Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Mar 1.
Published in final edited form as: Body Image. 2020 Dec 8;36:172–179. doi: 10.1016/j.bodyim.2020.11.009

A test of the Tripartite Influence Model of disordered eating among men

Lauren M Schaefer 1, Rachel F Rodgers 2, J Kevin Thompson 3, Scott Griffiths 4
PMCID: PMC7987734  NIHMSID: NIHMS1653523  PMID: 33307310

Abstract

Studies of the Tripartite Influence Model conducted with female samples suggest that thinness-oriented pressures, internalization, and appearance comparisons may be risk factors for disordered eating. However, limited work has investigated this model among men utilizing measures specifically designed to assess both muscularity- and thinness-oriented processes, which characterize male body image concerns. To address this gap, the current study examined the Tripartite Influence Model of disordered eating among men using a battery of such measures. A convenience sample of 265 male undergraduates completed validated, self-report measures of disordered eating, muscularity- and thinness-oriented appearance pressures, internalization, and appearance comparisons. Path analyses supported a slightly modified version of the Tripartite Influence Model, indicating direct and indirect paths from sociocultural pressures to disordered eating via internalization and appearance comparisons. These results provide support for a version of the Tripartite Influence Model among men. Specifically, muscularity- and thinness-oriented pressures appear to be related to disordered eating among college men via indirect pathways involving thin internalization, muscular internalization, and appearance comparisons. Future work is needed, however, to examine whether the model would generalize to muscularity-oriented disordered eating.

Keywords: Men, sociocultural influences, internalization, appearance pressures, appearance comparisons

1. Introduction

Although early clinical descriptions of eating pathology acknowledged the presence of eating concerns among boys and men (Morton, 1694), subsequent research examining associated risk factors has utilized predominantly female samples (Murray et al., 2017). As more recent work highlights the importance of studying eating concerns among boys and men (Brown, Griffiths, & Murray, 2018; Lipson & Sonneville, 2017; Murray et al., 2017), including potentially divergent etiological pathways (Allen, Byrne, Crosby, & Stice, 2016; Griffiths, Murray, & Touyz, 2013), examination of established etiological theories using appropriate measurement tools designed to capture the distinct experiences of boys and men is needed. Such work would help to inform intervention efforts for this currently underserved group (Coffino, Udo, & Grilo, 2019; Griffiths, Mond, Gunatilake, Murray, Sheffield, & Touyz, 2015).

Sociocultural models of disordered eating (e.g., the Tripartite Influence Model; Thompson Heinberg, Altabe, & Tantleff-Dunn, 1999) offer useful etiological frameworks for examining the ways in which social pressures to achieve appearance ideals are implicated in the development and maintenance of eating disturbance. The Tripartite model suggests that appearance pressures from three sociocultural agents (i.e., media, family members, and peers) lead to the internalization (or, personal acceptance) of socially-constructed appearance ideals, which then motivate comparisons of one’s own appearance to the idealized standards or to the appearance of others as a means of continuously evaluating one’s own appearance (Rodgers, McLean, & Paxton, 2015). As these comparison are often unfavorable (Rogers, Fuller-Tyszkiewicz, Lewis, Krug, & Richardson, 2017), they are theorized to increase the likelihood of experiencing weight and shape concerns, and engaging in disordered eating behaviors in order to modify one’s appearance (Thompson et al., 1999).

Consistent with the field’s historical focus on women’s body image and disordered eating patterns, sociocultural models, including the Tripartite Influence Model, have so far been disproportionately explored among girls and young women, with results providing converging support for their usefulness in these populations (Keery, Van den Berg, & Thompson, 2004; Papp, Urbán, Czeglédi, Babusa, & Túry, 2013; Rodgers et al., 2011; Shroff & Thompson, 2006). However, male appearance pressures and ideals differ considerably from female appearance pressures and ideals (Thompson & Cafri, 2007), raising the need for gender-appropriate adaptations to both theory and measurement for this group. Specifically, while female appearance ideals generally center on achieving a thin physique and primarily drive weight loss behaviors, appearance ideals for boys and men increasingly emphasize achieving a low percentage of body fat combined with large well-defined musculature (Griffiths et al., 2013). This dual emphasis on both muscularity and low body fat for boys and men is thought to drive a combination of problematic behaviors that include attempts at increasing muscle size (e.g., strenuous weight-lifting, high calorie/protein consumption) and reducing adiposity (e.g., extreme dietary restriction). Therefore, clarification of the role of both thinness- and muscularity-oriented appearance pressures, internalization, and comparison processes in contributing to eating pathology among boys and men is an important area of inquiry. While previous studies have provided partial support for sociocultural models of disordered eating among boys and men (e.g., Tylka, 2011), this work has been limited. Below, we review this existing literature base, and discuss key limitations, which we attempt to address in the current study.

A number of studies have utilized adapted versions of the Tripartite Influence Model to examine predictors of muscularity-oriented body dissatisfaction (e.g., Karazsia & Crowther, 2009), drive for muscularity (Stratton, Donovan, Bramwell, & Loxton, 2015), and muscle-building behaviors (e.g., Smolak, Murnen, & Thompson, 2005) among samples of men. However, only a subset of published studies has utilized the full model to identify predictors of disordered eating in males. In a series of studies by Tylka and colleagues (2011, 2012), researchers expanded the traditional “tripartite” model to include a fourth hypothesized sociocultural influence: romantic partners. Analyses among a sample of college men identified significant pathways from family and media pressures to obtain a mesomorphic appearance (i.e., a body which is both lean and muscular) to internalization of the mesomorphic ideal. Internalization then exhibited direct pathways to muscularity and body fat dissatisfaction, which ultimately were linked to muscularity enhancement behaviors and disordered eating behaviors, respectively (Tylka, 2011). Among a sample of gay men, Tylka and Andorka (2012), revealed direct pathways from partner and media pressure to mesomorphic ideal internalization, which in turn, predicted increased frequency of general appearance comparisons. Internalization and comparisons then exhibited differential relationships with body fat and muscularity dissatisfaction, which ultimately were linked to disordered eating behaviors and muscularity enhancement behaviors, respectively. Among French adolescent boys, Rodgers and colleagues (2012) illustrated that pressures to lose weight and gain muscle contributed to media ideal internalization and general appearance comparisons (modeled together as a single variable), which ultimately were linked to drives for thinness and muscularity, and finally disordered eating. Finally, using a sample of French college men, Girard and colleagues (2018) showed that media influences, as well as thinness and muscularity pressures from peers, were associated with general appearance comparisons, which in turn were associated with internalization of thinness and muscularity. Thinness internalization then was linked to disordered eating via body fat dissatisfaction, while muscularity internalization was directly linked to engagement in muscle enhancing behaviors.

As illustrated above, previous work suggests that adapted versions of the Tripartite Influence Model may offer a useful framework for understanding male body image and eating concerns. However, important gaps remain. Key among them is the fact that until recently, few measurement tools specifically designed to assess muscularity-oriented appearance pressures, internalization, and comparison processes had undergone strenuous psychometric evaluation among men. In lieu of such measures, each of the studies referenced above was required to either adapt preexisting measures that had originally been developed for female respondents to reflect male ideals, or to use the unadapted scales. In either case, the scales did not undergo rigorous empirical development/validation, raising questions about the psychometric integrity of the measures and, ultimately, the validity of the results. More recently, researchers have carefully developed and validated measures of appearance pressures, appearance ideal internalization, and appearance comparisons with the explicit intention of capturing both thin and muscular appearance ideals among men (Schaefer, Harriger, Heinberg, Soderberg, & Thompson, 2017; Schaefer & Thompson, 2018). The availability of these scales therefore raises the opportunity to test the Tripartite Influence Model among boys and men using measures that have been designed for and validated within male samples. In addition to potential problems of measurement, previous work frequently combined theoretically distinct constructs into a single latent variable (e.g., internalization and appearance comparisons; Rodgers et al., 2012), omitted key variables from the model (e.g., appearance comparison; Tylka, 2011), did not test hypothesized indirect pathways (e.g., Girard, 2018), or tested the model in samples of minority groups (e.g., Tylka & Andorka, 2012) that demonstrate elevated rates of disordered eating and unique patterns of risk (Calzo, Blashill, Brown, & Argenal, 2017; Miller & Luk, 2019), leaving questions about the strength of model pathways among broader samples of non-minority males.

Therefore, in the present study, we sought to build upon the existing literature base by testing a sociocultural model of disordered eating (see Figure 1) among men using recently validated assessment tools designed to capture both muscularity- and leanness-oriented interpersonal (i.e., sociocultural pressures) and intrapersonal (i.e., appearance ideal internalization, appearance comparisons) experiences. Building off of prior research in men (e.g., Girard et al., 2018; Tylka, 2011), we aimed to test a version of the Tripartite Influence Model of eating disturbance (Thompson et al., 1999), which suggests that perceived appearance pressures for low body fat and muscularity from four sources (e.g., peers, family, significant others, media) contribute to the internalization of thinness and muscularity. This internalization is then proposed to motivate appearance comparisons regarding both weight/shape and muscularity, which in turn is linked to eating disordered cognitions and behaviors.

Figure 1.

Figure 1.

Open in a new tab

Hypothesized path model.

2. Method

2.1. Participants

Participants were 265 male undergraduates ranging in age from 18 to 30 years (M = 20.79, SD = 2.67). Participants identified as Caucasian (48%), Hispanic (19.2%), Black (9.1%), Asian (12.8%), or multiracial/other (10.6%). Most of the sample identified as heterosexual (94.7%), followed by gay (3.8%) and bisexual (1.5%). Average body mass index (BMI; kg/m2) was 25.44 (SD = 5.09) and was calculated based on participants’ self-reported height and weight. Self-reported height and weight have been found to be highly correlated with objectively measured height and weight (rs = .98 to .99; Pursey et al., 2014).

2.2. Measures

2.2.1. Sociocultural Attitudes Towards Appearance-Questionnaire-4-Revised-Male (SATAQ-4R-Male).

The SATAQ-4R (Schaefer et al., 2017) contains seven subscales assessing appearance pressures and internalization of predominant appearance ideals. The authors offer separate versions of the scale for males and females, which were independently developed and validated in male and female samples, respectively. Although both versions share 20 identical items, the male scale, labeled SATAQ- 4R-Male, contains eight items not present in the female version; the female scale, labeled the SATAQ-4R-Female, contains 11 items not present in the male version. Consistent with male appearance ideals, the SATAQ-4R-Male version places a stronger emphasis on muscularity pressures than the SATAQ-4R-Female. However, both measures separately assess a desire for thinness/low body fat and muscularity. The current study utilized six SATAQ-4R-Male subscales reflecting a desire for a thinness (Internalization: Thin/Low Body Fat; example item: “I think a lot about looking thin”), a desire for muscularity (Internalization: Muscular; example item: “I want my body to look muscular”), and perceived pressures from family (Pressures: Family: example item: “I feel pressure from my family members to increase the size or definition of my muscles”), peers (Pressures: Peers; example item: “I feel pressure from my peers to be more muscular”), media (Pressures: Media; example item: “I feel pressure from the media to increase the size or definition of my muscles”), and significant others (Pressures: Significant Others; example item: “I feel pressure from significant others to decrease my level of body fat”) to improve one’s appearance and achieve a thin and muscular physique. Respondents indicate their level of agreement with each statement using a 5-point Likert scale ranging from 1 (definitely disagree) to 5 (definitely agree). Higher scores indicate greater internalization and perceived pressures. The measure has demonstrated good construct validity and reliability in previous work with men (Schaefer et al., 2017). In the current sample, Cronbach’s alpha for each of the subscales ranged from .74 (Internalization: Thin/Low Body Fat) to .96 (Pressures: Media).

2.2.2. Physical Appearance Comparison Scale-3 (PACS-3).

The PACS-3 (Schaefer & Thompson, 2018) comprises 27 items and 9 subscales assessing appearance comparison processes. The current study utilized three subscales assessing an individual’s frequency of engaging in comparisons of weight and shape to peers (Proximal: Frequency; example item: “When I’m out in public, I compare my weight/shape to the weight/shape of others”) and media figures (Distal: Frequency; example item: “When I watch television, I compare my weight/shape to the weight/shape of the actors/actresses”), as well as their frequency of engaging in comparisons of muscularity to peers and media figures (Muscular: Frequency; example item: “When I’m out in public, I compare my muscularity to the muscularity of others”). Respondents indicate their level of agreement with each statement using a 5-point Likert-type scale ranging from 1 (never) to 5 (almost always). Higher scores indicate greater frequency of appearance comparisons. Consistent with recommendations from the scale developers, the current study utilized a composite score (Total Frequency) reflecting the overall frequency of weight, shape, and muscularity comparisons (i.e., mean of Proximal: Frequency, Distal: Frequency, and Muscular: Frequency subscales). The PACS-3 has demonstrated good construct validity and reliability in previous work with men (Schaefer & Thompson, 2018). In the current sample, Cronbach’s alpha for the PACS-3 Total Frequency score was .93.

2.2.3. Eating Disorder Examination – Questionnaire (EDEQ).

The EDEQ (Fairburn & Beglin, 2008) is a 28-item measure assessing disordered eating attitudes and behaviors over the last 28 days (example item: “Have you tried to follow definite rules regarding your eating [e.g., a calorie limit] in order to influence your shape or weight [whether or not you have succeeded]?”). Items are rated on a 7-point Likert-type scale ranging from 0 (no days/not at all) to 6 (everyday/markedly). The current study utilized the global score, representing an average of the four subscale scores (Restraint, Eating Concern, Shape Concern, and Weight Concern). Therefore, the EDEQ global score may be thought of as comprising both the “body image” component of eating pathology, as well as the more behavioral components. Higher global scores signal greater levels of eating pathology. The EDEQ global score has demonstrated good reliability and construct validity in male samples (Schaefer et al., 2018). In the current sample, internal consistency for the global score was .94.

2.3. Procedure

Participants were recruited through the undergraduate research pool (i.e., students enrolled in psychology courses who were incentivized to participate in research studies in exchange for course credit) at a large university in the southeastern United States to participate in a study examining appearance attitudes. Eighty-three percent of participants responded to questionnaires online, while the remaining 17% completed paper and pencil measures in the presence of a research assistant. Participants who completed measures in-person versus online did not differ with regard to age, ethnicity, sexual orientation, or BMI (all p-values > .05). Upon completion of the study, all participants were debriefed and received course credit as compensation. The study received approval from the institutional review board at the University of South Florida.

2.4. Data Analyses

The average amount of missing item-level data across our study variables was low at 0.4%. Skewness for all variables was 1.02 or less, with most below 0.6. Model fit was nearly identical for path models using full information maximum likelihood estimation and using estimators with Satorra-Bentler corrections (i.e., maximum likelihood with robust standard errors), suggesting the data were multivariate normal (Byrne, 2012). Therefore, variables were not transformed. Means, standard deviations, and intercorrelations for all scales were calculated. Path analysis with maximum likelihood estimation was utilized to examine the hypothesized model. Model fit was evaluated using the comparative fit index (CFI), Tucker-Lewis index (TLI), standardized root mean square residual (SRMR), root mean square error of approximation (RMSEA), and chi-square. Guidelines for interpreting CFI and TLI values are as follows: values less than .85 indicate poor fit, values between .85 and .90 indicate mediocre fit, values between .90 and .95 indicate acceptable fit, values between .95 and .99 very good fit, values greater than .99 indicate outstanding fit (Little, 2013). Guidelines for interpreting RMSEA and SRMR values are as follows: values greater than .10 indicate poor fit, values between .10 and .08 indicate mediocre fit, values between .08 and .05 indicate acceptable fit, values between .05 and .02 indicate good fit, values less than .01 indicate outstanding fit (Little, 2013). Although chi-square is often significant with larger sample sizes, smaller chi-square values indicate improved fit with non-significant chi-square values indicating good fit (Tabachnick & Fidell, 2009). Chi-square difference testing was utilized to identify statistically significant improvements in model fit. Modification indices were used to examine sources of model misfit and identify potential improvements to the model. In the process of model adjustment, both the statistical and theoretical value of suggested modifications were considered. Indirect effects were tested using a bias-corrected bootstrapping method. Estimates of indirect effects were evaluated using bootstrapped confidence intervals based on 10,000 resamples. Indirect effects are considered statistically significant when the 95% confidence interval does not include zero. All analyses were conducted using MPlus version 8.1.

3. Results

Means, standard deviations, and intercorrelations among variables are presented in Table 1. In general, we observed small positive associations between the pressures subscales and the internalization subscales, with the exception that family pressures were not significantly associated with muscular internalization. Internalization subscales exhibited small to medium positive associations with appearance comparisons. Small to medium positive associations were also observed between disordered eating and appearance comparisons, muscular internalization, and thin internalization.

Table 1.

Means, Standard Deviations, and Intercorrelations Among Study Variables

Mean (SD) 1 2 3 4 5 6 7
1. Thin/Low Body Fat Internalization 2.59 (0.97) –
2. Muscular Internalization 3.61 (0.87) .18** –
3. Family Pressures 2.08 (1.00) .22*** .11 –
4. Peer Pressures 2.39 (1.11) .17** .26*** .53*** –
5. Significant Other Pressures 2.09 (1.04) .26*** .17** .53*** .57*** –
6. Media Pressures 2.61 (1.21) .26*** .26*** .30*** .35*** .38*** –
7. Appearance Comparisons 2.38 0.94) .15* .43*** .09 .22*** .18** .49*** –
8. Disordered Eating 1.14 (1.11) .40*** .25*** .39*** .33*** .39*** .36*** .32***
Open in a new tab

Note. Thin/Low Body Fat Internalization = SATAQ-4R-Male Internalization: Thin/Low Body Fat subscale; Muscular Internalization = SATAQ-4R-Male Internalization: Muscular subscale; Family Pressures = SATAQ-4R-Male Pressures: Family subscale; Peer Pressures = SATAQ-4R-Male Pressures: Peer subscale; Significant Other Pressures = SATAQ-4R-Male Pressures: Significant Others subscale; Media Pressures = SATAQ-4R-Male Pressures: Media subscale; Appearance Comparisons = Physical Appearance Comparison Scale-3 Total Frequency score; Disordered Eating = Eating Disorder Examination Questionnaire Global Score.

*

p < .05.

**

p < .01.

***

p < .001.

Results from the path analysis indicated that the model provided less than acceptable fit to the data: χ2(9) = 97.44, p < .001, CFI = .66, TLI = .18, RMSEA = .19 (90% confidence interval = .01, .11), SRMR = .10. The RMSEA statistic indicated a negligible probability of close fit (p RMSEA ≤ .05 = .000). Modification indices suggested the addition of two direct paths from (a) media pressures to appearance comparisons and (b) from family pressures to disordered eating. Importantly, these pathways were theoretically consistent with the Tripartite Influence Model’s assertion that cultural pressures increase the likelihood of appearance comparisons and disordered eating. Therefore, their inclusion in the modified model was both statistically and theoretically justified. The addition of these pathways significantly improved model fit according to the chi-square difference test, Δχ2(2, N = 265) = 83.28, p < .05, and resulted in acceptable to very good fit across all examined fit indices, χ2(7) = 14.16, p = .05, CFI = .97, TLI = .91, RMSEA = .06, SRMR = .03. The RMSEA statistic indicated a high probability of close fit (p RMSEA ≤ .05 = .284). No further adjustments were suggested by the modification indices. Figure 2 presents the standardized path coefficients for the final model.

Figure 2.

Figure 2.

Open in a new tab

Final path model with standardized regression coefficients. Solid lines indicate significant pathways, dashed lines indicate non-significant pathways. Bolded lines indicate significant indirect pathways. *p < .05.

Summary statistics for the total, direct, and indirect paths are provided in Table 2. Family pressures demonstrated a direct path to disordered eating, while peer pressures demonstrated a direct path to muscular internalization, and significant other pressures demonstrated a direct path to thin internalization. Media pressures demonstrated direct paths to thin internalization, muscular internalization, and appearance comparisons. Thin internalization demonstrated a direct path to disordered eating, while muscular internalization demonstrated a direct path to appearance comparisons. Finally, appearance comparisons exhibited a direct path to disordered eating. Peer pressures and media pressures demonstrated significant indirect paths to disordered eating. More specifically, there was a significant indirect path from peer pressures to disordered eating via muscular internalization in combination with appearance comparisons. Media pressures demonstrated an indirect path to disordered eating via (a) thin internalization, (b) muscular internalization in combination with appearance comparisons, and (c) appearance comparisons alone. The final model predicted 32% of the variance in disordered eating, 34% of the variance in appearance comparison frequency, and 10% of variance in both thin and muscular internalization (ps < .001).

Table 2.

Summary Statistics for the Total, Direct, and Indirect Paths from Sociocultural Pressures to Eating Disorder Symptoms

Pathway Total path p Direct path p Indirect path Bootstrapped○ 95% CI
Family pressures -> disordered eating .31 < .001 .30 < .001 .02 −.03, .08
Family pressures -> thin internalization -> disordered eating .31 < .001 .30 < .001 .03 −.02, .10
Family pressures -> muscular internalization -> disordered eating .31 < .001 .30 < .001 −.01 −.02, .01
Family pressures -> thin internalization -> appearance comparison -> disordered eating .31 < .001 .30 < .001 .00 −.01, .00
Family pressures -> muscular internalization -> appearance comparison -> disordered eating .31 < .001 .30 < .001 −.01 −.02, .01
Peer pressures -> disordered eating .02 .45 -- -- .02 −.04, .09
Peer pressures -> thin internalization -> disordered eating .02 .45 -- -- −.01 −.06, .05
Peer pressures -> muscular internalization -> disordered eating .02 .45 -- -- .02 −.01, .05
Peer pressures -> thin internalization -> appearance comparison -> disordered eating .02 .45 -- -- .00 .00, .00
Peer pressures -> muscular internalization -> appearance comparison - > disordered eating .02 .45 -- -- .02 .01, .04
Significant other pressures -> disordered eating .04 .12 -- -- .04 −.01, .10
Significant other pressures -> thin internalization -> disordered eating .04 .12 -- -- .05 −.01, .11
Significant other pressures -> muscular internalization -> disordered eating .04 .12 .00 −.02, .02
Significant other pressures -> thin internalization -> appearance comparison -> disordered eating .04 .12 -- -- .00 −.01, .00
Significant other pressures -> muscular internalization -> appearance comparison -> disordered eating .04 .12 -- -- .00 −.02, .01
Media pressures -> disordered eating .17 < .001 -- -- .17 .10, .25
Media pressures -> thin internalization -> disordered eating .17 < .001 -- -- .05 .01, .10
Media pressures -> muscular internalization -> disordered eating .17 < .001 -- -- .01 −.01, .04
Media pressures -> appearance comparison -> disordered eating .17 < .001 -- -- .09 .04, .14
Media pressures -> thin internalization -> appearance comparison -> disordered eating .17 < .001 -- -- .00 −.01, .00
Media pressures -> muscular internalization -> appearance comparison -> disordered eating .17 < .001 -- -- .02 .01, .03
Open in a new tab

Notes. CI = confidence interval. Estimates of the indirect path and their respective confidence intervals have been standardized.

The absence and presence of zero within the bounds of a bootstrapped 95% CI indicates a significant and non-significant indirect effect, respectively.

○

Bootstrapping was performed with 10,000 resamples.

Bold indicates a statistically significant indirect effect, based on bootstrapped 95% CIs

4. Discussion

The current study aimed to evaluate a version of the Tripartite Influence Model (Thompson et al., 1999) among adult men using muscularity- and thinness-oriented sociocultural influence measures explicitly designed for, and validated with, men. Specifically, we tested a model in which the relationships between thinness- and muscularity-focused pressures from four separate cultural influences (i.e., peers, family, significant others, media) and disordered eating were indirectly related via an internalized desire for thinness, an internalized desire for muscularity, and comparisons of one’s weight/shape and muscularity. Overall, results provided mixed support for the Tripartite Influence Model. While the original model demonstrated poor fit to the data, the addition of two theoretically justified pathways (i.e., from media pressures to appearance comparisons, from family pressures to disordered eating) resulted in acceptable to very good fit.

The findings from this study build upon previous research in this area (Girard et al., 2018; Rodgers et al., 2012; Tylka, 2011; Tylka & Andorka, 2012), providing additional support for the utility of sociocultural risk-models of body image and eating concerns among young men. Altogether, the variables in our model successfully explained 32% of the variance in eating disorder cognitions and behaviors among young men, which is comparable to previous research examining sociocultural influences among French college males (Girard et al., 2018). This study therefore adds to a growing body of literature, which suggests the importance of both muscularity- and thinness-oriented inter- and intra-personal processes (i.e., both an individuals’ endorsement of appearance ideals, as well as the perceived pressures from their environment) in the development and/or maintenance of eating pathology among men.

Consistent with the Tripartite Influence Model, all pressures variables exhibited direct pathways to at least one proposed downstream variable. As in previous empirical investigations of similar models (e.g., Tylka, 2011; Tylka & Andorka, 2012), media pressures evidenced the most robust relationships with hypothesized outcomes, highlighting this particular cultural agent as an especially pervasive influence on men’s eating and body image. Interestingly, pressures from peers exhibited the strongest association with muscular internalization, while pressures from significant others appeared to exert greater influence on thin internalization. Finally, family pressures, thin internalization, and appearance comparison each displayed direct relationships with eating disorder cognitions and behaviors, suggesting that these experiences may operate as powerful proximal influences on eating concerns among men.

Although not included in our original model, the data supported the addition of direct pathways from family pressures to disordered eating, and from media pressures to appearance comparisons. These findings suggest that the impact of appearance pressures on proposed negative outcomes may not be entirely mediated by appearance ideal internalization. Previous examinations of the Tripartite Influence Model among Australian young women have similarly found that parental influences were not mediated by internalization and comparison processes (Rodgers et al., 2011). Instead, family pressures may exert a direct influence on disordered eating (e.g., as a function of genetic influences or behavioral modeling), or these pressures may be mediated by mechanisms not specified within the current model (e.g., negative affect; Stice & Van Ryzin, 2019). Similarly, media pressures may uniquely evoke appearance comparisons, even in the absence of appearance ideal internalization. Indeed, much of the experimental literature examining the effects of exposure to media images of idealized bodies assumes that viewing such images inherently provokes a comparison of one’s own body to the idealized images (Myers & Crowther, 2009). The direct and indirect effects present here between media pressures and comparison processes may therefore reflect dual pathways by which media exposure increases appearance comparison frequency. That is, media pressures may result in both in-the-moment appearance comparisons, as well as gradual increases in internalization, that then increase the tendency to engage in appearance comparisons. Research investigating these relationships over different time frames (e.g., using ecological momentary assessment) will be important for furthering our understanding of these processes among men.

In addition to the observed direct pathways, analyses supported the presence of four indirect pathways from pressures variables to disordered eating. One indirect effect was observed from peer pressures to disordered eating via muscular internalization and appearance comparisons, while three indirect effects from media pressures to disordered eating were observed via thin internalization alone, appearance comparisons alone, and muscular internalization combined with appearance comparisons. These results suggest that appearance pressures from peer groups may primarily emphasize ideals of heightened muscularity, which ultimately contribute to disordered eating practices. In contrast, pressures from the media may more evenly promote ideals of both thinness and muscularity, contributing to unhealthy eating practices which may be intended to concurrently reduce adiposity and increase muscle tone. Again, it is notable that media was the only sociocultural source that exhibited either direct or indirect pathways to all downstream variables (i.e., thin internalization, muscular internalization, appearance comparisons, and disordered eating). This finding supports a large body of research highlighting media pressures as a particularly pernicious influence among women (Griffiths, Castle et al., 2018; Hausenblas et al., 2013), and adds to a growing body of literature indicating the detrimental effects of media messages among males (Griffiths, Murray, Krug, & McLean, 2018; Mayer-Brown, Lawless, Fedele, Dumont-Driscoll, & Janicke, 2016).

Notably, six direct pathways hypothesized by the model were not significant (i.e., family pressures to internalization of thinness and muscularity; peer pressures to thin internalization; significant others to muscular internalization; thin internalization to appearance comparisons; muscular internalization to disordered eating). Four of these pathways concern relationships between sources of sociocultural pressure and internalization of appearance ideals. It may be that among this age group, family influences on internalization are attenuated, while their role in modeling eating behaviors might prevail (Rodgers & Chabrol, 2009). With regard to pressures from peers and significant others, it is notable that prior work among college men failed to support relationships between peer and partner appearance pressures and internalization of the mesomorphic ideal (Tylka, 2011), while findings among gay men revealed significant relationships between mesomorphic ideal internalization and partner pressures, but not friend pressures (Tylka & Andorka, 2012). These mixed findings suggest the need for continued research interrogating these relationships and their possible moderators. Finally, while thin internalization demonstrated a small bivariate association with appearance comparisons, this relationship was attenuated in the path model. Notably, research among gay men demonstrated a direct path from mesomorphic ideal internalization to general appearance comparisons (Tylka & Andorka, 2012). This finding underscores the importance of modeling these constructs separately in order to clarify their relationship (and relative importance) among males.

Given mounting evidence supporting the role of inter- and intra-personal processes in the development or maintenance of disordered eating among men (Papp et al., 2013; Rodgers et al., 2012; Tylka, 2011; Tylka & Andorka, 2012), it is possible that intervention programs previously designed to target these constructs among females, such as The Body Project (Stice & Presnell, 2007) or media literacy approaches (e.g., McLean, Wertheim, Masters, & Paxton, 2017), may have utility among men. In particular, findings suggest the importance of adapting these interventions to address both muscularity- and thinness-oriented processes as significant contributors to disordered eating among males. Indeed, initial work in this area supports the application of cognitive dissonance-based approaches targeting internalization of thinness and muscularity among gay men (Brown & Keel, 2015) and media literacy programs targeting internalization of thinness and athleticism among adolescent boys (Wilksch, Tiggemann, & Wade, 2006).

Strengths of the current study include the use of a large and ethnically-diverse sample, as well as the use of measures of sociocultural constructs (appearance pressures, internalization, and appearance comparison) that were designed and validated for use among men, and therefore account for pressures towards both leanness and muscularity. In addition, the implementation of path analysis allowed for a comprehensive test of the Tripartite Influence Model, including comparison of the relative importance of individual influence variables and identification of specific mechanisms of influence through the examination of indirect effects. Finally, this study explored an expanded range of sociocultural influence variables, incorporating significant others as a potential source of appearance pressures. Findings support the continued inclusion of this influence variable in future work with males.

Limitations of the study are also noted, and provide guidance for future work in this area. While the SATAQ-4R captures media pressures from a range of sources (e.g., television, internet, movies, magazines), it does not specifically assess appearance pressures resulting from social media. As social media has emerged as a significant contributor to body image and eating concerns (Griffiths, Murray et al., 2018), future work examining the Tripartite Influence Model in males may consider highlighting this source of appearance pressure. Further, although all predictor variables in the examined model incorporated references to muscularity (e.g., muscularity pressures, muscularity comparisons, muscularity internalization), researchers have expressed concerns that extant measures of eating pathology may be limited in their ability to assess disordered eating patterns organized around the pursuit of both thinness (or, leanness) and muscularity (Murray et al., 2017). That is, in order to achieve the prevailing appearance ideal for men, males may intermittently pursue a caloric surplus (e.g., overconsumption of protein to build muscle), and then a caloric deficit (e.g., restriction of carbohydrates and fats to reveal or accentuate those muscles). As existing measures largely assess the pursuit of thinness, these measures may fail to assess the full spectrum of disordered eating practices common to males (Schaefer et al., 2018). Although the current study employed a widely-used measure of eating pathology, which has been well-validated among female samples (Berg, Peterson, Frazier, & Crow, 2012; Berg, Stiles et al., 2012), it is possible that the measure was not able to capture muscularity-oriented disordered eating, which may have attenuated observed relationships with predictor variables. As prior research among males suggests that sociocultural influences relate to muscularity-oriented body image and behaviors (e.g., Karazsia et al., 2009; Smolak and colleagues, 2005; Stratton et al., 2015), future work is encouraged to examine the predictive utility of the Tripartite Influence Model, utilizing measures of eating pathology that may more adequately reflect the full scope of disordered eating experiences among men (e.g., the Eating Pathology Symptoms Inventory, Forbush et al., 2013; Eating Disorder Assessment for Men, Stanford & Lemberg, 2012). In addition, although research suggests that internalization of thinness or low body fat is a relevant construct among males (Schaefer et al., 2017), a desire for leanness may also be key to male body image concerns (Smolak & Murnen, 2008). Therefore, examination of leanness-oriented constructs would be of great value. Further, the Tripartite Influence Model and considerable empirical evidence suggests the importance of body dissatisfaction as a key contributor to eating pathology onset and maintenance (Stice, 1994; Thompson et al., 1999). While the current study utilized a measure of eating pathology that incorporates body image disturbance as a core facet of the construct, future work should include dedicated measures of body image, examining body dissatisfaction as a mechanism relating sociocultural influences to disordered eating expression. Relatedly, the current study is not able to examine other hypothesized risk or maintenance factors that may interact with sociocultural influences to increase risk for negative outcomes. For example, extant work suggests the importance of genetic influences on sociocultural variables (Suisman et al., 2014), as well the role of emotional states in predicting disordered eating behaviors (Stice & Van Ryzin, 2019). Accordingly, the integration of sociocultural, psychological, and biological variables into a comprehensive model of disordered eating is an important goal for future work among men. Finally, the current study utilized cross-sectional data, limiting our ability to draw temporal or causal inferences. Continued work examining the relations between examined variables using experimental and longitudinal designs is strongly encouraged.

In sum, the current study provided partial support for a version of the Tripartite Influence Model in college men, suggesting that appearance pressures, internalization, and comparisons are each significant contributors to disordered eating in men. In addition, results highlighted the utility of considering both thinness- and muscularity-oriented appearance processes in predicting thinness-oriented disordered eating among men. Moving forward, future work is needed to extend these findings to the prediction of muscularity-oriented disordered eating in this group.

Highlights.

  • Results provide support for the Tripartite Influence Model among men

  • Thinness/muscularity pressures predicted thin/muscular internalization and comparisons

  • Thin/muscular internalization and comparisons predicted disordered eating

Acknowledgments

This work was supported by the National Institute of Mental Health (grant number T32 MH082761).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Contributor Information

Lauren M. Schaefer, Sanford Center for Bio-behavioral Research

Rachel F. Rodgers, Northeastern University

J. Kevin Thompson, University of South Florida.

Scott Griffiths, University of Melbourne.

References

  1. Allen KL, Byrne SM, Crosby RD, & Stice E (2016). Testing for interactive and non-linear effects of risk factors for binge eating and purging eating disorders. Behaviour Research and Therapy, 87, 40–47. doi: 10.1016/j.brat.2016.08.019 [DOI] [PubMed] [Google Scholar]
  2. Berg KC, Peterson CB, Frazier P, & Crow SJ (2012). Psychometric evaluation of the Eating Disorder Examination and Eating Disorder Examination-Questionnaire: A systematic review of the literature. International Journal of Eating Disorders, 45(3), 428–438. doi: 10.1002/eat.20931 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Berg KC, Stiles SEC, Swanson SA, Peterson CB, Lebow J, & Le Grange D (2012). Diagnostic concordance of the interview and questionnaire versions of the Eating Disorder Examination. International Journal of Eating Disorders, 45(7), 850–855. doi: 10.1002/eat.20948 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brown TA, Griffiths S, & Murray SB (2018). Eating disorders in males. In Anderson LK, Murray SB, & Kaye WH (Eds.), Clinical handbook of complex and atypical eating disorders. (pp. 309–326). New York, NY: Oxford University Press. [Google Scholar]
  5. Brown TA, & Keel PK (2015). A randomized controlled trial of a peer co-led dissonance-based eating disorder prevention program for gay men. Behaviour Research and Therapy, 74, 1–10. doi: 10.1016/j.brat.2015.08.008 [DOI] [PubMed] [Google Scholar]
  6. Byrne BM (2012). Structural equation modeling with Mplus: Basic concepts, applications and programming. New York, NY: Routledge. [Google Scholar]
  7. Calzo JP, Blashill AJ, Brown TA, & Argenal RL (2017). Eating disorders and disordered weight and shape control behaviors in sexual minority populations. Current Psychiatry Reports, 19(8), 49. 10.1007/s11920-017-0801-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Coffino JA, Udo T, & Grilo CM (2019). Rates of help-seeking in US adults with lifetime DSM-5 eating disorders: Prevalence across diagnoses and differences by sex and ethnicity/race. Mayo Clinic Proceedings, 94(8), 1415–1426. doi: 10.1016/j.mayocp.2019.02.030 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Fairburn CG, & Beglin SJ (2008). Eating Disorder Examination Questionnaire (EDE-Q 6.0). In Fairburn CG (Ed.) Cognitive behavior therapy and eating disorders (pp. 309–313) New York, NY: Guilford Press. [Google Scholar]
  10. Forbush KT, Wildes JE, Pollack LO, Dunbar D, Luo J, Patterson K, Petruzzi L, Pollpeter M, Miller H, Stone A, Bright A, & Watson D (2013). Development and validation of the Eating Pathology Symptoms Inventory (EPSI). Psychological Assessment, 25(3), 859–878. doi: 10.1037/a0032639 [DOI] [PubMed] [Google Scholar]
  11. Girard M, Chabrol H & Rodgers RF (2018). Support for a modified tripartite dual pathway model of body image concerns and risky body change behaviors in french young men. Sex Roles, 78, 799–809. doi: 10.1007/s11199-017-0850-z [DOI] [Google Scholar]
  12. Griffiths S, Castle D, Cunningham M, Murray SB, Bastian B, & Barlow FK (2018). How does exposure to thinspiration and fitspiration relate to symptom severity among individuals with eating disorders? Evaluation of a proposed model. Body Image, 27, 187–195. doi: 10.1016/j.bodyim.2018.10.002 [DOI] [PubMed] [Google Scholar]
  13. Griffiths S, Mond JM, Li Z, Gunatilake S, Murray SB, Sheffield J, & Touyz S (2015). Self-stigma of seeking treatment and being male predict an increased likelihood of having an undiagnosed eating disorder. International Journal of Eating Disorders, 48(6), 775–778. doi: 10.1002/eat.22413 [DOI] [PubMed] [Google Scholar]
  14. Griffiths S, Murray SB, Krug I, & McLean SA (2018). The contribution of social media to body dissatisfaction, eating disorder symptoms, and anabolic steroid use among sexual minority men. Cyberpsychology, Behavior, and Social Networking, 21(3), 149–156. doi: 10.1089/cyber.2017.0375 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Griffiths S, Murray SB, & Touyz S (2013). Disordered eating and the muscular ideal. Journal of Eating Disorders, 1(1), 15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hausenblas HA, Campbell A, Menzel JE, Doughty J, Levine M, & Thompson JK (2013). Media effects of experimental presentation of the ideal physique on eating disorder symptoms: A meta-analysis of laboratory studies. Clinical Psychology Review, 33(1), 168–181. doi: 10.1016/j.cpr.2012.10.011 [DOI] [PubMed] [Google Scholar]
  17. Karazsia BT, & Crowther JH (2009). Social body comparison and internalization: mediators of social influences on men’s muscularity-oriented body dissatisfaction. Body Image, 6(2), 105–112. 10.1016/j.bodyim.2008.12.003 [DOI] [PubMed] [Google Scholar]
  18. Keery H, Van den Berg P, & Thompson JK (2004). An evaluation of the Tripartite Influence Model of body dissatisfaction and eating disturbance with adolescent girls. Body Image, 1(3), 237–251. doi: 10.1016/j.bodyim.2004.03.001 [DOI] [PubMed] [Google Scholar]
  19. Lipson SK, & Sonneville KR (2017). Eating disorder symptoms among undergraduate and graduate students at 12 US colleges and universities. Eating Behaviors, 24, 81–88. doi: 10.1016/j.eatbeh.2016.12.003 [DOI] [PubMed] [Google Scholar]
  20. Little TD (2013). Longitudinal structural equation modeling. Guilford Press. [Google Scholar]
  21. Mayer-Brown S, Lawless C, Fedele D, Dumont-Driscoll M, & Janicke DM (2016). The effects of media, self-esteem, and BMI on youth’s unhealthy weight control behaviors. Eating Behaviors, 21, 59–65. doi: 10.1016/j.eatbeh.2015.11.010 [DOI] [PubMed] [Google Scholar]
  22. McLean SA, Wertheim EH, Masters J, & Paxton SJ (2017). A pilot evaluation of a social media literacy intervention to reduce risk factors for eating disorders. International Journal of Eating Disorders, 50(7), 847–851. doi: 10.1002/eat.22708 [DOI] [PubMed] [Google Scholar]
  23. Miller JM, & Luk JW (2019). A systematic review of sexual orientation disparities in disordered eating and weight-related behaviors among adolescents and young adults: Toward a developmental model. Adolescent Research Review, 4(2), 187–208. 10.1007/s40894-018-0079-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Morton R (1694). A treatise of consumptions. London: Princess Arms Press. [Google Scholar]
  25. Murray SB, Nagata JM, Griffiths S, Calzo JP, Brown TA, Mitchison D, … Mond JM (2017). The enigma of male eating disorders: A critical review and synthesis. Clinical Psychology Review, 57, 1–11. doi: 10.1016/j.cpr.2017.08.001 [DOI] [PubMed] [Google Scholar]
  26. Myers TA, & Crowther JH (2009). Social comparison as a predictor of body dissatisfaction: A meta-analytic review. Journal of Abnormal Psychology, 118(4), 683–698. doi: 10.1037/a0016763 [DOI] [PubMed] [Google Scholar]
  27. Papp I, Urbán R, Czeglédi E, Babusa B, & Túry F (2013). Testing the Tripartite Influence Model of body image and eating disturbance among Hungarian adolescents. Body Image, 10(2), 232–242. doi: 10.1016/j.bodyim.2012.12.006 [DOI] [PubMed] [Google Scholar]
  28. Pursey K, Burrows TL, Stanwell P, Collins CE (2014). How accurate is web-based self-reported height, weight, and body mass index in young adults? Journal of Medical Internet Research, 16(1), e4 doi: 10.2196/jmir.2909 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Rodgers R, & Chabrol H (2009). Parental attitudes, body image disturbance and disordered eating amongst adolescents and young adults: A review. European Eating Disorders Review, 17(2), 137–151. [DOI] [PubMed] [Google Scholar]
  30. Rodgers R, Chabrol H, & Paxton SJ (2011). An exploration of the Tripartite Influence Model of body dissatisfaction and disordered eating among Australian and French college women. Body Image, 8(3), 208–215. doi: 10.1016/j.bodyim.2011.04.009 [DOI] [PubMed] [Google Scholar]
  31. Rodgers RF, Ganchou C, Franko DL, & Chabrol H (2012). Drive for muscularity and disordered eating among French adolescent boys: A sociocultural model. Body Image, 9(3), 318–323. doi: 10.1016/j.bodyim.2012.03.002 [DOI] [PubMed] [Google Scholar]
  32. Rodgers RF, McLean SA, & Paxton SJ (2015). Longitudinal relationships among internalization of the media ideal, peer social comparison, and body dissatisfaction: Implications for the tripartite influence model. Developmental Psychology, 51(5), 706–713. 10.1037/dev0000013 [DOI] [PubMed] [Google Scholar]
  33. Rogers A, Fuller-Tyszkiewicz M, Lewis V, Krug I, & Richardson B (2017). A person-by-situation account of why some people more frequently engage in upward appearance comparison behaviors in everyday life. Behavior Therapy, 48(1), 19–28. 10.1016/j.beth.2016.09.007 [DOI] [PubMed] [Google Scholar]
  34. Schaefer LM, Harriger JA, Heinberg LJ, Soderberg T, & Thompson JK (2017). Development and validation of the Sociocultural Attitudes Towards Appearance Questionnaire-4-Revised (SATAQ-4R). International Journal of Eating Disorders, 50(2), 104–117. doi: 10.1002/eat.22590 [DOI] [PubMed] [Google Scholar]
  35. Schaefer LM, Smith KE, Leonard R, Wetterneck C, Smith B, Farrell N, … Thompson JK (2018). Identifying a male clinical cutoff on the Eating Disorder Examination-Questionnaire (EDE-Q). International Journal of Eating Disorders. doi: 10.1002/eat.22972 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Schaefer LM, & Thompson JK (2018). The development and validation of the Physical Appearance Comparison Scale–3 (PACS-3). Psychological Assessment, 30(10), 1330–1341. doi: 10.1037/pas0000576 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Shroff H, & Thompson JK (2006). The tripartite influence model of body image and eating disturbance: A replication with adolescent girls. Body Image, 3(1), 17–23. 10.1016/j.bodyim.2005.10.004 [DOI] [PubMed] [Google Scholar]
  38. Smolak L, & Murnen SK (2008). Drive for leanness: assessment and relationship to gender, gender role and objectification. Body Image, 5(3), 251–260. doi: 10.1016/j.bodyim.2008.03.004 [DOI] [PubMed] [Google Scholar]
  39. Smolak L, Murnen SK, & Thompson JK (2005). Sociocultural influences and muscle building in adolescent boys. Psychology of Men & Masculinity, 6(4), 227–239. https://doi/10.1037/1524-9220.6.4.227 [Google Scholar]
  40. Stanford SC, & Lemberg R (2012). Measuring eating disorders in men: Development of the Eating Disorder Assessment for Men (EDAM). Eating Disorders: The Journal of Treatment & Prevention, 20(5), 427–436. doi: 10.1080/10640266.2012.715522 [DOI] [PubMed] [Google Scholar]
  41. Stice E (1994). Review of the evidence for a sociocultural model of bulimia nervosa and an exploration of the mechanisms of action. Clinical Psychology Review, 14, 633–661. doi: / 10.1016/0272-7358(94)90002-7 [DOI] [Google Scholar]
  42. Stice E, & Presnell K (2007). The body project: Promoting body acceptance and preventing eating disorders: Facilitator guide. New York, NY: Oxford University Press. [Google Scholar]
  43. Stice E, & Van Ryzin MJ (2019). A prospective test of the temporal sequencing of risk factor emergence in the dual pathway model of eating disorders. Journal of Abnormal Psychology, 128(2), 119–128. doi: 10.1037/abn0000400 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Stratton R, Donovan C, Bramwell S, & Loxton NJ (2015). Don’t stop till you get enough: Factors driving men towards muscularity. Body Image, 15, 72–80. doi: 10.1016/j.bodyim.2015.07.002 [DOI] [PubMed] [Google Scholar]
  45. Suisman JL, Thompson JK, Keel PK, Burt SA, Neale M, Boker S, … Klump KL (2014). Genetic and environmental influences on thin-ideal internalization across puberty and preadolescent, adolescent, and young adult development. International Journal of Eating Disorders, 47(7), 773–783. doi: 10.1002/eat.22321 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Tabachnick BG, & Fidell LS (2007). Using multivariate statistics. Boston, MA: Pearson Education. [Google Scholar]
  47. Thompson JK, & Cafri G (2007). The muscular ideal: Psychological, social, and medical perspectives. Washington, DC: American Psychological Association. [Google Scholar]
  48. Thompson JK, Heinberg LJ, Altabe M, & Tantleff-Dunn S (1999). Exacting beauty: Theory, assessment, and treatment of body image disturbance. Washington, DC: American Psychological Association. [Google Scholar]
  49. Tylka TL (2011). Refinement of the tripartite influence model for men: Dual body image pathways to body change behaviors. Body Image, 8(3), 199–207. doi: 10.1016/j.bodyim.2011.04.008 [DOI] [PubMed] [Google Scholar]
  50. Tylka TL, & Andorka MJ (2012). Support for an expanded tripartite influence model with gay men. Body Image, 9(1), 57–67. doi: 10.1016/j.bodyim.2011.09.006 [DOI] [PubMed] [Google Scholar]
  51. Wilksch SM, Tiggemann M, & Wade TD (2006). Impact of interactive school-based media literacy lessons for reducing internalization of media ideals in young adolescent girls and boys. International Journal of Eating Disorders, 39(5), 385–393. doi: 10.1002/eat.20237 [DOI] [PubMed] [Google Scholar]

RESOURCES