Computers in Human Behavior 144 (2023) 107730

Contents lists available at ScienceDirect

Computers in Human Behavior

journal homepage: www.elsevier.com/locate/comphumbeh

Predicting the use of visually oriented social media: The role of

psychological well-being, body image concerns and sought
appearance gratifications
Johanna M.F. van Oosten a, *, Laura Vandenbosch b, Jochen Peter a
a
Amsterdam School of Communication Research (ASCoR), University of Amsterdam, Nieuwe Achtergracht 166, 1018, WV, Amsterdam, the Netherlands
b
School for Mass Communication Research, KU Leuven, Parkstraat 42, 3000, Leuven, Belgium

A R T I C L E I N F O A B S T R A C T

Handling Editor: Catalina L Toma The present study examined the antecedents of visually oriented social media use based on the Transactional
Model of Social Media and Body Image Concerns. Specifically, we studied how self-esteem, depressive symptoms,
Keywords: body dissatisfaction, and appearance consciousness within the social media context are, via two social media
Adolescents gratifications (i.e., appearance related self-validation and social comparison), related to the use of visually ori­
Emerging adults
ented social media platforms (i.e., Instagram and Snapchat). Based on data from a two-wave panel survey among
Body image
1852 social media users aged 13–25, we found that appearance consciousness and high self-esteem were sig­
Youth
Longitudinal research nificant predictors of the gratifications self-validation and social comparison. Increased body dissatisfaction also
predicted social comparison. Appearance consciousness and self-esteem indirectly predicted the selection of
visually oriented social media platforms via self-validation. These findings enhance our theoretical under­
standing of visually oriented social media use and its underlying mechanisms.

The past years have seen an increase in the use of visually oriented research in the field is cross-sectional (Appel, Marker, & Gnambs, 2020).
social media, such as Instagram and Snapchat, which can be defined as It is thus unclear whether body image concerns and an overall lower
social media platforms whose features emphasize engagement with well-being are the result of visually oriented social media use, as is often
images (Brown & Tiggemann, 2016; Meier & Gray, 2014; Tiggemann & assumed, or whether they (also) predict that type of social media use.
Zaccardo, 2018). Such platforms are especially popular among young The scarce available longitudinal research seems to indicate that the
people: Three quarters of teens and emerging adults use Instagram and latter is the case (e.g., Aalbers, Mcnally, Heeren, Wit, & Fried, 2019),
Snapchat, with Snapchat being at the top of the most often used plat­ thereby calling for more research on who may be prone to use visually
forms by teens (Anderson & Jiang, 2018; Pew, 2019). Against this oriented social media (Brown & Tiggemann, 2016; Meier & Gray, 2014).
background, research has started to focus on how the use of visually Second, research has rarely investigated indirect relationships between
oriented social media relates to users’ well-being and body image. This visually oriented social media use and psychological well-being or body
research showed that the use of highly visually oriented social media image concerns. Moreover, when such indirect relationships are studied
platforms (i.e., Instagram and Snapchat) was associated more strongly (e.g., de Lenne, Vandenbosch, Eggermont, Karsay, & Trekels, 2020),
with body image concerns and psychological well-being than the use of conclusions on indirect relationships are mostly based on cross-sectional
more textually oriented social media platforms such as Facebook research designs, which are subject to multiple methodological prob­
(Engeln, Loach, Imundo, & Zola, 2020; Marengo, Longobardi, Fabris, & lems (Maxwell & Cole, 2007).
Settanni, 2018). The present study aimed to fill these gaps by longitudinally investi­
Although existing studies are instructive, there are still two main gating antecedents of visually oriented social media use and the mech­
gaps in the literature on the relationship between visually oriented so­ anisms that underlie these relationships. The study is based on the
cial media use and well-being or body image concerns. First, little is Transactional Model of Social Media and Body Image Concerns (see
known about the directionality of this relationship because most Perloff, 2014, Fig. 1 on p. 368). This model posits that seeking

* Corresponding author. The Amsterdam School of Communication Research, ASCoR, University of Amsterdam, Postbus 15791, 1001, NG, Amsterdam, the
Netherlands.
E-mail address: j.m.f.vanoosten@uva.nl (J.M.F. van Oosten).

https://doi.org/10.1016/j.chb.2023.107730
Received 25 March 2021; Received in revised form 31 January 2022; Accepted 28 February 2023
Available online 2 March 2023
0747-5632/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
J.M.F. van Oosten et al. Computers in Human Behavior 144 (2023) 107730

appearance related social media gratifications, such as self-validation (e.g., each post on Instagram or Snapchat includes an image while an
and social comparison, predicts the use of visually oriented social image is not required for posts on Facebook or Twitter). The unique
media. Appearance-related social media gratifications are, according to features of visually oriented social media distinguish these platforms
the model, themselves driven by psychological well-being and body from other more textually oriented social media platforms (e.g., Face­
image concerns. Specifically, the present study thus investigated book, Twitter), which may be suitable for image sharing but are less
whether psychological well-being (i.e., self-esteem, depressive symp­ focused on appearance-related self-presentations. Accordingly, research
toms) and body image concerns (i.e., body dissatisfaction, appearance has shown that textually oriented platforms are used less for image
consciousness) indirectly predict the use of visually oriented social sharing and editing or for comparing one’s appearance with that of
media platforms through appearance-related social media gratifications others (Engeln et al., 2020; Mull & Lee, 2014; Sheldon & Bryant, 2016;
(i.e., self-validation and social comparison). Thelwall & Vis, 2017).
The study focused on adolescents and emerging adults because they The features of visually oriented media also allow for an optimized
are the most frequent users of visually oriented social media, such as visual self-expression and self-promotion (notably through the use of
Instagram and Snapchat (Alhabash & Ma, 2017; Anderson & Jiang, special filters), surveillance of the appearance of others, and obtaining
2018; Perrin, 2015; Pew, 2019). Adolescents, and young women and validation of one’s images in the form of ‘likes’ (Alhabash & Ma, 2017;
girls in particular, are more likely to seek appearance-related social Dumas et al., 2017; Sheldon & Bryant, 2016; Utz, Muscanell, & Khalid,
media gratifications compared to emerging adults and males (e.g., 2015). Against this background, scholars have suggested that each social
Choukas-Bradley, Nesi, Widman, & Galla, 2020; Perloff, 2014; Stefa­ media platform should be considered as a different type of context
none, Lackaff, & Rosen, 2011). Developmental status and gender were because its unique features elicit different types of norms, behaviors,
therefore studied as moderators in the relationships under investigation. and expectations (e.g., Phua et al., 2017; Waterloo, Baumgartner, Peter,
& Valkenburg, 2018). In line with this notion, research has shown that
1. Visually oriented social media use users select specific platforms because the features of these platforms
meet their needs (Alhabash & Ma, 2017; Dumas et al., 2017; Phua et al.,
Visually oriented social media have vastly grown in popularity 2017; Sheldon & Bryant, 2016).
(Newcom, 2021), probably because they have, like other social media, Finally, if users select platforms whose features meet their needs, the
unique features that meet the gratifications their users seek (Alhabash & use of multiple visually oriented social media platforms indicates a
Ma, 2017; Dumas, Maxwell-Smith, Davis, & Giulietti, 2017; Phua, Jin, greater need for the features of these platforms (Pittman & Reich, 2016).
Kim, & Jay, 2017; Sheldon & Bryant, 2016). Examples of such features Research has accordingly indicated positive associations between the
are plugins to edit images before posting a message (such a plugin is use of different visually oriented social media platforms (e.g., Pittman &
available, for instance, on Instagram but not on textually oriented social Reich, 2016). Moreover, the number of visually oriented social media
media such as Twitter); the central position of an image in a post (in an platforms selected may be related to well-being and body image con­
Instagram post, text appears below an image in a small font post while in cerns, as studies have found positive associations between the number of
a Facebook post, text is positioned above a potential image post in a social media platforms used and symptoms of depression and anxiety
larger font than the one on Instagram); and the necessity to post images (Primack et al., 2017; Vannucci, Ohannessian, & Gagnon, 2019).

Fig. 1. Hypothesized relationships between psychological well-being, body image concerns, social media gratifications, and visually oriented social media use,
moderated by gender and age.

2
J.M.F. van Oosten et al. Computers in Human Behavior 144 (2023) 107730

However, research on the selection of one or multiple visually oriented To support our tenet that it is mostly the selection of visually oriented
social media platforms is still scarce. In the present study, we therefore social media platforms that is predicted by self-validation and social
focused on the selection of one or multiple visually oriented social media comparison, we also tested hypotheses 1 and 2 for the selection on
platforms. textually oriented social media platforms (i.e., Facebook and Twitter).
Although Facebook and Twitter also allow for sharing appearance-
2. Gratifications sought on visually oriented social media related photos, it is not the main purpose of, and occurs less often on,
platforms: self-validation and social-comparison these platforms (especially among males; Thelwall, 2008; Utz et al.,
2015). We therefore expected self-validation and social comparison to
According to the Uses & Gratifications perspective (Blumler & Katz, be only weakly related to the selection of Facebook and Twitter.
1974), people select media for particular purposes (i.e., gratifications
sought) and desired outcomes (i.e., gratifications obtained; Palmgreen, 3. Gratifications sought and psychological well-being and body
Wenner, & Rayburn, 1980; Ruggiero, 2000; Sundar & Limperos, 2013). image concerns
Building on the Uses & Gratifications perspective, the Transactional
Model of Social Media and Body Image Concerns (Perloff, 2014) Based on the Transactional Model of Social Media and Body Image
responded to the need for a theoretical model that addresses the role of Concerns (Perloff, 2014), it can also be expected that self-validation and
gratifications sought in predicting appearance-focused social media in­ social comparison are underlying mechanisms in the relationships be­
teractions, including the use of visually oriented social media, such as tween psychological well-being or body image concerns and the selec­
Snapchat and Instagram. The model argues that the features of visually tion of visually oriented social media. More specifically, the extent to
oriented social media make these platforms particularly suitable for which people seek the gratifications of self-validation and social com­
seeking appearance related gratifications (Perloff, 2014). Users with a parison in social media may be associated with their psychological
high need for appearance related gratifications may thus be particularly well-being and body image concerns (Perloff, 2014). A rationale comes
likely to select one or more visually oriented social media platforms for from the self-affirmation and social comparison literature. The
use. self-affirmation literature posits that when self-worth is threatened and
The sought gratifications of appearance-related self-validation and depressive symptoms are felt, people have a need to restore their image
social comparison may play a particularly important role in the selection of themselves and thus a higher self-affirmation need emerges (Sherman
of visually oriented social media platforms (Perloff, 2014). The social & Cohen, 2006; Steele, 1988; Morling & Rusk, 2009). Moreover, the
media gratification of self-validation can be defined as seeking approval social comparison literature highlights that the need for social com­
of, and thus positive feedback on, one’s self-presented appearance on parisons is the highest among those with depressive symptoms and/or
social media (e.g., Toma & Hancock, 2013). According to low self-esteem, who seek both downward social comparisons as a
Self-Affirmation Theory (Sherman & Cohen, 2006; Steele, 1988), people strategy to enhance their self-worth and upward comparisons as a
have a fundamental need for self-worth and self-integrity, or for seeing “proof” of their lower self-worth (Wheeler, 2000).
themselves as valuable, and will be motivated to boost and protect their A reduced psychological well-being has been found in cross-sectional
sense of self-worth. Seeking validation, (positive) feedback, and atten­ (Cramer, Song, & Drent, 2016) and experimental (Johnson &
tion, mainly in the form of “likes”, is also one of the main motivations for Knobloch-Westerwick, 2014) research to be associated with engagement
young people to use social media (Jong & Drummond, 2016). This in social comparison on social media. For instance, experimental
validation motivation can take the form of uploading flattering, but also research has shown that individuals with mild depressive symptoms, or
altered or enhanced, photos of oneself (Dumas et al., 2017). related traits such as self-criticism or negative mood states, engage in
Social comparison, which is the tendency to base parts of our self- increased social comparison (Johnson & Knobloch-Westerwick, 2014;
perceptions on how we perceive ourselves in comparison to others Santor & Yazbek, 2006; Swallow & Kuiper, 1992). Similarly, research
(Festinger, 1954), is another important gratification sought in social has found significant positive correlations between low self-esteem and
media (e.g., Fox & Moreland, 2015; Haferkamp & Krämer, 2011a, the frequency of upward and downward social comparisons on social
2011b). It can take the form of looking at social media profiles of others media (Vogel, Rose, Roberts, & Eckles, 2014). It is likely that these
as a comparison standard for one’s own appearance. In fact, social findings extend to appearance related social comparisons on social
comparisons occur more frequently in social than in traditional media media, but research is lacking. Moreover, low self-esteem and increased
(Fardouly, Pinkus, & Vartanian, 2017). Moreover, social media invite depressive symptoms have been linked to self-validation behaviors in
social interaction with peers (Quan-Haase & Young, 2010), who are the context of social media (e.g., Lenton-Brym, Santiago, Fredborg, &
important targets of social comparisons for physical attributes (Carlson Antony, 2021; Shchebetenko, 2019). Therefore, we hypothesized:
Jones, 2001). Seeking self-validation in social media is predicted by lower levels of
The tenet of Perloff’s (2014) model that the two sought gratifications self-esteem (H3a) and higher levels of depressive symptoms (H3b).
of seeking self-validation or social comparison in social media predict Seeking social comparison in social media is predicted by lower
the selection of visually oriented social media has hardly been tested. levels of self-esteem (H4a) and higher levels of depressive symptoms
The few studies that did focus on motivations of using visually oriented (H4b).
social media (i.e., Instagram and Snapchat; Alhabash & Ma, 2017; Young people with more body dissatisfaction and greater appearance
Sheldon & Bryant, 2016) showed that such social media use was moti­ consciousness are also likely to seek the appearance related social media
vated by personal identity, self-promotion or -documentation, and sur­ gratifications of self-validation and social comparison (Perloff, 2014;
veillance of others, that is, motivations that are similar to the Rodgers, 2016). Body dissatisfaction occurs when an individual nega­
gratifications of self-validation and social comparison. However, these tively evaluates his/her body (Holland & Tiggemann, 2016) and has
studies were cross-sectional, which means the directional nature of these been shown to be both a predictor and an outcome of social comparison
relationships is still unknown. Moreover, the studies did not look at the in social media in longitudinal research (Rousseau, Eggermont, & Fri­
specific social media gratifications of self-validation and social com­ son, 2017). Cross-sectional studies have also reported correlations be­
parison as predictors of visually oriented social media use. To further tween body dissatisfaction and self-validation behavior on social media,
test the predictive value of social media gratifications in the selection of such as seeking likes and comments (e.g., Seekis, Bradley, & Duffy,
visually oriented social media platforms, we hypothesized: 2020).
Higher needs for self-validation (H1) and social comparison (H2) will In addition, research has recently started to focus on how young
predict the selection of a higher number of visually oriented social media people develop an ongoing (online and offline) awareness of whether
platforms. they look sufficiently attractive to an online audience, defined as

3
J.M.F. van Oosten et al. Computers in Human Behavior 144 (2023) 107730

appearance-related social media consciousness (Choukas-Bradley, Nesi, select visually oriented social media platforms because of needs for
Widman, & Higgins, 2018, 2020). This concept, which we call self-validation or social comparison.
‘appearance consciousness’ for linguistic ease, may be another relevant Second, we need to distinguish between adolescents and emerging
type of body image concern to study in relation to self-validation and adults when looking at the antecedents of self-validation and social
social comparison. Previous cross-sectional research has shown that comparison in social media and subsequent selection of visually oriented
traits that are similar to appearance consciousness, such as public social media platforms. Adolescence is a particularly sensitive period for
self-consciousness (Shim, Lee, & Park, 2008) and appearance related body image development and the role of social media herein (Chou­
contingencies of self-worth, predict visually oriented social media use (i. kas-Bradley et al., 2020), with a culture of appearance, and seeking
e., photo sharing in social media) (Stefanone et al., 2011). Moreover, appearance approval by peers, being characteristic of (early) adoles­
correlational research has linked this concept to appearance compari­ cence (Helfert & Warschburger, 2013; Jones & Crawford, 2006; Webb
sons and self-validation behaviors on social media (Burnell, George, et al., 2017). In addition, appearance-related self-consciousness on so­
Kurup, & Underwood, 2021; Choukas-Bradley et al., 2018), yet longi­ cial media seems to decrease when adolescents grow older (Stefanone
tudinal research is lacking. In sum, as a relationship between the grat­ et al., 2011). These findings suggest that adolescents, rather than
ifications sought of self-validation or social comparison and both body emerging adults, are most likely to show a need for seeking
dissatisfaction and appearance consciousness seems likely, we self-validation and social comparison and subsequently select more
hypothesized: visually oriented social media platforms for use.
Seeking self-validation in social media is predicted by higher levels Against this backdrop, we investigated whether the hypothesized
of body dissatisfaction (H5a) and appearance consciousness (H5b). indirect relationships between psychological well-being or body image
Seeking social comparison in social media is predicted by higher concerns and the selection of visually oriented social media, via self-
levels of body dissatisfaction (H6a) and appearance consciousness validation and social comparison, depended on gender and age (cf.
(H6b). Holbert & Park, 2020). Specifically, we hypothesized:
Fig. 1 visualizes hypotheses 1–6. Moreover, as a final test of the te­ The prediction of the selection of visually oriented social media
nets of the Transactional Model of Social Media and Body Image Con­ platforms by self-validation (H11a) or social comparison (H11b) will be
cerns (Perloff, 2014), we expected that the selection of visually oriented stronger among adolescent girls compared to adolescent boys, young
social media platforms would be indirectly predicted by psychological adult men, and young adult women.
well-being and body image concerns, through self-validation and social The indirect relationship between psychological well-being and the
comparison (see Fig. 1). We thus hypothesized that: selection of visually oriented social media platforms, via self-validation
There will be an indirect prediction of the selection of visually ori­ (H12a&b) and social comparison (H13a&b), will be stronger among
ented social media platforms by lower levels of self-esteem (H7a/H8a), adolescent girls compared to adolescent boys, young adult men, and
and higher levels of depressive symptoms (H7b/H8b) through seeking young adult women.
self-validation in social media (H7) and seeking social comparison (H8) The indirect relationship between body image concerns and the se­
in social media. lection of visually oriented social media platforms, via self-validation
There will be an indirect prediction of the selection of visually ori­ (H14a&b) and social comparison (H15&ab), will be stronger among
ented social media platforms by higher levels of body dissatisfaction adolescent girls compared to adolescent boys, young adult men, and
(H9a/H10a) and appearance consciousness (H9b/H10b) through young adult women.
seeking self-validation (H9) and seeking social comparison (H10) in To address spurious relationships, we controlled for sexual orienta­
social media. tion and ethnicity, as previous research has shown that these factors may
play a role in body image concerns, well-being, and media use (e.g.,
4. Moderating roles of gender and age Frederick & Essayli, 2016; Greenwood & Dal Cin, 2012; Markey &
Markey, 2014; Mcardle & Hill, 2009; Ricciardelli, McCabe, Williams, &
Visually oriented social media appeal to the entire population of Thompson, 2007; Sussman, Truong, & Lim, 2007; Tiggemann, 2015).
youth between the ages of 12 and 25 (Anderson & Jiang, 2018; Pew,
2019). Still, there may be differences within this population in terms of 5. Methods
the aforementioned hypotheses. First, differences can be expected based
on gender. Research on gender socialization has shown that girls, 5.1. Participants
compared to boys, are socialized to be more concerned with their
appearance (Zurbriggen et al., 2010). Against this backdrop, the A two-wave longitudinal panel survey was conducted among Dutch
Transactional Model of Social Media and Body Image Concerns (Perloff, adolescents (13–17 years old) and emerging adults (18–25 years old).
2014) and related empirical research (Brown & Tiggemann, 2016; The same questionnaire was administered in April/May 2015 (wave 1)
Choukas-Bradley et al., 2018; Gioia, Griffiths, & Boursier, 2020; Meier & and in June/July 2015 (wave 2), which meant there were approximately
Gray, 2014) consider female social media users more likely to select two months between waves. Ethical approval was granted by the Ethical
visually oriented social media platforms out of appearance related mo­ Review Board of the Communication Science Department of the Uni­
tivations and body image concerns. Similarly, girls have more appear­ versity of Amsterdam (project number 2015-CW-21). Respondents were
ance consciousness than boys (Choukas-Bradley et al., 2020; Stefanone randomly selected by Dutch research agency Veldkamp (now Kantar
et al., 2011) and use more visually oriented social media (Thelwall, Public) from their pool of respondents, which was originally sampled
2008). randomly among the Dutch population and continuously updated,
That said, the striving for an ideal appearance and related body reducing problems of self-selection biases. Participants received a
image concerns has been found among both genders (Tobin et al., 2010; voucher with a worth of approximately €5 (as was the common practice
Vandenbosch & Eggermont, 2013). Boys and men may thus also seek for members of the respondent pool) after completing each wave of the
appearance-related social media gratifications in social media (Van­ study. When completing both waves of the study, the participants ob­
denbosch & Eggermont, 2016) and use body image management in so­ tained an additional voucher of €5. Informed consent was asked from the
cial media to improve their sexual attractiveness (Gioia et al., 2020). parents of the adolescents before the adolescents were contacted, as well
Yet, these appearance-focused social media behaviors occur overall to a as from the adolescents and emerging adults themselves at the start of
lower extent among men and boys than among women and girls (Gioia the survey in each wave. Invitations for participation, as well as the
et al., 2020; Vandenbosch & Eggermont, 2016). Together, these findings survey itself, were administered online and in Dutch. The response rate
thus suggest that girls and young women may be particularly likely to was 68% for adolescents and 47% for emerging adults at wave 1 (i.e., the

4
J.M.F. van Oosten et al. Computers in Human Behavior 144 (2023) 107730

percentage of participants that filled out the survey in response to the to items used in the validated scale. Participants were presented with the
invitation to participate). Of the respondents that participated in wave following statements, and had to indicate on a scale from 1 (does not
1, 81.6% of the adolescents and 80.9% of the emerging adults also apply at all) to 7 (totally applies) whether the statements applied to them:
participated at wave 2. “When I post a photo of myself on a social media profile, I …”, a) “…
Only respondents that had participated in both waves and that used think it is important that I look beautiful in the photo”, b) “… think it’s
social media were included in the analytical sample of the current study, important to get enough likes”, and c) “… think it’s important to get
resulting in a final sample of 953 adolescents and 899 young adults positive comments about how I look.” The items loaded on one factor
(total N = 1852), of which 52.6% was female. The majority of our with an eigenvalue of 2.24 (2.28 in Wave 2), an explained variance of
sample had a heterosexual orientation (91%), and Dutch nationality 75% (76% in Wave 2) and a Cronbach’s alpha of .83 (0.84 in Wave 2)
(97% had parents who were born in the Netherlands). Data from this (M = 3.90, SD = 1.47 in Wave 1; M = 3.79, SD = 1.47 in Wave 2).
sample have been used in previous studies of the authors, and further
information about the sample can thus be found in (van Oosten & 6.3.2. Body dissatisfaction
Vandenbosch, 2017, 2020; Vandenbosch & van Oosten, 2017). Body dissatisfaction was measured with the ‘general dissatisfaction’
subscale of the Body Attitude Scale (Probst, Vandereycken, Coppenolle,
6. Measures & Vanderlinden, 1995), consisting of 4 items: “When I compare my body
with that of peers, I am dissatisfied with my body”, “I tend to hide my
6.1. Selection of social media platforms body (for example by wearing loose clothes)", “When I look at my body
in the mirror, I am dissatisfied”, and “I envy others for their physical
We created a measure of the number of visually oriented social media appearance”. Participants could indicate their agreement with the
platforms that an individual used. Respondents were asked if they used statements on a scale from 1 (totally disagree) to 7 (totally agree). The
certain types of social media platforms (i.e., Facebook, Twitter, Insta­ scale had an Eigen Value of 2.86 (2.89 in Wave 2), an explained variance
gram, Snapchat), coded as 1 (“yes”) or 0 (“no”). A sum score of the of 71% (72% in Wave 2), and a Cronbach’s alpha of .87 in both waves
platforms that were mainly photo-sharing- and appearance-oriented (i. (M = 3.27, SD = 1.43 in Wave 1; M = 3.28, SD = 1.41 in Wave 2).
e., Instagram, Snapchat) was created to measure the ‘selection of visu­
ally oriented social media platforms.’ A separate sum score was created 6.3.3. Psychological well-being
for other social media platforms (i.e., Facebook, Twitter) to measure the
‘selection of other social media platforms.’ A higher score meant that 6.3.3.1. Self-esteem. Self-esteem was measured with the Single-Item
respondents used both these platforms and a lower score indicated that Self-Esteem Scale (SISE) of Robins, Hendin, and Trzesniewski (2001),
they used only one or none of these platforms. Table 1 shows the per­ where participants had to indicate on a scale from 1 (does not apply at all)
centages for usage, for the total sample and distinguishing between the to 7 (totally applies) whether the following statement applied to them: “I
groups based on gender and age. have a high self-esteem” (M = 4.88, SD = 1.29 in Wave 1; M = 4.91, SD
= 1.27 in Wave 2).
6.2. Social media gratifications of self-validation and social comparison
6.3.3.2. Depressive symptoms. Depressive symptoms were measured
Participants were presented with questions about why they use social with the Mental Health Inventory (MHI-5, (Rumpf, Meyer, Hapke, &
media. This measure was created for the purpose of this study and John, 2001). Participants were asked: “How often in the previous two
inspired by literature on social media gratifications (sought) as well as months have you …” a) “… felt very tense?”; b) “… felt calm and quiet?”
self-presentation and social comparison in social media (e.g., Haferkamp (recoded); c) “… felt discouraged and sad?”; d) “… felt so down that
& Krämer, 2011a, 2011b; Krämer & Winter 2008; Palmgreen et al., nothing could cheer you up?”; e) “… felt like a happy person?” (reco­
1980; Perloff, 2014). As we were interested in two specific gratifications ded). Response options ranged from 1 (never) to 7 (always). The scale
(i.e., self-validation and social comparison) and as previous research has had an Eigen Value of 2.92 (3.00 in Wave 2), an explained variance of
shown that specific gratifications can be validly assessed by a single item 58% (60% in Wave 2) and a Cronbach’s alpha of .82 (0.83 in Wave 2).
(Quan-Haase & Young, 2010), we used single-item measures for both Higher scores indicated more depressive symptoms (M = 2.97, SD =
gratifications. 0.92 in Wave 1; M = 2.96, SD = 0.92 in Wave 2).
First, we wanted to know why participants engage with their own
profile on social media (i.e., posting photos and messages). Participants 6.3.4. Control variables
indicated on a scale from 1 (does not apply at all) to 7 (totally applies) In our main analyses, we controlled for gender (coded “0” for female
whether the following statement applied to them: “I engage with my and “1” for male) and age (in years), given their potential linkage with
profile in social media to get confirmation from others about my our main variables under study as explained in the theoretical frame­
appearance.” This was our measure of self-validation (M = 2.54, SD = work. Heterosexual orientation was coded as “1”, and non-heterosexual
1.45 in Wave 1; M = 2.62, SD = 1.46 in Wave 2). Next, we asked par­ was orientation coded as “0”. Dutch ethnicity was coded as “1”, and non-
ticipants why they looked at other people’s profiles in social media and Dutch ethnicity (i.e., one or both of the participant’s parents were born
to indicate on a scale from 1 (does not apply at all) to 7 (totally applies) outside of the Netherlands) was coded as “0”. To control for amount of
whether the following statement applied to them: “I look at profiles of general social media use, participants were asked how much time they
others to compare how I look with what others look like on photos on had spent on social media on an average weekday and weekend day in
social media.” This was our measure of social comparison (M = 2.93, SD the past 2 months, in hours and minutes. The number of hours was
= 1.64 in Wave 1; M = 2.99, SD = 1.64 in Wave 2). multiplied by 60 and added to the number of minutes for weekdays and
weekend days separately to obtain a total score of minutes per weekday
6.3. Body image concerns or weekend day. The total amount of minutes in a weekday was then
multiplied by 5 and the total amount of minutes in a weekend-day was
6.3.1. Appearance consciousness multiplied by 2. The sum of these amounts was divided by 7 to obtain an
The measure of appearance consciousness was created for the pur­ average amount of daily social media use (in minutes), resulting in an
pose of this study and based on previous research (Krämer & Winter average number of 168 min per day (SD = 170.62 in wave 1 and 169 min
2008; Perloff, 2014). The data-collection occurred before the validation per day (SD = 165.77) in wave 2. These use numbers are comparable to
of an appearance-related social media consciousness scale (Choukas-­ recent numbers of social media use by youth in the Netherlands
Bradley et al., 2020), but the items used in the current study are similar

5
J.M.F. van Oosten et al. Computers in Human Behavior 144 (2023) 107730

(Newcom, 2021; Verbeij, Pouwels, Beyens, & Valkenburg, 2021) and construct in wave 1 were regressed on the values of the construct in
other European countries (Marengo et al., 2018; Reer, Festl, & Quandt, wave 2. Covariances between error terms of the same manifest items of
2021). latent constructs between waves were included. Following Cheung and
Rensvold (2002), we tested for measurement equivalence by observing
7. Data analysis whether the difference in the CFI values between an unconstrained
model (i.e., factor loadings of the items of the latent construct vary freely
We conducted a confirmatory factor analysis in AMOS 25 for the between waves) and a constrained model (i.e., factor loadings are con­
measurement model of the latent variables (i.e., appearance con­ strained to be equal between waves) is lower than 0.01. Such a test
sciousness, depressive symptoms, and body dissatisfaction) in wave 1 determines whether the measurement of a latent constructs remains
and wave 2. Good model fit was determined by looking at the CFI and stable over time. None of the latent variables showed measurement
TLI (>0.90), the SRMR and RMSEA (<0.08) (Hu & Bentler, 1999; Kline, variance over time.
1998); convergent validity and the composite reliability were ensured.
The discriminant validity (i.e., square root of the AVE) had to be above 8. Results
0.70, and above the correlations between each latent variable and the
other latent or manifest variables, to ensure a good discriminant validity 8.1. Confirmatory factor analysis
(Cheung & Wang, 2017; Zaiţ & Bertea, 2011).
The hypothesized model was tested using structural equation The confirmatory factor analysis for the measurement model of the
modelling in AMOS 25. Latent constructs at wave 1 and 2 (i.e., latent variables (i.e., appearance consciousness, depressive symptoms,
appearance consciousness, body dissatisfaction and depressive symp­ and body dissatisfaction) in wave 1 and wave 2 showed that standard­
toms) were loaded on the manifest items used to measure each ized factor loadings were 0.60 or higher, except for one item of the
construct. Self-esteem, visually oriented social media platforms, other depressive symptoms scale at wave 1 (i.e., “… felt calm and quiet?”
social media platforms, self-validation and social comparison were [recoded]), which had a factor loading of 0.56. The model fit of the
entered as manifest variables at wave 1 and 2. Disturbance terms of all measurement models was deemed adequate as CFI = 0.95, TLI = 0.94,
endogenous variables at wave 2 were also allowed to correlate. Auto- RMSEA = 0.07 and SRMR = 0.04. The scales further showed good
regressions were added from wave 1 to wave 2 measures of all main convergent validity with an Average Variance Extracted (AVE) above
variables. Error terms of the same manifest items measuring the latent 0.50, except for the depressive symptoms scale in wave 1 (AVE = 0.495).
constructs were allowed to covary over time. All exogenous variables at A composite reliability of 0.80 and higher was found for all scales. The
wave 1 were correlated with each other. Age, gender, heterosexual discriminant value (square root of the AVE) was 0.70 and higher for all
orientation, ethnicity, and general social media use were entered as latent scales, which was also higher than the correlations between the
manifest control variables at wave 1, and modelled to correlate with separate latent variables, and between the latent and manifest variables,
exogenous variables at wave 1 and predict the endogenous variables at and thus showed good discriminant validity. Zero-order correlations
wave 2. In our moderation analyses, only heterosexual orientation, between the main variables, and means and standard deviations of the
ethnicity, and general social media use were entered as control variables, are shown in Table 2.
variables.
We tested all direct and indirect hypothesized relationships using 8.2. Direct relationships between social media gratifications and the
one model in which self-validation and social comparison at wave 1 and selection of visually oriented social media platforms
wave 2 were modelled to predict the selection of visually oriented social
media platforms at wave 2, as well as the textual social media platforms The model showed a good fit (χ 2 [556, N = 1852] = 2479.50, p <
(i.e., Facebook and Twitter) at wave 2 (to test our tenet that the hy­ .001, CFI = 0.95, TLI = 0.94, SRMR = 0.04, RMSEA = 0.043, χ 2/df, =
pothesized relationships would to a lesser extent hold for these plat­ 4.44). In contrast to what was stated in H1, self-validation at wave 1 did
forms). Psychological well-being and body image concerns at wave 1 not significantly predict visually oriented social media use at wave 2, β
predicted self-validation and social comparison at wave 2, which in turn = 0.03, B = 0.02, SE = 0.01, p = .11, 95% BCI: − 0.002/.04 (although
predicted the selection of visually oriented social media platforms at this relationship was significant within the same wave, see the indirect
wave 2. relationships below). In contrast to what H2 posited, social comparison
To investigate the moderating effect of gender and age, we created at wave 1 negatively predicted use of visually oriented social media at
groups based on gender (i.e., male and female) and age (i.e., 13–17 years wave 2, β = − 0.06, B = − 0.03, SE = 0.01, p < .001, 95% BCI: − 0.05/
old = adolescents; 18–25 years old = emerging adults), resulting in 4 − 0.02 (this prediction was not significant within the same wave, see the
groups (i.e., adolescent boys, N = 483, adolescent girls, N = 470, indirect relationships below) .1
emerging adult men, N = 395, and emerging adult women, N = 504). To
see whether the hypothesized relationships were significantly different
8.3. Indirect predictions by psychological well-being and body image
for our different age and gender groups, we conducted multiple group
concerns
comparisons (Rigdon, Schumacker, & Wothke, 1998). More specifically,
we created another model in which each hypothesized path was con­
As can be seen in Table 3, seeking both self-validation and social
strained to be equal across the four groups and compared this con­
comparison in social media at wave 2 were predicted by appearance
strained model to a model where the paths were allowed to vary for each
consciousness at wave 1. We thus found support for Hypotheses 5 b and
group. A significant change in model fit between the two models in­
6 b. Moreover, self-esteem and body dissatisfaction at wave 1 both
dicates a significant moderation of that hypothesized path by the
positively predicted seeking social comparison at wave 2, which was in
grouping variable.
Because our measures were not all normally distributed (e.g., our
social media gratifications measures were skewed to the left), we added 1
When the cross-sectional predictions of visually oriented social media use
a non-parametric test (i.e., bootstrapping) to our analyses. We calculated
by the social media gratifications within wave 2 are removed from the model,
the 95% bootstrap confidence interval of the unstandardized coefficients the social media gratification of self-validation at wave 1 positively predicted
of the predictors, based on 1000 bootstrap iterations (N = 1852). visually oriented social media use at wave 2, β = 0.05, B = 0.03, SE = 0.01, p <
Measurement invariance was tested by creating a structural equation .01, 95% BCI: 0.01/.05, and the social media gratification of social comparison
model for each latent construct (i.e., depressive symptoms, body at wave 1 negatively predicted visually oriented social media use at wave 2, β
dissatisfaction, appearance consciousness) in which values of the = − 0.05, B = − 0.03, SE = 0.01, p < .001, 95% BCI: − 0.04/− 0.01.

6
J.M.F. van Oosten et al. Computers in Human Behavior 144 (2023) 107730

Table 1 wave 2, β = 0.003, B = 0.002, SE = 0.001, p = .037, 95% BCI: 0.0001/

Descriptives of the main variables for gender, separately for adolescents and .007, supporting H7a. All other hypothesized indirect relationships
emerging adults, and for the total sample. (H7b, H8a&b, H9a, and H10a&b) were not supported. Fig. 2 shows an
Boys Girls Men Women Total overview of the direct and indirect relationships, with standardized
M(SD)/ M(SD)/ M(SD)/ M(SD)/ M(SD)/
regression coefficients (for clarity, only the significant predictions of the
% % % % % social media gratifications by wellbeing or body concern variables are
shown).
Self-validation 2.40 2.86 2.37 2.52 2.54
gratification (w1) (1.48) (1.52) (1.35) (1.40) (1.45)
Self-validation 2.60 2.90 2.33 2.60 2.62 8.4. Moderation effect of the gender and age groups
gratification (w2) (1.53) (1.55) (1.31) (1.37) (1.46)
Social comparison 2.73 3.23 2.74 3.00 2.93
The model fit for the unconstrained model testing H11a&b, dis­
gratification (w1) (1.65) (1.74) (1.54) (1.58) (1.64)
Social comparison 2.84 3.30 2.74 3.08 2.99
tinguishing between the four gender and age groups, was good, χ2
gratification (w2) (1.63) (1.74) (1.52) (1.61) (1.64) (2048, N = 1852) = 3890.15, p < .001, CFI = 0.94, TLI = 0.93, SRMR =
Appearance 3.46 4.39 3.50 4.18 3.90 0.04, RMSEA = 0.02, χ2/df = 1.90. As can be seen in Table 4, some of the
consciousness (w1) (1.54) (1.39) (1.43) (1.29) (1.47) found relationships, such as the negative prediction of visually oriented
Appearance 3.45 4.22 3.33 4.05 3.78
social media use (wave 2) by social comparison (wave 1), appeared to be
consciousness (w2) (1.54) (1.41) (1.41) (1.31) (1.47)
Body dissatisfaction 2.78 3.41 3.07 3.78 3.27 stronger for adolescent girls (β = − 0.14, p < .001, 95% BCI: 0.10/-0.03).
(w1) (1.28) (1.46) (1.33) (1.44) (1.43) However, model comparison did not suggest any significant differences
Body dissatisfaction 2.85 3.47 3.02 3.70 3.28 between the four groups based on gender and age in the relationship
(w2) (1.25) (1.43) (1.31) (1.45) (1.41) between self-validation (at wave 1) and visually oriented social media
Self-esteem (w1) 5.21 4.61 5.09 4.68 4.89
(1.15) (1.45) (1.18) (1.26) (1.29)
use (at wave 2), CMIN = 0.93, p = .82, nor in the relationship between
Self-esteem (w2) 5.22 4.60 5.13 4.71 4.91 social comparison (at wave 1) and visually oriented social media use (at
(1.16) (1.42) (1.15) (1.23) (1.27) wave 2), CMIN = 4.74, p = .19. Hypotheses 11a and 11 b, stating that
Depressive symptoms 2.75 3.00 2.94 3.15 2.97 these relationships would be stronger among adolescent girls, were thus
(w1) (.84) (.98) (.89) (.92) (.92)
not supported.
Depressive symptoms 2.74 3.06 2.88 3.14 2.96
(w2) (.87) (.95) (.86) (.93) (.92) In response to Hypotheses 12–15, the multiple group comparison
Social media usage based on constraining the relationships to be equal between the four
Visually oriented social media platforms gender and age groups did not show a significant change in model fit for
% using 0 platforms 42/ 18/ 65/ 57/54% 45/ the paths predicting self-validation or social comparison at wave 2 by
(w1/w2) 42% 18% 66% 44%
% using 1 platform 25/ 20/ 21/ 29/30% 24/
psychological well-being or body image concerns at wave 1, CMIN
(w1/w2) 22% 21% 20% 23% <5.24, p > .16. As there was thus no evidence of gender differences in
% using 2 platforms 34/ 62/ 14/ 14/16% 31/ the indirect relationships, hypotheses 12a&b through 15a&b were
(w1/2) 37% 61% 13% 32% rejected.
Other social media
platforms
% using 0 platforms 19/ 13/ 6/7% 3/3% 10/ 8.5. Additional analyses: other social media platforms and testing
(w1/w2) 19% 13% 11% reciprocal relationships
% using 1 platform 52/ 59/ 61/ 73/70% 61/
(w1/w2) 50% 59% 62% 60%
As for our additional test for the selection of other (more textual)
% using 2 platforms 29/ 28/ 33/ 25/26% 29/
(w1/2) 32% 28% 32% 29%
social media platforms, no significant prediction of the selection of
Social media usage per platform (% of respondents indicating using the platform) textual social media platforms (at wave 2) by self-validation (β = − 0.03,
Instagram (w1/w2) 50/ 75/ 26/ 36/39% 47/ B = − 0.01, SE = 0.01, p = .26) or social comparison (β = − 0.01, B =
51% 75% 26% 48% 0.00, SE = 0.01 p = .59) at wave 1 was found. There was, however, a
Snapchat (w1/w2) 42/ 69/ 24/ 21/23% 39/
significant prediction of the selection of the other social media platforms
44% 69% 21% 40%
Facebook (w1/w2) 77/ 86/ 93/ 96/95% 88/ at wave 2 by social comparison at wave 2 (β = 0.04, B = 0.02, SE = 0.01,
79% 85% 93% 88% p = .04, 95% BCI: 0.001/0.03), suggesting an indirect relationship via
Twitter (w1/w2) 32/ 29/ 34/ 26/27% 30/ this particular social media gratification. We indeed found an indirect
34% 30% 32% 31% prediction of the selection of other platforms at wave 2, via social
Note. w1 = wave 1; w2 = wave 2. comparison at wave 2, by body dissatisfaction (β = 0.004, B = 0.002, SE
= 0.001, p = .021, 95% BCI: 0.0002/.004), appearance consciousness (β
line with H6a but in contrast to hypothesis H4a, which stated that lower = 0.024, B = 0.013, SE = 0.005, p = .005, 95% BCI: 0.005/.022) and
self-esteem would predict social comparison. Similarly, greater self- self-esteem (β = 0.004, B = 0.002, SE = 0.001, p = .013, 95% BCI:
esteem at wave 1 was associated with a higher seeking of self- 0.0004/.005) at wave 1.
validation in wave 2. This contrasted with H3a, which stated that Furthermore, research on Uses & Gratifications (i.e., Palmgreen
lower self-esteem would predict seeking self-validation in social media. et al., 1980; Ruggiero, 2000; Sundar & Limperos, 2013) suggests that the
In terms of indirect relationships, visually oriented social media use effects of gratifications sought on users’ wellbeing occur through the use
at wave 2 was positively and significantly predicted by self-validation of media and the gratifications obtained as a result of such use. The
gratification at wave 2, β = 0.05, B = 0.03, SE = 0.01, p < .01, 95% present study did not include a measure of gratifications obtained. As a
BCI: 0.01/.05, but not by social comparison gratification at wave 2, β = result, we could not test whether obtaining the gratifications of
0.01, B = 0.01 SE = 0.01, p = .54, 95% BCI: − 0.01/03. Appearance self-validation or social comparison in social media would predict body
consciousness (at wave 1) thus indirectly predicted visually oriented image concerns and psychological wellbeing. However, as a first
social media use (at wave 2), via self-validation (but not social com­ exploration of transactional relationships (as also put forth by Perloff’s
parison) at wave 2, β = 0.019, B = 0.014, SE = 0.005, p = .002, 95% BCI: model, 2014), we also tested a model in which the social media grati­
0.007/.032, supporting H9b. A positive indirect relationship was also fications, as well as psychological well-being and body image concerns,
found between self-esteem (at wave 1) and visually oriented social at wave 2 were predicted by the selection of visually oriented social
media use (at wave 2), via self-validation (but not social comparison) at media platforms at wave 1. The model showed good fit (χ 2 [545, N =
1852] = 2469.08, p < .001, CFI = 0.95, TLI = 0.93, SRMR = 0.04,

7
 J.M.F. van Oosten et al.
Table 2
Zero-order correlations.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

1. Heterosexual orientation
2. Dutch ethnicity − .00
3. Gender (male) .05 − .03
4. Age (w1) − .10*** − .05 − .09***
5. General social media use − .00 − .04 − .10** − .16***
(w1)
6. Validation gratification − .02 − .04 − .10*** − .09*** .19***
(w1)
7. Validation gratification − .03 − .04 − .09*** − .10*** .17*** .58***
(w2)
8. Social comparison − .03 − .02 − .11*** − .04 .18*** .63*** .49***
gratification (w1)
9. Social comparison − .03 .00 − .12*** − .05* .15*** .48*** .64*** .58***
8

gratification (w2)
10. Appearance consciousness − .04 − .04 − .27*** − .03 .13*** .53*** .45*** .53*** .45***
(w1)
11. Appearance consciousness − .03 − .04 − .25*** − .05 .14*** .47*** .54*** .46*** .55*** .67***
(w2)
12. Body dissatisfaction (w1) − .11*** − .02 − .24*** .16*** .11*** .16*** .16*** .27*** .23*** .24*** .23***
13. Body dissatisfaction (w2) − .10*** .03 − .23*** .11*** .12*** .13*** .16*** .23*** .28*** .19*** .24*** .76***
14. Depressive symptoms (w1) − .15*** − .07** − .13*** .10*** .14*** .19*** .14*** .24*** .19*** .18*** .17*** .44*** .41***
15. Depressive symptoms (w2) − .10*** − .04 − .16*** .07** .11*** .19*** .18*** .22*** .20*** .18*** .18*** .38*** .40*** .71***
16. Self-esteem (w1) .10*** − .06* .20*** − .04 − .05 − .04 − .02 − .14*** − .08* − .11*** − .09*** − .44*** − .43*** − .48*** − .43***
17. Self-esteem (w2) .11*** − .05* .21*** − .02 − .04 − .05 − .03 − .14*** − .09*** − .10*** − .08* − .38*** − .42*** − .42*** − .46*** .75***
18. Visual social media use .05* .01 − .15*** − .39*** .25*** .22*** .20*** .21*** .19*** .26*** .24*** .05* .05* .04 .06* − .01 − .01
(w1)
19. Visual social media use .03 .01 − .16*** − .39*** .25*** .21*** .21*** .16*** .18*** .26*** .24*** .07* .07* .05* .07* − .02 − .02 .87***
(w2)

Computers in Human Behavior 144 (2023) 107730

Note. All scales are measured on a 7-point scale, social media use is measured in minutes per day; *p < .05,; **p < .01; **p < .001; 1 = wave 1; w2 = wave 2.
J.M.F. van Oosten et al. Computers in Human Behavior 144 (2023) 107730

Table 3 RMSEA = 0.043, χ 2/df, = 4.53). None of these relationships were sta­
Standardized regression weights and unstandardized confidence intervals for tistically significant.
the hypothesized relationships predicting social media gratifications, for the full
sample. 9. Discussion
Predictor (wave 1) Self-Validation (Wave 2) - Social Comparison (Wave
R2 = 37% 2) - R2 = 37% Given the increasing popularity of visually oriented social media
В p 95% BCI β p 95% BCI platforms among young people, the present study aimed to investigate
Psychological Well-Being
what drives the selection of these types of social media platforms for use.
Self-esteem .05 .04 .002/.10 .07 <.01 .03/.14 We focused on appearance related social media gratifications of self-
Depressive symptoms .01 .81 − .09/ .03 .24 − .04/ validation and social comparison as well as on psychological well-
.09 .16 being and body image concerns as their antecedents. In contrast with
Body Image Concerns
the Uses & Gratifications perspective (Blumler & Katz, 1974; Sundar &
Body dissatisfaction .03 .25 − .02/ .07 <.01 .03/.13
.08 Limperos, 2013), and the Transactional Model of Social Media and Body
Appearance .34 <.001 .35/.50 .32 <.001 .36/.54 Image Concerns (Perloff, 2014), we found that the selection of visually
consciousness oriented social media platforms was not significantly predicted by
Note. w1 = wave 1; w2 = wave 2. self-validation, and was negatively predicted by social comparison, over
time. Self-validation and social comparison at wave 2 were driven by
higher appearance consciousness and self-esteem at wave 1. Appearance
consciousness and self-esteem at wave 1 predicted the use of visually
oriented social media platforms at wave 2 indirectly through the

Fig. 2. Standardized Regression Coefficients of the Relationships Between Psychological Well-Being, Body Image Concerns, Social Media Gratifications, and Visually
Oriented Social Media Use Note: *p < .05, **p < .01, ***p < .001; w1 = wave 1; w2 = wave 2.

Table 4
Standardized regression weights and unstandardized confidence intervals for the relationships in the multiple group analysis based on age and gender.
Predictor: Adolescent boys Adolescent girls Young adult men Young adult women

β p 95% BCI β p 95% BCI β p 95% BCI β p 95% BCI

DV: Visually Oriented Social Media Use (wave 2)

Self-Validation (wave 1) .01 .79 − .04/.05 .06 .10 .00/.07 .03 .43 − .02/.05 .02 .42 − .02/05
Social Comparison (wave 1) − .02 .57 − .04/.03 − .14 <.001 − .10/-.03 − .08 .04 − .07/-.005 − .06 .04 − .06/-.001
DV: Self-Validation Gratification (wave 2)
Psychological Well-Being (wave 1)
Self-esteem .00 .94 − .11/.11 .08 .07 .00/.18 .04 .41 − .09/.14 .07 .13 − .03/.18
Depressive symptoms .00 .96 − .19/.23 .01 .90 − .19/.19 .00 .98 − .20/.22 .04 .39 − .12/.26
Body Image Concerns (wave 1)
Body dissatisfaction .02 .64 − .08/.13 .01 .87 − .09/.11 .00 .95 − .11/.10 .06 .19 − .04/.14
Appearance consciousness .37 <.001 .29/.63 .29 <.001 .27/.61 .39 <.001 .26/.62 .30 <.001 .29/.66
DV: Self-Comparison Gratification (wave 2)
Psychological Well-Being (wave 1)
Self-esteem .01 .70 − .10/.14 .11 .02 .01/.25 .09 .06 − .01/.24 .07 .12 − .02/.20
Depressive symptoms .00 .92 − .22/.28 .08 .20 − .08/.36 .03 .62 − .18/.32 .04 .35 − .12/.25
Body Image Concerns (wave 1)
Body dissatisfaction .07 .12 − .04/.20 .03 .59 − .09/.16 .06 .20 − .04/.19 .10 .02 .02/.19
Appearance consciousness .40 <.001 .35/.72 .25 <.001 .22/.60 .30 <.001 .22/.57 .29 <.001 .33/.78

9
J.M.F. van Oosten et al. Computers in Human Behavior 144 (2023) 107730

gratification of self-validation at wave 2. 2018). Over time, this may lead to an avoidance of visually oriented
social media platforms. Such avoidance is known in social media literacy
9.1. Gratifications sought and the selection of visually oriented social literature as a coping strategy for dealing with social media pressures
media platforms: the role of psychological well-being and body image (Schreurs & Vandenbosch, 2021). Our interpretation also dovetails with
concerns previous findings that individuals with a higher tendency for
self-objectification watched less sexually objectifying television over
The present findings point to the importance of appearance con­ time, most likely as a form of selective avoidance and self-protection
sciousness for seeking the appearance-related social media gratifications (Aubrey, 2006). Moreover, our additional analyses showed that this
of self-validation and social comparison. This construct has only recently may be different on less visually oriented social media platforms, such as
been included in research on social media use (Choukas-Bradley et al., Facebook and Twitter, where body dissatisfaction, appearance con­
2018, 2020), but was a particularly strong predictor of young people’s sciousness, and higher self-esteem predicted, via increased social com­
self-validation and social comparison, as well as an indirect predictor of parison, a greater selection of such platforms.
visually oriented social media use via self-validation. Furthermore, To better understand differential predictions by social media grati­
while research has often studied body dissatisfaction or body image fications, future research should distinguish between different (visually
concerns as an outcome of social comparison (e.g., Fardouly, Diedrichs, oriented) social media platforms, in terms of the number of platforms
Vartanian, & Halliwell, 2015; Fardouly et al., 2017; Holland & Tigge­ chosen, their content and unique features, as well as the types of (active
mann, 2016), the present study showed that higher body dissatisfaction and passive) use of such platforms. In a cross-sectional study by Sheldon
predicted seeking social comparison gratifications in social media. This and Bryant (2016), for instance, the motivation to look at others only
is in line with previous research that suggested that youth with low body predicted the number of hours spent on Instagram. The motivation to
satisfaction use social media to look for inspiration and set standards for become popular and promote oneself, however, also predicted the
their own body and appearance (Rousseau et al., 2017). amount of editing or using hashtags (Sheldon & Bryant, 2016).
The current study further showed that youth with higher self-esteem Finally, although we added depressive feelings as a predictor
have a higher need for appearance related self-validation in social following Perloff’s (2014) model, our findings suggest that this psy­
media, which in turn predicted the selection of more visually oriented chological wellbeing variable predicts neither seeking appearance
social media platforms. This was in contrast with our expectations as we related self-validation or social comparison nor the selection visually
predicted that individuals with lower self-esteem would have a stronger oriented social media platforms. An explanation could be that our social
need for self-validation in social media. Potentially, visually oriented media gratifications measures both focused on one’s appearance.
social media platforms may especially attract high self-esteem in­ Depressive feelings may more likely to predict other social media grat­
dividuals to protect their already high self-worth, as they may expect to ifications that have to do with overall quality of life or social connect­
receive positive feedback in such a platform (e.g., through likes or edness (e.g., Elhai et al., 2018; Grieve, Indian, Witteveen, Anne Tolan, &
positive comments) given its unique features. When individuals do not Marrington, 2013).
receive the right amount of likes or seem to miss likes from meaningful
ties, their post will not result in the aimed positive outcomes (Hayes, 9.2. Implications for the Transactional Model of Social Media and Body
Wesselmann, & Carr, 2018). As such, individuals with low self-esteem Image Concerns
may consider visually oriented social media platforms as inappro­
priate tools to boost their low self-worth. They may rather prefer other, Our additional analyses suggest that the selection of (multiple)
more private social media to gratify their self-validation needs. Future visually oriented social media platforms does not predict body image
research should further explore this reasoning, also with respect to concerns and well-being over time, similar to recent cross-sectional
Self-Affirmation Theory (Sherman & Cohen, 2006; Steele, 1988). research (Jarman, Marques, McLean, Slater, & Paxton, 2021). Still, we
Moreover, we found that individuals with higher self-esteem were need better insights into the specific gratifications people seek and obtain
also more likely to show a need for social comparison in social media. on such platforms to fully understand the differential effects of social
One explanation may be that, in the case of high self-esteem, particular media use on well-being and body image concerns. For instance, when
types of social comparison gratifications are sought in social media. For the sought gratification of self-validation is obtained by receiving posi­
instance, research has shown that for people with high self-esteem, tive feedback or when the sought gratification of social comparison is
upward social comparison is used for further self-improvement or obtained with downward social comparison to less attractive peers,
inspiration (i.e., social comparison to successful others in order to psychological well-being and body image may be positively influenced.
improve oneself; Cramer et al., 2016; Meier & Schäfer, 2018). At the In contrast, when these gratifications are not obtained due to negative
same time, individuals with high self-esteem may also be more likely to feedback or when they result in upward social comparison with negative
use (downward) social comparison as a means to maintain their positive self-evaluations, well-being and body image may be negatively influ­
self-view (e.g., Gonzales & Hancock, 2011; Vohs & Heatherton, 2004). enced. The important role of feedback herein also seems to be supported
This suggests that social media gratifications sought may have to be by research on self-effects, suggesting that receiving and anticipating
further specified (e.g., need for inspiration versus need for feedback from others can amplify the effects of self-presentation (e.g.,
self-improvement, maintaining a positive self-view versus boosting a Valkenburg, 2017).
negative self-view) to better understand their relationship with Previous research has found associations between appearance con­
self-esteem. sciousness and decreased psychological well-being as well as higher
Unexpectedly, social comparison gratification negatively predicted body surveillance, higher body comparison, and lower body esteem
the number of visually oriented social media platforms used over time (Choukas-Bradley et al., 2018). Moreover, appearance feedback moti­
(from wave 1 to wave 2). One explanation could be that seeking the vation, a construct similar to the gratification of self-validation, has been
gratification of social comparison in visually oriented social media may associated with decreased body satisfaction and well-being (Jarman
not lead to a positive outcome. This is supported by studies showing that et al., 2021). Our findings suggest that these constructs may be indi­
the use of visually oriented social media, such as Snapchat and Insta­ rectly linked: Young people with greater appearance consciousness seek
gram, results in more appearance related social comparison as well as certain gratifications, such as self-validation, in social media, and as a
lower body satisfaction and more internalizing problems (Engeln et al., result select more visually oriented social media platforms for use. This
2020; Marengo et al., 2018). Moreover, subjectively negative experi­ may subsequently make young people more susceptible to effects on
ences on Instagram or Snapchat in particular impact body dissatisfaction their psychological well-being and body image concerns based on
through social comparison and body surveillance (Saunders & Eaton, whether their gratification of self-validation is obtained. Hence, future

10
J.M.F. van Oosten et al. Computers in Human Behavior 144 (2023) 107730

research will find a fruitful task in testing transactional relationships the selection of visually oriented social media platforms. As such, these
between well-being or body image concerns and social media use, findings contribute to a more nuanced and theory-driven understanding
focusing specifically on social media gratifications sought and how these of the use of visually oriented social media, and specifically a better
gratifications are obtained on visually oriented social media platforms. understanding of the processes that underlie such use.

9.3. Limitations Credit author statement

Several limitations of our study may have affected the validity of our Johanna M.F. van Oosten: Conceptualization; Methodology; Formal
findings. First, the two-wave longitudinal design of our survey did not analysis; Investigation; Writing - Original Draft; Laura Vandenbosch:
allow for fully testing the hypothesized indirect effects. Future research Conceptualization; Methodology; Investigation; Jochen Peter: Concep­
using three- or four-wave longitudinal designs is needed to fully test the tualization; Methodology; Investigation; Supervision; Project adminis­
modelled relationships between well-being and visually oriented social tration; Funding acquisition.
media use via gratifications sought. Moreover, such designs would help
testing the proposition that subsequent gratifications obtained may in­ Funding
fluence well-being and body image concerns. Second, we were unable to
establish causality given the correlational design of the study. That said, This research was supported in part by grants from the Netherlands
this is one of the few longitudinal studies on this topic and is thus able to Organization for Scientific Research (NWO) to the third author.
show that well-being and body image concerns precede appearance
related social media gratifications and use.
Third, our social media gratifications measures consisted of a single- Declaration of competing interest
item measure. Several short measures have been shown to perform as
good as longer measures and are a good way to prevent survey fatigue The authors declare no potential conflicts of interest with respect to
and burden (Bergkvist & Rossiter, 2007; Fisher, Matthews, & Gibbons, the research, authorship, and/or publication of this article.
2016; Petrescu, 2013; Robins et al., 2001; Schreurs & Vandenbosch,
2021; Zimmerman et al., 2006). In addition, single-item measures may
References
have more face validity (Nagy, 2002) and may facilitate interpretation
by respondents (Fisher et al., 2016). Although we believe that each item Aalbers, G., Mcnally, R. J., Heeren, A., Wit, S. De, & Fried, E. I. (2019). Social media and
did cover the meaning of the construct, our measures need to be vali­ depression symptoms: A network perspective. Journal of Experimental Psychology:
General, 148, 1454–1462. https://doi.org/10.1037/xge0000528.supp
dated in future research. Fourth, we focused on a general measure of
Alhabash, S., & Ma, M. (2017). A tale of four platforms: Motivations and uses of
social comparison. As mentioned above, social comparison gratifications Facebook, Twitter, Instagram, and Snapchat among college students. Social Media
can entail upward, downward, and lateral comparisons, and may need to and Society, 3, 1–13. https://doi.org/10.1177/2056305117691544
be specified to include different purposes of social comparison (i.e., Anderson, K. E. (2020). Getting acquainted with social networks and apps: It is time to
talk about TikTok. Library Hi Tech News, 37, 7–12. https://doi.org/10.1108/LHTN-
inspiration, self-enhancement). As a result, we still have a limited un­ 01-2020-0001
derstanding of the social media gratification of social comparison and Anderson, M., & Jiang, J. (2018). Teens, social media & technology 2018. In Pew research
how it explains the relationship between well-being or body image center (issue 31). http://assets.pewresearch.org/wpcontent/uploads/sites/14/2
018/05/31102617/PI_ 2018.05.31_TeensTech_FINAL.pdf.
concerns and visually oriented social media use. Appel, M., Marker, C., & Gnambs, T. (2020). Are social media ruining our lives? A review
Fifth, we only included two visually oriented social media platforms of meta-analytic evidence. Review of General Psychology, 24, 60–74. https://doi.org/
in our measure of selection of such platforms as these platforms were 10.1177/1089268019880891
Aubrey, J. S. (2006). Exposure to sexually objectifying media and body self-perceptions
frequently used among our target population and relevant for the grat­ among college women: An examination of the selective exposure hypothesis and the
ifications of self-validation and social comparison. However, this role of moderating variables. Sex Roles, 55(3–4), 159–172. https://doi.org/10.1007/
resulted in limited variance in our outcome measure. In the years since s11199-006-9070-7
Bergkvist, L., & Rossiter, J. R. (2007). The predictive validity of multiple-item versus
our data-collection, more visually oriented social media platforms have
single-item measures of the same constructs. Journal of Marketing Research, 44,
emerged, notably TikTok, which has become one of the most popular 175–184. https://doi.org/10.1509/jmkr.44.2.175
platforms among adolescents and young adults (Anderson, 2020; New­ Blumler, J. G., & Katz, E. (1974). The uses of mass communications: Current perspectives on
gratifications research. Sage Publications, Inc.
com, 2021). Future research should therefore include a greater number
Brown, Z., & Tiggemann, M. (2016). Attractive celebrity and peer images on Instagram:
of visually oriented social media platforms, such as TikTok, to further Effect on women’s mood and body image. Body Image, 19, 37–43. https://doi.org/
our knowledge on the relationship between visually oriented social 10.1016/j.bodyim.2016.08.007
media use and psychological well-being or body image concerns. Sixth Burnell, K., George, M. J., Kurup, A. R., & Underwood, M. K. (2021). “Ur a freakin
goddess!”: Examining appearance commentary on Instagram. Psychology of Popular
and finally, self-reported measures of time spent on social media are Media, 10, 422–433. https://doi.org/10.1037/ppm0000341
often problematic and young people, in particular, are not always able to Carlson Jones, D. (2001). Social comparison and body image: Attractiveness comparisons
correctly report their frequency of use (e.g., Scharkow, 2019; Verbeij to models and peers among adolescent girls and boys. Sex Roles, 45, 645–664.
https://link.springer.com/content/pdf/10.1023/A:1014815725852.pdf.
et al., 2021). Still, it may be valuable to combine the selection of visually Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing
oriented social media platforms with measures of time spent on each of measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 9,
those platforms in future research. 233–255. https://doi.org/10.1207/S15328007SEM0902_5
Cheung, G. W., & Wang, C. (2017). Current approaches for assessing convergent and
discriminant validity with SEM: Issues and solutions. Academy of Management
10. Conclusion Proceedings.
Choukas-Bradley, S., Nesi, J., Widman, L., & Galla, B. M. (2020). The appearance-related
social media consciousness scale: Development and validation with adolescents.
As young people spend more time online, they may create a constant Body Image, 33, 164–174. https://doi.org/10.1016/j.bodyim.2020.02.017
awareness of how they look to others online and may be constantly Choukas-Bradley, S., Nesi, J., Widman, L., & Higgins, M. K. (2018). Camera-ready: Young
aware of how their looks are judged by their online audience, resulting women’s appearance-related social media consciousness. Psychology of Popular
Media Culture, 8, 473–481. https://doi.org/10.1037/ppm0000196
in a greater need for self-validation by, or social comparison to, others in
Cramer, E. M., Song, H., & Drent, A. M. (2016). Social comparison on Facebook:
visually oriented social media platforms. The present study responds to a Motivation, affective consequences, self-esteem, and Facebook fatigue. Computers in
call for a better understanding of the relationship between social media Human Behavior, 64, 739–746. https://doi.org/10.1016/j.chb.2016.07.049
use and well-being or body image, and in particular to better understand Dumas, T. M., Maxwell-Smith, M., Davis, J. P., & Giulietti, P. A. (2017). Lying or longing
for likes? Narcissism, peer belonging, loneliness and normative versus deceptive like-
people’s antecedents of social media use (Vandenbosch & Eggermont, seeking on Instagram in emerging adulthood. Computers in Human Behavior, 71,
2016). Thereby, it initially fills a gap in the literature on what predicts 1–10. https://doi.org/10.1016/j.chb.2017.01.037

11
J.M.F. van Oosten et al. Computers in Human Behavior 144 (2023) 107730

Elhai, J. D., Levine, J. C., Alghraibeh, A. M., Alafnan, A. A., Aldraiweesh, A. A., & Lenton-Brym, A. P., Santiago, V. A., Fredborg, B. K., & Antony, M. M. (2021).
Hall, B. J. (2018). Fear of missing out: Testing relationships with negative affectivity, Associations between social anxiety, depression, and use of mobile dating
online social engagement, and problematic smartphone use. Computers in Human applications. Cyberpsychology, Behavior, and Social Networking, 24, 86–93. https://
Behavior, 89, 289–298. https://doi.org/10.1016/j.chb.2018.08.020 doi.org/10.1089/cyber.2019.0561
Engeln, R., Loach, R., Imundo, M. N., & Zola, A. (2020). Compared to Facebook, Marengo, D., Longobardi, C., Fabris, M. A., & Settanni, M. (2018). Highly-visual social
Instagram use causes more appearance comparison and lower body satisfaction in media and internalizing symptoms in adolescence: The mediating role of body image
college women. Body Image, 34, 38–45. https://doi.org/10.1016/j. concerns. Computers in Human Behavior, 82, 63–69. https://doi.org/10.1016/j.
bodyim.2020.04.007 chb.2018.01.003
Fardouly, Jasmine, Diedrichs, P. C., Vartanian, L. R., & Halliwell, E. (2015). Social Markey, C. N., & Markey, P. M. (2014). Gender, sexual orientation, and romantic partner
comparisons on social media: The impact of Facebook on young women’s body influence on body image: An examination of heterosexual and lesbian women and
image concerns and mood. Body Image, 13, 38–45. https://doi.org/10.1016/j. their partners. Journal of Social and Personal Relationships, 31, 162–177. https://doi.
bodyim.2014.12.002 org/10.1177/0265407513489472
Fardouly, J., Pinkus, R. T., & Vartanian, L. R. (2017). The impact of appearance Maxwell, S. E., & Cole, D. A. (2007). Bias in cross-sectional analyses of longitudinal
comparisons made through social media, traditional media and in person in mediation. Psychological Methods, 12, 23–44. https://doi.org/10.1037/1082-
women’s everyday lives. Body Image, 20, 31–39. https://doi.org/10.1016/j. 989X.12.1.23
bodyim.2016.11.002 Mcardle, K. a, & Hill, M. S. (2009). Understanding body dissatisfaction in gay and
Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, heterosexual men. The roles of self-esteem, media, and peer influence. Men and
117–140. Masculinities, 11, 511–532. https://doi.org/10.1177/1097184X07303728
Fisher, G. G., Matthews, R. A., & Gibbons, A. M. (2016). Developing and investigating the Meier, E. P., & Gray, J. (2014). Facebook photo activity associated with body image
use of single-item measures in organizational research. Journal of Occupational Health disturbance in adolescent girls. Cyberpsychology, Behavior, and Social Networking, 17,
Psychology, 21, 3–23. https://doi.org/10.1037/a0039139 199–206. https://doi.org/10.1089/cyber.2013.0305
Fox, J., & Moreland, J. J. (2015). The dark side of social networking sites: An exploration Meier, A., & Schäfer, S. (2018). Positive side of social comparison on social network sites:
of the relational and psychological stressors associated with Facebook use and How envy can drive inspiration on Instagram. Cyberpsychology, Behavior, and Social
affordances. Computers in Human Behavior, 45, 168–176. https://doi.org/10.1016/j. Networking, 21, 411–417. https://doi.org/10.1089/cyber.2017.0708
chb.2014.11.083 Mull, I. R., & Lee, S. E. (2014). PIN” pointing the motivational dimensions behind
Frederick, D. A., & Essayli, J. H. (2016). Male body image: The roles of sexual orientation Pinterest. Computers in Human Behavior, 33, 192–200. https://doi.org/10.1016/j.
and body mass index across five national U.S. studies. Psychology of Men and chb.2014.01.011
Masculinity, 17, 336–351. https://doi.org/10.1037/men0000031 Nagy, M. S. (2002). Using a single-item approach to measure facet job satisfaction.
Gioia, F., Griffiths, M. D., & Boursier, V. (2020). Adolescents’ body shame and social Journal of Occupational and Organizational Psychology, 75, 77–86. https://doi.org/
networking sites: The mediating effect of body image control in photos. Sex Roles, 10.1348/096317902167658
83, 773–785. https://doi.org/10.1007/s11199-020-01142-0 Newcom. (2021). Nationale social media onderzoek 2020 [National Social Media Research
Gonzales, A. L., & Hancock, J. T. (2011). Mirror, mirror on my Facebook wall: Effects of 2020] https://www.newcom.nl/socialmediaonderzoek/.
exposure to Facebook on self-esteem. Cyberpsychology, Behavior, and Social Palmgreen, P., Wenner, L. A., & Rayburn, J. D. (1980). Relations between gratifications
Networking, 14, 79–83. https://doi.org/10.1089/cyber.2009.0411 sought and obtained: A study of television news. Communication Research, 7,
Greenwood, D. N., & Dal Cin, S. (2012). Ethnicity and body consciousness: Black and 161–192. https://doi.org/10.1177/009365028000700202
White American women’s negotiation of media ideals and others’ approval. Perloff, R. M. (2014). Social media effects on young women’s body image concerns:
Psychology of Popular Media Culture, 1(4), 220–235. https://doi.org/10.1037/ Theoretical perspectives and an agenda for research. Sex Roles, 71(11–12), 363–377.
a0029411 https://doi.org/10.1007/s11199-014-0384-6
Grieve, R., Indian, M., Witteveen, K., Anne Tolan, G., & Marrington, J. (2013). Face-to- Perrin, A. (2015). Social networking usage: 2005-2015. http://www.pewinternet.org/2
face or Facebook: Can social connectedness be derived online? Computers in Human 015/10/08/2015/Social-Networking-Usage-2005-2015/.
Behavior, 29, 604–609. https://doi.org/10.1016/j.chb.2012.11.017 Petrescu, M. (2013). Marketing research using single-item indicators in structural
Haferkamp, N., & Krämer, N. C. (2011a). Social comparison 2.0: Examining the effects of equation models. Journal of Marketing Analytics, 1, 99–117. https://doi.org/
online profiles on social-networking sites. Cyberpsychology, Behavior, and Social 10.1057/jma.2013.7
Networking, 14(5), 309–314. https://doi.org/10.1089/cyber.2010.0120 Pew. (2019). Share of U.S. adults using social media, including Facebook. is mostly unchanged
Haferkamp, N., & Krämer, N. C. (2011b). Social comparison 2.0: Examining the effects of since 2018 https://www.pewresearch.org/fact-tank/2019/04/10/share-of-u-s-adu
online profiles on social-networking sites. Cyberpsychology, Behavior, and Social lts-using-social-media-including-facebook-is-mostly-unchanged-since-2018/.
Networking, 14(5), 309–314. https://doi.org/10.1089/cyber.2010.0120 Phua, J., Jin, S. V., Kim, J., & Jay, ). (2017). Uses and gratifications of social networking
Hayes, R. A., Wesselmann, E. D., & Carr, C. T. (2018). When nobody “likes” you: sites for bridging and bonding social capital: A comparison of Facebook, twitter,
Perceived ostracism through paralinguistic digital affordances within social media. Instagram, and Snapchat. Computers in Human Behavior, 72, 115–122. https://doi.
Social Media+ Society, 4, 1–12. https://doi.org/10.1177/2056305118800309 org/10.1016/j.chb.2017.02.041
Helfert, S., & Warschburger, P. (2013). The face of appearance-related social pressure: Pittman, M., & Reich, B. (2016). Social media and loneliness: Why an Instagram picture
Gender, age and body mass variations in peer and parental pressure during may be worth more than a thousand Twitter words. Computers in Human Behavior,
adolescence. Child and Adolescent Psychiatry and Mental Health, 7, 7–9. https://doi. 62, 155–167. https://doi.org/10.1016/j.chb.2016.03.084
org/10.1186/1753-2000-7-16 Primack, B. A., Shensa, A., Escobar-Viera, C. G., Barrett, E. L., Sidani, J. E., Colditz, J. B.,
Holbert, R. L., & Park, E. (2020). Conceptualizing, organizing, and positing moderation et al. (2017). Use of multiple social media platforms and symptoms of depression and
in communication research. Communication Theory, 30, 227–246. https://doi.org/ anxiety: A nationally-representative study among U.S. Young adults. Computers in
10.1093/ct/qtz006 Human Behavior, 69, 1–9. https://doi.org/10.1016/j.chb.2016.11.013
Holland, G., & Tiggemann, M. (2016). A systematic review of the impact of the use of Probst, M., Vandereycken, W., Coppenolle, H. Van, & Vanderlinden, J. (1995). The body
social networking sites on body image and disordered eating outcomes. Body Image, attitude test for patients with an eating disorder: Psychometric characteristics of a
17, 100–110. https://doi.org/10.1016/j.bodyim.2016.02.008 new questionnaire. Eating Disorders, 3, 133–144. https://doi.org/10.1080/
Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure 10640269508249156
analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, Quan-Haase, A., & Young, A. L. (2010). Uses and gratifications of social media: A
6(1), 1–55. https://doi.org/10.1080/10705519909540118 comparison of Facebook and instant messaging. Bulletin of Science, Technology &
Jarman, H. K., Marques, M. D., McLean, S. A., Slater, A., & Paxton, S. J. (2021). Society, 30, 350–361. https://doi.org/10.1177/0270467610380009
Motivations for social media use: Associations with social media Engagement and Reer, F., Festl, R., & Quandt, T. (2021). Investigating problematic social media and game
body satisfaction and well-being among adolescents. Journal of Youth and use in a nationally representative sample of adolescents and younger adults.
Adolescence. https://doi.org/10.1007/s10964-020-01390-z. advance on. Behaviour & Information Technology, 40, 776–789. https://doi.org/10.1080/
Johnson, B. K., & Knobloch-Westerwick, S. (2014). Glancing up or down: Mood 0144929X.2020.1724333
management and selective social comparisons on social networking sites. Computers Ricciardelli, L. A., McCabe, M. P., Williams, R. J., & Thompson, J. K. (2007). The role of
in Human Behavior, 41, 33–39. https://doi.org/10.1016/j.chb.2014.09.009 ethnicity and culture in body image and disordered eating among males. Clinical
Jones, C. J., & Crawford, J. (2006). The peer appearance culture during adolescence: Psychology Review, 27(5), 582–606. https://doi.org/10.1016/j.cpr.2007.01.016
Gender and body mass variations. Journal of Youth and Adolescence, 35, 257–269. Rigdon, E. E., Schumacker, R. E., & Wothke, W. (1998). A comparative review of
https://doi.org/10.1007/s10964-005-9006-5 interaction and nonlinear modeling. In R. E. Schumacker, & G. A. Marcoulides (Eds.),
Jong, S., & Drummond, M. (2016). Hurry up and ’ like ’ me : Girls, social media and Interaction and nonlinear efects in structural equation modeling (pp. 1–16). Mahwah,
immediate feedback. Asia-Pacific Journal of Health, Sport and Physical Education, 7, NJ: Lawrence Erlbaum.
251–267. https://doi.org/10.1080/18377122.2016.1222647 Robins, R. W., Hendin, H. M., & Trzesniewski, K. H. (2001). Measuring global self-
Kline, R. B. (1998). Principles and practice of structural equation modeling. Guilford. esteem: Construct validation of a single-item measure and the Rosenberg Self-Esteem
Krämer, N. C., & Winter, S. (2008). Impression management 2.0: The relationship of self- Scale. Personality and Social Psychology Bulletin, 27(2), 151–161. https://doi.org/
esteem, extraversion, self-efficacy, and self-presentation within social networking 10.1177/0146167201272002
sites. Journal of Media Psychology, 20, 106–116. https://doi.org/10.1027/1864- Rodgers, R. F. (2016). The relationship between body image concerns, eating disorders
1105.20.3.106 and internet use, part II: An integrated theoretical model. Adolescent Research Review,
de Lenne, O., Vandenbosch, L., Eggermont, S., Karsay, K., & Trekels, J. (2020). Picture- 1, 121–137. https://doi.org/10.1007/s40894-015-0017-5
perfect lives on social media: A cross-national study on the role of media ideals in Rousseau, A., Eggermont, S., & Frison, E. (2017). The reciprocal and indirect
adolescent well-being. Media Psychology, 23, 52–78. https://doi.org/10.1080/ relationships between passive Facebook use, comparison on Facebook, and
15213269.2018.1554494 adolescents’ body dissatisfaction. Computers in Human Behavior, 73, 336–344.
https://doi.org/10.1016/j.chb.2017.03.056

12
J.M.F. van Oosten et al. Computers in Human Behavior 144 (2023) 107730

Ruggiero, T. E. (2000). Mass communication and society uses and gratifications theory in Tobin, D. D., Menon, M., Menon, M., Spatta, B. C., Hodges, E. V. E., & Perry, D. G. (2010).
the 21st Century. Mass Communication & Society, 3, 3–37. https://doi.org/10.1207/ The intrapsychics of gender: A model of self-socialization. Psychological Review, 117,
S15327825MCS0301 601–622. https://doi.org/10.1037/a0018936
Rumpf, H. J., Meyer, C., Hapke, U., & John, U. (2001). Screening for mental health: Toma, C. L., & Hancock, J. T. (2013). Self-affirmation underlies Facebook use. Personality
Validity of the MHI-5 using DSM-IV Axis I psychiatric disorders as gold standard. and Social Psychology Bulletin, 39, 321–331. https://doi.org/10.1177/
Psychiatry Research, 105, 243–253. https://doi.org/10.1016/S0165-1781(01)00329- 0146167212474694
8 Utz, S., Muscanell, N., & Khalid, C. (2015). Snapchat elicits more jealousy than Facebook:
Santor, D. A., & Yazbek, A. A. (2006). Soliciting unfavourable social comparison: Effects A comparison of Snapchat and Facebook use. Cyberpsychology, Behavior, and Social
of self-criticism. Personality and Individual Differences, 40, 545–556. https://doi.org/ Networking, 18, 141–146. https://doi.org/10.1089/cyber.2014.0479
10.1016/j.paid.2005.06.029 Valkenburg, P. M. (2017). Understanding self-effects in social media. Human
Saunders, J. F., & Eaton, A. A. (2018). Snaps, selfies, and shares: How three popular Communication Research, 43, 477–490. https://doi.org/10.1111/hcre.12113
social media platforms contribute to the sociocultural model of disordered eating van Oosten, & Vandenbosch, L. (2017). Sexy online self-presentation on social network
among young women. Cyberpsychology, Behavior, and Social Networking, 21, sites and the willingness to engage in sexting: A comparison of gender and age.
343–354. https://doi.org/10.1089/cyber.2017.0713 Journal of Adolescence, 54, 42–50. https://doi.org/10.1016/j.
Scharkow, M. (2019). The reliability and temporal stability of self-reported media adolescence.2016.11.006
exposure: A meta-analysis. Communication Methods and Measures, 13, 198–211. van Oosten, & Vandenbosch, L. (2020). Predicting the willingness to engage in non-
https://doi.org/10.1080/19312458.2019.1594742 consensual forwarding of sexts: The role of pornography and instrumental notions of
Schreurs, L., & Vandenbosch, L. (2021). The development and validation of measurement sex. Archives of sexual behavior, 49(4), 1121–1132. https://doi.org/10.1007/s10508-
instruments to address interactions with positive social media content. Media Psychology. 019-01580-2
https://doi.org/10.1080/15213269.2021.1925561. Advance on. Vandenbosch, L., & Eggermont, S. (2013). Sexualization of adolescent boys: Media
Seekis, V., Bradley, G. L., & Duffy, A. L. (2020). Appearance-related social networking exposure and boys’ internalization of appearance ideals, self-objectification, and
sites and body image in young women: Testing an objectification-social comparison body surveillance. Men and Masculinities, 16, 283–306. https://doi.org/10.1177/
model. Psychology of Women Quarterly, 44, 377–392. https://doi.org/10.1177/ 1097184X13477866
0361684320920826 Vandenbosch, L., & Eggermont, S. (2016). The interrelated roles of mass media and social
Shchebetenko, S. (2019). Do personality characteristics explain the associations between media in adolescents’ development of an objectified self-concept: A longitudinal
self-esteem and online social networking behaviour? Computers in Human Behavior, study. Communication Research, 43, 1116–1140. https://doi.org/10.1177/
91, 17–23. https://doi.org/10.1016/j.chb.2018.09.017 0093650215600488
Sheldon, P., & Bryant, K. (2016). Instagram: Motives for its use and relationship to Vandenbosch, L., & van Oosten, J. M. (2017). The relationship between online
narcissism and contextual age. Computers in Human Behavior, 58, 89–97. https://doi. pornography and the sexual objectification of women: The attenuating role of porn
org/10.1016/j.chb.2015.12.059 literacy education. Journal of Communication, 67(6), 1015–1036. https://doi.org/
Sherman, D. K., & Cohen, G. L. (2006). The psychology of self-defense: Self-affirmation 10.1111/jcom.12341
theory. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 38). Vannucci, A., Ohannessian, C. M. C., & Gagnon, S. (2019). Use of multiple social media
Academic Press, 183–142. platforms in relation to psychological functioning in emerging adults. Emerging
Shim, M., Lee, M. J., & Park, S. H. (2008). Photograph use on social network sites among Adulthood, 7, 501–506. https://doi.org/10.1177/2167696818782309
South Korean college students: The role of public and private self-consciousness. Verbeij, T., Pouwels, J. L., Beyens, I., & Valkenburg, P. M. (2021). The accuracy and
CyberPsychology and Behavior, 11(4), 489–493. https://doi.org/10.1089/ validity of self-reported social media use measures among adolescents. Computers in
cpb.2007.0104 Human Behavior Reports, 3, Article 100090. https://doi.org/10.1016/j.
Steele, C. M. (1988). The psychology of self-affirmation: Sustaining the integrity of the chbr.2021.100090
self. In L. Berkowtz (Ed.), Advances in experimental social psychology (Vol. 21, pp. Vogel, E. A., Rose, J. P., Roberts, L. R., & Eckles, K. (2014). Social comparison, social
261–302). Academic Press. media, and self-esteem. Psychology of Popular Media Culture, 3, 206–222. https://doi.
Stefanone, M. A., Lackaff, D., & Rosen, D. (2011). Contingencies of self-worth and social- org/10.1037/ppm0000047
networking-site behavior. Cyberpsychology, Behavior, and Social Networking, 14, Vohs, K. D., & Heatherton, T. F. (2004). Ego threat elicits different social comparison
41–49. https://doi.org/10.1089/cyber.2010.0049 processes among high and low self-esteem people: Implications for interpersonal
Sundar, S. S., & Limperos, A. M. (2013). Uses and grats 2.0: New gratifications for new perceptions. Social Cognition, 22, 168–191. https://doi.org/10.1521/
media. Journal of Broadcasting & Electronic Media, 57, 504–525. https://doi.org/ soco.22.1.168.30983
10.1080/08838151.2013.845827 Waterloo, S. F., Baumgartner, S. E., Peter, J., & Valkenburg, P. M. (2018). Norms of
Sussman, N. M., Truong, N., & Lim, J. (2007). Who experiences “America the beautiful”?: online expressions of emotion: Comparing Facebook, twitter, Instagram, and
Ethnicity moderating the effect of acculturation on body image and risks for eating WhatsApp. New Media & Society, 20(5), 1813–1831. https://doi.org/10.1177/
disorders among immigrant women. International Journal of Intercultural Relations, 31 1461444817707349
(1), 29–49. https://doi.org/10.1016/j.ijintrel.2006.03.003 Webb, H. J., Zimmer-Gembeck, M. J., Waters, A. M., Farrell, L. J., Nesdale, D., &
Swallow, S. R., & Kuiper, N. A. (1992). Mild depression and frequency of social Downey, G. (2017). Pretty pressure” from peers, parents, and the media: A
comparison behavior. Journal of Social and Clinical Psychology, 11, 167–180. longitudinal study of appearance-based rejection sensitivity. Journal of Research on
Thelwall, M. (2008). Social networks, gender and friending: An analysis of MySpace Adolescence, 27, 718–735. https://doi.org/10.1111/jora.12310
member profiles. Journal of the American Society for Information Science and Zaiţ, A., & Bertea, P. S. P. E. (2011). Methods for testing discriminant validity.
Technology, 59, 1321–1330. https://doi.org/10.1002/asi.20835 Management and Marketing Journal, 9, 217–224. https://core.ac.uk/download/p
Thelwall, M., & Vis, F. (2017). Gender and image sharing on Facebook, twitter, df/6277904.pdf.
Instagram, Snapchat and WhatsApp in the UK: Hobbying alone or filtering for Zimmerman, M., Ruggero, C. J., Chelminski, I., Young, D., Posternak, M. A.,
friends? Aslib Journal of Information Management, 69, 702–720. https://doi.org/ Friedman, M., et al. (2006). Developing brief scales for use in clinical practice: The
10.1108/AJIM-04-2017-0098 reliability and validity of single-item self-report measures of depression symptom
Tiggemann, M. (2015). Considerations of positive body image across various social severity, psychosocial impairment due to depression, and quality of life. Journal of
identities and special populations. Body Image, 14, 168–176. https://doi.org/ Clinical Psychiatry, 67, 1536–1541. https://doi.org/10.4088/JCP.v67n1007
10.1016/j.bodyim.2015.03.002 Zurbriggen, E., Collins, R., Lamb, S., Roberts, T., Tolman, D., Ward, L., et al. (2010).
Tiggemann, M., & Zaccardo, M. (2018). ‘Strong is the new skinny’: A content analysis of# Report of the APA task force on the sexualization of girls. American Psychological
fitspiration images on Instagram. Journal of Health Psychology, 23, 1003–1011. Association. http://www.apa.org/pi/women/programs/girls/report-full.pdf.
https://doi.org/10.1177/1359105316639436

13