BoPopriation: How self-promotion and corporate commodification can undermine the body positivity (BoPo) movement on Instagram

ABSTRACT

Proponents of the body positivity (BoPo) movement prominently use social media to promote body appreciation and normalize marginalized bodies. However, companies and social media users have increasingly commodified the movement for self-serving reasons or economic gain. Providing a unique test of the persuasion knowledge model, this experiment examined (a) how the commodification of a prosocial movement can undermine its efficacy and (b) how the symmetry between visual and text-based messaging can influence viewer reactions. Results indicated that body positive posts on Instagram that contained self-promotion or promoted products were viewed as less morally appropriate and were less effective at promoting body appreciation and inclusivity. Practical implications are discussed and a novel boundary condition for the persuasion knowledge model is presented.

KEYWORDS:

• Body positivity
• Instagram
• persuasion knowledge
• prosocial movements
• social media
• body image

Research has illustrated how narrow beauty standards perpetuated by mass media content can contribute to body image disturbance in women, including increased body dissatisfaction, negative mood, and restrictive eating behavior (Grabe et al., 2008; Roberts et al., 2018). Compounding problems associated with mass media depictions of women, concern exists that female users of newer media such as Instagram are regularly sharing and viewing images that cohere to the same narrow beauty standards (Tiggeman & Zaccardo, 2015). However, some social media users are employing online platforms to further a more inclusive view of beauty. The body positivity movement, organized primarily through social media, seeks to challenge narrow appearance ideals by encouraging acceptance of bodies of any age, shape, size, color, or ability (Cohen et al., 2019a; Cwynar-Horta, 2016). By proliferating social media with images of women who do not conform to status quo beauty norms, the body positivity movement seeks to normalize other types of bodies and diversify the qualities that society deems beautiful.

Given the millions of body-positive images posted on Instagram, researchers have begun to explore the degree to which the movement is achieving its intended outcomes (see Cohen et al., 2020). For instance, a recent study (Cohen et al., 2019b) indicated that women who viewed body-positive images on Instagram were found to have significantly higher state body appreciation scores and body satisfaction when compared with women who viewed images consistent with the thin ideal. However, there is still a lot that is unknown about the efficacy of the body positivity movement, in particular, and more broadly how user-generated movements on social media can achieve their goals.

This study examined two prominent factors that might affect the ability of prosocial movements on social media to fulfill their intended aims. First, the popularity of the body positivity movement on social media has attracted the attention of companies who have commodified the movement for commercial gain (Cwynar-Horta, 2016; Lazuka et al., 2020). We investigated how various promotional cues that regularly accompany body-positive images on Instagram affected how viewers reacted to body-positive content online, providing a novel test of the persuasion knowledge model (Friestad & Wright, 1994). Second, we examined how the symmetry between visual and text-based messaging influences evaluations of body-positive posts on Instagram by exploring how viewers reacted to body-positive posts shared by women who vary in the degree to which they deviate from traditional beauty norms.

Body positivity movement

Photo-sharing social media platforms provide an environment for the transmission of potentially harmful images. For example, Instagram has become a breeding ground for physique and appearance-focused photo communities that have been organized through hashtags such as #thinspiration and #fitspiration. Empirical studies have confirmed that both thinspiration and fitspiration images shared on social media have negative effects on body satisfaction, mood, and self-esteem (Slater et al., 2017; Tiggeman & Zaccardo, 2015). However, a new community has formed that seeks to combat the negative effects of dominant beauty ideals perpetuated over social media by using the power of user-generated content to promote body appreciation.

The body positivity movement on social media challenges traditional appearance ideals – such as needing to be young, thin, and without physical flaws – and encourages the increased representation and normalization of people with traditionally marginalized bodies (Cohen et al., 2019a; Cwynar-Horta, 2016). Unlike fitspiration and thinspiration content on social media, body-positive content positively affects women’s body image and self-worth (Cohen et al., 2019b; Ghaznavi & Taylor, 2015; Slater et al., 2017; Tiggeman & Zaccardo, 2015). The beneficial impact that body-positive content has on women is due to its broader conceptualization of beauty and appreciation for the body. This helps overcome negative body image outcomes that result from feeling unable to attain narrow beauty standards (see Clayton et al., 2017). A content analysis of body-positive posts on Instagram indicated that posts are largely incongruent with traditional beauty ideals (Cohen et al., 2019a). Body-positive posts often showcase body characteristics such as cellulite, stomach rolls, and stretch marks; these perceived imperfections are celebrated and embraced as beautiful (Cohen et al., 2019a; Lazuka et al., 2020). Cohen et al. (2019a) also reported that many body-positive posts emphasize body appreciation. Although body-positive content appears to have some beneficial effects on viewers, its positive influence may be mitigated by the increasing commercial appropriation of the movement.

Commodification of body positivity

Given the rise in popularity of the body-positive movement, there has been a noticeable shift in the source of body-positive content produced online (Cohen et al., 2019a; Cwynar-Horta, 2016). Companies have begun to recognize the power of the body-positive movement and capitalize off of its imagery and messaging (Lazuka et al., 2020). Cwynar-Horta (2016) has described how companies claim to be body positive by posting make-up free social media ad campaigns, launching plus-size clothing lines, and including plus-sized models in advertising campaigns. Although the body-positive themes that companies promote may seem noble, often such ad campaigns are not inclusive and still push a narrative of what acceptable beauty is. Other ways corporations undermine body-positive values include Photoshopping flaws out of images or using plus-sized models who are closer to the “normative ideal body” (Cwynar-Horta, 2016, p. 42). Furthermore, Cwynar-Horta (2016) suggested that body-positive advocates online have begun to focus less on promoting body inclusivity and more on using body-positive tropes to sell products and promote themselves. Content analyses supported this assertion with over a third of “body positive” posts being classified as promoting commercial products or being self-promotional, indicating a change of focus from body-positive ideals to consumption practices (Cohen et al., 2019a). A shift in the body-positive narrative on social media from a more exclusive focus on body self-acceptance to a focus that includes using the popularity of the movement for economic or personal gain might hinder the ability of the movement to achieve its original goals.

Persuasion knowledge model

The persuasion knowledge model (Friestad & Wright, 1994) provides a framework to understand how people interpret and respond to persuasive messages. The persuasion knowledge model seeks to explain how targets (i.e., receivers of persuasive content) identify and respond to persuasion agents (i.e., those who seek to persuade). Once targets recognize a persuasive attempt, they seek to cope or respond to the attempt in a way that allows them to maintain control over the outcome of the attempt (Boerman et al., 2018; Friestad & Wright, 1994).

Recognition of sponsored content refers to the ability of audience members to differentiate sponsored content from other content seen on social media. Boerman et al. (2018) have argued that the recognition of sponsored content is an important prerequisite to activating other aspects of persuasion knowledge. Recent work has indicated that persuasion knowledge can be triggered in a variety of ways (Boerman et al., 2017; Zarouali et al., 2021). For instance, focusing on persuasion knowledge recognition in adolescents, Zarouali and colleagues (2021) documented the ability of norm-based labels (e.g., “your friends know this is sponsored”) to trigger persuasion knowledge. Their work illustrated how the population of interest (adolescents) and the communicative context (social networking sites) should be taken into account when studying the factors that can trigger persuasion knowledge.

In the context of body positivity posts on Instagram, all posts could be recognized as containing persuasion due to the fact that users challenge dominant beauty ideals and promote body appreciation in their posts. As such, it is essential to identify and isolate how messages can trigger the recognition of specific forms of persuasion in particular online spaces. This is especially important in online environments like Instagram wherein influencer and corporate-sponsored posts can blend in with user posts, making it difficult to recognize self-promotion or advertising (Boerman et al., 2021; Boerman et al., 2017; Hudders et al., 2017). Accordingly, in this study, we examined how messages can trigger the recognition of self-promotion or commercial advertising and the effects that recognizing these specific forms of persuasion have for the body positivity movement.

Existing content analyses of body-positive content on Instagram have documented several distinct forms of persuasion that accompany posts using body-positive hashtags. First, users who share body-positive content might engage in self-promotion (Lazuka et al., 2020). For instance, users encourage viewers to like and follow their content on Instagram and other online platforms (e.g., private blog, personal YouTube channel). Even when such messages do not contain an explicit connection to companies, products, or profit, they still contain self-benefiting persuasive appeals that are distinguishable from messaging aimed at promoting body appreciation.

More explicit forms of persuasion regularly accompany body-positive content such as when Instagram users directly promote products (Cohen et al., 2019a; Lazuka et al., 2020). The promotion of products in body-positive posts is highlighted as companies – often through user posts – seek to garner a favorable public response from associating their brand/product with the body-positive movement and having seemingly unbiased users promote their products. Given that users visibly promote themselves and/or products in body-positive posts, we expected viewers to be able to readily differentiate persuasive appeals designed to promote body appreciation from persuasive appeals designed to promote users, products, and services.

H_1a: Viewers are more likely to recognize body positive posts as self-promotional persuasion when users promote their online profiles relative to when they do not.

H_1b: Viewers are more likely to recognize body positive posts as promotional advertising when users promote products relative to when they do not.

One implication of increased recognition of promotional persuasion in body-positive posts is that viewers’ overall attributions for why users shared the posts might be affected. Specifically, the motives viewers ascribe to users for sharing body-positive posts should shift more toward self-promotion and less toward the promotion of inclusive beauty norms. Past research applying the persuasion knowledge model across a variety of contexts has demonstrated how the recognition of persuasion can meaningfully influence viewer attributions (e.g., Ham & Kim, 2019). Bae (2018) documented how the presentation of cause-related marketing on social networking sites can have important implications for the attributions viewers ascribe to company motives and thus how their messages are evaluated. Kapitan and Silvera (2016) argued that the attributions viewers make for the posts social media endorsers share online is critical to consider and that greater recognition of situational factors that spur posts (e.g., companies paying endorsers) can engender negative attributions. Thus, we proposed the following hypotheses:

H_2a: The more body positive posts are recognized as containing self-promotional persuasion, the less viewers will ascribe body positive reasons for users’ posts.

H_2b: The more body positive posts are recognized as containing promotional advertising, the less viewers will ascribe body positive reasons for users’ posts.

Consistent with the persuasion knowledge literature, the recognition of persuasion should have direct implications for how people ultimately evaluate body-positive posts. Boerman et al. (2018) differentiated specific evaluative dimensions that the recognition of persuasion can affect. One evaluative persuasion knowledge dimension we sought to examine is the perceived moral appropriateness of body-positive posts (Friestad & Wright, 1994; Hudders et al., 2017). When body-positive posts on Instagram contain promotional persuasion, users are appropriating a prosocial movement for personal or corporate gain. Whereas some viewers might find it appropriate for a user or company to promote themselves on Instagram (see Johnson et al., 2019), they might deem the tactic of co-opting a prosocial movement for self-promotion less morally acceptable (see Hudders et al., 2017). In general, the perceived commercialization and appropriation of prosocial or cultural messaging has garnered criticism and negative reactions in different contexts such as the commercialization and appropriation of hip-hop (Motley & Henderson, 2008; Rodriquez, 2006) and the commodification of social issues for corporate profit (Tinic, 1997). Similar critiques have been leveled against the commodification of the body-positive movement specifically (Cwynar-Horta, 2016; Miller, 2016). With this in mind, we proposed the following hypotheses:

H_3a: The more body positive posts are recognized as containing self-promotional persuasion, the less morally appropriate viewers will find the posts.

H_3b: The more body positive posts are recognized as containing promotional advertising, the less morally appropriate viewers will find the posts.

In addition to investigating how viewers evaluate the moral appropriateness of body-positive posts, we also examined what these differences may mean for the efficacy of the body-positive movement. The purpose of the body positivity movement is to challenge the predominant thin-ideal messages perpetuated by the media by encouraging inclusivity, self-acceptance, and self-love (Cwynar-Horta, 2016). To achieve these goals, body-positive posts are largely incongruent with traditional beauty ideals, highlighting body characteristics that are traditionally considered flawed (Cohen et al., 2019a). Social media users may find promotional body-positive posts to be less morally appropriate as they seek to benefit the user rather than exclusively promote body appreciation. Furthermore, research has consistently indicated that perceptions of self-interest can affect the credibility and persuasiveness of sources and that people can even infer self-interested motives for seemingly altruistic acts (e.g., Critcher & Dunning, 2011). Benoit and Kennedy (1999) documented how objective (no self-interest) and reluctant (arguing against self-interest) sources are more trustworthy and persuasive than self-interested sources. More recent work in online environments has documented similar results, indicating how perceptions of self-interest affect the credibility of sources, the perceived authenticity of their messages, and ultimately the degree to which their messages are influential (DeAndrea & Vendemia, 2019). If social media users view promotional body-positive posts to be less morally appropriate and inherently self-serving, it stands to reason that social media viewers would be less inclined to shift their attitudes and beliefs in a manner consistent with the aims of the body-positive messaging. Therefore, we expected the following:

H_4a: The more body positive posts are recognized as containing self-promotional persuasion, the less viewers report the cultural embracement of inclusive beauty standards.

H_4b: The more body positive posts are recognized as containing promotional advertising, the less viewers report the cultural embracement of inclusive beauty standards.

H_5a: The more body positive posts are recognized as containing self-promotional persuasion, the less viewers report personally embracing inclusive beauty standards.

H_5b: The more body positive posts are recognized as containing promotional advertising, the less viewers report personally embracing inclusive beauty standards.

Image and messaging symmetry

One way body-positive content aims to challenge narrow appearance ideals is by promoting size inclusivity. Cohen et al. (2019a) suggested a clear majority of body-positive posts from prominent Instagram accounts feature women who are either overweight or obese (as determined by the Current Female Body Image Table; Pulvers et al., 2013). However, Lazuka et al. (2020) found in their content analysis of body-positive posts from the broader Instagram community that the majority of individuals featured in body-positive posts were of normal weight (54.9%) when compared with ∼25% being obese and 17% being overweight. Thus, there is a population of body-positive images that feature women who are more consistent with prominent sociocultural appearance ideals related to body size. This begs the question, do viewers react in the same manner to body-positive images featuring women who visually challenge traditional appearance ideals as compared to images containing women who appear closer to traditional body size ideals?

Past body image research has suggested that audiences are receptive to content that highlights bodies that do fit the thin ideal (see Clayton et al., 2017; Hendrickse et al., 2020). For example, Rogers et al. (2019) found that when presented with images from the body-positive Aerie Real campaign, female respondents expressed greater purchase intentions for company products. Similarly, Paraskeva et al. (2015) found that consumers perceive representatively sized models to be effective in promoting more inclusive appearance ideals and improved body image – illustrating that audiences find representations of female bodies that counter mainstream ideals to be a positive attribute of media content. More broadly, research has suggested that inconsistencies between visuals and text in product descriptions can negatively influence product evaluations (e.g., Lee & Choi, 2019). Thus, the inconsistency between the body-positive messaging presented in some posts and the images of women who fit traditional body and appearance ideals may cause audience members to consider the motives of the posters and subsequently prompt (relatively) more negative reactions.

However, some may not view the inclusion of thin or average-sized women as inconsistent with the central tenets of the body-positive movement. A core focus of the body-positive movement is the celebration and acceptance of all bodies (Cwynar-Horta, 2016). Lazuka et al. (2020) in their content analysis found that individuals of normal weight were featured in the majority of body-positive posts on Instagram. Together, this may suggest to viewers that it is standard for the body-positive movement to highlight bodies of all shapes and sizes. As such, viewing images of women of average body size with messaging about body positivity may not seem incongruent, and thus, would not be less effective in promoting body appreciation. It is possible, however, that body-positive posts that are promotional and feature women who appear more similar in size to traditional weight ideals might be especially problematic for the efficacy of the movement. Compared to users who exclusively promote body positivity and display bodies that deviate from size ideals, users who promote themselves/products and do not deviate from cultural size ideals might come across as especially insincere. In addition, users who proudly display bodies that challenge body size ideals might not be judged as harshly for also promoting themselves or products. To examine these possibilities, we offered the following research questions:

RQ₁: Does the body size of women in body positive posts influence how viewers react to the posts?

RQ₂: Does the effect of self-promotional persuasion or promotional advertising recognition on the outcome measures vary depending on the body size of the women in the body positive posts?

Method

Research design overview

A 4 (post type) × 2 (body size) between-subjects experiment was designed to test our hypotheses. The post type factor had four conditions: user posts, user posts with self-promotion (hereafter called self-promotional posts), corporate-sponsored non-appearance posts, and corporate-sponsored appearance posts. The body size factor varied the size of the women in the images to be either average or plus-sized. We also included an offset control condition consisting of landscape pictures (e.g., forest, mountains, lakes).

Participants

A sample of 890 female participants was recruited from TurkPrime (see Litman et al., 2017). A modified version of a validated attention check by Berinsky et al. (2014) was used; 39 participants failed the attention check and were excluded from the analyses for a final sample of 851 self-identified females. All participants received $2.50 in compensation in exchange for their participation. Participants ranged from age 18 to 89 (M = 39.4, SD = 12.6), had an average body mass index (BMI) of 26.3 (SD = 7.7), and identified as “White/Caucasian” (n = 636), “African American/Black” (n = 82), “Hispanic/Latino” (n = 35), “Asian/Asian American” (n = 65), “American Indian” (n = 6), “Multiracial” (n = 25), and “Other” (n = 2).

Procedure

All procedures were submitted for approval by the host university’s Institutional Review Board. Participants who chose to take part in the study were directed to an online Qualtrics survey where consent was obtained. After participants provided consent, participants were randomly assigned to one of nine conditions where they viewed 10 Instagram posts indicative of the condition to which they were assigned. After random assignment, participants were required to view each post for at least 10 seconds; after 10 seconds passed, participants could move on at their own discretion. After viewing all 10 Instagram posts in their randomly determined condition, participants then completed the questionnaire. After submitting their responses, participants were directed back to the original host site for compensation.

Stimuli

All body-positive images were sourced from public Instagram accounts that were not verified, had no more than 20,000 followers, and contained the hashtag “bopo.” Across all conditions, participants viewed 10 body-positive posts. Cohen et al. (2019a) was used as a guide to enhance the ecological validity and generalizability of the stimuli. Cohen et al. (2019a) indicated that ∼50% of body-positive posts feature a White individual and ∼50% feature minority women. From this we opted to have five posts feature White women and the remaining posts feature minority women. With respect to age, Cohen et al. (2019a) indicated that the majority (65.0%) of images contained individuals in their 20s, followed by ∼22.0% appearing to be in their 30s, 4.0% in their 40s, and 3.0% appearing older than 40. Again, we sought to mirror this age range in our selected images.

With respect to body-related attributes, Cohen et al. (2019a) used the Current Female Body Image Table (Pulvers et al., 2013) to determine the range of body sizes in body-positive posts. They determined that only roughly 2.0% of the women in body-positive images were perceived as underweight (Current Female Body Image Table images 1–2), 26.0% as normal weight (Table images 3–4), 33.0% as overweight (Table image 5), and 35.0% as obese (Table images 6–9). For the average size conditions, we selected 10 images that were consistent with body types 3 and 4 of the body image table. For the plus-size condition, we selected 10 images ranging from sizes 5 to 9 of the body image table. With respect to body flaws, Cohen and colleagues indicated that just under half (39.6%) of body-positive images contain a person showing at least one flaw. In their work, body flaws included cellulite, stomach rolls/soft belly, stretch marks, acne/skin blemishes, body hair, facial hair, scars, wrinkles, and back fat rolls. Again, we opted to mirror these ratios by having four women with at least one body flaw and six women without any of the aforementioned flaws in each condition.

Lastly, Cohen et al. (2019a) indicated that body-positive images vary in their degree of sexual objectification. Their content analysis indicated that ∼32.0% of body-positive images showed a human in extremely or very revealing clothing, 30.0% in moderately revealing clothing, 25.0% in slightly revealing clothing, and 14.0% in clothing that was not revealing at all. Thus, we selected three images of women in extremely revealing clothing, three in moderately revealing clothing, three in slightly revealing clothing, and one image of a woman in clothing that was not revealing at all in each condition. In sum, the body size of the women in the images varied across conditions while seeking to hold the age, race/ethnicity, presence of body flaws, and sexualization of the images constant across size conditions. Thus, images were selected that reflected a representative range of body-positive posts that naturally appear on Instagram.

Captions

For examples of captions, see Table 1. Each body-positive image was accompanied with text known as a caption to be representative of Instagram posts. Captions served two purposes. First, captions served as a cue that the posts were body positive in nature. Captions included common body-positive statements such as critiquing beauty ideals and celebrating diverse body size, as well as popular hashtags such as #bopo, #bodyposi, #effyourbeautystandards, #plussize, and #losehatenotweight (Cohen et al., 2019a). The body positive messaging in the captions was held constant across all conditions. Second, captions were used to manipulate the post type (user-generated, self-promotional, corporate-sponsored). In addition to the body-positive messaging that was held constant, captions in the self-promotional condition included language encouraging audience members to “follow” the poster on Instagram, engage with them by commenting or liking the posts, and/or link to their other online profiles/websites. Additional hashtags for this condition focused on increasing engagement (e.g., #likeforlike, #followforfollow). Captions in the corporate-sponsored conditions included additional brief language that encouraged the purchasing of products or services (e.g., “Check out our website, link in bio!”). The language mirrored the type of promotional statements that regularly accompany body-positive posts on Instagram. For the corporate-sponsored non-appearance enhancing product condition, users promoted a variety of different clothing types. Finally, for the corporate-sponsored appearance enhancing condition users promoted products and services designed to enhance traditional beauty standards by promoting diet programs, exercise programs, make-up, or shapewear.

Table 1. Sample captions for each condition.

User-generated	User-generated self-promotion	Corporate-sponsored appearance	Corporate-sponsored non-appearance
(1) We won’t be distracted by comparison if we are captivated by purpose. #allbodiesaregoodbodies #bodypositive #ootd #instastyle #bopo	(1) We won’t be distracted by comparison if we are captivated by purpose. Like and comment below, and don’t forget to check out my YouTube! #allbodiesaregoodbodies #bodypositive #ootd #instastyle #bopo #instagrammer #youtuber	(1) We won’t be distracted by comparison if we are captivated by purpose. I’ve found my purpose. Due to our new fitness plan I’m refocused and getting fit. Check out their plan today with my 20% off link in bio! # allbodiesaregoodbodies #bodypositive #ootd #instastyle #bopo	(1) We won’t be distracted by comparison if we are captivated by purpose. Be captivated our new line. Shop 20% off all SALE items now!, #allbodiesaregoodbodies #bodypositive #ootd #instastyle #bopo
(2) Be done shrinking your mind body and soul. Take up space. #allbodiesaregoodbodies #instagood #bopo #picoftheday #fun	(2) Be done shrinking your mind body and soul. Take up space. Like this post and drop a if you want to see more from me! #likeforlike #allbodiesaregoodbodies #instagood #bopo #picoftheday #fun #instablogger	(2) Be done shrinking your mind body and soul. Take up space. STANDOUT in a room with our new line of flaw blurring facial products, swipe up in our stories to learn more!, #allbodiesaregoodbodies #instagood #bopo #picoftheday	(2) Be done shrinking your mind body and soul. Take up space. Like you, our fabrics don’t shrink. Shop now!, #allbodiesaregoodbodies #instagood #bopo #picoftheday #fun

Measures

All items were measured using 7-point Likert scales (1 = strongly disagree; 7 = strongly agree) unless otherwise noted.

Persuasion knowledge – recognition of promotional advertising

One item was adopted from Boerman and colleagues (2012) (i.e., “The Instagram posts you viewed were advertising products or services”) and four items were adopted from Tutaj and van Reijmersdal (2012) to measure recognition of promotional advertising (e.g., “The aim of the Instagram posts you viewed was to make people like certain products or services”; ω = 0.99 [95% CI: 0.983, 0.988]). A confirmatory factor analysis (CFA) yielded factor loadings ranging from 0.95 to 0.97 with the following fit statistics: χ²(5, N = 756) = 234.36, CFI = 0.97, SRMR = 0.01, RMSEA = 0.25 [90% CI: 0.22, 0.27].

Persuasion knowledge – recognition of self-Promotion

One item was created to measure the degree to which the posts contained self-promotion: “The Instagram posts you viewed were self-promotional.”

Persuasion knowledge – moral appropriateness of sponsored content

This subscale established by Boerman and colleagues (2018) asked participants to indicate how morally appropriate they found the Instagram posts. Participants viewed the statement, “I think that showing body positive content in this manner on Instagram is … ” followed by a series of seven bipolar adjectives (Inappropriate/Appropriate; Wrong/Right; Unacceptable/Acceptable; Bad taste/Good taste; Undesirable/Desirable; Unfair/Fair; Illegitimate/Legitimate). Each item was rated on a 7-point semantic differential scale (ω = 0.97 [95% CI: 0.966, 0.975]).

Body positive reasons for posting

Six items were constructed to assess how participants viewed the users’ reasons for sharing the posts on Instagram (ω = 0.90 [95% CI: 0.886, 0.920]). The stem for all items read, “The women in the photos that you viewed shared these images on Instagram to … ” followed by a series of potential reasons. Sample items included, “to display diverse bodies,” “to promote inclusivity,” and “to challenge traditional beauty standards.” A CFA yielded factor loadings ranging from 0.62 to 0.89 with the following fit statistics: χ²(9, N = 748) = 151.11, p < .001; CFI = 0.95; SRMR = 0.03; RMSEA = 0.15 [90% CI: 0.125, 0.166].

Ideal body stereotype scale

The Ideal Body Stereotype Scale (IBSS-R) subscale (Stice et al., 2016) was employed to measure participants’ awareness about sociocultural ideals and the desirability of certain bodily attributes (e.g., “Slender women are more attractive”; “Women who are in shape are more attractive”). Given the nature of our study, an a priori decision was made to drop one item from the scale (i.e., “Shapely women are more attractive”). The purpose of the scale is to measure perceptions of cultural appearance ideals and given the nature of our study, we anticipated shapely would be interpreted as referring to weight/size rather than a curvaceously thin body ideal (e.g., Harrison, 2003). Higher scores indicated higher stereotype endorsement (ω = 0.91 [95% CI: 0.891, 0.918]).

SATAQ-4R internalization: thin/low body fat subscale

The female version of the four-item SATAQ-4R (Schaefer et al., 2016) was used to measure participants’ desire for a thin physique; higher scores indicated a greater desire for a thin body. A sample item was, “I want my body to look thin” (ω = 0.92 [95% CI: 0.906, 0.927]). See Thompson et al. (2018) for the theoretical value and empirical appropriateness of independently measuring both the IBSS-R (for awareness of cultural appearance ideals) and the SATAQ-4R (for the personal pursuit of thinness).

Image evaluations

Two items were included to measure the perceived weight of the women in the images. The stem read, “The women in the photos were … ” with two sets of bipolar adjectives (Skinny/Overweight; Thin/Heavy) evaluated on 7-point scales (Spearman’s rho = 0.95). Three items were included to measure the perceived sexualization of the images as a control variable. The stem read, “The women in the photos appeared in a way that was … ” with three sets of bipolar adjectives (Not sexual/Sexual; Not revealing/Very revealing; Fully clothed/Minimally clothed) evaluated on 7-point scales (ω = 0.81 [95% CI: 0.782, 0.835]). Finally, a single item was used to measure the perceived attractiveness of the women in the images. The stem read, “The women in the photos were … ” with the bipolar adjectives of unattractive/attractive.

Results

Analysis plan

Our post type factor contained four conditions (user posts, self-promotional posts, corporate-sponsored non-appearance posts, and corporate-sponsored appearance posts). The size factor contained two conditions (average, plus size) and we had an offset control for a total of nine conditions. We first compared via ANOVAs the user-post and self-promotional conditions on recognition of self-promotional persuasion and the user-post and corporate-sponsored conditions on recognition of promotional advertising (H_1a and H_1b). Next, we tested via PROCESS Model 4 (Hayes, 2017) the effect of each unique form of persuasion recognition (i.e., self-promotion or advertising) on the outcomes of body-positive reasons for posting (H_2a and H_2b), moral appropriateness (H_3a and H_3b), and the two beauty ideal scales (IBSS-R for H_4a and H_4b; SATAQ-4R for H_5a and H_5b). Finally, we examined the effect of the size factor on the outcome measures (RQ₁) and if the size factor moderated the effects of persuasion recognition on the outcomes (RQ₂).

H_1a stated that viewers are more likely to recognize body-positive posts as self-promotional persuasion when users promote their online profiles relative to when they do not. A two-way ANOVA was conducted with the post type factor (user vs. self-promotion condition) and size factor as independent variables and the recognition of self-promotional persuasion item as the dependent variable. There was a significant main effect for post type, F(1, 379) = 29.47, p < .001, η² = 0.07; there was not a significant main effect for the size factor, F(1, 379) = 0.005, p = .95, η² < 0.001, and there was not a significant interaction effect, F(1, 379) = 0.82, p = .37, η² = 0.002. Thus, H_1a was supported. Next, a two-way ANOVA was conducted with the post factor (user vs. corporate-sponsored post conditions) and the size factor as independent variables and the recognition of promotional advertising as the outcome. There was a significant main effect for post type, F(2, 556) = 526.26, p < .001, η² = 0.65; there was not a significant main effect for the size factor, F(1, 556) = 2.43, p = .13, η² = 0.004, and there was not a significant interaction effect, F (2, 556) = 0.78, p = .46, η² = 0.003. Thus, H_1b was supported. See Table 2 for descriptive statistics and a zero-order correlation matrix and Table 3 for means and standard deviations across all post conditions (including the offset control) for each outcome, as well as the results of Bonferroni adjusted pairwise comparisons.

Table 2. Descriptive statistics and correlations for study variables.

Variable	n	M	SD	1	2	3	4	5	6
1. Self-promotion, recognition	756	5.07	1.65	–
2. Advertising recognition	756	4.19	2.27	0.08*	–
3. Reasons: body positive	748	5.67	1.20	−0.05	−0.31*	–
4. Moral appropriateness	851	5.72	1.45	−0.08*	−0.25**	0.52**	–
5. IBSS-R	851	4.35	1.26	0.05	0.09*	−0.06	−0.14**	–
6. SATAQ-4R	851	3.99	1.63	0.06	0.06	−0.02	−0.09**	0.46**	–

*p < .05.

**p < .001.

Table 3. Dependent variables by post type condition.

	User (n = 189)		Self-promotion (n = 194)		Corporate non-appearance (n = 186)		Corporate, appearance (n = 187)		Control (n = 95)
Variable	M	SD	M	SD	M	SD	M	SD	M	SD
Self-promotion, recognition	4.67a	1.75	5.53b	1.35	5.14bc	1.67	4.92c	1.69	–	–
Advertising, recognition	2.27a	1.53	2.55a	1.61	5.93b	1.16	6.08b	1.15	–	–
Body positive, reasons	6.06a	.96	5.90a	1.06	5.58b	1.23	5.11c	1.30	–	–
Moral, appropriateness	5.94a	1.35	5.83a	1.15	5.58a	1.52	5.05b	1.74	6.65c	0.51
IBSS-R	4.32	1.38	4.32	1.21	4.31	1.22	4.44	1.20	4.31	1.34
SATAQ-4R	3.93	1.72	4.00	1.57	3.82	1.69	4.18	1.55	4.04	1.63

Note. For each row, means with a different subscript indicate significant differences at p < .05 after the Bonferroni adjustment. All variables were measured on 1–7 scales. Participants did not vary by age or race across conditions.

H₂ stated that the more body-positive posts are recognized as containing self-promotional persuasion (H_2a) or promotional advertising (H_2b), the less viewers will ascribe body-positive reasons for users’ posts. Model 4 in the PROCESS macro (Hayes, 2017) was used to estimate the indirect effect of the post factor (X) on the body-positive reason outcome (Y) through the respective mediator (M) of persuasion recognition. For all mediation analyses, effect estimates were reported with their associated 95% bias-corrected bootstrap confidence intervals based on 10,000 resamples.

Supporting H_2a, participants reported believing the posts in the user condition were shared for the reason of promoting body positivity to a greater extent than the posts in the self-promotional condition, b = − 0.08 (SE = 0.03), 95% CI = [−0.14, −0.03], through the recognition of self-promotional persuasion. Supporting H_2b, participants reported believing the posts in the user condition were shared for the reason of promoting body positivity to a greater extent than those in the corporate-sponsored non-appearance condition, b = −0.53 (SE = 0.16), 95% CI = [−0.85, −0.23], and the corporate-sponsored appearance condition, b = − 0.55 (SE = 0.17), 95% CI = [−0.88, −0.24], through the recognition of promotional advertising.

H₃ predicted that the more body-positive posts are recognized as containing self-promotional persuasion (H_3a) or promotional advertising (H_3b), the less morally appropriate viewers will find them. Supporting H_3a, relative to the user condition, participants evaluated the posts as less morally appropriate in the self-promotional condition, b = − 0.08 (SE = 0.04), 95% CI = [−0.17, −0.01], through the recognition of self-promotional persuasion. Supporting H_3b, relative to the user condition, participants evaluated the posts as less morally appropriate in the corporate-sponsored non-appearance condition, b = − 0.59 (SE = 0.21), 95% CI = [−1.01, −0.16], and the corporate-sponsored appearance condition, b = − 0.61 (SE = .22), 95% CI = [−1.05, −0.17], through the recognition of promotional advertising.

The final hypotheses stated that the more body-positive posts are recognized as containing self-promotional persuasion (H_4a and H_5a) or promotional advertising (H_4b and H_5b), the less viewers are expected to report that their culture embraces inclusive beauty standards (H₄) and the less viewers are expected to personally embrace inclusive beauty standards (H₅). Again, Model 4 in PROCESS was used to estimate the indirect effects. The post condition served as the IV, recognition of self-promotional persuasion or promotional advertising served as the mediator, with the ideal body stereotype scale (IBSS-R) serving as the outcome for H₄ and the thin/low body fat subscale (SATAQ-4R) serving as the outcome for H₅.

The results indicated that, relative to the user condition, participants in the self-promotional condition did not score significantly higher on the IBSS-R, b = 0.02 (SE = 0.04), 95% CI = [−0.07, 0.11], or the SATAQ-4R, b = −0.001 (SE = .05), 95% CI = [−0.11, 0.11], due to increased recognition of self-promotional persuasion. H_4a and H_5a were not supported. However, relative to the user condition, participants in the non-appearance corporate-sponsored condition, b = 0.39 (SE = 0.15), 95% CI = [0.09, 0.70], and the appearance corporate-sponsored condition, b = 0.41 (SE = 0.16), 95% CI = [0.09, 0.72], scored higher on the IBSS-R through increased recognition of promotional advertising, supporting H_4b. Relative to the user-post condition, participants did not score significantly higher on the SATAQ-4R in the non-appearance corporate-sponsored condition, b = 0.33 (SE = 0.18), 95% CI = [−0.04, 0.67], and the appearance corporate-sponsored condition, b = 0.34 (SE = 0.19), 95% CI = [−0.04, 0.69]. H_5b was not supported.

RQ₁ asked if the body size of the women in the posts influenced how viewers reacted to the posts. A two-way ANOVA with the size and post type factors as independent variables confirmed that participants viewed the women in the plus-size condition (M = 6.47, SD = 0.80) as weighing significantly more than the women in the average size condition (M = 3.53, SD = 1.15), F(1, 748) = 1669.76, p < .001, η² = 0.69. There was not a significant main effect for the post condition, F(3, 748) = 0.081, p = .97, η² < 0.001, or a significant interaction effect, F(3, 748) = 0.45, p = .72, η² = 0.002. To probe RQ₁, mediation analyses were conducted with the size factor as the IV, perceived weight as the mediator, and each outcome reported above as a dependent variable. Although attempts were made to only vary perceived body size across conditions, perceptions of sexualization and attractiveness were included as covariates in all analyses. The indirect effect of the body size factor, through perceptions of weight, was significant for the body-positive reasons for posting outcome. The size factor positively influenced perceptions of weight, which, in turn, increased perceptions that the posts were shared to promote body positivity, b = 0.76 (SE = 0.13), 95% CI = [0.51, 1.04]. Similarly, the size factor, through increased perceptions of weight, significantly influenced how morally appropriate participants evaluated the posts to be, b = 0.80 (SE = 0.17), 95% CI = [0.47, 1.13]. The size factor, through perceptions of weight, did not influence either of the body image outcomes (i.e., IBSS-R or SATAQ-4R). See Table 4 for detailed results of each test.

Table 4. Indirect effect estimates: size (X), perceived weight (M), outcome (Y).

Outcome (Y)	b	SE	LLCI	ULCI
Moral appropriateness Body positive reasons IBSS-R SATAQ-4R	0.80 0.76 0.24 −0.12	0.17 0.13 0.14 0.18	0.47 0.51 −0.03 −0.48	1.13 1.04 0.52 0.24

Note. Estimates control for perceived sexualization and attractiveness.

RQ₂ asked whether the body size of the women featured in the posts interacts with the presence of promotional persuasion to exacerbate or mitigate the negative effects promotional persuasion were anticipated to have. To explore this question, Model 14 in PROCESS was used to examine whether the size factor moderated the relationship between self-promotional and promotional advertising recognition and the outcome variables. Thus, we tested whether the indirect effects reported above in our hypothesis tests varied depending on whether participants viewed images of women in the average size condition or the plus-size condition of the size factor. The significant effects reported in support of H_2a, H_3a, and H_4b were not moderated by the size factor. However, the effects reported in support of H_2b and H_3b were moderated by the size factor, controlling for perceived sexualization and attractiveness. For the outcome of body-positive reasons for posting (H_2b), the index of moderated mediation (i.e., the difference between the indirect effect in the average size and plus-size condition) was significant for each comparison: user vs. non-appearance corporate-sponsored condition, index = 0.34 (SE = 0.16), 95% CI = [0.01, 0.65]; user vs. appearance corporate-sponsored condition, index = 0.35 (SE = 0.16), 95% CI = [0.01, 0.66]. The same pattern of contingent, moderated mediation emerged (see Holbert & Park, 2020). Increased recognition of promotional advertising only reduced perceptions that the images were shared for body-positive reasons in the average size condition; the recognition of promotional advertising did not influence the degree to which participants thought the posts were shared for body-positive reasons in the plus-size condition. Likewise, contingent moderated mediation was detected for the moral appropriateness outcome (H_3b) in a manner that followed a similar pattern: user vs. non-appearance corporate-sponsored condition, index = 0.58 (SE = 0.19), 95% CI = [0.21, 0.94], user vs. appearance corporate-sponsored condition, index = 0.60 (SE = 0.19), 95% CI = [0.21, 0.97]. Increased recognition of promotional advertising reduced the perceived moral appropriateness of the posts in the average size condition, but not the plus-size condition. Thus, posts that contained images featuring women who visually deviate from traditional beauty ideals eliminated the negative effects recognition of promotional advertising had on the body-positive reason for posting and moral appropriateness outcomes. See Online Supplementary Tables 1–5 for complete results of the moderated mediation analyses conducted to probe RQ₂.

Discussion

The results of our study indicate how self-promotion and corporate commodification can undermine the body positivity movement online. First, participants readily recognized cues that naturally appear on Instagram and function to promote users or products. The more participants recognized the presence of self-promotional persuasion or advertising, the more they changed their attributions for why users shared the body-positive posts – decreasing explanations related to promoting inclusivity and body appreciation. Further, increased recognition of both forms of persuasion led participants to evaluate the posts as less morally appropriate. Notably, increased recognition of promotional advertising led participants to report that others in society embrace traditional beauty norms to a greater extent (as assessed by the IBSS-R). Although recognizing promotional persuasion did not significantly influence the degree to which participants reported desiring a thinner body (as assessed by the SATAQ-4R), the estimates approached statistical significance and warrant future research. Overall, the recognition of promotional advertising produced relatively more negative effects than the recognition of self-promotion; whether the advertising posts marketed clothes or products/services more directly opposed to body-positive ideals (e.g., make-up, diets) was inconsequential.

Two research questions were posed to examine whether the physical appearance (specifically the body size) of the women in the posts influenced the outcome measures directly or interacted with the recognition of promotional persuasion to affect the outcome measures. Controlling for the perceived sexualization and attractiveness of the posts, the more participants perceived the women to physically deviate from traditional size ideals that value thinness, the more they attributed body-positive motives to the users’ posts and the more morally appropriate they rated the posts. Perceptions of source body size did not influence either body image outcome measure. Finally, there was evidence of conditional indirect effects for the outcome measures of body-positive reasons for posting and moral appropriateness. The significant decrease in each outcome, due to the increased recognition of promotional advertising, only held for the average size condition. Put differently, the recognition of promotional advertising did not induce adverse effects in the plus-size condition.

Theoretical implications

Although the persuasion knowledge model has been widely used to examine how people respond to persuasive appeals in a variety of contexts (including Instagram; Boerman, 2021; Johnson et al., 2019), examining how the recognition of persuasion influences the perceived moral appropriateness of persuasive appeals has received relatively less attention. Recent conceptual (Hudders et al., 2017) and measurement (Boerman et al., 2018) work has encouraged researchers to focus on this distinct and less studied evaluative dimension of the persuasion knowledge model. van Dam and van Reijmersdal (2019) and Boerman et al. (2021) did not find evidence connecting an increased recognition of persuasion with a reduction in the moral appropriateness of persuasive appeals. In fact, van Dam and van Reijmersdal (2019) reported evidence that adolescents actually find it perfectly appropriate for influencers on YouTube to market products. Likewise, Boerman and colleagues (2021) found that advertising disclosures during television programs can lead to short- and long-term recognition of persuasion. However, this recognition of persuasion was not associated with viewing this tactic as more or less appropriate.

Moving forward, the persuasion knowledge model can be advanced by establishing formal boundary conditions for when the recognition of persuasion influences perceptions of moral appropriateness. In their foundational work, Friestad and Wright (1994) discussed how the perceived appropriateness of certain persuasive tactics is likely conditional; the same tactic may be deemed appropriate in one context (e.g., using fear-appeals to get adults to refrain from drinking and driving) and not appropriate in a different context (e.g., using fear-appeals to sell ice cream). The source, message, context, and target audience can all serve as conditional factors that moderate the relationship between persuasion recognition and moral appropriateness judgments. Our work suggests that various forms of promotional persuasion were deemed less appropriate (albeit to different degrees) when accompanying prosocial messaging on social media. From our findings and recent work (Boerman et al., 2021; van Dam & van Reijmersdal, 2019), we can speculate that the relationship between persuasion recognition and moral appropriateness is contingently moderated by the presence of altruistic persuasion: there is a negative relationship between the recognition of promotional persuasion and moral appropriateness when promotional persuasive appeals are presented in conjunction with altruistic persuasive appeals. However, this relationship is non-significant in the absence of altruistic persuasive appeals (ceteris paribus). Future work is needed to test this inductively derived proposition.

Practical implications

Content analyses and critiques of the body-positive movement detail, respectively, the ubiquity of marketing in body-positive posts online and concerns associated with the corporate commodification of the body-positive movement (Cwynar-Horta, 2016; Lazuka et al., 2020; Miller, 2016). Overall, the findings of this study indicate that body-positive posts that contain self-promotion are less effective at promoting body appreciation than posts without such promotion. However, the detriments of self-promotion are less severe in comparison to the detriments associated with body-positive posts that market products and services.

Our results indicate that viewers are not blind to the selling tactics of content creators despite the inclusion of body-positive messaging. From a content creator’s perspective, individuals and companies need to be more strategic with the quantity and type of branded body-positive content on their pages if they truly desire to promote body positivity as well as their brands. Pöyry et al. (2019) suggest that the perceived authenticity of sponsored photos on social media can affect how favorably viewers react to the images. For posts to be effective, they need to be in alignment with the content creators’ perceived values (Pöyry et al., 2019). Thus, in the context of the body-positive movement, a viewer’s negative attribution of content creators’ intent for posting may lead them to become less engaged with content creators whom they feel are inauthentic and are using body-positive messaging for personal gain.

As prosocial movements on social media become more commercialized, content creators should be wary of the role they are potentially playing in the misappropriation of such movements. The frequent pairing of prosocial messaging and product promotion may dull the intended effect of these movements, making them more recognizable as marketing ploys rather than movements promoting admirable causes. Such effects may be heightened if content creators are promoting products that are inconsistent with the values of the prosocial movement. However, our results indicate that the marketing of any product might have the same adverse effects. Future research should examine further how the type of products Instagram users market in prosocial posts detracts from the effectiveness of any prosocial messaging. A limitation of this work is that the corporate non-appearance condition contained advertising for apparel that could be viewed as related to appearance or counter to the body-positive movement. Although we sought to differentiate appearance from non-appearance product advertising, we did so in a manner that was relatively subtle. Future studies should better maximize the experimental variance across product type conditions to truly detect whether marketing, in general, produces negative effects or if specific forms of marketing are especially deleterious. Furthermore, the use of landscape images as an offset control should be considered in future work as such images might aid message processing and function as a limitation.

Future experimental studies might also seek to vary the presence of promotional persuasion in different ways. A central way we manipulated the presence of self-promotion and advertising was through adding additional text to accompany the Instagram posts. As such, all posts had the same body-positive messaging; however, images in the self-promotion condition (e.g., “Like and comment below, and don’t forget to check out my YouTube!”) and corporate-sponsored conditions (e.g., “Like you, our fabrics don’t shrink. Shop now!”) contained brief additional statements. It is possible that the lack of equivalence in the amount of messaging attracted/detracted attention from the visual stimuli unevenly across conditions (see Harrison et al., 2006). Although measuring the causal mechanism of persuasion recognition helps reduce this concern in our study, it is nonetheless a limitation for future researchers to consider.

Although a central focus of this work was examining the negative effects self-promotion and corporate commodification can have on the body-positive movement, there are some positive aspects of our findings. Overall, participants viewed the body-positive posts as morally appropriate. In terms of moral appropriateness, it would be more accurate to state that promotional persuasion reduces the perceived moral appropriateness of posts rather than makes viewers evaluate the posts as inappropriate (see Table 3; means ranged from 5.05 to 5.94). Thus, whereas participants tended to view posts as morally appropriate, this sentiment was depreciated when viewing body-positive content that included promotional persuasion.

Furthermore, images in the plus-sized condition were not viewed as less morally appropriate as recognition of promotional advertising increased. As such, body-positive posts that display bodies that deviate from traditional beauty ideals (see Afful & Ricciardelli, 2015) might have a greater ability to include messaging extraneous to the body positivity movement without undermining efforts to promote body appreciation. The central aim of the body-positive movement is to highlight and uplift bodies that are not historically celebrated by mainstream society. Accordingly, body-positive posts that contain images of women with bodies that do not conform to traditional beauty ideals might be perceived as particularly authentic and genuine, which a degree of self-promotion or advertising does not attenuate. Nonetheless, it is important to note that empirical evidence documenting how viewers do react to posts is somewhat divorced from the issue of whether Instagram users and companies ought to use the body-positive movement to advance their own interests from an ethical (see Cwynar-Horta, 2016; Miller, 2016; Rodgers et al., 2019) or corporate social responsibility perspective (see Hayes & Carr, 2021).

In addition to exploring issues germane to the commodification of the body-positive movement, future work is needed that investigates the role-specific platforms, such as Instagram, play in the evaluation of prosocial movements online. Given that image-centric social media platforms such as Instagram are commonly used for self-promotion and attention seeking (Lee et al., 2015; Sung et al., 2016) and viewers by default often assume that Instagram users digitally modify the pictures they post (Marwick, 2015; Vendemia & DeAndrea, 2018), it may be difficult for users to escape the appearance that they are sharing content to (primarily) benefit themselves. As such, the promotion of prosocial messaging on Instagram may be ineffective or counterproductive because viewers see self-promotion as an intrinsic component of the platform. Future studies should explore how normative perceptions regarding user behavior on specific platforms impact how viewers interpret and respond to messages hosted on the platform.

Finally, researchers have discussed (Cohen et al., 2020) and documented (e.g., Cohen et al., 2019a) the degree to which body-positive posts are sexualized and appearance-oriented. In general, women who appear in a sexualized manner can experience dehumanization (see Roberts et al., 2018) and recent work found that, for women, seeing sexualized images of other women on Instagram was associated with endorsing gender stereotypes that objectify women and evaluating the women who appeared in a sexualized manner as less warm and competent (see Biefeld et al., 2021). Vendemia et al. (2021) experimentally varied the degree to which body-positive posts were sexualized and found that the more viewers perceived body-positive images on Instagram to be sexualized, the more they (a) objectified themselves and the women in the images, (b) thought the images were shared for self-serving reasons, and (c) endorsed traditional beauty ideals. Although the perceived sexualization of the images was a covariate in this study and not a central focus of our work, greater perceived sexualization was associated with lower scores on the body-positive reason for sharing scale and the moral appropriateness scale. Furthermore, perceived sexualization was positively associated with the beauty ideal scales (see Online Supplementary Tables 3–5). Future research is needed to investigate the unintended negative effects sexualized body-positive content can have on viewers.

Conclusion

As prosocial movements become more commercialized on social media, it is important to understand the effects such appropriation has on their efficacy. Our findings demonstrate that both the source of body-positive content as well as the symmetry between visual and text-based body-positive messages matter in shaping audience perceptions. Misalignment between the features of body-positive posts and the perceived values of the movement has negative consequences for the promotion of body appreciation. Thus, the present study helps to elucidate how the appropriation of body-positive content by both average users and companies can hinder the movement's ultimate goals. Given the rise of user-generated movements on social media, this work can more broadly inform how the commercialization of prosocial movements produces adverse effects.

Acknowledgements

The authors would like to acknowledge Dr. Teresa Lynch and Dr. Megan Vendemia for their very helpful feedback and encouragement during the development of this study.