The Ouroboros of Seeking Validation? Exploring the Interconnection of Appearance (Dis)satisfaction and Content Creation on Social Media

Appearance Satisfaction and Content Creation

Previous empirical studies on content creation are focused mainly on selfies and their relation to self-esteem. In a study by Pounders et al. (2016), participants revealed that, rather than striving to appear authentic and real, the informants indicated a desire to portray themselves as living happy lives and displaying positive physical self-images, regardless of their internal feelings. The initial motivator for posting selfies was impression management, and informants disclosed posting both genuine and non-genuine posts to manage the impression of happiness and physical appearance. Findings also showed that self-esteem plays an integral role in understanding selfie-posting, and the number of ‘likes’ can positively or negatively affect the person's self-esteem.

In other words, taking and posting happy selfies can provide a self-esteem boost, regardless of the initial mindset of the person posting them (Kedzior et al., 2016), pointing to the empowering potential of social media and selfies. As networked technologies allow for capturing and sharing embodied experience (Tiidenberg & Gómez Cruz, 2015), social media can empower its users. Moreover, taking and sharing selfies offers groups that have previously been objectified and denied agency possibilities to challenge their public portrayal (Barnard, 2016). However, the very same studies have also suggested that such empowerment includes a threat to self-esteem, as it requires constant validation from others in the form of likes, retweets and comments. This conditional provision of validation may start to arouse the need to perform in specific ways to meet the audience's expectations to receive the aforementioned comments and likes (Tiidenberg & Gómez Cruz, 2015). Seeking validation from peers and its effect on self-esteem can be seen as a perfect illustration of what is described in Fredrickson and Roberts's (1997) objectification theory. The theory claims that the pervasiveness of sexual objectification gradually socialises women and girls to internalise an observer's perspective of their bodies and view themselves as an object to be looked at and evaluated based on appearance.

Despite the possibilities for self-presentation, as well as empowerment and broader appearance ideals, the online presentations of other people in general predominantly promote such attributes as youthfulness and thinness. Fulfilling these appearance ideals is impossible to achieve for most women, who are said to be vulnerable to body dissatisfaction, particularly concerning body shape and weight (Groesz et al., 2002; Tiggemann & Slater, 2013). The most often used and generally accepted framework for understanding the prevailing body dissatisfaction of women is the sociocultural theory, which claims that contemporary beauty ideals are reinforced and sustained by sociocultural influences, most notably parents, peers and media (e.g. Thompson et al., 1999). These sociocultural ideals are hard to overcome individually: if the source is flawed, it cannot be fixed by retouching one's photos. A study by Shome et al. (2020) established that the opportunity to retake and modify selfies before posting them on social media predicted a decrease in the participant's social anxiety. However, other negative outcomes, like decreased feelings of self-confidence and the desire to undergo cosmetic surgery, were higher among the participants who had an option to retouch their selfies. These findings were partly consistent with the conclusion of Mills et al. (2018), who correspondingly observed that taking and uploading selfies resulted in negative feelings: decreased confidence and feelings of anxiety, irrespective of the opportunity to edit the selfie.

The above-cited studies focus on selfie-related behaviours and appearance concerns but speak in terms of general appearance. Other studies have distinguished between facial and body dissatisfaction, and claim that individuals who have greater body satisfaction are, to begin with, more likely to engage in posting selfies (Ridgway & Clayton, 2016). Cohen et al. (2018) found similar associations in their study. However, as selfies predominantly include portrait photos rather than full-bodied images, it may be that the relationships between appearance concerns and selfie-related behaviours may contribute more to facial appearance dissatisfaction than to body dissatisfaction. Moreover, it has been stated that taking and editing of selfies leads to experiences of more negative mood and resulted especially in facial dissatisfaction (Tiggemann et al., 2020b). Similarly, viewing idealised selfies with makeup on social media specifically influenced women's face appearance concerns, rather than concerns about their overall physical appearance (Fardouly & Rapee, 2019). A more extensive approach was found in Wang et al. (2019), establishing a reciprocal relationship among selfie-related behaviours, associating them with self-objectification and appearance concerns among adolescents. Their results indicate that selfie-editing, but not selfie-posting, predicted an increase in adolescents' self-objectification and appearance concerns (both body and face) over time. However, selfie-viewing predicted increases in self-objectification and facial dissatisfaction, but not body dissatisfaction, over time. Moreover, participants' facial dissatisfaction positively predicted selfie-viewing and selfie-editing, whereas body dissatisfaction did not influence subsequent selfie-related behaviours.

To our knowledge, only a few studies (e.g. Ridgway & Clayton, 2016; Wang et al., 2019) have explicitly examined the interconnection of distinguished appearance dissatisfaction and selfie practices. However, despite stressing the central role of appearance dissatisfaction in selfie-related behaviour, only bodily dissatisfaction has been associated with an increased tendency for selfie-posting in these studies. Thus, we use our data to investigate whether body dissatisfaction or facial dissatisfaction predicts content creation in overall, not just selfies, differently on social media.

With this previous research and their findings in mind, we assume that social media increases self-objectification and social comparison that results in appearance (dis)satisfaction, which predicts content creation. Thus, we propose the following hypotheses: People who are less satisfied with their facial appearances create more content on social media than those who are satisfied with their facial appearances (H1.1). Second, people who are less satisfied with their bodily appearances create more content on social media, compared to those who are satisfied with their bodily appearances (H1.2).

Appearance Satisfaction and Content Creation on Different Social Media Platforms

Previous research has found several differences among social networking sites, in terms of the method of interaction, self-presentation and importantly, the types of social ties that those different platforms allow. As social media platforms have been designed for specific user purposes and are structured in varied ways, their user profiles differ accordingly. Like other countries, in Finland, active social media use and content generation are more common among women and young people (Ertiö et al., 2018). Young people on Instagram are highly focused on self-presentation through visual means and aim to control how others receive them or their tastes (Sheldon & Bryant, 2016).

When it comes to platform-specific features, Instagram, compared to other social media sites, is based on embodying one's identity. ‘Selfies’, i.e. photographs people take of themselves, are abundant on Instagram (Ridgway & Clayton, 2016). Based on its visuality, studies have found negative consequences for engaging in the appearance-related features of social media. For example, a meta-analysis by Mingoia et al. (2017) found that internalisation of sociocultural ideals, such as thinness, is mainly associated with posting and viewing images on social media.

Moreover, social media platforms are also distinguished by their social nature. Facebook is based more on social interaction, which focuses on reciprocity between individuals, as well as consent to social bonding. Facebook is more like an extension of social life outside the Internet, as its users form social ties with people who they typically have known from real life (Manzi et al., 2018). Social interaction on the platform is also more social and reciprocal: on Facebook, people belong to groups with like-minded people or those who are interested in the same things. Moreover, Facebook has become increasingly popular as a market platform for peer-to-peer trade with other users, peers, on Facebook. (Lee et al., 2014).

In contrast to Facebook, on Instagram, a profile can be public and thus, not requiring approval from the person followed. Hence, Instagram invites being followed by and, in particular, following of strangers, including celebrities (cf. Lup et al., 2015.). As a result, increasing the size of different networks, as well as increased visibility, is easier on Instagram Phua et al. (2017). However, the quality of networks may be different on Instagram, where one-sided following is more likely to occur, which also means that information flows in only one direction. Instagram is also more performative than Facebook; it is not based on information-sharing or other social reciprocity in the same way as Facebook. Given its visual nature, it can be seen as focusing more specifically on selective self-presentation and performative identities (e.g. boyd, 2008; Dobson, 2016), as well as more niched performances, such as ‘doing pregnancy’ (Tiidenberg & Baym, 2017). Moreover, the content can also be tagged and thus made available to the so-called ‘attention economy’ (Marwick, 2015, see also Åberg & Salonen in this book) prominent on social media. Besides tagging one's content, tagging, but also sharing publicly available content, for example, cute animals or funny memes, can be seen as a type of performance in and of itself (e.g. Tiidenberg & Baym, 2017).

As previous studies have indicated, Instagram allows for a selective self-presentation, more performative use and, potentially, a more extensive network. Thus, we hypothesise that the relationship between appearance (dis)satisfaction and content creation is more prominent among Instagram users when compared to the users of other social media platforms (H2).

Appearance Satisfaction, Content Creation, Social Media Platforms and Gender

Most of the previously presented findings are consistent with the prediction of Fredrickson and Roberts's objectification theory (1997) that engaging in a self-objectifying activity (like selfie posting) results in worsened body image outcomes for women. This theory claims that sexual objectification in contemporary society socialises girls and women to treat themselves as objects to be looked upon and evaluated based upon bodily appearance, leading to four negative subjective experiences: body shame, appearance anxiety, diminished peak motivational status and reduced connection with one's internal bodily states (Fredrickson & Roberts, 1997). In addition to self-objectification, body image disturbances and appearance concerns of women are often approached with social comparison theory (Festinger, 1954). According to social comparison theory, people seek to compare their appearances with prevailing social standards, including the appearance standards presented in the media. Making evaluative comparisons between their appearance and the idealised and artificially created appearance of women in the media, most women are likely to feel they need to meet that standard. Due to that shortage, female viewers typically end up feeling dissatisfied with their appearance.

From subjective experiences, Terán et al. (2019) identified body shame and appearance anxiety as the most relevant to the potential consequences of selfies for women. Similarly, Tiggemann and Barbato (2018) claimed that objectification theory may be particularly relevant in the context of a photographic social media site like Instagram, with its primary function being posting photos of oneself for others to look and comment. According to their study, women's body dissatisfaction increased in response to viewing images and posted comments on Instagram. In addition to self-objectification, it has been established that social networking sites are associated with upward social comparison and lowered feelings for women (e.g. Fox & Vendemia, 2016). Moreover, contemporary body ideals on social media have become even more unattainable for most women: relatively thin, but also toned and strong (Tiggemann & Zaccardo, 2018). Although most of these previous studies have focused on women, social media has an impact on men also, reproducing stereotypical norms of masculinity in the form of embodying strength and power (Fox & Vendemia, 2016), and it forces men to increasingly rely on their bodies for personal value creation (Hakim, 2015, see also Sarpila in this book).

Moreover, men and women differ in their social media practices. Whereas women typically use social media more actively and are more likely to use social media to view others' photos (Smith, 2014), as well as compare themselves with others, men are more likely to use social media to find or contact friends (Haferkamp et al., 2012). Social networking sites allow for self-conspicuousness, and young women, in particular, present idealised images of themselves on social media (Dhir et al., 2016). Consequently, other young women see these idealised images of their peers and compare themselves with them, causing appearance-related distress (e.g. Holland & Tiggemann, 2016). A recent study found that active social media usage with attractive peers on appearance-based social media resulted in worsened body image for young adult women, whereas interacting with members of the family did not affect body image (Hogue & Mills, 2019).

Moreover, living up to prevailing appearance norms is effectively ensured on social media, where followers, ‘likes’ and positive appearance comments on uploaded content can be seen as rewarding appearance ideals. Prior research reveals a tendency for women to both receive and give compliments on appearance (Åberg, Koivula, Kukkonen, Sarpila, & Pajunen, 2020; Holmes, 1998) and well as place a higher premium on receiving those comments. These gendered commenting norms tend to reinforce the norm of placing even more emphasis on and directing the viewer's focus to women's appearances.

Following these strands of literature, we formulate the next hypotheses: Women who are less satisfied with their appearances create more content on Instagram than equally satisfied men (H3.1). Second, appearance (dis)satisfaction is not associated with men's content creation on social media (H3.2).

Participants

Our data were derived from the survey ‘Finland in the digital age’ (Sivonen et al., 2018), that included a total of 3,724 respondents. The survey data were collected in two parts, using two different sampling methods. First, the survey reached a total of 2,470 participants aged 18–74 years, who were from the initial group of 8,000 Finnish-speakers sampled randomly from the Finnish census. Additionally, the data were improved with 1,254 participants (also aged 18–74 years) from a nationally representative online panel of volunteer respondents administrated by a market research company.

Although the total share represents relatively well the share of the Finnish population, older users were slightly overrepresented. Consequently, the data were post-stratified in terms of age and gender distribution to correspond with the official population distribution in Finland according to the Official Statistics of Finland.

In this study, the analyses were based on social media users, totalling 2,761 respondents who comprise 74 % of the total data. Our final study sample included 1, 319 women (47%) and the mean age was 47.7 (SD = 15.8).

Measures

The applied measures are shown in Table 1. The dependent variable measured how often respondents shared their own created content on social media. The item was a part of a question battery in which we asked a total of 14 sub-questions, with a question: ‘How often do you do the following?’. The responses and their distribution were 1 ‘Never’ (22.9%), 2 ‘Less than weekly’ (51.4%), 3 ‘Weekly’ (22.5%) and 4 ‘Daily’ (3.2%). Only 0.7% of the answers were missing. In the analysis, we focused on predicting active content creation. We dichotomised the variable into a binary, in which 1 differentiates those who created content at least weekly (25.7%) from those who created content less than weekly (74.3%).

The main predictor variables were satisfaction with bodily appearance and satisfaction with facial appearance. The dimensions were explored separately with the same main question, namely, ‘How satisfied are you with (a) your face (b) your body?’ The responses were scored with a 5-point interval scale, in which 1 was labelled as ‘Not content at all’ and 5 was ‘Very content’.

The moderating variables were the preferred social network site (SNS) and gender. Initially, gender was queried via three categories. However, we had just one observation from individuals other than men or women social media users, and consequently, we excluded the ‘other’ category from further analyses. In terms of SNS, we paid especially attention to differences between the users of Facebook and Instagram by differentiating the respondents who used both Facebook and Instagram or just Instagram, from those who only used Facebook. We also filtered those who did not use Facebook or Instagram at all but use some other platform(s) into the category ‘Else’. Accordingly, the final variable labelled as ‘The preferred SNS’ had three categories, namely, 1 ‘Only Facebook’, 2 ‘Instagram’, 3 ‘Else’.

Analysis Strategy

The analysis was performed with Stata 16. First, we conducted logit models to predict the probability of creating content on social media at least weekly, according to satisfaction with facial and body appearance. Next, we focused on the moderating effect of the preferred SNS. We added an interaction term ‘the preferred SNS × appearance satisfaction’ into the base model. Finally, we continued the procedure and added the interaction term ‘the preferred SNS × appearance satisfaction × gender’ to determine whether the effect of appearance satisfaction is potentially gendered.

Instead of presenting odds ratios from the logit models, we estimated the predicted probabilities that describe the adjusted percentage point estimates of the differences in the outcome variables, according to the different levels of predicting variables (Muller & MacLehose, 2014). We illustrated the predicted probabilities into the figures by utilising the user-written coefplot-package (Jann, 2014). In each figure, the y-axis describes the probability of content creation at the different levels of appearance satisfaction indicated at the x-axis. The results of interaction analyses were presented by plotting the relationship between appearance satisfaction and content creation for each moderating variable.