V No covariates 4.76 (1.04) a 4.61 (1.06) a 3.91 (1.04) b 4.49 ( 0 .95) c 4.36 (1.05) 4.10 (1.17) d 4.59 ( 0 .98) e 4.37 (1.02) e 3.80 (1.18) f F(2,636) = 3.46, p = .009, partial 2 = .02 BMI 4.79 ( 0 .09) a 4.60 ( 0 .10) a 3.91 ( 0 .12) b 4.50 ( 0 .09) c 4.36 ( 0 .10) 4.20 ( 0 .12) d 4.61 ( 0 .09) e 4.36 ( 0 .10) e 3.78 ( 0 .12) f F (2,632) = 3.85, p = .004, partial 2 = .02 DMS 4.78 ( 0 .10) a 4.60 ( 0 .11) a 3.30 ( 0 .12) b 4.51 ( 0 .09) c 4.40 ( 0 .10) 4.21 ( 0 .11) d 4.60 ( 0 .09) e 4.39 ( 0 .11) e 3.83 ( 0 .11) f F (2,626) = 3.28 p = .012, partial 2 = .02 PACS 4.76 ( 0 .10) a 4.61 ( 0 .10) a 3.95 ( 0 .12) b 4.50 ( 0 .10)c 4.35 ( 0 .10) 4.19 ( 0 .12) d 4.61 ( 0 .10) e 4.36 ( 0 .10) e 3.80 ( 0 .12) f F (2,632) = 3.21, p = .013, partia l 2 = .02

PAGE 255

243 Table M.4. Repeated measure s effects: Significant main effects of target race Target Race Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure C No covariates 3.09 (1. 62) a 3.51 (1.64) b 2.99 (1.59) a F (2, 635) = 19.11, p < .001 partial 2 = .06 SATAQ 3.08 ( 0 .09) a 3.50 ( 0 .10) b 3.01 ( 0 .09) a F(2,629) = 4.85, p = .008, partial 2 = .02 Figure D No covariates 5.23 (1.26) a 5.30 (1.16) a 5.44 (1.26) b F (2,637) = 5.77, p = .003, partial 2 = .02 EDI BD 5.26 ( 0 .07) a 5.31 ( 0 .07) a 5.47 ( 0 .07) b F (2,632) = 5.67, p = .004, partial 2 = .02 Figure H No covariates 4.89 (1.20) a 5.11 (1.04) b 5.19 (1.03) b F (2,625) = 10.88, p < .001 partial 2 = .03 Figure L No covariates 4.12 (1.30) a 4.53 (1.38) b 4.08 (1.23) a F (2,632) = 19.37, p < .001 partial 2 = .06 Figure O No covariates 3.00 (1.06) a 3.20 (1.09) b 3.15 (1.14) b F(2,641) = 4.86, p = .008, partial 2 = .02 Figure P No covariates 5.93 (1.07) a 5.74 (1.21) b 5.94 (1.13) a F (2,606) = 7. 70, p = .001, partial 2 = .02

PAGE 256

244 Target Race Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure R No covariates 4.65 (1.24) a 4.53 (1.22) 4.42 (1.16) b F (2,640) = 4.33, p = .014, partial 2 = .01 SATAQ 4.65 ( 0 .07) a 4.53 ( 0 .07) 4.43 ( 0 .07) b F (2,634) = 5.32, p = .005, partial 2 = .02 Figure W No covariates 5.66 (1.15) a 5.34 (1.05) b 5.62 (1.18) a F (2,613) = 13.23, p < .001 partial 2 = .04 Figure Z No covariates 3.57 (0.99) a 3.44 (0.98) a 3.70 (1.09) b F (2,642) = 8.38, p < .001 partial 2 = .03 Figure AA No covariates 2.64 (1.37) a 2.83 (1.42) b 3.28 (1.41) c F (2 ,637) = 28.21, p < .001 partial 2 = .08 SATAQ 2.65 ( 0 .08) a 2.85 ( 0 .08) b 3.32 ( 0 .08) c F (2,633) = 5.37, p = .005, partial 2 = .02 Figure DD BMI 2.74 (1.20) 2.73 (1.14) 2.73 (1.31) F (2,628) = 4.48, p = .012, partial 2 = .01

PAGE 257

245 Table M.5. Between subjec ts effects: Significant main effects of rater gender Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure L No covariates 4.43 ( 0 .97) a 4.11 ( 0 .99) b F (1,321) = 10.14, p = .002, partial 2 = .03 BMI 4.41 ( 0 .09) a 4.12 ( 0 .07) b F (1,319) = 6.26, p = .013, partial 2 = .02 EDI BD 4.47 ( 0 .09) a 4.09 ( 0 .07) b F (1,320) = 10.52, p = .001, partial 2 < .03 PACS 4.46 ( 0 .09) a 4.09 ( 0 .07) b F (1,319) = 9.94, p = .002, partial 2 = .03 SATAQ 4.44 ( 0 .09) a 4.10 ( 0 .07) b F (1,3 19) = 8.97, p = .003, partial 2 = .03 Figure Q PACS 5.85 ( 0 .07) a 5.62 ( 0 .16) b F (1,324) = 6.46, p = .012, partial 2 = .02 Figure R No covariates 4.74 ( 0 .93) a 4.40 ( 0 .97) b F (1,320) = 8.37, p = .004, partial 2 = .03 BMI 4.69 ( 0 .08) a 4.40 ( 0 .07) b F(1,3 18) = 6.83, p = .009, partial 2 = .02 DMS 4.68 ( 0 .08) a 4.41 ( 0 .07) b F (1,315) = 10.71, p = .001, partial 2 = .03 PACS 4.76 ( 0 .09) a 4.34 ( 0 .08) b F(1,318) = 7.27, p = .007, partial 2 = .02 SATAQ 4.69 ( 0 .08) a 4.40 ( 0 .07) b F(1, 317) = 6.94, p = .009, parti al 2 = .02

PAGE 258

246 Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure V DMS 4.50 ( 0 .08) a 4.22 ( 0 .06) b F(1, 318) =6.66, p = .010, partial 2 = .02 Figure AA No covariates 3.11 (1.09) a 2.78 (1.05) b F (1,323) = 8.25, p = .004, partial 2 = .03 EDI BD 3.14 ( 0 .10) a 2.77 ( 0. 08) b F (1,322) = 8.50, p = .004, par tial 2 = .03 PACS 3.13 ( 0 .10) a 2.78 ( 0 .08) b F (1,321) = 7.70, p = .006, partial 2 = .02 SATAQ 3.10 ( 0 .10) a 2.78 ( 0 .08) b F(1, 320) = 6.78, p = .010, partial 2 = .02 Table M.6. Between subjects effects: Significant main effects of rater race Rater Ra ce Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure G No covariates 4.84 ( 0 .85) a 4.76 ( 0 .90) a 4.40 ( 0 .99) b F(2, 324) = 4.72, p = .010, partial 2 = .03 BMI 4.85 ( 0 .08) a 4.76 ( 0 .90) a 4.47 ( 0 .10) b F (2,322) = 4.43, p = .013, partial 2 = .03 DMS 4.85 ( 0 .08) a 4.77 ( 0 .09) a 4.47 ( 0 .10) b F (2,319) = 4.60, p = .011, partial 2 = .03 SATAQ 4.86 ( 0 .08) a 4.76 ( 0 .09) a 4.44 ( 0 .10) b F(2 321) = 5.05, p = .007, partial 2 = .03

PAGE 259

247 Rater Ra ce Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure Q No covari ates 5.89 ( 0 .77) a 5.86 ( 0 .81) a 5.35 ( 0 .86) b F (2,326) = 10.29, p < .001 partial 2 = .06 BMI 5.88 ( 0 .07) a 5.87 ( 0 .08) a 5.40 ( 0 .09) b F (2,324) = 10.73, p < .001 partial 2 = .06 EDI BD 5.88 ( 0 .07) a 5.87 ( 0 .08) a 5.43 ( 0 .09) b F (2,325) = 9.38, p < .001 part ial 2 = .06 DMS 5.87 ( 0 .07) a 5.87 ( 0 .08) a 5.43 ( 0 .09) b F (2,321) = 9.64, p < .001 partial 2 = .06 PACS 5.85 ( 0 .07) a 5.89 ( 0 .08) a 5.47 ( 0 .09) b F (2,324) = 7.56, p = .001, partial 2 = .05 SATAQ 5.88 ( 0 .07) a 5.87 ( 0 .08) a 5.44 ( 0 .09) b F (2,323) = 8.24, p < .001 partial 2 = .05 Figure R No covariates 4.76 ( 0 .92) a 4.66 ( 0 .92) a 4.14 ( 0 .96) b F (2,320) = 11.58, p < .001 partial 2 = .07 BMI 4.79 ( 0 .08) a 4.66 ( 0 .09) a 4.17 ( 0 .10) b F (2,318) = 11.70, p < .001 partial 2 = .07 EDI BD 4.80 ( 0 .08) a 4.66 ( 0 .09) a 4.17 ( 0 .10) b F (2,319) = 11.94, p < .001 partial 2 = .07 DMS 4.79 ( 0 .08) a 4.69 ( 0 .09) a 4.17 ( 0 .10) b F (2,315) = 12.06, p < .001 partial 2 = .07 PACS 4.80 ( 0 .09) a 4.66 ( 0 .09) a 4.17 ( 0 .10) b F (2,318) = 11.94, p < .001 partial 2 = .07 SATAQ 4.80 ( 0 .08) a 4.66 ( 0 .09) a 4.15 ( 0 .10) b F (2,317) = 12.15, p < .00 1 partial 2 = .07 Figure V No covariates 4.61 ( 0 .71) a 4.44 ( 0 .79) a 5.62 (1.16) b F (2,323) = 17.53, p < .001 partial 2 = .10 BMI 4.63 ( 0 .07) a 4.34 ( 0 .08) a 3.96 ( 0 .09) b F (2,321) = 18.54, p < .001 par tial 2 = .10

PAGE 260

248 Rater Ra ce Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values EDI BD 4.62 ( 0 .07) a 4.44 ( 0 .08) a 4.00 ( 0 .09) b F (2,322) = 15.97, p < .00 1 partial 2 = .09 DMS 4.63 ( 0 .07) a 4.47 ( 0 .08) a 3.99 ( 0 .08) b F (2,318) = 17.78, p < .001 partial 2 = .10 PACS 4.63 ( 0 .07) a 4.44 ( 0 .08) a 3.98 ( 0 .09) b F (2,321) = 16.79 p < .001 partial 2 = .10 SATAQ 4.62 ( 0 .07) a 4.44 ( 0 .08) a 3.96 ( 0 .09) b F (2,320) = 17.09, p < .001 partial 2 = .10 Table M.7. Between subjects effects: Significant covariate effects Covariate F p and partial 2 values Figure A DMS F (1, 318) = 9. 92, p = .002, partial 2 = .03 Figure O SATAQ F (1,322) = 11.83, p = .001, partial 2 = .04 Figure Q PACS F(1, 324) = 7.29, p = .007, partial 2 = .02 Figure S DMS F (1,312) = 7.97, p = .005, partial 2 = .03 Figure AA BMI F (1,321) = 7.64, p = .006, p artial 2 = .02

PAGE 261

249 Appendix N: Significant F p and 2 values with Means and Standard Deviations for Attractiveness Analyses General Notes: 1. Findings in bold are significant at the p < .006 level. Italicized findings represent a trend towards significance (.006 < p < .01). Significant pairwise differences are indicated for values up to and including p = .014. 2. For analyses without covariates, raw means are presented with standard deviations. Adjusted means are presented for all ANCOVAs with adjusted standard errors. 3. Superscripts denote means that differ significantly from each other. Subscripts denote means that differ significantly from each other. Table N.1. Repeated measures effects: Significant two way interactions (target race X covariate) Covariate F p and partial 2 values Figure G SATAQ F(2, 607) = 4.85, p = .009, partial 2 = .02

PAGE 262

250 Table N.2. Repeated measures effects: Significant two way interactions (target race X rater gender) Target R ace with Rater Gender Caucasian Hispanic African Americ an Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values Figure G EDI BD 4.63 ( 0 .13) a 4.15 ( 0 .10) b 4.53 ( 0 .13) c 4.09 ( 0 .10) d 4.25 ( 0 .13) 4.26 ( 0 .11) F (2,610) = 5.37, p = .0 06, partial 2 = .02 PACS 4.65 ( 0 .12) a 4.15 ( 0 .10) b 4.51 ( 0 .12) c 4.10 ( 0 .10) d 4.28 ( 0 .10) 4.24 ( 0 .11) F (2,609) = 4.64, p = .011, partial 2 = .01 SATAQ 4.70 ( 0 .12) a 4.15 ( 0 .10) b 4.51 ( 0 .12) c 4.10 ( 0 .10) d 4.28 ( 0 .13) 4.25 ( 0 .11) F (2,607) = 5.36, p = .006, partial 2 = .02 Figure H EDI BD 3.18 ( 0 .12) 3.22 ( 0 .10) 3.24 ( 0 .12) 3.30 ( 0 .10) 3.13 ( 0 .12) a 3.60 ( 0 .10) b F (2,641) = 4.45, p = .012, partial 2 = .01

PAGE 263

251 Target R ace with Rater Gender Caucasian Hispanic African Americ an Covariate Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) Male M (SD/SE) Female M (SD/SE) F p and partial 2 values SATAQ 3.21 ( 0 .15) 3.14 ( 0 .13) 3.08 ( 0 .15) 3.42 ( 0 .15) 3.23 ( 0 .19) 3.59 ( 0 .14) F(2 638) = 5.04, p = .0 07, partial 2 = .02 Figure Q SATAQ 4.84 ( 0 .13) 4.94 ( 0 .11) 4.76 ( 0 .13) a 5.17 ( 0 .11) b 4.68 ( 0 .13) c 5.26 ( 0 .10) d F (2,631) = 4.52, p = .012, partial 2 = .01

PAGE 264

252 Table N.3. Repeated measures effects: Significant two way interactions (target race X rater rac e) Target Race with Rater Race Caucasian Hispanic African American Covariate Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) Af rican American M (SD/SE) F p and partial 2 values Figure C No covariates 2.86 (1.70) a 2.43 (1.44) b 2.31 (1.43) b 3.24 (1.66) c 2.84 (1.61) 2.60 (1.53) d 2.46 (1.44) 2.36 (1.43) 2.40 (1.42) F(4,633) = 3.51, p = .008, partial 2 = .02 BMI 2.85 ( 0 .14) a 2.4 1 ( 0 .15) b 2.41 ( 0 .17) b 3.22 ( 0 .14) c 2.84 ( 0 .16) 2.68 ( 0 .17) d 2.43 ( 0 .13) 2.34 ( 0 .14) 2.52 ( 0 .15) F(4,629) = 3.31, p = .011, partial 2 = .02 EDI BD 2.89 ( 0 .14) a 2.41 ( 0 .15) b 2.30 ( 0 .17) b 3.26 ( 0 .15) c 2.84 ( 0 .16) 2.58 ( 0 .18) d 2.47 ( 0 .13) 2.33 ( 0 .14) 2.41 ( 0 .16) F (4,631) = 3.19, p = .013, partial 2 = .02 DMS 2.86 ( 0 .14) a 2.39 ( 0 .16) b 2.31 ( 0 .17) b 3.24 ( 0 .15) c 2.82 ( 0 .16) 2.58 ( 0 .17) d 2.46 ( 0 .13) 2.36 ( 0 .15) 2.42 ( 0 .15) F(4,620) = 3.39, p = .010, partial 2 = .02

PAGE 265

253 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) Af rican American M (SD/SE) F p and partial 2 values Figure G No covariates 4.64 (1.37) a 4.39 ( 1.33) 4.01 (1.51) b 4.30 (1.35) 4.51 (1.35) c 3.94 (1.44) d 4.33 (1.37) 4.17 (1.55) 4.20 (1.39) F (4,612) = 3.28, p = .013, partial 2 = .02 BMI 4.64 ( 0 .13) a 4.40 ( 0 .14) 4.11 ( 0 .15) b 4.33 ( 0 .12) 4.51 ( 0 .13) c 4.08 ( 0 .15) d 4.34 ( 0 .13) 4.17 ( 0 .14) 4.28 ( 0 .16) F ( 4,607) = 3.30, p = .012, partial 2 = .02 Figure H No covariates 3.32 (1.36) 3.06 (1.33) 3.19 (1.40) 3.09 (1.29) 3.29 (1.23) 3.40 (1.38) 3.13 (1.24) a 3.43 (1.34) 3.69 (1.45) b F (4,643) = 3.65, p = .006, partial 2 = .02 BMI 3.34 ( 0 .12) 3.05 ( 0 .14) 3.18 ( 0 .15) 3.13 ( 0 .12) 3.28 ( 0 .13) 3.39 ( 0 .14) 3.15 ( 0 .12) a 3.42 ( 0 .13) 3.53 ( 0 .14) b F(4, 639) =3.45, p = .008, partial 2 = .02 EDI BD 3.32 ( 0 .12) 3.05 ( 0 .14) 3.23 ( 0 .15) 3.10 ( 0 12) 3.28 ( 0 .13) 3.42 ( 0 .14) 3.12 (0 .12) a 3.43 ( 0 .13) 3.55 ( 0 .15) b F(4, 641) = 3.38, p = .010, partial 2 = .02

PAGE 266

254 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) Af rican American M (SD/SE) F p and partial 2 values DMS 3.32 ( 0 .12) 3.04 ( 0 .14) 3.17 ( 0 .15) 3.10 ( 0 .12) 3.24 ( 0 .13) 3.40 ( 0 .14) 3.11 ( 0 .12) a 3.41 ( 0 .13) 3.54 ( 0 .14) b F(4,633) = 3.57, p = .007, partial 2 = .02 PACS 3.34 ( 0 .13) 3.05 ( 0 .14) 3.16 ( 0 .15) 3.11 ( 0 .12) 3.28 ( 0 .13) 3.3 7 ( 0 .14) 3.12 ( 0 .12) a 3.43 ( 0 .13) 3.54 ( 0 .14) b F(4,639) = 3.72, p = .005, partial 2 = .02 SATAQ 3.31 ( 0 .13) 3.05 ( 0 .14 ) 3.24 ( 0. 15) 3.08 ( 0 .12) a 3.27 ( 0 .13) 3.48 ( 0 .14) b 3.14 ( 0 .12) 3.43 ( 0 .13) 3.51 ( 0 .15) F (4,638) = 3.18, p = .013, partial 2 = .02 Fi gure Q DMS 5.12 ( 0 .14) a 4.71 ( 0 .15) b 4.62 ( 0 .16) b 4.90 ( 0 .13) 5.17 ( 0 .15) 4.66 ( 0 .16) c 5.02 ( 0 .13) d 4.96 ( 0 .15) 4.78 ( 0 .16) F (2,626) = 3.35, p = .011, partial 2 = .02 Figure R No covariates 4.10 (1.44) a 3.91 (1.46) a 3.30 (1.21) b 3.71 (1.40) c 3.87 (1.44 ) c 3.27 (1.39) d 3.61 (1.39) 3.51 (1.30) 3.49 (1.39) F (2,637) = 4.68, p = .001, partial 2 = .03

PAGE 267

255 Target Race with Rater Race Caucasian Hispanic African American Covariate Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) African American M (SD/SE) Caucasian M (SD/SE) Hispanic M (SD/SE) Af rican American M (SD/SE) F p and partial 2 values BMI 4.12 ( 0 .13) a 3.91 ( 0 .14) a 3.27 ( 0 .15) b 3.71 ( 0 .13) c 3.86 ( 0. 14) d 3.25 ( 0 .16) d 3.63 ( 0 .12) 3.52 ( 0 .14) 3.52 ( 0 .15) F (2,632) = 5.14, p < .001 partial 2 = 03 EDI BD 4.12 ( 0 .13) a 3.91 ( 0 .14) a 3.28 ( 0 .15) b 3.72 ( 0 .13) c 3.86 ( 0 .14) c 3.23 ( 0 .15) d 3.64 ( 0 .12) 3.52 ( 0 .14) 3.47 ( 0 .15) F (2,635) = 4.43, p = .002, partial 2 = .03 DMS 4.12 ( 0 .13) a 3.93 ( 0 .14) a 3.28 ( 0 .15) b 3.71 ( 0 .14) c 3.90 ( 0 .14) c 3.26 ( 0 .15) d 3.64 ( 0 .12) 3.57 ( 0 .14) 3.47 ( 0 .15) F (2,628) = 4.36, p = .002, partial 2 = .03 PACS 4.13 ( 0 .13) a 3.91 ( 0 .14) a 3.27 ( 0 .15) b 3.72 ( 0 .13) c 3.86 ( 0 .14) c 3.25 ( 0 .16) d 3.65 ( 0 .13) 3.51 ( 0 .14) 3.48 ( 0 .15) F (2,633) = 4.51, p = .001, partial 2 = .03 SATAQ 4.13 ( 0 .1 3) a 3.91 ( 0 .14) a 3.34 ( 0 .15) b 3.73 ( 0 .13) 3.86 ( 0 .14) c 3.32 ( 0 .16) d 3.63 ( 0 .12) 3.52 ( 0 .12) 3.55 ( 0 .15) F (2,631) = 4.39, p = .002, partial 2 = .03

PAGE 268

256 Table N.4. Repeated measure s effects: Significant main effects of target race Target Race Ca ucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure A No covariates 2.73 (1.41) a 2.60 (1.43) b 2.86 (1.53) a F (2, 644) = 5.67, p = .004, partial 2 = .02 Figure C No covariates 2.56 (1.56) a 2.92 ( 1.62) b 2.41 (1.43) a F (2,633) = 20.85, p < .001 partial 2 = .06 Figure D No covariates 2.83 (1.40) a 3.09 (1.43) b 3.00 (1.42) b F (2,633) = 6.24, p = .002, partial 2 = .02 Figure K No covariates 3.68 (1.37) a 4.13 (1.44) b 3.93 (1.36) c F (2,590) = 15.21, p < .001 partial 2 = .05 SATAQ 3.72 ( 0 .08) a 4.14 ( 0 .08) b 3.94 ( 0 .08) c F (2,589) = 7.12, p = .001, partial 2 = .02 Figure L No covariates 2.59 (1.18) a 2.72 (1.26) a 3.03 (1.31) b F (2,634) = 19.32, p < .001 partial 2 = .06 Figure O No covariates 2.63 ( 1.19) a 2.78 (1.24) b 2.67 (1.24) F (2,646) = 17.45, p < .001 partial 2 = .05

PAGE 269

257 Target Race Ca ucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure P No covariates 4.73 (1.66) a 4.68 (1.61) a 4.93 (1.64) b F(2,610) = 4.74, p = .010, partial 2 = .01 Figure R No covariates 3.80 (1.42) a 3.63 (1.43) b 3.54 (1.36) b F (2,637 ) = 4.51, p = .011, partial 2 = .01 Figure V No covariates 3.74 (1.37) a 3.27 (1.33) b 3.48 (1.28) c F (2,634) = 16.88, p < .001 partial 2 = .05 Figure AA No covariates 2.15 (1.08) a 2.33 (1.27) b 2.64 (1.26) c F (2,643) = 26.10, p < .001 partial 2 = .08

PAGE 270

258 Table N.5. Between subjects effects: Significant main effects of rater gender Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure B EDI BD 2.68 ( 0 .11) a 3.04 (.09) b F (1, 321) = 6.30, p = .013, partial 2 = .02 DMS 2.59 ( 0 .12) a 3.12 ( 0 .10) b F (1,317) = 9.54, p = .002, partial 2 = .03 Figure D DMS 2.71 ( 0 .11) a 3.18 ( 0 .09) b F (1,318) = 8.82, p = .003, partial 2 = .03 Figure L No covariates 2.56 (1.00) a 2.93 ( 0 .94) b F (1,320) = 10.87, p = .001, partial 2 = .03 BMI 2.54 ( 0 .09) a 2.98 ( 0 .07) b F (1,318) = 16.04, p < .001 partial 2 = .05 EDI BD 2.60 ( 0 .09) a 2.93 ( 0 .07) b F (1,319) = 8.01, p = .005, partial 2 = .03 DMS 2.56 ( 0 .10) a 2.95 ( 0 .08) b F (1,315) = 8.45, p = .004, partial 2 = .03 PACS 2.57 ( 0 .09) a 2 .94 ( 0 .07) b F (1,318) = 11.46, p = .001, partial 2 = .04 SATAQ 2.58 (.09) a 2.95 ( 0 .07) b F (1,318) = 11.05, p = .001, partial 2 = .03 Figure O No covariates 2.45 (1.01) a 2.87 (1.00) b F (1,325) = 12.58, p < .001 partial 2 = .04

PAGE 271

259 Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values BMI 2.43 ( 0 .09) a 2.92 ( 0 07) b F (1,323) = 18.74, p < .001 partial 2 = .06 EDI BD 2.52 ( 0 .09) a 2.86 ( 0 .07) b F(1,324) = 8.41, p = .004, partial 2 = .03 DMS 2.47 ( 0 .10) a 2.89 ( 0 .08) b F (1 320) = 10.03, p = .002, partial 2 = .03 PACS 2.45 ( 0 .09) a 2.90 ( 0 .07) b F (1,323) = 15.65, p < .001 partial 2 = .05 SATAQ 2.47 ( 0 .09) a 2.89 ( 0 .07) b F (1,322) = 14.13, p < .001 partial 2 = .04 Figure P DMS 4.41 ( 0 .14) a 5.06 ( 0 .11) b F (1,318) = 11.42, p = .001, partial 2 = .04 Figure Q No covariates 4.66 (1.56) a 5.15 ( 0 .99) b F (1,326) = 12.92 p < .001 partial 2 = .04 BMI 4.65 ( 0 .11) a 5.14 ( 0 .09) b F (1,324) = 10.69, p = .001, partial 2 = .03 EDI BD 4.65 ( 0 .11) a 5.15 ( 0 .09) b F (1,325) = 10.80, p = .001, partial 2 = .03 DMS 4.51 ( 0 .12) a 5.25 ( 0 .10) b F (1,321) = 17.81, p < .001 partial 2 = .05 PACS 4.70 ( 0 .11) a 5.12 ( 0 .09) b F (1,324) = 8.79, p = .003, partial 2 = .03 SATAQ 4.76 ( 0 .11) a 5.13 ( 0 .09) b F(1, 323) = 7.01, p = .009, partial 2 = .02 Figure S DMS 4.02 ( 0 .14) a 4.65 ( 0 .11) b F (1,312) = 10.26, p = .001, partial 2 = .03

PAGE 272

260 Rater G ender Male Female Covariate Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure W DM S 3.89 ( 0 .12) a 4.35 ( 0 .10) b F(1,313) = 7.43, p = .007, partial 2 = .02 Table N.6. Between subjects effects: Significant main effects of rater race Rater R ace Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure A SATAQ 2.68 ( 0 .11) a 2.50 ( 0 .12) a 3.05 ( 0 .13) b F(2, 320) = 4.68, p = .010, partial 2 = .03 Figure O No covariates 2.51 ( 0 .87) a 2.57 ( 0 .98) a 3.06 (1.15) b F (2,325) = 7.10, p = .001, partial 2 = .04 BMI 2.53 ( 0 .09) a 2.55 ( 0 .10) a 2.94 ( 0 .11) b F (2,323) = 5.22, p = .006, partial 2 = .03 EDI BD 2.49 ( 0 .09) a 2.56 ( 0 .10) a 3.02 ( 0 .11) b F(2,324) = 7.91, p < .001 partial 2 = .05 DMS 2.49 ( 0 .09) a 2.55 ( 0 .10) a 2.99 ( 0 .11) b F(2,320) = 7.20, p = .001, partial 2 = .04 PACS 2.54 ( 0 09) a 2.54 ( 0 .10) a 2.94 ( 0 .11) b F (2,323) = 5.20, p = .006, partial 2 = .03 SATAQ 2.54 ( 0 .09) a 2.57 ( 0 .10) a 2.93 ( 0 .11) b F (2,322) = 4.33, p = .014, partial 2 = .03

PAGE 273

261 Rater R ace Caucasian Hispanic African American Covariate Mean (SD/SE) Mean (SD/SE) Mean (SD/SE) F p and partial 2 values Figure R No covariates 3.81 (1.17) a 3.76 (1.16) a 3.35 (1.08) b F(2,320) = 4.91, p = .008, partial 2 = .03 BMI 3.82 ( 0 .10) a 3.76 ( 0 .11) a 3.35 ( 0 .13) b F(2,318) = 4.70, p = .010, partial 2 = .03 EDI BD 3.83 ( 0 .10) a 3.76 ( 0 .11) a 3.33 ( 0 .13) b F (2,319) = 5.21, p = .006, partial 2 = .03 DMS 3.82 ( 0 .10) a 3.80 ( 0 .12) a 3.34 ( 0 .12) b F (2,315) = 5.27 p = .006, partial 2 = .03 PACS 3.84 ( 0 .11) a 3.76 ( 0 .11) b 3.33 ( 0 .13) c F(2,318) = 5.08, p = .007, partial 2 = .03 Figure V No covariates 3.74 (1.08) a 3.51 (1.01) a 3.13 (1.10) b F (2,323) = 8.23, p < .001 partial 2 = .05 BMI 3.75 ( 0 .10) a 3.51 ( 0 .11) a 3.14 ( 0 .12) b F (2,321) = 7.98, p < .001 partial 2 = .05 EDI BD 3.74 ( 0 .10) a 3.51 ( 0 .11) a 3.15 ( 0 .12) b F (2,322) = 7.45, p = .001, partial 2 = .04 DMS 3.75 ( 0 .10) a 3.53 ( 0 .11) a 3.13 ( 0 .12) b F (2,318) = 8.16, p < .001 partial 2 = .05 PACS 3.73 ( 0 .10) a 3.52 ( 0 .11) a 3.15 ( 0 .12) b F (2,321) = 6.97, p = .001, partial 2 = .04 SATAQ 3.72 ( 0 .10) a 3.51 ( 0 .10) 3.21 ( 0 .12) b F (2,320) = 5.47, p = .005, partial 2 = .03

PAGE 274

262 Table N.7. Between subjects effects: Significant covariate effects Covariate F p and partial 2 values Figure B DMS F (1, 317) = 6.23, p = .013, partial 2 = .02 Figure C BMI F (1,318) = 7.89, p = .005, partial 2 = .02 SATAQ F(1,317) = 7.02, p = .008, partial 2 = .21 Figure L BMI F (1,318) = 12.59, p < .001 partial 2 = .04 Figure O BMI F (1,323) = 11.76, p = .001, partial 2 = .04 Figure P DMS F (1,318) = 13.42, p < .001 partial 2 = .04 PACS F (1,321) = 6.43, p = .012, partial 2 = .02 SATAQ F (1,320) = 16.53, p < .001 partial 2 = .05 Figure Q PACS F (1,324) = 11.03, p = .001, partia l 2 = .03 SATAQ F (1,323) = 18.80 p < .001 partial 2 = .06

PAGE 275

263 Covariate F p and partial 2 values Figure S DMS F (1,312) = 15.45, p < .001 partial 2 = .05 SATAQ F (1,314) = 12.21, p = .001, partial 2 = .04 Figure W SATAQ F (1,316) = 11.37, p = .001, partial 2 = .04 Figure AA BMI F (1,321) = 16.33, p < .001 partial 2 = .05

PAGE 276

264 Appendix O: Additional Analyses Results Health Analyses: Three Way Interactions For the health ratings, no figures displayed a three way interaction. Health Analyses: Covariate X Target Race Interactions Ther e were no significant two way interactions with target race and the covariates, although several figures showed a trend in this direction. For figure D, the EDI BD showed a trend towards an interaction with target race ( F (2, 632) = 4.70, p = .010, partial 2 = .01). For figure R, the SATAQ showed this same trend ( F (2, 634) = 4.81, p = .008, partial 2 = .02). Health Analyses: Target Race X Rater Gender Interactions One figure, figure D, showed a two way interaction between target race and rater gender, but on ly when the EDI BD was entered as a covariate ( F (2, 632) = 5.33, p = .005, partial 2 = .02). Post hoc testing revealed that when the target was Caucasian, male raters ( adjusted M = 5.42, SE = 0 .11 ) provided a higher health rating than did female raters ( ad justed M = 5.11, SE = 0 .09 ). Health Analyses: Target Race X Rater Race Interactions A two way interaction between target race and rater race was found for two male figures. For figure Q, this interaction effect showed a trend towards significance with no c ovariates in the model ( F (2,629) = 3.56, p = .008, partial 2 = .02). Post hoc examination of the rater race means at each level of target race revealed that when the target was Caucasian, the rating given by Caucasian raters ( M = 5.91, SD = 0 .98 ) was sign ificantly higher than the mean ratings given by both Hispanic raters ( M = 5.62, SD = 1.13 ) and African American raters ( M = 5.29, SD = 1.13 ). When the target was Hispanic, the mean ratings given by both Caucasian raters ( M = 5.86, SD = 1.07 ) and Hispanic r aters ( M = 6.05, SD = 1.06 ) were significantly higher than the mean rating assigned by African American raters ( M = 5.27, SD = 1.03 ). When the target was African American, once again the mean ratings given by both Caucasian raters ( M = 5.90, SD = 0 .96 ) and

PAGE 277

265 Hispanic raters ( M = 5.94, SD = 0 .96 ) were significantly higher than the mean rating assigned by African American raters ( M = 5.49, SD = 1.09 ). This finding continued to display a trend towards significance when BMI and the PACS were entered as covariates The effect reached significance when the EDI BD, the DMS, and the SATAQ were covaried. For figure V, the interaction between target race and rater race also showed a trend towards significance with no covariates entered ( F (2,636) = 3.46, p = .009, partia l 2 = .02). Inspection of the rater race means at each level of target race revealed that when the target was Caucasian, the mean ratings given by both Caucasian raters ( M = 4.76, SD = 1.04 ) and Hispanic raters ( M = 4.61, SD = 1.06 ) were significantly hig her than the mean rating assigned by African American raters ( M = 3.91, SD = 1.04 ). When the target was Hispanic, the rating given by Caucasian raters ( M = 4.49, SD = 1.09 ) was significantly higher than the mean rating assigned by African American raters ( M = 4.10, SD = 1.17 ). When the target was African American, the mean ratings assigned by both Caucasian raters ( M = 4.59, SD = 0 .98 ) and Hispanic raters ( M = 4.37, SD = 1.02 ) were significantly higher than the mean rating assigned by African American rater s ( M = 3.80, SD = 1.18 ). This finding reached significance when BMI was entered as a covariate. Health Analyses: Main Effects of Target Race Female Stimuli There was a significant main effect of target race for a number of female figures. There was a sig nificant main effect of target race for figure C ( F (2, 635) = 19.11, p < .001 partial 2 = .06) such that the mean ratings provided for Caucasian ( M = 3.09, SD = 1.62 ) and African American ( M = 2.99, SD = 1.59 ) targets were significantly lower than that pr ovided for the Hispanic target ( M = 3.51, SD = 1.64 ). This finding showed a trend towards significance when the SATAQ was covaried but failed to reach significance when each of the other covariates was entered into the equation. For figure D, there was a s ignificant main effect of target race ( F (2, 637) = 5.77, p = .003, partial 2 = .02) that remained significant when the EDI BD was covaried but disappeared when each of the other covariates was entered. This effect is qualified by the two way interaction be tween target race and rater gender. Post hoc examination of the means showed that the mean ratings given to the Caucasian target ( M = 5.23, SD = 1.26 ) and Hispanic target ( M = 5.30, SD = 1.16 ) were significantly lower than the mean rating assigned to the A frican

PAGE 278

266 American target ( M = 5.44, SD = 1.26 ). For figure H, there was also a main effect of target race ( F (2, 625) = 10.88, p < .001 partial 2 = .03) Post hoc LSD tests showed that the mean rating given to the Caucasian target ( M = 4.89, SD = 1.20 ) was significantly lower than those provided for the Hispanic ( M = 5.11, SD = 1.04 ) and African American ( M = 5.19, SD = 1.03 ) targets. This significant effect disappeared when each of the covariates was entered into the equation. There was a significant main effect of target race for figure L ( F (2, 632) = 19.37, p < .001 partial 2 = .06). Pairwise comparisons revealed that the mean ratings provided for the Caucasian ( M = 4.12, SD = 1.30 ) and African American ( M = 4.08, SD = 1.2 3 ) targets did not differ signif icantly but both were significantly lower than the rating provided for the Hispanic target ( M = 4.53, SD = 1.38 ). This main effect failed to reach significance when each of the covariates was present in the model. For figure O, there was a strong trend to wards significance for the main effect of target race ( F (2, 641) = 4.86, p = .008, partial 2 = .02) that disappeared when each of the covariates was entered. Post hoc LSD tests showed that the mean ratings provided for the Caucasian ( M = 3.00, SD = 1.06 ) and African American ( M = 3.15, SD = 1.14) targets did not differ significantly but both were significantly lower than the mean rating provided for the Hispanic target ( M = 3.20, SD = 1.09). Health Analyses: Main Effects of Target Race Male Stimuli There were also several male figures that displayed a main effect of target race. For figure P there was a main effect of target race ( F (2, 606) = 7.70, p = .001, partial 2 = .02) such that the mean ratings given to the Caucasian ( M = 35.93, SD = 1.07) and Af rican American ( M = 5.94, SD = 1.13) targets were both significantly higher than the mean rating given to the Hispanic target ( M = 5.74, SD = 1.21) This significant effect disappeared when each of the covariates was entered into the model with the excepti on of the SATAQ. There was a significant main effect of target race for figure W ( F (2, 613) = 13.23, p < .001 partial 2 = .04). Post hoc LSD tests revealed that the mean rating for the Caucasian ( M = 5.66, SD = 1.15 ) target was not significantly different from the mean rating for the African American ( M = 5.62, SD = 1.18 ) target. However, both were significantly higher than the mean rating given to the Hispanic target ( M = 5.34, SD = 1.18 ). This effect was no longer significant when each of the covariates was entered.

PAGE 279

267 There was a significant main effect of target race for figure Z ( F (2, 642) = 8.38, p < .001 partial 2 = .03). Post hoc LSD tests revealed that the mean rating for the Caucasian ( M = 3.57, SD = 0 .99 ) target was not significantly different from the mean rating for the Hispanic ( M = 3.44, SD = 0 .98 ) target. However, both were significantly lower than the mean rating given to the African American target ( M = 3.70, SD = 1.09 ). This main effect failed to reach significance when each of the covariate s was entered into the equation. There was a significant main effect of target race for figure AA ( F (2, 637) = 28.21, p < .001 partial 2 = .08). Post hoc tests showed that the mean rating given to the African American target ( M = 3.28, SD = 1.41) was sign ificantly higher than the mean rating given to the Hispanic target ( M = 2.83, SD = 1.42 ), which was significantly higher than the mean rating given to the Caucasian target ( M = 2.64, SD = 1.37 ). This effect remained significant when the SATAQ was covaried but not when each of the other covariates was entered. Health Analyses: Between Subjects Interactions No figures displayed a two way interaction between rater race and rater gender. Health Analyses: Main Effects of Rater Gender Several figures showed a ma in effect of rater gender, collapsed across level of target race. For figure L, this main effect of gender was significant with no covariates in the model ( F (1, 321) = 10.14, p = .002, partial 2 = .03). Pairwise comparisons showed that male raters ( M = 4.4 3, SD = 0 .97 ) gave a higher rating than female raters ( M = 4.11, SD = 0 .99 ). This effect remained significant when the EDI BD, the PACS, and the SATAQ were covaried but not when each of the other covariates was entered. For figure R, again there was a main effect of rater gender with no covariates entered ( F (1, 320) = 8.37, p = .004, partial 2 = .03). Post hoc LSD tests revealed that, once again, male raters ( M = 4.74, SD = 0 .93 ) provided higher ratings than females raters ( M = 4.40, SD = 0 .97 ). This effect remained significant when the DMS was covaried and showed a strong trend towards significance when BMI, the PACS, and the SATAQ were covaried. The effect was no longer significant when the EDI BD was covaried. For figure V, there was a strong trend for th e main effect of rater gender to reach significance only when the DMS

PAGE 280

268 was covaried ( F (1, 318) = 6.66, p = .010, partial 2 = .02). Post hoc tests showed that male raters ( adjusted M = 4.50, SE = 0 .08 ) provided higher ratings than female raters ( adjusted M = 4.22, SE = 0 .06 ). For figure AA, again there was a main effect of rater gender with no covariates entered ( F (1, 323) = 8.25, p = .004, partial 2 = .03). Examination of the cell means again revealed that male raters ( M = 3.11, SD = 1.09 ) provided higher ra tings than female raters ( M = 2.78, SD = 1.05 ). This main effect remained significant when the EDI BD and the PACS were covaried and showed a strong trend when the SATAQ was covaried. It disappeared when BMI and the DMS were covaried. Health Analyses: Main Effects of Rater Race Several figures showed a significant main effect of rater race collapsed across levels of target race. The only female figure in this list, figure G, showed a strong trend towards a significant main effect of rater race with no covar iates in the model ( F (2, 324) = 4.72, p = .010, partial 2 = .03) Pairwise comparisons revealed that the mean rating given by Caucasian raters ( M = 4.84, SD = 0 .85 ) was not significantly different from the mean rating given by Hispanic raters ( M = 4.76, SD = 0 .90 ). However, both were significantly higher than the mean rating assigned by African American raters ( M = 4.40, SD = 0 .99 ). This trend towards significance was also seen when the SATAQ was covaried but not when each of the other covariates was entere d. Figure Q showed a significant main effect of rater race with no covariates entered ( F (2, 326) = 10.29, p < .001 partial 2 = .06) that remained significant when each of the covariates was entered in the model. This effect must be considered in light of the trend for an interaction between target race and rater race. Post hoc tests showed that the mean rating given by African American raters ( M = 5.35, SD = 0 .86 ) was significantly lower than the mean ratings given by both Caucasian ( M = 5.89, SD = 0 .77 ) a nd Hispanic ( M = 5.86, SD = 0 .86 ) raters. Figure R showed the same pattern as figure Q with a significant main effect of rater race ( F ( 2, 320) = 11.58, p < .001 partial 2 = .07) that remained significant when each covariate was entered. Once again, this e ffect is tempered by the trend towards a significant interaction between target race and rater race. Again, post hoc tests showed that the mean rating given by African American raters ( M = 4.14, SD = 0 .96 ) was significantly lower than the mean ratings

PAGE 281

269 give n by both Caucasian ( M = 4.76, SD = 0 .92 ) and Hispanic ( M = 4.66, SD = 0 .92 ) raters. Figure V also showed this same pattern of a significant main effect of rater race ( F (2, 323) = 17.53, p < .001 partial 2 = .10) that remained significant with each covari ate entered. For this figure, examination of the means revealed that the mean rating given by African American raters ( M = 5.62, SD = 1.16 ) was significantly higher than the mean ratings given by both Caucasian ( M = 4.61, SD = 0 .71 ) and Hispanic ( M = 4.44, SD = 0 .79 ) raters. Health Analyses: Significant Covariates There were several figures with significant covariates in these analyses. For figure A, the DMS was a significant covariate ( F (1, 318) = 9.92, p = .002, partial 2 = .03). For figure O, the SATAQ w as a significant covariate ( F (1, 322) = 11.83, p = .001, partial 2 = .04).There was a strong trend for the PACS to be a significant covariate for figure Q ( F (1,324) = 7.29, p = .007, partial 2 = .02). The DMS was a significant covariate for figure S ( F (1, 312) = 7.97, p = .005, partial 2 = .03). BMI was a significant covariate for figure AA ( F (1, 321) = 7.64, p = .006, partial 2 = .02). Attractiveness Analyses: Three Way Interactions For the attractiveness analyses, no figures showed a three way interactio n. Attractiveness Analyses: Covariate X Target Race Interactions One figure, figure G, showed a trend for the SATAQ to interact with target race ( F (2, 607) = 4.85, p = .009, partial 2 = .02). Attractiveness Analyses: Target Race X Rater Gender Interaction s Several figures displayed a two way interaction between target race and rater gender. For figure G, this interaction was significant when the EDI ( F (2, 610) = 5.37, p = .006, partial 2 = .02 ) and the SATAQ ( F (2, 607) = 5.36, p = .006, partial 2 = .02 ) we re covaried. Post hoc tests showed that when the target was Caucasian, male raters ( adjusted M = 4.63, SE = 0 .13 ) provided higher ratings than female raters ( adjusted M = 4.15, SE = 0 .10 ). When the target was Hispanic, male raters ( adjusted M = 4.53, SE = 0 .13 ) also provided higher ratings than female raters ( adjusted M = 4.09, SE = 0 .10 ). There was a

PAGE 282

270 trend for a significant interaction for figure H only when the SATAQ was covaried ( F (2, 638) = 5.04, p = .007, partial 2 = .02 ). Post hoc tests did not reveal significant differences between specific pairs of means. Attractiveness Analyses: Target Race X Rater Race Interactions Several figures showed a two way interaction between target race and rater race. For figure C, this interaction showed a trend towards significance with no covariates in the model ( F (4,633) = 3.51, p = .008, partial 2 = .02). Examination of the rater race mean ratings at each level of target race revealed that when the target was Caucasian, Caucasian raters ( M = 2.86, SD = 1.70 ) provide d a higher rating than did Hispanic raters ( M = 2.43, SD = 1.44 ) and African American raters ( M = 2.31, SD = 1.43 ). When the target was Hispanic, Caucasian raters ( M = 3.24, SD = 1.66 ) provided a higher rating than did African American raters ( M = 2.60, SD = 1.53 ). This trend remained when the DMS was entered as a covariate but not when each of the other covariates was entered. For figure H, there was a significant interaction with no covariates entered ( F (4,643) = 3.65, p = .006, partial 2 = .02 ). Post ho c analyses showed that when the target was African American, African American raters ( M = 3.69, SD = 1.45 ) provided higher ratings than did Caucasian raters ( M = 3.13, SD = 1.24 ). The interaction remained significant when the PACS was covaried and showed a strong trend when BMI, the EDI BD, and the DMS were covaried. The effect was no longer significant when the SATAQ was covaried. Figure R showed a significant interaction between target race and rater race ( F (2,637) = 4.68, p = .001, partial 2 = .03) that remained significant when each of the covariates was entered. Post hoc tests showed that when the target was Caucasian, Caucasian raters ( M = 4.10, SD = 1.44 ) and Hispanic raters ( M = 3.91, SD = 1.46 ) provided higher mean ratings than did African American raters ( M = 3.30, SD = 1.21 ). When the target was Hispanic, the mean rating provided by Caucasian raters ( M = 3.71, SD = 1.40 ) was lower than the mean rating provided by Hispanic raters ( M = 3.27, SD = 1.39 ). Attractiveness Analyses: Main Effects of Targe t Race Female Stimuli There was a significant main effect of target race for several female figures. For figure A, there was a significant main effect of target race ( F (2, 644) = 5.67, p = .004,

PAGE 283

271 partial 2 = .02) such that the mean rating provided for the Caucasian target ( M = 2.73, SD = 1.41 ) did not differ significantly from the rating given to the African American target ( M = 2.86, SD = 1. 53). These means were both significantly higher than the mean rating given to the Hispanic target ( M = 2.60, SD = 1. 43 ). This main effect was no longer significant when each of the covariates was entered into the model. There was a significant main effect of target race for figure C ( F (2, 633) = 20.85, p < .001 partial 2 = .06) but this effect must be considered in lig ht of the significant two way interaction between target race and rater race. Post hoc tests for this main effect revealed that mean ratings assigned to the Caucasian ( M = 2.56, SD = 1.56 ) and African American ( M = 2.41, SD = 1.43 ) targets were significant ly lower than the rating assigned to the Hispanic target ( M = 2.92, SD = 1.62 ). This main effect disappeared when each of the covariates was entered into the equation. For figure D, there was also a significant main effect of target race with no covariates entered ( F (2, 633) = 6.24, p = .002, partial 2 = .02). Post hoc tests showed that the mean rating given to the Caucasian target ( M = 2.83, SD = 1.40 ) was significantly lower than the mean rating given to both the Hispanic ( M = 3.09, SD = 1.43) and African American ( M = 3.00, SD = 1.42) targets. This main effect was no longer significant when each of the covariates was entered. For figure K, the main effect of target race was significant with no covariates in the model ( F (2, 590) = 15.21, p < .001 partial 2 = .05). Pairwise comparisons of the means showed that the mean rating given to the Caucasian target ( M = 3.68, SD = 1.37 ) was lower than the mean rating given to the African American target ( M = 3.93, SD = 1.36 ) which, in turn, was significantly lower th an the rating given to the Hispanic target ( M = 4.13, SD = 1.44 ). The effect remained significant when the SATAQ was covaried but disappeared when each of the other covaria tes was entered. For figure L, t he main effect of target race was significant only w ith no covariates in the model ( F (2, 634) = 19.32, p < .001 partial 2 = .06). Post hoc tests revealed that the Caucasian ( M = 2.59, SD = 1.18 ) and Hispanic ( M = 2.72, SD = 1.26) targets received significantly lower ratings than did the African American ta rget ( M = 3.03, SD = 1.31). For figure O, the main effect of target race was also significant only with no covariates in the model ( F ( 2, 646) = 17.45, p < .001 partial 2 = .05). Post hoc

PAGE 284

272 tests revealed that the Hispanic target ( M = 2.63, SD = 1.19 ) receiv ed higher ratings than did the Caucasian target ( M = 2.78, SD = 1.24 ). Attractiveness Analyses: Main Effects of Target Race Male Stimuli For the male figures, several showed a main effect of target race with no covariates in the model that disappeared wh en each of the covariates was entered. For figure P, the main effect showed a strong t rend towards significance ( F (2, 610) = 4.74, p = .010, partial 2 = .01). Examination of the means revealed that the Caucasian ( M = 4.73, SD = 1.66 ) and Hispanic ( M = 4.68 SD = 1.61) targets received significantly lower ratings than did the African American target ( M = 4.93, SD = 1.64). For figure V, the main effect of target race ( F (2, 634) = 16.88, p < .001 partial 2 = .05 ) was such that the mean rating given to the His panic target ( M = 3.27, SD = 1.33 ) was lower than the mean rating given to the African American target ( M = 3.48, SD = 1.28 ) which, in turn, was significantly lower than the rating given to the Caucasian target ( M = 3.74, SD = 1.37 ). For the main effect of target race for figure AA ( F (2, 643) = 26.10, p < .001 partial 2 = .08) the mean rating given to the Caucasian target ( M = 2.15, SD = 1.08 ) was lower than the mean rating given to the Hispanic target ( M = 2.33, SD = 1.27 ) which, in turn, was significantl y lower than the rating given to the African American target ( M = 2.64, SD = 1.26 ). Attractiveness Analyses: Between Subjects Interactions No figures displayed a two way interaction between rater race and rater gender. Attractiveness Analyses: Main Effect s of Rater Gender Several figures showed a main effect of rater gender, collapsed across levels of target race. Figure B showed a main effect of rater gender only when the DMS was covaried ( F (1, 317) = 9.54, p = .002, partial 2 = .03). Post hoc tests showe d that male raters ( adjusted M = 2.59, SE = 0 .12 ) gave lower ratings than did female raters ( adjusted M = 3.04, SE = 0 .09 ). For figure D, the main effect of rater gender was also only significant when the DMS was covaried ( F (1, 318) = 8.82, p = .003, partia l 2 = .03). Again, post hoc tests showed that male raters ( adjusted M = 2.71, SE = 0 .11) gave lower

PAGE 285

273 ratings than did female raters ( adjusted M = 3.18, SE = 0 .09). Figure L showed a significant main effect of rater gender with no covariates in the model ( F (1, 320) = 10.87, p = .001, partial 2 = .03 ) that remained significant when each covariate was entered. Post hoc tests once again revealed that male raters ( M = 2.56, SD = 1.00) assigned lower ratings than female raters ( M = 2.93, SD = 0 .94). Figure O also showed a main effect of rater gender without covariates ( F (1, 325) = 12.58, p < .001 partial 2 = .04 ) that remained significant when each covariate was entered. Post hoc tests showed that male raters ( M = 2.45, SD = 1.01) provided lower ratings than fema le raters ( M = 2.87, SD = 1.00). Figure P showed a main effect of rater gender only when the DMS was covaried ( F (1, 318) = 11.42, p = .001, partial 2 = .04 ). Post hoc tests again revealed that male raters ( adjusted M = 4.41, SE = 0 .14) provided lower ratin gs than did female raters ( adjusted M = 5.06, SE = 0 .11). Figure Q showed a main effect of rater gender with no covariates in the model ( F (1, 326) = 12.92, p < .001 partial 2 = .04) Post hoc tests revealed that male raters ( M = 4.66, SD = 1.56) provided lower ratings than did female raters ( M = 5.15, SD = 0 .99). This effect remained significant with BMI, the EDI BD, the DMS, and the PACS as covariates. There was a strong trend for a significant effect with the SATAQ covaried. Figure S showed a significant main effect of rater gender only when the DMS was covaried ( F (1, 312) = 10.26, p = .001, partial 2 = .03 ). Post hoc tests revealed that male raters ( adjusted M = 4.02, SE = 0 .14) provided lower ratings than did female raters ( adjusted M = 4.65, SE = 0 .11) Figure W showed a strong trend for a significant main effect of rater gender only when the DMS was covaried ( F (1, 313) = 7.43, p = .007, partial 2 = .02) Again, post hoc tests showed that male raters ( adjusted M = 3.89, SE = 0 .12) assigned lower ratings than did female raters ( adjusted M = 4.35, SE = 0 .10). Attractiveness Analyses: Main Effects of Rater Race Several figures showed a significant main effect of rater race collapsed across levels of target race. Figure A showed a strong trend for a signif icant main effect of rater race only when the SATAQ was entered as a covariate ( F (2, 320) = 4.68, p = .010, partial 2 = .03 ). Post hoc examination of the means showed that the mean ratings provided by Caucasian ( adjusted M = 2.68, SE = 0 .11 ) and Hispanic ( adjusted M = 2.50, SE = 0 .12 )

PAGE 286

274 raters were both significantly lower than the mean rating provided by African American raters ( adjusted M = 3.05, SE = 0 .13 ). Figure O showed a significant main effect of rater race with no covariates in the model ( F (2, 325) = 7.10, p = .001, partial 2 = .04 ). Post hoc tests revealed that the mean rating provided by Caucasian raters ( M = 2.51, SD = 0 .87) did not differ significantly from the mean rating provided by Hispanic raters ( M = 2.57, SD = .98). These were both significa ntly lower than the ratings assigned by African American raters ( M = 3.06, SD = 1.15). This effect remained significant with each covariate in the model with the exception of the SATAQ. There was a strong trend for a significant main effect of rater race f or figure R ( F (2, 320) = 4.91, p = .008, partial 2 = .03 ). This trend must be considered in light of the significant target race by rater race interaction. Pairwise comparisons among the means revealed that Caucasian raters ( M = 3.81, SD = 1.17) and Hispan ic raters ( M = 3.76, SD = 1.16) provided mean ratings that were significantly higher than those provided by African American raters ( M = 3.35, SD = 1.08). Figure V showed a main effect of rater race with no covariates entered ( F (2, 323) = 8.23, p < .001 pa rtial 2 = .05 ). Post hoc tests revealed that Caucasian raters ( M = 3.74, SD = 1.08) and Hispanic raters ( M = 3.51, SD = 1.01) provided mean ratings that were significantly higher than those provided by African American raters ( M = 3.13, SD = 1.10). This e ffect remained significant when each of the covariates was entered. When the SATAQ was covaried, only the mean ratings given by Caucasian and African American raters differed. Attractiveness Analyses: Significant Covariates For the attractiveness analyses there were several figures with significant covariates. For figure C, BMI wa s a significant covariate ( F (1, 318) = 7.89, p = .005, partial 2 = .02) and the SATAQ showed a strong trend to b e a significant covariate ( F (1, 317) = 7.02, p = .008, partial 2 = .21). For figure L, BMI was a significant covari ate ( F (1,318) = 12.59, p < .001 partial 2 = .04). For figure O, BMI was also a sig nificant covariate ( F (1, 323) = 11.76, p = .001, partial 2 = .04). For figure P, both the DMS ( F (1,318) = 13.42, p < .001 partial 2 = .04 ) and the SATAQ ( F (1,320) = 16.53, p < .001 partial 2 = .05 ) were significant covariates. For figure Q, both the PACS ( F (1, 324)

PAGE 287

275 = 11.03, p = .001, partial 2 = .03 ) and the SATAQ ( F (1,323) = 18.80 p < .001 partial 2 = .06 ) were signific ant covariates. Variables acting as significant covariates for figure S were the DMS ( F (1, 312) = 15.45, p < .001 partial 2 = .05 ) and the SATAQ ( F (1, 314) = 12.21, p = .001, partial 2 = .04 ). The SATAQ ( F (1, 316) = 11.37, p = .001, partial 2 = .04 ) was a significant covariate for figure W. Finally, BMI was a significant covariate for figure AA ( F (1,321) = 16.33, p < .001 partial 2 = .05).

PAGE 288

About the Author Tovah Yanover earned a degree in Scholar's Electives: French and Psychology from the University of Western Ontario in June 2003. The following September, she began her graduate studies in Clinical Psychology at the University of South Florida as the recipient of a Presidential Doctoral Scholarship. During her tenure at the University of South Florida, s he won the Stefanie Gilbert Endowed Scholarship in Psychology for her Master's thesis. Tovah is currently completing her predoctoral internship at St. Joseph's Healthcare in Hamilton, Ontario, Canada. She is working primarily in the Eating Disorders Clinic and she hopes to build a career in this exciting and important field.