Bias Against Associates of the Obese

Research conducted by: Mikki Hebl and Laura Mannix

Case study prepared by: Emily Zitek


Obesity is a major stigma in our society. People who are obese face a great deal of prejudice and discrimination. For example, Roehling (1999) showed that obese people experience a lot of discrimination in the workplace (e.g., they are less likely to be hired and get lower wages). We know that people who are obese are stigmatized, but what about people who are somehow associated with an obese person? Neuberg et al. (1994) found that friends of gay men and lesbians suffer from "stigma by association". Perhaps the negative effects of the obesity stigma can also spread to other people. This study seeks to examine how the stigma of obesity can spread to a job applicant of average weight.

As part of a larger study, participants had to rate how qualified a particular job applicant was. This applicant was sitting by a woman. The researchers manipulated the following two variables: the weight of the woman and the relationship between the woman and the applicant. The woman was either obese or of average weight. This woman was also portrayed as being the applicant's girlfriend or a woman simply waiting to participate in a different experiment.

Questions to Answer
Are male applicants who are seated next to an obese woman rated as less qualified for a job? Are applicants who are seated next to their girlfriend rated differently from applicants seated next to a woman with whom they do not have an intimate relationship? Finally, does the effect of the type of relationship differ depending on the weight of the woman?

Design Issues
This study only looked how at how an obese woman seated next to a male job applicant could affect qualification ratings. Future research could address other gender combinations.

Descriptions of Variables
Variable Description
Weight The weight of the woman sitting next to the job applicant
1 = obese, 2 = average weight
Relate Type of relationship between the job application and the woman seated next to him
1 = girlfriend, 2 = acquaintance (waiting for another experiment)
Qualified Larger numbers represent higher professional qualification ratings


Hebl, M. R., & Mannix, L. M. (2003). The weight of obesity in evaluating others: A mere proximity effect. Personality and Social Psychology Bulletin, 29, 28-38.

Neuberg, S. L., Smith, D. M., Hoffman, J. C., & Russell, F. J. (1994). When we observe stigmatized and "normal" individuals interacting: Stigma by association. Personality and Social Psychology Bulletin, 20, 196-209.

Roehling, M. (1999). Weight-based discrimination in employment: Psychological and legal aspects. Personnel Psychology, 52, 969-1016.

  1. What is/are the independent variable(s) in this study? What is/are the dependent variable(s)?
  2. What is the mean rating of job qualification? What is the standard deviation? Does "qualified" appear to be normally distributed?
  3. If a participant from this sample were picked at random, what is the probability that he/she would have rated the qualification of the applicant as 7 or higher?
  4. Pretend the distribution of "qualified" is normally distributed (with same mean and standard deviation you computed in #2). Now what is the probability of selecting a participant who gave a rating of 7 or higher?
  5. Describe the design of this study. What type of ANOVA would you use? How many factors are there, and how many levels does each factor have?
  6. Make an interaction plot of the data. Based on this graph, do you think there are main effects? Do you think there is an interaction?
  7. Conduct a 2-way ANOVA.
    1. What is the best estimate of the population variance?
    2. Look at the main effect of "weight".
      • What are the marginal means?
      • What is the sum of squares for "weight"?
      • Is the main effect of "weight" significant?
    3. Look at the main effect of "relate".
      • What are the marginal means?
      • What is the sum of squares for "relate"?
      • Is the main effect of "relate" significant?
    4. Look at the interaction between "weight" and "relate".
      • How many degrees of freedom does the interaction have?
      • Is the interaction between "weight" and "relate" significant?
    5. What do these results mean?