This is an open access article distributed under the terms of the Creative Commons
Attribution License (Weight is a stigmatizing issue within romantic relationships. A large body of research reveals a general weight-bias, such that overweight peoplei are often devalued as potential romantic partners (see Puhl & Heuer, 2009). Indeed, overweight people are perceived as less physically (Chen & Brown, 2005) and relationally attractive (Kurzban & Weeden, 2007) than healthy-weight people. In an online dating paradigm, fewer men responded to a dating profile for an overweight woman than a recovering drug addict (Sitton & Blanchard, 1995). Furthermore, when asked to rank order a potential romantic partner’s desirability, people ranked an overweight person as the least sexually attractive partner, lower than partners who required the use of a wheelchair, had only one arm, had a history of mental illness, or had tested positive for a sexually transmitted disease (Chen & Brown, 2005).
As evidenced by the previous findings, weight-bias within romantic relationships has largely been approached from an outside perspective, assessing perceptions of people who are not in a relationship with an overweight partner. This is problematic because it reveals only a single perspective. Being a member of a marginalized relationship (i.e., relationships that are socially devalued and unapproved) may be stigmatizing in itself. Indeed, couples within other marginalized relationships, such as interracial or age-gap relationships, report less commitment than their non-marginalized counterparts (Lehmiller & Agnew, 2006). Assessing the stigma experienced by individuals within marginalized relationships is important because marginalized couples often report poorer health and lower self-esteem than non-marginalized couples (Lehmiller, 2012). Given the research on weight-bias and the stigma of being in a marginalized relationship, it is surprising that little research has assessed the stigma associated with being in a romantic relationship with an overweight person. To fill this gap in the literature, we assess people’s perceptions of romantic relationships comprised of one or more overweight partners (i.e., an outside perspective) as well as actual reports of people within these types of relationships (i.e., an inside perspective).
According to research on weight bias, relationship stigma may be greater among romantic couples comprised of at least one overweight partner, as compared to two healthy-weight partners.ii This weight bias hypothesis is consistent with a large literature suggesting that overweight people are the targets of greater prejudice and discrimination than healthy-weight people (Crocker, Major, & Steele, 1998). If weight-bias is directed towards overweight individuals, relationships with two overweight individuals should elicit greater weight bias than relationships with one overweight individual and one healthy-weight individual.
A robust literature suggests that a pro-thin, anti-fat weight-bias exists on an individual person level (Crandall, 1994). Two similar-weight partners essentially double the opportunities for people to express their approval or disapproval of each partner’s weight and relationship. More interactions, especially with close friends and family, may highlight the weight-bias experienced by similar-weight couples because close friends and family members are the greatest source of weight-related bias and stigma for obese people (Puhl & Heuer, 2009).
According to the weight-bias hypothesis, a romantic relationship consisting of two overweight individuals should be more stigmatizing than a relationship consisting of only one overweight individual. However, it is possible that a relationship comprised of two individuals with dissimilar body mass indexes (BMI), hereafter referred to as mixed-weight relationships, may be more stigmatizing than matched-weight relationships (i.e., similarly healthy-weight or overweight individuals). According to social comparison theory (Festinger, 1954), weight may be a particularly salient issue for mixed-weight couples. Such comparisons among mixed-weight couples reveal dissimilarity, such that the overweight partner is larger than the healthy-weight partner.
Other relationships marked by physical dissimilarities (e.g., interracial and age-gap relationships) elicit greater prejudice and discrimination than their similarly-matched counterparts (e.g., intraracial or age-similar relationships; Banks & Arnold, 2001; Gaines & Leaver, 2002). Recent research on mixed-weight couples reveal that people express greater prejudice and offer more discriminatory dating advice to mixed-weight couples than matched-weight couples, regardless of whether the matched-weight couple is healthy-weight or overweight (Collisson, Howell, Rusbasan, & Rosenfeld, 2016). People advised both healthy-weight and overweight men and women going on a date with a mixed-weight, rather than similar-weight, partner to avoid physical affection, delay introductions to friends and family, and invest less financially in the date. Furthermore, when people act as matchmakers they avoid pairing mixed-weight couples together and express greater dislike for mixed-weight, than similar-weight, relationships (Collisson et al., 2016). Given people’s negative attitudes towards mixed-weight relationships, it is not surprising that mixed-weight couples experience greater relationship conflict than similar-weight couples (Burke, Randall, Corkery, Young, & Butler, 2012).
According to social comparison theories, relationship stigma may be greater among mixed-weight, rather than matched-weight, couples because the relationship partners are comparing their own body to that of their partner’s. This mixed-weight hypothesis is consistent with recent findings suggesting that mixed-weight couples elicit greater prejudice and discrimination than matched-weight couples, regardless of whether the matched-weight couple is healthy-weight or overweight (Collisson et al., 2016). In the current research, we assess weight stigma within romantic relationships and pit two rival hypotheses against each other. According to the weight-bias hypothesis, relationships consisting of one or more overweight individuals should more stigmatizing than relationships between two healthy-weight individuals. According to the mixed-weight hypothesis, mixed-weight relationships should be more stigmatizing than matched-weight relationships, regardless of whether the matched-weight couple is healthy-weight or overweight.
The current research is important for a variety of reasons. First, numerous studies have documented the negative consequences of being the target of stigma (Crocker, Major, & Steele, 1998) as well as being within in a marginalized, stigmatizing relationship (Lehmiller & Agnew, 2006). Second, people’s perceptions of stigma may influence whether they wish to form a relationship with someone who is overweight. Given the importance of assessing weight-stigma among mixed-weight and matched-weight couples, we proposed two studies. Study 1 seeks to assess people’s perceptions of weight-stigma within romantic relationships. Study 2 seeks to assess actual reports of weight-stigma for people who are in mixed or matched-weight relationships. To provide convergent validity for our findings regarding stigma within romantic relationships, we assessed both relational stigma (i.e., the degree to which a relationship is socially devalued and unapproved) as well as weight-related distress (i.e., the degree to which weight is a concerning issue within a relationship) in both studies.
Study 1: Others’ Perceptions of Stigma
Method
Participants
Three hundred seventy-five online participants (216 men, 158 women, 3 no response, Mage = 31.40) were recruited from Amazon’s Mechanical Turk and paid $0.26 as remuneration. Amazon’s Mechanical Turk is a crowdsourcing website used to recruit a high-quality, diverse sample of American participants (Buhrmester, Kwang, & Gosling, 2011). The sample varied in regard to ethnicity, with 49% of participants identifying as Asian, 41% as Caucasian, 5% as Hispanic, and 4% as African American. Participants BMI was calculated using participant height and weight information (MBMI = 26.5). Fifty percent were healthy-weight (BMI < 25), 30% were overweight (BMI 25-29.99) and 29% were obese (BMI >30). Approximately 52% of the sample was married, 24% was single, and 23% was in a relationship but not currently married.
Materials and Procedure
Online participants were presented with an “Inferences about Romantic Couples” survey, which was ostensibly a follow-up to a previous study of men and women in romantic relationships. Using this minor deception paradigm, participants were asked to infer how a person in the previous study had answered his or her questionnaire about being in a relationship. For confidentiality reasons, participants were not provided with real names or pictures of the couple. Instead, participants were presented with basic demographic information (e.g., gender, height, weight) and avatar pictures that the couple had presumably chosen to represent themselves. See Figure 1 for the avatar pictures used to represent the similar or mixed-weight couple.
Figure 1
Avatar pictures representing similar and mixed-weight romantic couples.
Note. The target person participants were asked to make inferences about was circled in red.
Participants were randomly assigned to make inferences about a male or female target person whose height, weight, and avatar suggested that he or she was healthy-weight (BMI = 18.5) or overweight (BMI > 25; Stunkard, Sørensen, & Schulsinger, 1983). Specific BMI numbers for each person in the relationship were not provided to participants. Rather, participants deduced that each person in the relationship was healthy-weight or overweight by looking at each person’s weight, height, and picture. The target person, circled in the avatar picture, was randomly presented to be in a relationship with a partner that was healthy-weight or overweight. Thus, participants made inferences about a target person who was in a similar-weight or mixed-weight relationship.
After reviewing the couple’s information, participants were asked to infer how the target person completed his or her questionnaire. Participants first inferred relational stigma (Gamarel, Reisner, Laurenceau, Nemoto, & Operario, 2014). Relational stigma, the degree to which a person’s relationship is devalued by society, was assessed by eight items, on a five-point scale (1 – Never, 5 – Always; α = .93). An example item is, “To what extent did he or she report feeling uncomfortable going out with his/her partner in public?” We also created five items to assess weight-related discomfort with the relationship. Each item asked participants to indicate the degree to which he or she feels uncomfortable being in a relationship because of weight while, (1) exercising/engaging in physical activity, (2) eating/dining out, (3) buying/trying on clothes, (4) wearing bathing suits/swimming, (5) discussing healthy food choices.iii After inferring how the target person answered his or her questionnaire, participants completed basic demographic information (e.g., age, gender, ethnicity, relationship status) about themselves including height and weight.
Results
First we reverse-coded appropriate items and created average scores for the dependent variables of relational stigma and weight-related discomfort. Then, we conducted a 2 (target gender: male, female) X 2 (target weight: healthy-weight, overweight) X 2 (partner weight: healthy-weight, overweight) between-persons factorial ANCOVA, with participant BMI as a covariate, for each dependent variable. Participant BMI was included as a covariate because relative weight may have influenced participants’ inferences of targets with varying weights.
As seen in Table 1, relational stigma significantly varied as a function of target gender, target weight, and partner weight, F(8,281) = 4.86, p < .001, partial ɳ2 = .12. Participant BMI did not emerge as a significant covariate, F(1,281) = 0.41, p = .52, partial ɳ2 < .01.There was no main effect of target gender, F(1,281) = 0.56, p = .46, partial ɳ2 < .01. However, there were significant main effects of target weight, F(1,281) = 12.56, p < .001, partial ɳ2 = .04, and partner weight, F(1,281) = 18.62, p < .001, partial ɳ2 = .06. People inferred that overweight targets (M = 2.73, SD = 0.80) experienced more stigma than healthy-weight targets (M = 2.45, SD = 0.97). They also inferred that a person in a relationship with an overweight partner (M = 2.78, SD = 0.79) experienced more stigmatized than a person with a healthy-weight partner (M = 2.39, SD = 0.97). Importantly, there was a significant interaction between target and partner weight, F(1,281) = 10.45, p < .01, partial ɳ2 = .04. People inferred that a healthy-weight target experienced less stigma when in a relationship with a healthy-weight partner (M = 2.50, SD = 1.21) than overweight partner (M = 2.80, SD = 0.88), t(194) = -2.01, p < .05, d = 0.28. People did not infer that the stigma experienced by an overweight person differed when in a relationship with a healthy-weight (M = 2.78, SD = 0.87) or overweight partner (M = 2.89, SD = .83), t(179) = -0.85, p = .39, d = 0.13. There was no three-way interaction between gender, target weight, and partner weight, F(1,281) = 1.03, p = .31, partial ɳ2 < .01.iv
Table 1
Study 1 - Relational Stigma as a Function of Target and Partner Weight.
| Partner-Weight | Target-Weight
|
|||
|---|---|---|---|---|
| Healthy-Weight
|
Overweight
|
|||
| M | SD | M | SD | |
| Healthy-Weight | 2.50 | 1.21 | 2.78 | 0.87 |
| Overweight | 2.80 | 0.88 | 2.89 | 0.83 |
Furthermore, weight-related discomfort did significantly vary as a function of target gender, target weight, and partner weight, F(8, 281) = 9.36, p < .001, partial ɳ2 = .21. Participant BMI did not emerge as a significant predictor, F(1,281) = 0.21, p = .65, partial ɳ2 < .01. There was no significant main effect of gender, F(1, 281) = .65, p = .42, partial ɳ2 < .01. There was a significant main effect of target weight, F(1, 281) = 40.89, p < .001, partial ɳ2 = .13, such that people inferred that an overweight target experienced more weight-related discomfort (M = 3.19, .87) than a healthy-weight target (M = 2.57, 1.10). There was also a significant main effect of partner weight, F(1, 281) = 18.38, p < .001, partial ɳ2 = .06, such that a person in a relationship with an overweight partner experienced more weight-related discomfort (M = 3.06, SD = .90) than someone with a healthy-weight partner (M = 2.69, SD = 1.14). There was also a significant interaction between target and partner weight, F(1, 281) = 23.88, p < .001, partial ɳ2 = .08. Further analyses revealed that people inferred that a healthy-weight target experienced less weight-related discomfort when in a relationship with a healthy-weight partner (M = 2.50, SD = 1.25) than overweight partner (M = 2.99, SD = .97), t(194) = -3.07, p < .01, d = .44. However, people did not infer that an overweight target experience less weight-related discomfort when in a relationship with an overweight partner (M = 3.49, SD = .86) than a healthy-weight partner (M = 3.26, SD = .89), t(179) = .50, p = .62, d = .26. There was no three-way interaction between target gender, weight, and partner weight, F(1, 281) = 1.71, p = .19, partial ɳ2 < .01.
Discussion
Perceptions of stigma largely supported the weight-bias hypotheses. People inferred that a healthy-weight person experiences less relational stigma and weight-related discomfort than an overweight person. People inferred that an individual dating a healthy-weight, rather than overweight, person experienced less relational stigma and weight-related discomfort. The mixed-weight hypothesis was only partially supported. People inferred that a healthy-weight person experienced less relational stigma and weight-related discomfort when dating a healthy-weight, as compared to overweight, partner. People did not infer that an overweight person experienced a different amount of stigma as function of partner weight. Although Study 1 supports the weight-bias hypothesis more than the mixed-weight hypothesis, it is possible that people’s inferences do not match reality. To address whether people are accurate in their perceptions, Study 2 assessed the actual relational stigma and weight-related discomfort experienced by actual couples who are in similar or mixed-weight relationships.
Study 2: Self-Reported Stigma
Method
Participants
Three hundred ninety-five online participants (241 men, 150 women, four no response, Mage = 40.75) were recruited from Amazon’s Mechanical Turk and paid $0.26 as remuneration. The study was limited to participants currently in a romantic relationship. Approximately 53% of the sample was married and 47% was in a relationship but not currently married. The sample varied in regard to ethnicity, with 50% of participants identifying as Asian, 38% as Caucasian, 5% as African American, and 4% as Hispanic. Participants BMI was calculated using participant height and weight information (MBMI = 25.90). Fifty-two percent were healthy-weight (BMI < 25), 30% were overweight (BMI 25-29.99) and 28% were obese (BMI >30). The average partner’s BMI score was 24.94.
Materials and Procedure
Participants currently in a romantic relationship were asked to participate in an online survey regarding romantic relationships. They were asked to complete a questionnaire that included the same scales used in Study 1 to assess the amount of relational stigma (Gamarel, Reisner, Laurenceau, Nemoto, & Operario, 2014) and weight-related discomfort that they personally experience.v After completing the questionnaire, participants answered basic demographic information (e.g., gender, age, ethnicity, relationship status) and then indicated the height and weight of themselves and their partner.
Results
As in Study 1, average composite scores were created for the dependent variables of relational stigma and weight-related discomfort. Using the recommendations of the Center for Disease Control, we created BMI scores for each participant and his or her relationship partner. Next, we created a dichotomous target weight variable by categorizing participants’ BMI scores as healthy-weight (BMI < 24.99) or overweight (BMI > 25). We also created a dichotomous partner weight variable by categorizing participants’ partners’ BMI scores as healthy-weight or overweight. Finally, we conducted a 2 (target gender: male, female) X 2 (target weight: healthy-weight, overweight) X 2 (partner weight: healthy-weight, overweight) between-persons factorial ANOVA for each dependent variable.
First, we tested whether relational stigma depended on target gender, target weight, and partner weight. As seen in Table 2, results confirmed our hypothesis, F(7, 276) = 3.49, p < .01, partial ɳ2 = .08. There was a main effect of gender, F(1, 276) = 9.57, p < .01, partial ɳ2 = .03, such that men reported greater relational stigma (M = 1.93, SD = .88) than women (M = 1.58, SD = .710). There was no main effect of target weight, F(1, 276) = 0.67, p = .42, partial ɳ2 < .01, nor partner weight, F(1, 276) = .827, p = .36, partial ɳ2 < .01. As hypothesized, there was a significant interaction between target and partner weight, F(1, 276) = 4.38, p < .05, partial ɳ2 = .01. Further analyses reveal that the relational stigma experienced by a healthy-weight target does not significantly differ with an overweight partner (M = 1.69, SD = .81) or healthy-weight partner (M = 1.90, SD = .81), t(176) = 1.36, p = .18, d = 0.26. However, the stigma experienced by an over-weight target is significantly more with an overweight partner (M = 1.87, SD = .94) than a healthy-weight partner (M = 1.58, SD = .68), t(131) = -2.05, p < .05, d = 0.35. There was no significant three-way interaction between target gender, target weight, and partner weight, F(1, 246) = 1.42, p = .24, partial ɳ2 = .01.
Table 2
Study 2 - Relational Stigma as a Function of Target and Partner Weight.
| Partner-Weight | Target-Weight
|
|||
|---|---|---|---|---|
| Healthy-Weight
|
Overweight
|
|||
| M | SD | M | SD | |
| Healthy-Weight | 1.90 | 0.81 | 1.58 | 0.68 |
| Overweight | 1.69 | 0.81 | 1.87 | 0.94 |
Second, we tested whether weight-related discomfort varied as a function of target gender, target weight, and partner weight. Results confirmed our hypothesis, F(1, 276) = 3.13, p < 014, partial ɳ2 = .07. There was a main effect of target gender, F(1, 276) = 6.72, p = .01, partial ɳ2=.02, such that men experienced more weight-related discomfort (M = 2.00, SD = .93) than women (M = 1.69, SD = .83). There was no main effect of target weight, F(1, 276) = 0.04, p = .85, partial ɳ2 < .01. There was no main effect of partner weight, F(1, 276) = 3.43, p = .06, partial ɳ2 = .02. There was a significant interaction between target weight and partner weight, F(1, 276) = 4.68, p < .05, partial ɳ2 = .02. Follow up analyses reveal that a healthy-weight person does not experience any more discomfort when in a relationship with an overweight partner (M = 1.81, SD = .90) than healthy-weight partner (M = 1.92, SD = .90), t(150) = 0.68, p = .50, d = 0.12. However, an overweight person experiences significantly more weight-related discomfort when in a relationship with an overweight partner (M = 2.06, SD = .98) than healthy-weight partner (M = 1.63, SD = .79), t(131) = -2.72, p < .01, d = 0.48. There was not a significant three way interaction between target gender, target weight, and partner weight, F(1, 276) = 1.15, p = .28, partial ɳ2 < .01.
Discussion
Findings from Study 2 were mixed but generally supportive of the weight-bias hypothesis. People with an overweight, rather than healthy-weight, partner reported greater weight-related discomfort when they are in a relationship. Overweight people dating another overweight, rather than healthy-weight, partner reported greater relational stigma and weight-related discomfort.
General Discussion
In two studies, we assessed the perceived and actual stigma experienced by similar and mixed-weight romantic couples. To predict the amount of weight-related stigma experienced by romantic couples, we pitted two hypotheses against each other. According to the weight-bias hypothesis, relationship stigma is predicted to be greater among romantic couples comprised of at least one overweight partner, as compared to two healthy-weight partners. Consistently, the weight-bias hypothesis also predicts that two overweight individuals should experience greater relational stigma than individuals within mixed-weight relationships. However the mixed-weight hypothesis offers a starkly different prediction. According to the mixed-weight hypothesis, mixed-weight couples should experience greater relational stigma than matched-weight couples, regardless of whether the matched-weight couple is healthy-weight or overweight.
Across two studies, the weight-bias hypothesis received greater support than the mixed-weight hypothesis. In Study 1, people inferred that overweight people and their partners experienced greater relational stigma and weight-related discomfort than healthy-weight people. Furthermore, people inferred that an overweight person in a relationship with an overweight, as compared to healthy-weight, partner would experience the greatest relational stigma and weight-related discomfort. In Study 2, only an overweight person in a relationship with an overweight, as compared to healthy-weight, partner reported greater relational stigma and weight-related discomfort.
These novel and important findings may have clear implications regarding the formation and maintenance of romantic relationships. Study 1 suggests that a healthy-weight person may avoid or dislike the idea of forming a relationship with an overweight partner because of perceived stigma and discomfort regarding weight. People believe that situations such as trying on clothing, engaging in physical exercise, or discussing food will be uncomfortable for any couple that consists of at least one overweight person. People’s negative inferences may help explain people’s apprehension at forming a relationship with an overweight partner (see Puhl & Heuer, 2009).
Contrary to people’s inferences, actual couples consisting of at least one overweight person often did not report greater stigma and weight-related discomfort, as predicted in Study 1. Indeed, only an overweight person in a relationship with an overweight, as compared to healthy-weight, partner reported more stigma and discomfort. This finding may be explained by two possibilities. First, people may simply be incorrect in their inferred level of stigma. It is unclear whether people are basing their inferences upon negative weight-stereotypes or personal dating experiences. It is possible that if people actually formed a mixed-weight relationship, they would learn the true amount of stigma and discomfort these couples may experience. Second, people may be correct in their inferences, but only for couples with at least one overweight individual who have already chosen to break up (i.e., end their relationship). In Study 2, the stigma and discomfort reported was limited to people currently in romantic relationships. Therefore, the stigma and discomfort reported by actual mixed-weight couples who are still committed may have been disproportionately lower than that reported by mixed-weight couples who have already chosen to end their relationship. Future research can address this limitation and speak to the degree to which similar and mixed-weight relationships differ in regard to relationship commitment.
Furthermore, additional studies are needed to better understand the social comparison and prejudicial mechanisms underlying why people may misperceive the weight stigma and discomfort experienced within relationships. We speculate that people’s weight biases lead them to infer that any couple that consists of at least one overweight partner experiences stigma and weight-related discomfort. This explanation suggests that people’s inferences may be a better reflection of their own personal biases and prejudices than actual representations of the couple. Future research that assesses implicit and explicit biases against mixed-weight couples may be a fruitful avenue for future research. Furthermore, the current studies did not assess participants own relationship quality, duration, satisfaction, or ethnic composition. Future research may choose to further identify whether aspects of participants’ own relationships and match of partner ethnicity may affect their perceptions of others’ relationships.
Finally, perceptions of, and actual reported, weight stigma have clear implications for people who are currently, or who have the opportunity to be, in relationships with an overweight person. For instance, mixed-weight couples should be cognizant that others outside of their relationship perceive that the couple is a poor fit (Collisson et al., 2016). As shown in the current research, people infer that mixed-weight couples, and similarly overweight couples, are the target of stigma and weight-related discomfort. As a result, people may be less supportive of mixed-weight relationships, at best, or active in their dissolution, at worst. Future research that addresses the factors that contribute to relationship conflict (Burke et al., 2012), commitment, and dissolution may shed more light on stigma of weight within relationships.