Boost Your Body: Self-Improvement Magazine Messages Increase Body Satisfaction in Young Adults

ABSTRACT

The verbal messages that contextualize exposure to idealized body imagery may moderate media users’ body satisfaction. Such contextualizing verbal messages often take the form of social comparison motives in fashion magazines, while body dissatisfaction is an important mechanism underlying various body image–related health issues like depression and unbalanced weight status. Hence, the present study applied social comparison motives as induced through magazine cover messages. Hypotheses were tested in an experimental design with social comparison motives (self-improvement vs. self-evaluation vs. control) and recipient gender as between-subjects factors and body satisfaction as within-subjects factor (N = 150). Results showed that self-improvement messages accompanying ideal body media models increased body satisfaction, compared to control messages and baseline measures. In contrast, the self-evaluation messages did not impact body satisfaction. Results imply that inconsistencies regarding effects from exposure to idealized body imagery are explained by the context in which media images are portrayed, evoking differential social comparison motives. Moreover, the findings imply that health communication interventions can use verbal messages on body improvement as helpful tools, if they draw on social comparison motives effectively.

Notes

^1. Body Mass Index (BMI) was calculated by dividing self-reported weight (in kilograms) by squared height (in meters).

^2. A trait measure of physical appearance comparison was measured with Physical Attractiveness Comparison Scale (PACS; Thompson, Heinberg, & Tantleff, 1991). These five items (four indicative) on a five-point scale (1 = totally disagree; 5 = totally agree) formed a reliable scale (Cronbach’s alpha = .72).

^3. Magazine use was assessed with one item on how many magazines were read during the past week (11-point scale from “not at all”, “1”, “2”, [. . .], “10, or more”).

^4. In the main study, respondents evaluated the portrayed models’ thinness (10-point scale: very thin, fit–very big, out of shape) and attractiveness (10-point scale: very unattractive, ugly-very attractive, beautiful); measurements were in line with previous studies (e.g., Martin & Gentry, 1997; Veldhuis et al., 2014b). Scores for both covers within an experimental condition added up to sum scores for model thinness (r = .369, p < .01) and model attractiveness (r = .342, p < .01). A MANOVA revealed that the portrayed models’ thinness and attractiveness were evaluated similarly across experimental conditions (multivariate effect: Wilks’s λ = .974, F(4,292) = 1.03, p = .392, ηp² = .01; univariate F-tests: p = .381 for model thinness, and p = .338 for model attractiveness).

^5. The pattern of (non)significance of this repeated measures ANOVA as well as subsequent pairwise comparisons (Sidak) did not change when BMI was included as a covariate.

^6. Senn (2006) discusses the (favorable) use of ANCOVA, compared to the simple analysis of change scores, for estimates of treatment effects and changes from baseline: including the baseline measure as a covariate and the pretreatment measure as the dependent variable.

^7. This pattern of (non)significance of this ANCOVA and subsequent pairwise comparisons (Sidak) did not change when BMI was included as a covariate.