Fat phobia scale revisited: the short form

Abstract

OBJECTIVE: To develop a shortened form of the original 50-item fat phobia scale.

METHOD: The first factor from the original fat phobia scale—undisciplined, inactive and unappealing—was identified as a potential short form of the scale. A new sample of 255 people completed the original 50-item scale. The reliability of a shortened 14-item version of the scale was tested and compared to that of the full scale using both the new sample and the original sample of 1135 study participants.

RESULTS: The fat phobia scale—short form demonstrated excellent reliability in both samples and was strongly correlated with the 50-item scale. Mean and 90th percentile scores are given for both the long and short versions of the scale.

CONCLUSION: The shortened fat phobia scale is expected to increase the utility of the measure in a diverse array of research and clinical settings. Future research should focus on developing scale norms for the general population and conducting research on fat phobia in males and among different ethnic groups.

Introduction

Widespread stereotypical and prejudicial attitudes toward fat people have been well documented among the general population1,2,3,4,5,6,7,8,9,10,11,12,13 as well as among several types of healthcare students and professionals, including doctors, nurses, dietitians and nutritionists.14,15,16,17,18,19,20 Such prejudicial attitudes can create negative consequences in the lives of fat children and adults. For example, fat children are often teased,1 and other children are less likely to want them as friends.21 Women who are fatter than average are less likely to receive financial support from their parents for college.22 Men are less likely to want to date fat women,23,24 and fat women are less likely to have had the opportunity to engage in sexual intercourse or other sexual activities.25 Negative attitudes toward fat people can also affect their employment opportunities26,27,28,29,30 and chances of promotion.31 In the clinical setting, psychologists have described fat people as more unattractive and embarrassed than non-obese people,32 and a group of mental health practitioners rated obese clients more negatively on a number of aspects of psychological functioning.33

Because critical attitudes towards fat people are so prevalent, some fat people internalize negative societal messages.1,3,8,9,10,34 This can cause fat people—or even average weight people who feel fat—to place restrictions on important aspects of their lives, such as going to school, changing jobs, buying stylish clothes, dating or enjoying a sexual relationship, or even seeking medical care.8 Such restriction of activities is also associated with higher levels of depression.35

These examples illustrate the importance of countering negative attitudes toward and stereotypes about fat people, which we have termed ‘fat phobia’.9 A necessary precursor to determining if fat phobic attitudes can be changed—and evaluating the effectiveness of methods for doing so—is a means of reliably measuring such attitudes. Thus, in 1984 we developed the 50-item, five-point semantic differential fat phobia scale.9 In constructing the scale, we drew from clinical experience and from adjectives used to describe fat people as gathered from a small sample of people who entered a motor vehicle license bureau in a Minnesota suburb. We then conducted a principal components factor analysis on a large sample (n=1135) of people interested in body image and weight, which yielded six distinct fat phobia factors or subscales: undisciplined, inactive and unappealing; grouchy and unfriendly; poor hygiene; passivity; emotional/psychological problems; and stupid and uncreative. Initial evidence of the fat phobia scale's construct validity was provided by a quasi-experimental study that showed that obese women who originally held very negative attitudes toward fatness and toward their own bodies had significantly improved attitudes toward fat persons (as measured by their total score and their scores on each of the six subscales) after completing a treatment program designed to improve body image and self-esteem. Additional analyses demonstrated that those who were average weight, female, younger, more educated and nonmedical professionals were more likely to have fat phobic attitudes than those who were overweight, male, older, less educated and students.8,9 Finally, a comparison of the adjectives used on the scale with those of other instruments indicated that the scale includes most of the stereotypical concepts specified in the literature, and that the six dimensions of fat phobia defined by the subscales are consistent with other described concepts of fat phobia.

Since its inception, the fat phobia scale has been cited or reproduced in several publications36,13 and requested for use in clinical settings by other psychologists and physicians, as well as by students conducting research and writing theses to study, measure, and treat fat phobic attitudes, fat prejudice and body image, and stigmatization caused by obesity.37 In a recent review of 23 measures used to assess attitudes toward fat people, Yuker et al13 concluded that the fat phobia scale was one of three scales that ‘demonstrate the best psychometric properties and should be incorporated in future research.’ Further, they noted there was no reason to develop new instruments unless there was a compelling theoretical reason to do so, advocating instead that it would be better to modify current scales to make them more reliable and valid. Since one of the limitations of the original 50-item fat phobia scale was its length, this report describes the development, validation and potential uses of a shortened version of the scale designed to make it easier to use in clinical and research settings.

Methods

Scale development

Using factor analysis, six factors were identified in the original fat phobia scale,9 the first of which—undisciplined, inactive and unappealing—was identified as a potential basis for a short form (Appendix 1). This 14-item subscale accounted for the most variance in the 50-item scale, and scores on this factor showed the greatest improvement for participants completing a program designed to lower fat phobia. Data were collected on all 50 items in 1999 from a new sample of respondents to verify that both the original 50-item and the shortened 14-item version of the scale retained the high reliability demonstrated in the original sample.9

Sample

For the original 1984–1991 sample, data were collected from 1135 primarily white, female (90%) participants recruited from weight loss groups, members of a national health and fitness organization, body image workshops, college classes and psychotherapy clients with body image issues. Recruitment efforts were focused on these groups because they were readily available and because they were the most likely target audience for the future use of the scale. Mean values and ranges for various demographic, weight, and body mass index (BMI) parameters for the sample are provided in Table 1. (See Robinson et al9 for a more detailed sample description.)

Table 1 Demographic, weight and body mass index characteristics for study samples

For this study, data on the complete 50-item scale were also collected from a new but similar 1999 sample of 255, primarily white, female (98%) participants. Of these, 207 (81%) completed and returned a copy of the scale which had been inserted in the December 1998 issue of a locally produced women's sports and health membership newsletter (national distribution of approximately 2125); this organization had originally published the F-scale in its membership newsletter in 1984. The remaining 48 respondents (19%) were recruited from three TOPS (Take Off Pounds Sensibly) weight loss groups in the Minneapolis–St Paul metropolitan area. Demographics for the 1984–1991 and the 1999 samples (Table 2) were quite similar in terms of sex, education, weight and BMI. The participants in the 1999 sample tended to be older.

Table 2 F-scale: total mean scores and reliability coefficients on 50-item and 14-item versions for 1984–1991 and 1999 samples

Reliability of both the short form and long form was assessed in the 1984–1991 and 1999 samples using Cronbach's α. In addition, average item-total correlations were calculated for each scale in both samples. Pearson's correlations were calculated to see how the long form correlated with the short form. The total mean scores in the 1984–1991 and 1999 samples were then compared to each other using independent sample t-tests to see if mean scores differed on the long and short forms. Because of the large sample size (n=1135) in the 1984–1991 sample, the α was set at 0.01 for these comparisons.

Results

The 14-item shortened scale demonstrated excellent reliability in both the 1984–1991 sample (Cronbach's α=0.87) and the 1999 sample (Cronbach's α=0.91). For both samples, there were no items, which, if deleted from the scale, would increase reliability. The short form also showed good correlation with the entire 50-item scale (r=0.82 in the 1984–1991 sample and 0.90 in the 1999 sample). Item-total correlations ranged from 0.42–0.65 (average=0.54) in the 1984–1991 sample and from 0.28–0.77 (average=0.62) in the 1999 sample (Table 2).

In addition to conducting reliability analyses, we made several score comparisons between the 1984–1991 and 1999 samples (Table 2). Using the long form, the mean total scores for the two samples did not differ significantly (t(308.36)=2.33, P=0.021), with an average score of 3.1 for both samples (s.d.=0.38 for the 1984–1991 sample and s.d.=0.50 for the 1999 sample). In the 1984–1991 sample, as well as in the 1999 sample, 3.6 marked the 90th percentile. Using the short form, the mean total scores for the 1984–1991 sample and the 1999 samples differed significantly (t(326.31)=2.64, P=0.009), with an average score of 3.7 (s.d.=0.54) for the 1984–1991 sample and 3.6 (s.d.=0.64) for the 1999 sample. In both samples, 4.4 marked the 90th percentile.

Finally, analysis of a subgroup of women from the original 1984–1991 sample who took part in a treatment program designed to improve body image and self-esteem (n=40),8,9 significantly reduced their scores (pretest mean=3.3, s.d.=0.28; posttest mean=2.6, s.d.=0.47; t(39)=8.88, P≤0.001) on the full 50-item scale as well as on the 14-item subscale (pretest mean=3.8, s.d.=0.43; posttest mean=2.8, s.d.=0.46; t(39)=10.79, P≤0.001).

Discussion

The fat phobia scale—short form maintains the excellent psychometric properties of the original fat phobia scale while reducing the number of items from 50 to 14. The short form demonstrated excellent reliability and correlated quite highly with the original 50-item F-scale in both the 1984–1991 and 1999 samples, thus providing evidence of concurrent validity. Moreover, evidence of construct validity for both long and short versions of the scale was demonstrated by the fact that a subgroup of women who took part in a treatment program designed to improve body image and self-esteem8,9 significantly reduced their mean scores from pretest to posttest on both versions of the scale. The new scale's brevity will simplify its use in research efforts and as a screening tool in clinical settings.

Although the 1984–1991 sample and the 1999 sample did differ significantly in their scores on the short form of the F-scale (3.7 vs 3.6 for the 1984–1991 and 1999 samples, respectively), the P-value (0.009) barely made the α cutoff of 0.01. Thus, while there was a statistically significant difference, the magnitude of the difference in mean scores (0.1) is clinically and practically insignificant. On the long form of the F-scale, there was no significant difference between the two samples. The fact that different samples, during different time periods, obtained essentially equal mean scores on each version (50 and 14 items) of the scale, speaks convincingly about the F-scale's reliability and provides some additional evidence of construct validity.

Based on the scores of all participants who took the 14-item F-scale, a score of 3.6 could be seen as an indication of an average amount of fat phobia and a score of 4.4 (the 90th percentile) or above would indicate a high level of fat phobia. When using the 50-item F-scale, a score of 3.1 would be average with 3.6 (the 90th percentile) or above indicating a high level of fat phobia. These findings are supported by a recent study using a probability cluster sampling method to sample 527 undergraduate dietetic students. Their mean score on the 14-item scale was 3.4 (s.d.=0.55) and 3.1 (s.d.=0.44) on the 50-item F-Scale.37 These scores, on yet another type of sample, are very similar to the scores found in this study.

The mean score on the 14-item F-scale was clearly higher than on the 50-item F-scale in both the 1984–1991 (mean=3.7 vs 3.1, respectively) and 1999 (mean=3.6 vs 3.1, respectively) samples. We believe that this difference reflects the fact that factor 1, which comprises the short form, focuses on the most negative stereotypes about fat people. On the other hand, the remaining five factors eliminated from the short version contained items reflecting more positive and less strongly endorsed negative stereotypes about fat people. The omitted factors9 measured positive stereotypical beliefs that fat people are good natured, humorous and warm (factor 2), as well as negative stereotypical beliefs that fat people have poor hygiene (factor 3), are more passive (factor 4), have more emotional and psychological problems (factor 5), and are unintelligent and uncreative (factor 6). Researchers and clinicians needing information on these other stereotypes should continue to use the longer version of the scale.13,36

Other limitations need to be acknowledged. There are no validity indicators in the scale to control for socially desirable answers. Research in social psychology suggests that when prejudice is assessed using obvious methods, people can hide their true attitudes, but when social desirability is accounted for by using more subtle and indirect techniques, prejudice reappears.40 In an effort to de-emphasize the true purpose of the scale, we have named it the F-scale rather than the fat phobia scale, but this is unlikely to disguise the true purpose of the instrument. Since fat phobia is still so widely accepted, however, the obvious nature of the fat phobia scale may not be as much of a drawback.

Another shortcoming of the scale is that in both the 1984–1991 and 1999 samples the respondents were primarily white, female, more highly educated than a representative sample of adults, and had enough interest in weight control, body image, health, or obesity to attend a lecture or meeting, join an organization, or to read an article on the topic. Norms on the fat phobia scale for the general population still need to be established. Moreover, in light of evidence that fatness is more acceptable among blacks than whites, as well as other cultural and ethnic groups,4,34 we recommend that further research using the fat phobia scale be conducted in the African-American population, in other ethnic groups, and in males.

In conclusion, we expect this new, shortened form of the fat phobia scale to be a valuable tool for researchers and clinicians seeking to build on our current understanding of fat phobia and to develop ways to eliminate it. One promising area of research is an exploration of the complex relationship between fat phobia and other personality and sociodemographic characteristics such as body mass index, education, occupation, age and sex.9,17,33,41 Other studies might investigate whether fat phobia is related to the ability to lose weight and maintain weight loss or the ability of fat people to make healthy lifestyle changes, such as developing healthier eating patterns, exercising more, and increasing their range of daily activities, thereby living life ‘as if they were thin’.8 The extent of fat phobia in parents could be compared to levels of fat phobia in their children, positive or negative body image in their children, and the nutritional quality of food served to their children. Because excessive concern about weight and body-image are defining characteristics of eating disorders, assessing fat phobic attitudes could be part of the routine evaluation in such treatment programs. Even when clients do not raise weight and body image concerns, such issues may contribute significantly to a variety of psychological, relational and social problems. Thus the scale could be used as a simple screening device to alert a therapist to explore such issues more thoroughly. Interventions designed to lower fat phobia8 could be used with managers, teachers and medical and mental health professionals who score high on the scale.

Studying fat phobia currently provides a ‘venue for investigating stereotyping and prejudice more generally’ because social disapproval is rarely ‘invoked against those who express anti-fat attitudes’.41 Research in social psychology examining prejudice against racial minorities and women has shown that overt prejudice against these groups is much less widespread today than 50 years ago.40 A similar change is occurring in attitudes towards gay, lesbian, bisexual and transgendered individuals.42,43,44 In part these changes are due to organizations and movements which have fought against discrimination in legal, social and economic venues. Today such work is being done on behalf of fat people and it is our hope this shortened and easier-to-use version of the fat phobia scale will be useful in measuring and ultimately, eliminating, fat phobic attitudes.

References

  1. 1

    Aronson D . No laughing matter Teaching Tolerance 1997 Fall: 21–23.

  2. 2

    Cowan GSM Jr . What do patients, families and society expect from the bariatric surgeon? Obes Surg 1998 8: 77–85.

  3. 3

    Friedman MA, Brownell KD . Psychological correlates of obesity: moving to the next research generation Psychol Bull 1995 117: 3–20.

  4. 4

    Hebl MR, Heatherton TF . The stigma of obesity in women: the difference is Black and White Personality Social Psychol Bull 1998 24: 417–426.

  5. 5

    Hill AJ, Silver EK . Fat, friendless and unhealthy: 9-year old children's perception of body shape stereotypes Int J Obes Relat Metab Disord 1995 19: 423–430.

  6. 6

    Kuss CL . Absolving a deadly sin: a medical and legal argument for including obesity as a disability under the Americans with Disabilities Act J Contemp Health Law Policy 1996 12: 563–605.

  7. 7

    Lewis RJ, Cash TF, Jacobi L, Bubb-Lewis C . Prejudice toward fat people: the development and validation of the Antifat Attitudes Test Obes Res 1997 5: 297–307.

  8. 8

    Robinson BE, Bacon JG . The ‘If Only I Were Thin...’ Treatment Program: decreasing the stigmatizing effects of fatness Prof Psychol Res Pract 1996 27: 175–183.

  9. 9

    Robinson BE, Bacon JG, O'Reilly J . Fat phobia: measuring, understanding, and changing anti-fat attitudes Int J Eat Disord 1993 14: 467–480.

  10. 10

    Robinson BE, Gjerdingen DK, Houge, DR . Obesity: a move from traditional to more patient-oriented management J Am Board Fam Pract 1995 82: 99–108.

  11. 11

    Senior K . Obesity: how can it be controlled? Mol Med Today 1997 May: 204–206.

  12. 12

    Stunkard AJ . Current views on obesity Am J Med 1996 100: 230–236.

  13. 13

    Yuker HE, Allison DB, Faith MS . Methods for measuring attitudes and beliefs about obese people. In: Allison DB (ed) Handbook of Assessment Methods for Eating Behaviors and Weight-Related Problems: Measures, Theory, and Research Sage Publishing: Thousand Oaks, CA 1995 81–105.

  14. 14

    Breytspraak LM, McGee J, Conger JC, Whatley JL, Moore JT . Sensitizing medical students to impression formation processes in the patient interview J Med Educ 1997 52: 47–54.

  15. 15

    Maddox GL, Liderman, V . Overweight as a social disability with medical implications J Med Educ 1969 44: 214–220.

  16. 16

    Price JH, Desmond SM, Krol RA, Snyder FF, O'Connell JK . Family practice physicians' beliefs, attitudes, and practices regarding obesity Am J Prev Med 1987 3: 339–345.

  17. 17

    Bagley CR, Conklin DN, Isherwood RT, Pechiulis DR, Watson LA . Attitudes of nurses toward obesity and obese patients Percept Motor Skills 1989 68: 954.

  18. 18

    Peternelj-Taylor C . The effects of patient weight and sex on nurses' perceptions: a proposed model of nurse withdrawal J Adv Nurs 1989 14: 744–754.

  19. 19

    Oberrieder H, Walker R, Monroe D, Adeyanju UM . Attitude of dietetics students and registered dietitians toward obesity J Am Diet Assoc 1995 95: 914–916.

  20. 20

    Maiman L, Wang VL, Becker M, Finlay J, Simonson M . Attitudes toward obesity and the obese among professionals J Am Diet Assoc 1978 74: 331–336.

  21. 21

    Goldfield A, Chrisler JC . Body stereotyping and stigmatization of obese persons by first graders Percept Motor Skills 1995 81: 909–910.

  22. 22

    Crandall CS . Do parents discriminate against their heavyweight daughters? Personality Social Psychol Bull 1995 21: 724–735.

  23. 23

    Sitton S, Blanchard S . Men's preferences in romantic partners: obesity vs addiction Psychol Rep 1995 77: 1185–1186.

  24. 24

    Sobal J, Nicolopoulos V, Lee J . Attitudes about overweight and dating among secondary school students Int J Obes Relat Metab Disord 1995 19: 376–381.

  25. 25

    Wiederman MW, Hurst SR . Body size, physical attractiveness, and body image among young adult women: relationships to sexual experience and sexual esteem J Sex Res 1998 35: 272–281.

  26. 26

    Bellizzi JA, Klassen ML, Belonax JJ . Stereotypical beliefs about overweight and smoking and decision-making in assignments to sales territories Percept Motor Skills 1989 69: 419–429.

  27. 27

    Jasper CR, Klassen ML . Stereotypical beliefs about appearance: implications for retailing and consumer issues Percept Motor Skills 1990 71: 519–528.

  28. 28

    Pagan JA, Davila A . Obesity, occupational attainment, and earnings Social Sci Q 1997 78: 757–770.

  29. 29

    Popovich PM, Everton WJ, Campbell KL, Godinho RM, Kramer KM, Mangan MR . Criteria used to judge obese persons in the workplace Percept Motor Skills 1997 85: 859–866.

  30. 30

    Rothblum ED, Miller CT, Garbutt B . Sterotypes of obese female job applicants Int J Eat Disord 1988 7: 277–283.

  31. 31

    Bordieri JE, Drehmer DE, Taylor DW . Work life for employees with disabilities: recommendations for promotion Rehab Counseling Bull 1997 40: 181–191.

  32. 32

    Agell G, Rothblum ED . Effects of clients' obesity and gender on the therapy judgments of psychologists Prof Psychol Res Pract 1991 22: 223–239.

  33. 33

    Young LM, Powell B . The effects of obesity on the clinical judgments of mental health professionals J Health Soc Behav 1985 26: 233–246.

  34. 34

    Kimm SYS, Barton BA, Berhane K, Ross JW, Payne GH, Schreiber GB . Self-esteem and adiposity in Black and White girls: the NHLBI Growth and Health Study Ann Epidemiol 1997 7: 550–560.

  35. 35

    Lewinsohn PM, Graf M . Pleasant activities and depression J Consult Clin Psychol 1973 41: 261–268.

  36. 36

    Lutter JM, Jaffe L . The Bodywise Woman, 2nd edn Human Kinetics: Champaign, IL 1996.

  37. 37

    Grassi AT . Dietetic students' attitudes toward obese individuals based on the student's self-perceptions of weight Dissertation, Northern Illinois University: DeKalb, IL 1999.

  38. 38

    Rowland ML . A nomogram for computing body mass index Diet Curr 1989 16: 1–12.

  39. 39

    Nunnally JC . Psychometric Theory, 2nd edn McGraw-Hill: New York 1978.

  40. 40

    Myers DG . Social Psychology, 2nd edn McGraw-Hill: New York 1987.

  41. 41

    Crandall C, Biernat M . The ideology of anti-fat attitudes J Appl Social Psychol 1990 20: 227–243.

  42. 42

    Gelman D, Rosenberg D, Quade V, Roberts E, Senna D . Tune in, come out Newsweek 1993 122: 70–72.

  43. 43

    Oppenheimer M . ‘The inherent worth and dignity’: gay unitarians and the birth of sexual tolerance in liberal religion J Hist Sex 1996 7: 73–101.

  44. 44

    Stark P . Gays see gains in country radio Billboard 1997 109: 68.

Download references

Acknowledgements

Preparation of this article was supported in part by resources and assistance from the Program in Human Sexuality, Department of Family Practice and Community Health, University of Minnesota Medical School. Credit goes to Deb Finstad for data management, to Libby Frost and Priscilla Palm for manuscript preparation, and to Anne Marie Weber-Main for her comments on manuscript flow, readability, and content. Thanks also to Michael Wiederman for his advice and comments. We acknowledge the assistance of Melpomene Institute, St Paul, Minnesota (especially Lynn Jaffe), TOPS Chapters 507, 794 and 1282, and TOPS leaders and coordinators, Olive Kunz, Mavis Hornby and Diane Stoetzel, for allowing us to survey their members.

Author information

Correspondence to BE Robinson.

Appendix 1. F-Scale short form

Appendix 1. F-Scale short form

Appendix 1 F-Scale short form

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Bacon, J., Scheltema, K. & Robinson, B. Fat phobia scale revisited: the short form. Int J Obes 25, 252–257 (2001). https://doi.org/10.1038/sj.ijo.0801537

Download citation

Keywords

  • obesity
  • fat phobia
  • fat prejudice

Further reading