SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Appendix
  8. SUPPLEMENTARY MATERIAL
  9. Acknowledgments
  10. Disclosure
  11. REFERENCES
  12. Supporting Information

Obese individuals are evaluated negatively and attributed negative trait characteristics in several contexts including employment, health care, and education. The current experimental study of college students examined the effect of body mass on the evaluation of political candidates and examined whether the gender of the candidate moderated the relationship. A series of ordinary least squares regression analyses found an interactive effect between candidate obesity and candidate gender for global evaluation and for several trait characteristics. Specifically, obese female candidates were evaluated more negatively than nonobese female candidates and nonobese male candidates were evaluated more negatively than were obese male candidates. This interaction persisted even after controlling for standard political and demographic characteristics of the evaluator. These findings suggest that weight bias exists for obese female political candidates, but that larger body size may be an asset for male candidates. The ability of candidates to be successful may depend less on their policy positions or even party affiliation and more on their physical attributes than has been previously assumed.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Appendix
  8. SUPPLEMENTARY MATERIAL
  9. Acknowledgments
  10. Disclosure
  11. REFERENCES
  12. Supporting Information

Throughout the 2008 US presidential campaign season, former Arkansas Governor Mike Huckabee reminded voters of his ability to rid himself of his excess weight through “getting rid of fried foods and sugar, and starting to learn how to eat right and then slowly developing an exercise routine that went from zero exercise to eventually running four marathons1.” Similarly, President Obama's pick for Surgeon General, Dr Regina Benjamin, was criticized because she is overweight2. Given Huckabee's emphasis on weight and the public reaction to Obama's choice for Surgeon General, one wonders if individuals evaluate obese political candidates more negatively than nonobese political candidates.

Political researchers have examined the way in which individuals evaluate political candidates and the factors that are critical to that evaluation. Using survey data from the American National Election Studies (ANES)3, researchers have argued that the way individuals feel toward candidates and the trait characteristics they attribute to candidates are driven by party affiliation, issue positions, ideology, and demographic characteristics (e.g., gender, race, religion). Increasingly, electoral choices have been found to be influenced by individuals' assessments of the candidates' character or personal attributes (1,2,3,4,5,6,7).

Despite the increasing attention paid to the role of personal qualities, the physical appearance of a candidate has received less attention, beyond such basic factors as gender and race. For example, in discussing why a voter might consider the personal qualities of the candidate, Glass (2) explicitly stated that certain characteristics are relevant to the role of president (e.g., competence, integrity) whereas others are not (e.g., hair color, religion, smile).

A few studies have considered the relationship between physical characteristics like height and facial appearance on electoral success. For example, research indicates that height influences the likelihood of electoral success for presidential candidates and ratings of presidential greatness (8,9,10,11). More recent experimental research notes the connection between facial appearance and electoral success. Todorov and colleagues (12) had participants evaluate the competence of congressional candidates based only on a head shot image of the candidate. They demonstrated that the competence ratings were significantly related to the actual electoral success of such candidates. Little and colleagues (13) expanded on this work by examining the effect of facial shape on voting decisions and found that differences in the shape of candidates' faces influenced the likelihood that participants would vote for them.

Additionally, in laboratory experiments, the physical attractiveness of political candidates has been studied with mixed findings. Budesheim and DePaola (14) found that physical attractiveness influenced candidate evaluation despite the provision of information about the candidate's policy stances and personality characteristics. In contrast, Rosenberg et al. (15) found no relationship between physical attractiveness and beliefs that the candidate would make a reasonable political leader. Similarly, Sigelman and colleagues (16) found no relationship between physical attractiveness and vote choice. Finally, Riggle and colleagues (17) examined the conditions under which physical attractiveness can have an effect on evaluation. They found that physical attractiveness had an effect when no other candidate information was present, but failed to have an effect when policy information about the candidate was provided.

The research on physical appearance, however, has not focused on candidate weight. This lack of attention to weight in candidate evaluation is especially problematic as many studies have documented bias against obese persons in a variety of contexts (18). Obesity bias is pervasive among laypersons, employers, teachers, and even health-care professionals (19,20,21,22,23,24,25,26,27). Although there have been several studies documenting bias against obese persons in a variety of contexts (18), surprisingly, none have examined prejudice against political candidates. This is of interest because political candidates are a group for which “perfection” is expected by their constituents. The extensive research on weight bias argues that individuals evaluate obese targets more negatively than nonobese targets. As a result, in the present study, we consider whether this extends to political candidates and argue that individuals should be more likely to rate obese candidates less favorably than nonobese candidates.

Although the existence of weight bias is a pervasive finding, researchers have also considered the extent to which gender of the obese individual may moderate the role of weight bias in evaluation. In their recent review of research on obesity bias, Puhl and Heuer (18) noted that obese female job applicants are often evaluated more negatively than obese male applicants. The research on this question is not conclusive, however, in part because many studies have focused on perceived rather than actual weight bias. Roehling and colleagues (27) found that 27% of “very obese” women (body mass >35 kg/m2) compared to only 12% of “very obese” men perceived weight as a cause for employment discrimination. Among women and men of all weight classes, the women were 16 times more likely to perceive weight-based employment discrimination than men. Given these findings, we considered whether the gender of the candidate moderates the effect of obesity on global evaluation and trait assessments.

Finally, researchers have also considered the extent to which characteristics of the evaluator influence weight bias in evaluations of others. For example, Crandall (19) studied antiobesity attitudes among college students and found a positive relationship between measures of authoritarian attitudes, the protestant work ethic, a belief in a just world, ideology, and antiobesity attitudes. In a second study, Crandall (28) found that College Republicans reported more antiobesity attitudes than College Democrats. Political scientists have found these same predispositions to be important factors in explaining political attitudes. As a result, we controlled for the relationship between political and social predispositions (ideology, partisanship, authoritarianism, the protestant work ethic, and just world beliefs), demographic characteristics, participant BMI, and candidate evaluation.

Aim of the current study

The aim of this study was to determine: (i) whether obese candidates are evaluated more negatively than nonobese candidates (global evaluation), (ii) whether individuals are more likely to attribute negative trait characteristics to obese candidates than nonobese candidates, (iii) whether certain characteristics of the candidate moderate weight bias, and (iv) whether these effects endure once we control for certain characteristics of the evaluator?

Our expectation was that individuals exposed to obese candidates would evaluate them more negatively on the global evaluation measure than would those viewing nonobese candidates. We also anticipated that obese candidates would be described more often in terms of negative trait characteristics (e.g., unreliable) and nonobese candidates would be described more often in terms of positive trait characteristics (e.g., intelligent). Furthermore, we anticipated an interactive effect between obesity and gender such that obese female candidates would be evaluated more negatively than nonobese female candidates, obese male candidates, and nonobese male candidates. Finally, we considered the possibility that controlling for characteristics of the participant including demographic characteristics, BMI, political attitudes, and social attitudes might minimize the relationship between obesity and evaluation.

Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Appendix
  8. SUPPLEMENTARY MATERIAL
  9. Acknowledgments
  10. Disclosure
  11. REFERENCES
  12. Supporting Information

Participants and recruitment

One hundred and twenty participants (mean age = 24.5 years; % female = 75; % Caucasian = 64.7; mean BMI = 27.9 kg/m2) were recruited from undergraduate psychology and political science courses at a medium-sized midwestern university. Participants were told that they would read information about candidates on a computer screen and answer a series of questions about them. For their full participation, participants were awarded course credit. This study was approved by the University Institutional Review Board.

Research design

In order to test our hypotheses, we used a 2 (gender of candidate) × 2 (obese vs. nonobese) experimental design. An experimental design allowed us to use hypothetical candidates to control information about the candidates and to hold each candidate's demographic, political, and policy information constant while varying gender and weight across the treatment groups. One-half of the participants were randomly assigned to read about female candidates and one-half were assigned to read about male candidates. These groups were then further divided into two groups, with one-half of participants reading about obese candidates and one-half of participants reading about nonobese candidates. Each condition (obese male candidates, nonobese male candidates, obese female candidates, and nonobese female candidates) included 30 participants.

Procedures

Upon providing informed consent, participants completed a prestimulus questionnaire. Participants were asked to provide their age, gender, race (White, Black, Asian, Native American, Hispanic or Latino, or other), social class (working class, middle class, or upper class), and year in school (freshman, sophomore, junior, senior, or other).

Additionally, participants answered a series of questions regarding their political attitudes. Participants placed themselves on an ideological scale ranging from extremely liberal (1) to extremely conservative (7). A political knowledge quiz was also administered to participants and included questions regarding the offices held by various foreign and domestic political leaders. The questionnaire also elicited the partisanship of the participant and participants were given the following options: republican, democrat, or no preference. We created a measure of partisan match coded one if the candidate's partisanship matched that of the participant and zero otherwise.

Next, participants were asked to read four vignettes describing the background, issue positions, and political party affiliation of four hypothetical political candidates who varied based on political ideology (see Supplementary Appendix online for an example of the stimulus materials). Candidate 1 was a democrat with liberal issue positions. Candidate 2 was a democrat with conservative issue positions. Candidate 3 was described as a republican with very conservative positions on the issues. Finally, candidate 4 was also described as a republican, but a more liberal republican.

Participants in each condition read the same four vignettes modified for gender and obesity status depending on their group assignment (i.e., obese male candidates, nonobese male candidates, obese female candidates, and nonobese female candidates). Of note, candidate ages ranged from the late 40s to the late 60s. Information on the obesity status of the candidate was provided both through a picture of the candidate and by providing the candidate's height and weight. The individuals used for the obese and nonobese photographs were the same: one picture was natural and the other was morphed to appear obese using AlterImage morphing software (29). Figure 1 presents an example of the photographs used in the experiment. Figure 1a represents the natural photograph and Figure 1b represents the morphed photograph.

image

Figure 1. An example of the photographs used in this experiment. (a) The natural photograph of the candidate used in the male, nonobese condition. (b) The morphed photograph of the candidate used in the male, obese condition.

Download figure to PowerPoint

After viewing each vignette, participants in all groups were asked to provide a global evaluation of the candidate and a series of trait assessments of the candidate. Global evaluation was measured with a feeling thermometer asking participants to rate the candidate from 0 to 100° with 0° representing very cold feelings toward the candidate, 50° neither cold nor warm feelings, and 100° representing very warm feelings toward the candidate. Trait assessments were evaluated by asking participants to decide how much they agreed with various statements describing the candidate. The statements included positive traits (moral, competent, strong leader, cares, likeable, energized, intelligent, able to inspire, and able to perform a strenuous job) and negative traits (dishonest, lazy, unreliable, undependable, and lacking in self-discipline). Participants were asked to rate whether the statement described the candidate not well at all (1) to extremely well (4).

Next, participants completed a poststimulus questionnaire. This questionnaire included the Anti-Fat Attitudes Questionnaire, the Protestant Ethic Values Scale, the Right-Wing Authoritarianism Scale, and the Just World Scale. Mean scores and α reliabilities for the current sample for each measure are presented in Table 1. To measure the Protestant Ethic Values Scale, we used Katz and Haas' (30) version with 13 items (α = 0.78). Our measure of authoritarianism consisted of Altemeyer's (31) 22-item scale (α = 0.93). The Just World Scale (α = 0.61) consisted of the 6-item measure of a general belief in a just world (32). Finally, our measure of antifat bias relied on Crandall's (19) 13-item measure of dislike of fat people, fear of fat, and belief that obesity is a function of willpower (α = 0.80). Crandall separates these 13-items into three scales representing dislike, fear of fat, and willpower. We combined these 13-items into a single scale because doing so simplified the results and did not alter the conclusions presented here.

Table 1.  Social attitude mean scores and reliabilities for the current sample
inline image

Participants were then asked to indicate what they believed the purpose of the experiment was4, their height and weight were obtained, and they were debriefed.

Manipulation check

To verify that the participants perceived the candidates' obesity status accurately, a similar group of participants (college students) viewed the photographs of all 16 hypothetical candidates (four obese female candidates, four obese male candidates, four nonobese female candidates, and four nonobese male candidates) four times in random order. While viewing each image, the participants were asked to determine whether the candidate was “not overweight” or “overweight/obese”. We then averaged each participant's rating for each candidate.

The assessment of obesity offered by these participants was similar to our classification with a few exceptions. Participants could choose “not overweight” (coded as zero) and “overweight/obese” (coded as one) to label each candidate. Averaging across all eight candidates and all participants, candidates categorized as nonobese were rated as not overweight by the participants (mean = 0.08, s.d. = 0.029) whereas the eight candidates we categorized as obese were rated as such by the participants (mean = 0.78, s.d. = 0.08), P < 0.0001. Despite the fact that the natural photographs of the hypothetical candidates were morphed to appear obese, two “obese” candidates were ranked on average as 0.35 and 0.51. Such a score indicates that on average, participants did not find these candidates to be overweight/obese. As a result, we conducted the analysis by dropping these two candidates5.

Statistical analyses

A series of ordinary least squares regressions were conducted modeling global evaluation and trait assessment as a function of the interactive effect of the candidate's obesity and gender, and including controls for participant demographic characteristics, BMI, political attitudes, and social attitudes. All participants evaluated four candidates and assessed the trait characteristics of these four candidates. As a result, there were four evaluations and four sets of trait assessments for each participant (n = 472). Therefore, a case in the dataset equates to a participant's evaluation of a specific candidate rather than a participant. Because the candidates vary in terms of partisanship and ideology, we also included a variable representing the candidates.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Appendix
  8. SUPPLEMENTARY MATERIAL
  9. Acknowledgments
  10. Disclosure
  11. REFERENCES
  12. Supporting Information

Candidate global evaluation

As Table 2 illustrates, the obesity status of the candidate was positively related to candidate global evaluation (model R2 = 0.158, F(14,318) = 4.27, P < 0.001). The interaction between the candidate's obesity and gender was significant and negatively related to global evaluation. These results indicate that obese female candidates were evaluated more negatively than nonobese female candidates. In contrast, obese male candidates were evaluated more positively than nonobese male candidates. Of the control variables, only participant obesity, ideology, and partisan match predicted variance in the model.

Table 2.  OLS regression model predicting global evaluation of candidates
inline image

Candidate trait evaluations

Separate ordinary least squares regression analyses were conducted for each of 14 candidate trait evaluations: competence, reliability, strong leadership, dependability, laziness, likeability, morality, honesty, self-discipline, energy, caring, able to inspire, intelligence, and ability to perform a strenuous job. The candidate's obesity, gender, and the interaction between candidate obesity and candidate gender were statistically significant predictors of reliability (model R2 = 0.102, F(14, 206) = 1.68, P = 0.063), dependability (model R2 = 0.085, F(14, 191) = 1.26, P = 0.237), honesty (model R2 = 0.112, F(14, 181) = 1.63, P = 0.075), able to inspire (model R2 = 0.105, F(14, 229) = 1.91, P = 0.026), and able to perform a strenuous job (model R2 = 0.079, F(14, 252) = 1.54, P = 0.098). The coefficients and s.e. for each of these models are presented in Table 3. Candidate obesity, candidate gender, and the interaction between candidate obesity and candidate gender were not statistically significant predictors of the remaining trait characteristics. Space prevents the presentation of full regression results for all models; these can be made available upon request of the authors.

Table 3.  OLS regression models predicting trait characteristics of candidates
inline image

Similar to the model for global evaluation, the control variables had negligible effects on the trait assessments. Not surprisingly, the partisan match variable was positively and significantly related to some of the trait assessment measures.

To further examine the impact of the interaction between the candidate's obesity and gender on evaluation and trait assessment, we simulated the effect of the candidate's gender and obesity on global evaluation and the trait assessments when all other variables were set at their mean value (33). The results for global evaluation, reliability, dependability, honesty, able to inspire, and able to perform a strenuous job are displayed in Figure 2.

image

Figure 2. The relationship between gender and obesity for global evaluation and trait characteristics for which obesity, gender, and the interaction were significant predictors. The upper left-hand panel illustrates the interactive effect of the candidate's obesity and gender on global evaluation when all other predictors were set at their mean, which ranged from 0 to 100 with 0 representing very negative feelings toward the candidate and 100 representing very positive feelings toward the candidate. The remaining panels display the interactive effect of the candidate's obesity and gender on the trait characteristics when all other predictors were set at their mean, which ranged from 1 (describes the candidate not too well) to 4 (describes the candidate extremely well).

Download figure to PowerPoint

Obese female candidates were evaluated more negatively overall and assessed more negatively in terms of reliability, dependability, honesty, dependability, able to inspire, and able to perform a strenuous job than nonobese female candidates. In contrast, obese male candidates were rated more positively than nonobese male candidates.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Appendix
  8. SUPPLEMENTARY MATERIAL
  9. Acknowledgments
  10. Disclosure
  11. REFERENCES
  12. Supporting Information

In support of our hypotheses, these results indicate that obesity and gender of political candidates can influence global evaluation and the trait characteristics individuals use to describe political candidates. The results of this study suggest that the consistently negative effect of obesity on evaluation and trait assessments of employees, patients, students, and others does not necessarily extend to both male and female political candidates within the confines of our experiment. An interactive effect between candidate obesity and candidate gender was found for global evaluation and for some of the trait characteristics. Of our four candidates (obese male, nonobese male, obese female, and nonobese female), obese female candidates were evaluated the most negatively and were least likely to be described with positive trait characteristics. In contrast, obese male candidates scored high on both global evaluation and the trait measures.

Although weight bias exists in a variety of settings for both men and women, the interaction between gender and obesity found in this study may not be all that surprising. There is significant pressure for women in western society to be thin, but for men there is pressure to have muscle mass (34). This was not explicitly tested in our study design, however, as none of the images were presented as more muscular.

Contrary to our expectations, participant demographic characteristics and indexes for authoritarianism, protestant work ethic, antifat attitudes, and belief in a just world, did not significantly influence overall evaluation or the attribution of trait characteristics to the candidates. Partisan match did predict additional variance for global evaluation and the trait characteristics “honesty,” “dependability,” and “ability to inspire.” This is not surprising given that partisanship is generally the strongest predictor of both candidate evaluation and trait assessment of political candidates.

The implications of these findings are intriguing, though in need of further exploration. For example, in this study participants evaluated and ascribed traits to a series of hypothetical candidates after reading physical and political descriptions of the candidates. Participants were not asked to choose between nonobese and obese candidates. Our suspicion is that the effects of physical characteristics as well as other personal characteristics are most influential in primary elections or when candidates share similar issue preferences. This study was also limited in that the participants were college students. To expand the generalizability of these results, we plan to replicate these results with a nonstudent population and explore the electoral success of actual political candidates. In this latter project, we will examine the vote totals for candidates to the US House by BMI and gender.

Despite these limitations, we believe the results of this study are important for a variety of reasons. The most obvious role for research of this kind is to inform the way researchers think about candidate evaluation and trait assessment. Our current models of public opinion either ignore candidates' physical attributes as important predictors of evaluation or minimize the theoretical importance of such attributes. Individuals in this experiment offered different evaluations and assessments of trait characteristics of the “same” candidates based on the candidate's gender and obesity. This suggests that failing to consider the physical characteristics and the connection between those characteristics and evaluation may lead researchers to overestimate the ability of candidates to be successful in elections.

Studying weight bias in this context also offers the opportunity to provide answers to practical political questions of interest to political candidates and their campaign managers in their efforts to construct campaign strategies. If there is a weight-bias underlying the public's evaluation of political candidates, the implications for potential candidates cannot be understated. Individuals contemplating a run for public office may be able to alter their issue positions or partisanship to fit their constituents and even portray their personality in an appealing way (e.g., appear less arrogant than they actually are), but many physical characteristics are less controllable though potentially important. These results provide further support for the possibility that political decisions, commonly thought to be influenced by issues, partisanship, and ideology, are in fact influenced by stereotypes associated with physical attributes. As a result, the ability of obese female candidates to be successful may depend less on their policy positions or even party affiliation and more on their physical attributes than previously assumed. Importantly, additional characteristics of the candidate that may be correlated with obesity (e.g., socioeconomic status) may further increase the likelihood that the candidate is evaluated negatively and described with negative trait terms. Future studies are needed to explore the interactive effect of gender, obesity, and other variables, such as socioeconomic status or race, on candidate evaluation.

Appendix

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Appendix
  8. SUPPLEMENTARY MATERIAL
  9. Acknowledgments
  10. Disclosure
  11. REFERENCES
  12. Supporting Information

1Interview with Mike Huckabee by ABC News entitled “Huckabee's Marathon: From Rotund to Runner GOP Candidate's Weight Struggle Shapes Health-Policy Views”, 19 December 2007 http:abcnews.go.comHealthDiabetesResourcestoryid4027975&page1.

2ABC news article written by Susan Donaldson James entitled “Critics Slam Overweight Surgeon General Pick, Regina Benjamin,” 21 July 2009 http:abcnews.go.comprintid8129947.

3The ANES is a national representative sample of the American adult population carried out every presidential and congressional election years. For more information about the ANES, please consult http:www.electionstudies.org.

4Because of the between participants design, no participant guessed the purpose of the experiment.

5We also ran the analysis using the average obesity scores for the candidates offered by this group of participants in place of our dichotomous obesity measure. The substantive results were similar to those presented here.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Appendix
  8. SUPPLEMENTARY MATERIAL
  9. Acknowledgments
  10. Disclosure
  11. REFERENCES
  12. Supporting Information

The authors thank Drs Patrick O'Neil, Gary Foster, Cindy Bulik, Marsha Marcus, Catana Brown, Karen Williams, Bill Brooks, and Cary Savage for volunteering their photos for use in this study.

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Appendix
  8. SUPPLEMENTARY MATERIAL
  9. Acknowledgments
  10. Disclosure
  11. REFERENCES
  12. Supporting Information
  • 1
    Funk CL. Bringing the candidate into models of candidate evaluation. J Polit 1999; 61: 700720.
  • 2
    Glass DP. Evaluating presidential candidates: Who focuses on their personal attributes? Public Opin Q 1985; 49: 517534.
  • 3
    Keeter S. The illusion of intimacy television and the role of candidate personal qualities in voter choice. Public Opin Q 1987; 51: 344358.
  • 4
    Kelly S, Mirer T. The simple act of voting. Am Polit Sci Rev 1974; 68: 572591.
  • 5
    Markus GB, Converse PE. A dynamic simultaneous equation model of electoral choice. Am Polit Sci Rev 1979; 73: 10551070.
  • 6
    Miller A, Wattenberg M, Malanchuk O. Schematic assessments of presidential candidates. Am Polit Sci Rev 1986; 80: 521540.
  • 7
    Miller AH. Public judgments of Senate and House candidates. Legis Stud Q 1990; 15: 525542.
  • 8
    Holmes JE, Elder RE. Our best and worst presidents: Some possible reasons for perceived performance. Pres Stud Q 1989; 19: 529557.
  • 9
    McCann SJH. Alternative formulas to predict the greatness of U.S. presidents: Personological, situational, and zeitgeist factors. J Per Soc Psycol 1992; 62: 469479.
  • 10
    Simonton DK. Putting the best leaders in the White House: Personality, policy, and performance. Polit Psychol 1993; 14: 537548.
  • 11
    Sommers PM. Is presidential greatness related to height? College Math J 2002; 33: 1416.
  • 12
    Todorov A, Mandisodza AN, Goren A, Hall CC. Inferences of competence from faces predict election outcomes. Science 2005; 308: 16231626.
  • 13
    Little AC, Burriss RP, Jones BC, Roberts SC. Facial appearance affects voting decisions. Evol Hum Behav 2007; 28: 1827.
  • 14
    Budesheim TL, DePaola SJ. Beauty or the beast? The effects of appearance, personality, and issue information on evaluations of political candidates. Pers Soc Psych Bull 1994; 20: 339348.
  • 15
    Rosenberg SW, Kahn S, Tran T, Lee MT. Creating a political image: Shaping appearance and manipulating the vote. Polit Behav 1991; 13: 345367.
  • 16
    Sigelman L, Sigelman CK, Fowler C. A bird of a different feather? An experimental investigation of physical attractiveness and the electability of female candidates. Soc Psych Quart 1987; 50: 3243.
  • 17
    Riggle ED, Ottati VC, Wyer RS, Kuklinski J, Schwarz N. Bases of political judgments: The role of stereotypic and nonstereotypic information. Polit Behav 1992; 14: 6787.
  • 18
    Puhl RM, Heuer CA. The stigma of obesity: a review and update. Obesity (Silver Spring) 2009; 17: 941964.
  • 19
    Crandall CS. Prejudice against fat people: ideology and self-interest. J Pers Soc Psychol 1994; 66: 882894.
  • 20
    Rudolph CW, Wells CL, Weller MD, Baltes BB. A meta-analysis of empirical studies of weight-bias in the workplace. J Vocat Behav 2009; 74: 110.
  • 21
    Foster GD, Wadden TA, Makris AP et al. Primary care physicians' attitudes about obesity and its treatment. Obes Res 2003; 11: 11681177.
  • 22
    Greenleaf C, Weiller K. Perceptions of youth obesity among physical educators. Soc Psych Education 2005; 8: 407423.
  • 23
    Neumark-Sztainer D, Story M, Harris T. Beliefs and attitudes about obesity among teachers and school health care providers working with adolescents. J Nutr Educ 1999; 31: 39.
  • 24
    Paul RJ, Townsend JB. Shape up or ship out? Employment discrimination against the overweight. Employee Responsibilities Rights J 1995; 8: 133145.
  • 25
    Pingitoire R, Dugoni R, Tindale S, Spring, B. Bias against overweight job applicants in a simulated employment interview. J Appl Psychol 1994; 79: 909917.
  • 26
    Puhl R, Brownell KD. Bias, discrimination, and obesity. Obes Res 2001; 9: 788805.
  • 27
    Roehling MV, Roehling PV, Pichler S. The relationship between body wait and perceived weight-related employment discrimination: The role of sex and race. J Voc Behav 2007; 71: 300318.
  • 28
    Crandall CS, Biernat M. The ideology of anti-fat attitudes. J App Soc Psychol 1990; 20: 227243.
  • 29
    Seattle Software Design. Alterimage. Seattle Software Design: Seattle, 2007.
  • 30
    Katz I, Hass RG. Racial ambivalence and American value conflict: Correlational and priming studies of dual cognitive structures. J Pers Soc Psychol 1988; 55: 893905.
  • 31
    Altemeyer R. Right-wing Authoritarianism. University of Manitoba Press: Winnipeg, Canada, 1984.
  • 32
    Rubin Z, Peplau LA. Who believes in a just world? J Soc Issues 1975; 31: 6589.
  • 33
    King G, Tomz M, Wittenberg J. Making the most of statistical analyses: Improving interpretation and presentation. Am J Polit Science 2000; 44: 347361.
  • 34
    Pope H, Phillips KA, Olivardia R. Rise of the Adonis Complex: The Secret Crisis of Male Body Obsession. The Free Press: New York, 2000, pp 2661.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Appendix
  8. SUPPLEMENTARY MATERIAL
  9. Acknowledgments
  10. Disclosure
  11. REFERENCES
  12. Supporting Information

supporting Information

FilenameFormatSizeDescription
oby_1687_sm_1.doc38KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.