Discussion questions

lable at ScienceDirect

Computers in Human Behavior 82 (2018) 63e69

Contents lists avai

Computers in Human Behavior

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

Full length article

Highly-visual social media and internalizing symptoms in adolescence: The mediating role of body image concerns

D. Marengo, C. Longobardi*, M.A. Fabris, M. Settanni Department of Psychology, University of Turin, Italy

a r t i c l e i n f o

Article history: Received 15 September 2017 Received in revised form 29 December 2017 Accepted 3 January 2018 Available online 3 January 2018

Keywords: Social media Mental health Adolescents Body image concern Internalizing symptoms Facebook

* Corresponding author. Department of Psychology, Po 14, 10123 TO, Italy.

E-mail address: [email protected] (C. Lo

https://doi.org/10.1016/j.chb.2018.01.003 0747-5632/© 2018 Elsevier Ltd. All rights reserved.

a b s t r a c t

Highly-visual social media (HVSM), such as Instagram and Snapchat, have experienced a significant increase in popularity among adolescents in recent years. Findings indicate use of social media is related to body image concerns and poorer mental health in adolescence. However, previous research on HVSM is scant and mainly focus on female samples. In this view, the present study investigated the association between time spent on HVSM, body image concerns and internalizing symptoms, in sample of adoles- cents attending grades 6e11 in Northern Italy. Data for this study were based on 523 students, 54.2% female; Mean age (SD)¼ 14.82 (1.52). Multiple linear regression was used to examine the associations between time spent using social media, body image concerns, and internalizing symptoms. Overall, students reporting using Facebook and HVSM for more than 2 h were 7.1% and 28.9% of the sample. Students reporting frequent use of HVSM (>2 h/day) reported significantly higher body image concerns and internalizing symptoms than peers reporting no use of HVSM. Further, we found the positive link between use of HVSM and internalizing symptoms to be mediated by participants’ body image concerns problems. These findings suggest that adolescents reporting high use of HVSM might be at risk for increased body image concerns, which in turn might lead to poorer psychological adjustment.

© 2018 Elsevier Ltd. All rights reserved.

1. Introduction

1.1. Body image concerns and internalizing symptoms in adolescence

Probably due to socio-cultural influences and socialization processes, body image is a central feature in adolescent self- concept. Adolescents can develop negative feelings about their body and may be dissatisfied with their body image. Girls tend to seek a thin figure, in line with the socio-cultural model of aesthetic perfection in force in western society; boys are divided between those that perceive themselves as overweight and those that feel they are too thin andmay therefore feel dissatisfied with their body and decide to lose weight in order to achieve a lean, muscular body shape (i.e, drive for muscularity) (Hargreaves & Tiggemann, 2004; Longobardi, Prino, Fabris, & Settanni, 2017).

Various studies have found a correlation between body dissat- isfaction and internalizing symptoms, such as anxiety and

University of Turin, Turin, Via

ngobardi).

depression, in adolescents, among whom girls appear more at risk than boys (Bucchianeri et al., 2016; Rawana & Morgan, 2014; Stice & Bearman, 2001). Further, findings from longitudinal studies indicate that body image act as a prospective predictor of inter- nalizing symptoms in adolescence: Over time, persistently negative body image perceptions can favor the development of depressed mood in both adolescent boys and girls (e.g., Holsen, Kraft, & Røysamb, 2001; Morin, Maïano, Scalas, Janosz, & Litalien, 2017; Paxton, Neumark-Sztainer, Hannan, & Eisenberg, 2006). The explanation for the relation between the two aspects is not straightforward. Comparisons with the models presented by the mass media and with their peers are likely to increase body image concerns in the young who perceive a gap between their own body image and the aesthetic ideal internalized. Internalizing symptoms may correlate with body dissatisfaction due to the sense of in- adequacy experienced by adolescents with negative body image and because of the difficulty in social interaction and concerns about peer acceptance often experienced by adolescents (Colunga- Rodríguez et al., 2016). As a result, adolescents trying to cope with these negative feelings are more at risk of being involved in un- healthy behaviors such as restrictive diet, excessive exercise, ste- roid or illicit drug consumption, excessive tanning etc.

D. Marengo et al. / Computers in Human Behavior 82 (2018) 63e6964

1.2. Use of social media, internalizing symptoms and body image concerns

Due to the increasing popularity of social media among young people, many studies have investigated its role as factor in their psychological adjustment. Findings indicate that heavy use of social media, such as Facebook, Twitter or Instagram, is associated with poor mental health and increased internalizing symptoms in adolescence (Sampasa-Kanyinga & Lewis, 2015) and young adult- hood (Jelenchick, Eickhoff, & Moreno, 2013; Moreno et al., 2011; Rosenthal, Buka, Marshall, Carey, & Clark, 2016; Simoncic, Kuhlman, Vargas, Houchins, & Lopez-Duran, 2014). A recent sur- vey conducted in the UK among young people aged 14e24 found social media platforms, including Facebook, Twitter, Instagram, to have a detrimental effect on many health-related psychological outcomes including sleep problems, anxiety, and depression (RSPH, 2017).

One of the key aspects of social media that has been under the scrutiny of researchers is that of their role in promoting negative social comparison, and its link with subjective well-being and un- healthy behaviors (Gerson, Plagnol, & Corr, 2016; Walker et al., 2015; Yang, 2016). This link is theoretically based on the notion that upward social comparison decreases happiness. Findings indicate that social media users are prone to impression manage- ment practices (Manago, Graham, Greenfield, & Salimkhan, 2008; Settanni & Marengo, 2015; Vogel & Rose, 2016; Zhao, Grasmuck, & Martin, 2008). For example, they may underreport negative ex- periences (Settanni&Marengo, 2015; Zhao et al., 2008), or digitally edit or limit publication of unflattering images on their online profiles (Lang & Barton, 2015; Manago et al., 2008; Toma & Hancock, 2010). This manipulation of their social media presenta- tion may increase the possibility that posted content reflect a highly idealized image (Manago et al., 2008), facilitating upward appearance comparisons among users browsing their social media profiles, and ultimately undermining their emotional wellbeing (Liu et al., 2017). This process might be partially interpreted in light of changes in body image perception. Regular exposure to the vast amount of visual content (image, videos) posted on social media platforms provides users with many situations in which they are led to compare their appearance with that of other users, which in turn can have negative effects on their personal body image, in particular during adolescence (Fardouly& Vartanian, 2016). Indeed, an ever-growing number of findings exist indicating a positive link between social media usage and body image concerns among ad- olescents and young adults (Kim & Chock, 2015; Manago, Ward, Lemm, Reed, & Seabrook, 2015), in particular among female users (Fardouly, Diedrichs, Vartanian, & Halliwell, 2015a, 2015b; Meier & Gray, 2014; Smith, Hames, & Joiner, 2013; Tiggemann & Miller, 2010; Tiggemann & Slater, 2013, 2014; Vandenbosch & Eggermont, 2012). The majority of these researches focus on Facebook, mostly due its great popularity compared with other social media platforms (Facebook has more than 2 billion active users, rendering it the most used social media in the world, Statista, 2017a). However, highly visual social media (HVSM) platforms, such as Instagram and Snapchat, have recently experienced a rise in popularity, particularly among young people (RSPH, 2017). These social media platforms mainly focus on sharing user-generated visual content, such images and short videos, and allow the use of filters tomodify and improve users’ looks before upload. Findings concerning the impact of these image-focused platforms on body image concerns and internalizing symptoms are scant, and mostly focus on female samples. Recent studies on young women have shown that exposure to other users’ Instagram profiles and photos can result in an increase in negative mood (Brown & Tiggemann, 2016) and body dissatisfaction (Hendrickse, Arpan, Clayton, &

Ridgway, 2017). Consistent with these findings, usage of two pop- ular HVSM, Instagram and Snapchat, has been linked with wors- ened body image and mental health by a recent UK-wide survey conducted among adolescents and young adults of both genders (RSPH, 2017). Still, as noted by Fardouly and Vartanian (2016), there is a need for research on more diverse samples to clarify the impact these social media platforms have on users’ body image concerns, and ultimately on their psychological well-being.

1.3. The present study

Guided by previous findings, the present study aims at evaluate the association between social media use, and in particular that of HVSM, with body image concerns and internalizing symptoms in a sample of adolescents attending grades 6e11. In particular, in light of the Social Comparison Theory (Festinger, 1954), we hypothesize that frequent use of HVSM could favor the development of inter- nalizing symptoms as a consequence of increased opportunities for upward comparisons with idealized body images posted by other users. Furthermore, we expect that the negative impact of HVSM usage on internalizing symptoms is partially explained by a wors- ening body image perception. Guided by findings of longitudinal studies (e.g., Holsen et al., 2001; Morin et al., 2017; Paxton et al., 2006), in the analyses we treat body image concerns as a predic- tor of internalizing symptoms. Specifically, we evaluate the role of body image concerns as mediator of the relationship between time spent on HVSM and internalizing symptoms, controlling for age, gender, and time spent on Facebook. We control for gender and age differences due to their expected impact on key study variables in adolescence as recognized by previous studies (i.e., use of social media, e.g., Lenhart, 2015; body image concerns, e. g, O’Dea & Caputi, 2001; and internalizing symptoms, e.g., Natsuaki, Biehl, M & Ge, 2009). Next, the rationale for controlling time spent on Facebook relates to the need to evaluate whether a negative effect of HVSM over body image concerns and internalizing symptoms exists, and if this effect is distinct from the effect of Facebook use.

Based on previous findings, our hypotheses are the following: h1) Adolescents’ HVSM use is linked to an increase in both body image concerns and internalizing symptoms; h2) The effect of HVSM use on internalizing symptoms is mediated by body image concerns.

2. Methods

2.1. Sample

Initial sample consisted of 598 adolescents - 54.2% female, mean age (SD)¼ 14.82 (1.52) - attending grades 6e11 in two large sec- ondary schools in Northern Italy. After removal of participants with missing data on studymeasures (n¼ 75), our final sample consisted of 523 participants - 53.5% female, mean age (SD)¼ 14.82 (1.52).

2.2. Materials and procedure

Data used in this study were collected from schools in May of 2017. Participants were asked to fill in a questionnaire using paper and pencil. Participation in the study required informed consent from both parents and students. Ethical approval to conduct research was obtained from the University of Turin IRB (protocol no. 256071).

2.3. Instruments

2.3.1. Use of social media We asked participants to report about daily use (hours per day)

D. Marengo et al. / Computers in Human Behavior 82 (2018) 63e69 65

of Facebook and highly visual social media (HVSM, i. e, Instagram, Snapchat). Based on participant reports, we computed two in- dicators of use of Facebook and HVSM. In order to distinguish be- tween adolescents reporting different patterns of social media use (i.e., no use, moderate and frequent social media use), participants’ use of social media platform was categorized based on the rec- ommended 2 h cut-off for identifying high social media use in adolescence (O’Keeffe, Clarke-Pearson, & Media, 2011; Sampasa- Kanyinga & Lewis, 2015). For the purpose of this study, we group- ed participants according to the following categories: no use, moderate use (�2 h/day), frequent use (>2 h/day). Distinguishing betweenmoderate and frequent use of social media has been found to be relevant for this kind of research: Findings indicate that moderate use may in some cases prove beneficial for adolescents’ mental health, while frequent use is generally expected to have a negative impact (Frith, 2017).

2.3.2. Body image concerns We assessed body image concerns using the Italian version of

BSQ (Body Shape Questionnaire; Cooper, Taylor, Cooper,& Fairbum, 1987), a self-report instrument evaluating weight and shape pre- occupations. Sample items include: “Have you been so worried about your shape that you have been feeling you ought to diet?”; “Have you noticed the shape of others and felt that your own shape compared unfavorable?”. The questions are answered on a six- point Likert scale (1¼ never, 6¼ always); the score ranges from 34 to 204. In our sample, reliability was excellent (a¼ 0.94).

2.3.3. Internalizing symptoms We used the Italian self-rated version (Marzocchi et al., 2004) of

the Strength and Difficulties Questionnaires (SDQ (Goodman, Meltzer, & Bailey, 1998); to assess students’ perception of their own difficulties and strengths. Adolescents in the present study compiled the Italian version of the SDQ2, which consists of 25 items measuring five dimensions. The five dimensions are: Emotional Symptoms Conduct Problems, Hyperactivity/Inattention, Peer Relationship Problems, and Prosocial Behavior. All items are rated on a three-point Likert scale (0¼Not True, 2¼ Certainly True), and raw scores are used to compute the five subscale scores; a higher score indicates more difficulties or strengths, depending on the subscale. In the present study, the SDQ emotional symptoms sub- scale was used in the analysis (e.g., “I am often unhappy”). Cron- bach’s alpha of the SDQ subscale emotional symptoms was adequate (a¼ 0.72).

2.3.4. Analysis strategy As noted above, we found a certain amount of observations with

missing data in our initial dataset (n¼ 75), i.e., observations for which only incomplete information about BSQ (n¼ 65) and SDQ (n¼ 10) was available, thus not allowing computation of the mea- sures’ scores. Prior to subsequent analyses, we tested if the missing data met the Missing Completely at Random (MCAR) assumption. If missing data meets the MCAR assumption, removal of observations with missing data is expected to produce unbiased estimates of coefficient in regression models (Allison, 2002). Results of Little’s MCAR test (c2¼17.34, p¼ 0.63) and the nonparametric MCAR test proposed by Jamshidian and Jalal (2010; as implemented in the MissMech R package, Jamshidian, Jalal, & Jansen, 2014) (Hawkins test: p¼ 0.12; non-parametric test of homoscedasticity: p¼ 0.31) as performed on the study variables indicated that missing data met the MCAR assumption, thus supporting our decision not to include observations with missing data in the following analyses.

Association between time (hours/day) spent on Facebook and HVSM, and the other studymeasures was evaluated using Pearson’s correlation. Next, Student’s t-test was used to explore mean

differences between users and non-users of Facebook and HVSM on continuous variables (i.e., age, body image concerns, and internal- izing symptoms). Chi-Square tests were used to investigate differ- ences in use of Facebook and HVSM across gender.

Multiple regression analyses were then performed to test the mediating role of body image concerns in the relationship between time spent on HVSM and internalizing symptoms. Specifically, we employed the bootstrapping technique for testing mediation using the Hayes’ Process macro for SPSS (2008), creating 5.000 boot- strapped samples for each analysis. This approach required the estimation of three regression models. In each model, the effect of time spent on HVSM was included by using two dummy coded variables (Moderate, and Frequent HVSM use), treating no HSVM use as a reference category in the analyses (Preacher & Hayes, 2004). First, a regression model was run to test the association between time spent on HVSM and the outcome variable, i.e. internalizing symptoms. A second regression model tested the impact of time spent on HVSM on the mediator, i.e., body image concerns. In a third, final model, time spent on HVSM and body image concerns are both entered in the samemodel as predictors of the outcome, i.e., internalizing symptoms. In each model, we controlled for students’ age, gender, and Time spent on Facebook (Moderate and Frequent use). After running the models, mediation was examined by testing significance of indirect effects, repre- senting the multiplication of the unstandardized effect of the in- dependent variable (i.e., time spent on HVSM) on the mediator (i.e. body image concerns, with the effect of the mediator (i.e., body image concerns) on the dependent variable (i.e., internalizing symptoms). Significance of the indirect effect was assessed using 95% bias-corrected confidence intervals. Indirect effects were deemed significant if the associated confidence interval did not span zero.

Except where indicated, all analyses were performed using SPSS version 18.

3. Results

3.1. Correlations among study measures

Correlations among study measures are reported in Table 1. We found a significant positive correlation between age and both time spent on Facebook and HVSM. Being female showed a positive correlation with the amount of time participants spent on HVSM, and reported body image concerns, and internalizing symptoms. Both time spent on Facebook and HVSM were positively correlated with body image concerns and internalizing symptoms. Finally, there was a positive correlation between body image concerns and internalizing symptoms.

3.2. Participant characteristics by social media platform

In our sample, Instagram emerged as themost used social media (79.9% of participants), followed by Facebook (57%), and Snapchat (49.5%). Users of HVSM (Instagram, Snapchat) were 82.4% of the sample. Table 2 reports participants’ characteristics by use of social media platform. We found use of Facebook to be significantly more common among males than females. In turn, females were more likely to use HVSM than males. Use of Facebook and HVSM was significantly more common among high school students. Coher- ently, Facebook and HVSM users were significantly older than non- users.

Concerning body image concerns and internalizing symptoms, we found HVSM users to report significantly higher scores than non-users on both measures, while no significant differences were found when comparing Facebook users and non-users.

Table 1 Correlations among study measures (N¼ 523).

1 2 3 4 5

1 Age e

2 Gender (1¼ Female; 0¼Male) -0.02 e

3 Time spent on Facebook (hours/day) 0.31** 0.05 e

4 Time spent on HVSM (hours/day) 0.13** 0.26** 0.38** e

5 Body image concerns (BSQ) 0.07 0.41** 0.13** 0.17** e

6 Internalizing symptoms (SDQ-SE) 0.05 0.44** 0.09* 0.21** 0.52**

*p < 0.05, **p < 0.01

Table 2 Demographic characteristics and prevalence of body shape concerns and internalizing symptoms by type of social media platform (N¼ 523).

Uses facebook Uses HVSM

Yes No Yes No

N (%) 298 (57.0%) 225 (43.0%) 431 (82.4%) 92 (17.6%) % % P % % p

Sex <0.05 <0.01 Female 52.5% 47.5% 87.5% 12.5% Male 62.1% 37.9% 76.5% 23.5%

M (SD) M (SD) P M (SD) M (SD) p Age 15.29 (1.38) 14.20 (1.45) <0.01 14.91 (1.44) 14.39 (1.75) <0.01 Body image concerns 66.99 (32.07) 61.75 (27.97) 67.28 (31.66) 52.78 (20.33) <0.01 Internalizing symptoms 3.72 (2.57) 3.56 (2.39) 3.78 (2.53) 3.02 (2.19) <0.01 Average daily use (Minutes) 73.33 (91.23) 154.12 (208.76) Frequent use (>2 h/day) N¼ 40 (7.6%) N¼ 152 (29.1%)

D. Marengo et al. / Computers in Human Behavior 82 (2018) 63e6966

3.3. Effects of time spent on HVSM, and body image concerns on internalizing symptoms

Multiple regression analyses were run to examine body- dissatisfaction as a mediator in the relation between time spent on HVSM and each internalizing symptoms, controlling for gender, age, and time spent on Facebook (No use/Moderate use/Frequent use) (see Fig. 1). The first regression (R2¼ 0.20) aimed at examining the impact of time spent on HVSM on adolescents’ internalizing symptoms. Results showed that frequent use of HVSM positively predicted internalizing symptoms (B¼ 0.69, SE¼ 0.32, p¼ 0.03, 95% CI [0.06, 1.33]), while moderate use (<2 h/day) did not emerge as a significant predictor (B¼ 0.13, SE¼ 0.28, p¼ 0.64, 95% CI [-0.42, 0.68]). The second regression model (R2¼ 0.20) tested the impact of HVSMuse on body image concerns. Results showed that frequent use of HVSM (>2 h/day) positively predicted body image concerns (B¼ 0 0.33, SE¼ 0.13, p¼ 0.01, 95% CI [0.07, 0.59]); again, moderate use (�2 h/day) showed a positive non-significant effect (B¼ 0.22, SE¼ 0.11, p¼ 0.06, 95% CI [-0.01, 0.44]). Further analyses were performed to verify whether the variance in internalizing symp- toms explained by frequent use of visual social media decreases when body image concerns is included in the model as a predictor. As shown in Fig. 1, when body-dissatisfaction was included in the

Fig. 1. Parameter estimates of mediation analyses, controlling for demographic characterist use). Note. *p < 0.05, **p < 0.01. a Effects controlling for body image concerns. Effects of co

model, frequent use of visual social media no longer predicted internalizing symptoms. In this final model (R2¼ 0.33), the path coefficient between frequent use of HVSM and internalizing symptoms decreased from 0.69 (p< 0.05) to 0.37 (p¼ 0.22). Moreover, body image concerns positively predicted internalizing symptoms (B¼ 0.99, SE¼ 0.10, p< 0.01, 95% [0.80, 1.19]). As a final step, the results for the indirect effects analysis were significant for both moderate (effect¼ 0.22, SE¼ 0.09, 95% CI [0.04, 0.40]) and frequent use (effect¼ 0.33, SE¼ 0.12, 95% CI [0.11, 0.57]) of visual social media, confirming the mediating role of body image concerns.

All tested effects were controlled for students’ age, gender, and time spent on Facebook (No use/Moderate use/Frequent use). Only gender showed significant effects, indicating female had higher body image concerns (B¼ 0.79, SE¼ 0.08, p< 0.001, 95% CI [0.63, 0.95]), and internalizing problems (B¼ 2.05, SE¼ 0.21, p< 0.001, 95% CI [1.64, 2.45], even after including body image concerns in the model, B¼ 1.26, SE¼ 0.21, p< 0.001, 95% CI [0.86, 1.66]).

4. Discussion

Social media is now an integral part of adolescents’ social life. In particular, in the last few years social media focusing on visual

ics (age, sex, school-level) and time spent on Facebook (No use/Moderate use/Frequent ntrol variables are not shown to improve readability.

D. Marengo et al. / Computers in Human Behavior 82 (2018) 63e69 67

content, such as Instagram and Snapchat, have been outshining older social media platforms in terms of popularity (e.g., Facebook, Twitter) especially among adolescents (Statista, 2017b). This was echoed in our study, as recruited adolescents were significantly more likely to report use of HVSM than Facebook.

Based on findings from previous studies conducted on different social media platforms, the first aim of this study was to investigate the relationship between use of HVSM and both body image con- cerns and internalizing symptoms. Our results are in line with previous findings indicating that frequent use of social media may have a negative impact on both body image (Fardouly, Diedrichs, Vartanian, & Halliwell, 2015b, 2015a; Kim & Chock, 2015; Manago et al., 2015; Meier & Gray, 2014; Smith et al., 2013; Tiggemann & Miller, 2010; Tiggemann & Slater, 2013, 2014; Vandenbosch & Eggermont, 2012) and mental health (Fabris, Longobardi, Prino, & Settanni, 2017; Fardouly et al., 2015a; Jelenchick et al., 2013; Moreno et al., 2011; Rosenthal et al., 2016; Sampasa-Kanyinga & Lewis, 2015; Simoncic et al., 2014) in adolescence and young adulthood. In our study, we found that adolescents reporting use of social media focusing on visual con- tent (i.e., Instagram, Snapchat) express significantly greater dissatisfaction with their body image and report greater levels of emotional symptoms. Further, these effects appear to be stronger than the effects linked to the use of Facebook, and to be stronger when use is frequent (>2 h per day).

Next, we tested for the presence of a mediating effect of body- image concerns on the link between HVSM use and internalizing symptoms. As hypothesized, we found that the effect of HVSM was fully mediated by body image concerns, supporting the hypothesis that frequent use of HVSM favors an increase of body image con- cerns in adolescents, and that in turn the greater worry for physical appearance results in higher levels of emotional symptoms among HVSM users.

These results are interpretable in the theoretical framework of Social Comparison Theory: As it happens in mass-media in general, the contents of HVSM become a source of comparison both for boys and girls. A more frequent use and interaction on HVSM increases the opportunities for comparison with images posted by other users, pictures that often emulate socially accepted “idealized” body image models (Kim & Chock, 2015). The perception of a gap between the assessment of their own body image and these pic- tures favors the increase of dissatisfaction and concerns about body image, which in turn, makes the development of internalizing symptoms, such as depression and anxiety, more likely.

Even though these results bring a new perspective to the rela- tionship between social media use and internalizing symptoms, they must be interpreted with caution, and the specific research design should be taken into account. We employed a cross- sectional design. Although most studies measuring a mediating effect have adopted this type of research design, longitudinal studies would allow a better understanding of the associations between use of social media, internalizing symptoms, and body image concerns. The next limitation is theoretical and has to do with the hypothesized direction of causality of the link between body image concerns and internalizing symptoms. Longitudinal studies suggest that in adolescence the direction of causality among these two constructs goes from body image concerns to internal- izing symptoms, i.e., body image concerns affect adolescents’ internalizing symptoms, while the reverse link does not generally hold (e.g., Holsen et al., 2001; Morin et al., 2017; Paxton et al., 2006). Guided by these findings, in our analyses we treated body image concerns as a predictor of internalizing symptoms. However, it is plausible that internalizing symptoms might also play a role in predicting body image concerns (e.g., Aderka, et al. 2014). Collect- ing longitudinal data, future studies exploring the impact of social

use on both body image concerns and internalizing symptoms might help establish the direction of the relationships linking these constructs in adolescence (e.g., by examining different influencing mechanisms using cross-lagged paths).

Another limitation relates to the use self-report to assess time spent on social media, which may have introduced potential biases due to social desirability effects and possible lack of awareness. In particular, findings indicate that when users are asked to self- report estimates of social media usage, reported data tend to differ from actual usage measures collected by directly tracking users’ online activity (e.g., Junco, 2013). Alternative data collection designs allowing for the assessment of adolescents’ actual use of social media (e.g., information collected by directly accessing par- ticipants’ social media profiles) would have provided us with a more reliable estimation of the frequency of participants’ online activity on social media. Furthermore, with respect to the assess- ment of the effect of time spent on social media, in the analyses we distinguished the impact of different patterns of use (no vs. mod- erate vs. frequent use) using an arbitrary, but widely used, cut-off (>2 h) for identifying frequent social media use. Although limited, this approach can be viewed as a simplified alternative to more complex regression approaches, such as the estimation of quadratic terms (i.e., polynomial regression), and the segmentation of main effects using breakpoint values (i.e., segmented regression), which can be useful when continuous independent variables are expected to exhibit non-linear relationships with outcomes. Finally, the present study was performed on a sample of Italian adolescents, thus results may not be fully generalizable to other national con- texts. However, it is worthy to note that Italy is similar to other Western industrialized countries both in terms of social media usage (Statista, 2017c) and standards of physical attractiveness (McElhone, Kearney, Giachetti, Zunft, & Martínez, 1999; Zaccagni, Masotti, Donati, Mazzoni, & Gualdi-Russo, 2014). In particular, attractiveness standards are different for males and females, and consist of a cultural ideal of thinness for women (Striegel-Moore & Franko, 2002), and leanness andmuscularity for men (Thompson& Cafri, 2007), with no substantial differences emerging between different geographic areas (i.e., North-vs. South-Italy, Jaeger et al., 2002). In light of this, we do expect to find a similar pattern of relationships among study variables in other Western countries, but further studies are needed to test this hypothesis.

5. Conclusions

Our study highlights the mediating effect of body image con- cerns on the relationship between use of image- and video-based social media (HVSM) and internalizing symptoms among adoles- cents. In doing this, our study fills a gap in the literature on the impact use of highly-visual social media is having on adolescents regarding body image and mental-health (Fardouly & Vartanian, 2016; RSPH, 2017). This is particularly important given the impressing percentages of social media use among adolescents, and in particular that of social-media platforms focusing on visual content. Given the scarcity of literature on these new social media platforms, it is important that further research is carried out into the consequences of their use on adolescents’ health (Brown & Tiggemann, 2016).

The study has practical implications for the development of interventions aimed at preventing or reducing the negative impact of social media on adolescent health: interventions should aim at promoting critical awareness of the idealized images that are pre- sented on online social media, as well as of the potential impact that exposure to these kind of images can have on adolescents’ body image, and consequently on their psychological well-being. This kind of intervention could be targeted not only at

D. Marengo et al. / Computers in Human Behavior 82 (2018) 63e6968

adolescents, but also at parents and education professionals in or- der to favor their involvement in monitoring and supporting ado- lescents in their social media life.

Finally, from a more general point of view, our study underlines that it is important for research to focus on the differences between the various kinds of social media. This not only allows us tomonitor the preferences of adolescents and adults in the wide range of existing social media platforms, but also enables us to make a more adequate study of the psychological and psychopathological phe- nomena associated with online activity, especially with social networks.

References

Aderka, I. M., Gutner, C. A., Lazarov, A., Hermesh, H., Hofmann, S. G., & Marom, S. (2014). Body image in social anxiety disorder, obsessiveecompulsive disorder, and panic disorder. Body image, 11(1), 51e56.

Allison, P. D. (2002). Missing data: Quantitative applications in the social sciences. British Journal of Mathematical and Statistical Psychology, 55(1), 193e196.

Brown, Z., & Tiggemann, M. (2016). Attractive celebrity and peer images on Insta- gram: Effect on women’s mood and body image. Body Image, 19, 37e43. https:// doi.org/10.1016/j.bodyim.2016.08.007.

Bucchianeri, M. M., Fernandes, N., Loth, K., Hannan, P. J., Eisenberg, M. E., & Neu- mark-Sztainer, D. (2016). Body dissatisfaction: Do associations with disordered eating and psychological well-being differ across race/ethnicity in adolescent girls and boys? Cultural Diversity and Ethnic Minority Psychology, 22(1), 137e146. https://doi.org/10.1037/cdp0000036.

Colunga-Rodríguez, C., Orozco-Solis, M. G., Flores-Villavicencio, M. E., de-la-Roca- Chiapas, J. M., G�omez-Martínez, R., Mercado, A., et al. (2016). Body image perception and internalization problems indicators in mexican adolescents. Psychology, 07(13), 1671e1681. https://doi.org/10.4236/psych.2016.713158.

Cooper, P. J., Taylor, M. J., Cooper, Z., & Fairbum, C. G. (1987). The development and validation of the body shape questionnaire. International Journal of Eating Dis- orders, 6(4), 485e494. https://doi.org/10.1002/1098-108X(198707)6:4<485:: AID-EAT2260060405>3.0.CO;2-O.

Fabris, M. A., Longobardi, C., Prino, L. E., & Settanni, M. (2017). Attachment style and risk of muscle dysmorphia in a sample of male bodybuilders. Psychology of Men & Masculinity. https://doi.org/10.1037/men0000096. Advance online publication.

Fardouly, J., Diedrichs, P. C., Vartanian, L. R., & Halliwell, E. (2015a). Social com- parisons on social media: The impact of Facebook on young women’s body image concerns and mood. Body Image, 13, 38e45. https://doi.org/10.1016/ j.bodyim.2014.12.002.

Fardouly, J., Diedrichs, P. C., Vartanian, L. R., & Halliwell, E. (2015b). The mediating role of appearance comparisons in the relationship between media usage and self-objectification in young women. Psychology of Women Quarterly, 39(4), 447e457. https://doi.org/10.1177/0361684315581841.

Fardouly, J., & Vartanian, L. R. (2016). Social media and body image Concerns: Current research and future directions. Current Opinion in Psychology, 9, 1e5. https://doi.org/10.1016/j.copsyc.2015.09.005.

Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117e140. https://doi.org/10.1177/001872675400700202.

Frith, E. (2017). Social media and children’s mental health: A review of the evidence. Education Policy Institute. Available at: https://epi.org.uk/report/social-media- and-childrens-mental-health-a-review-of-the-evidence/.

Gerson, J., Plagnol, A. C., & Corr, P. J. (2016). Subjective well-being and social media use: Do personality traits moderate the impact of social comparison on Face- book? Computers in Human Behavior, 63, 813e822. https://doi.org/10.1016/ j.chb.2016.06.023.

Goodman, R., Meltzer, H., & Bailey, V. (1998). The strengths and difficulties ques- tionnaire: A pilot study on the validity of the self-report version. European Child & Adolescent Psychiatry, 7(3), 125e130.

Hargreaves, D. A., & Tiggemann, M. (2004). Idealized media images and adolescent body image: “comparing” boys and girls. Body Image, 1(4), 351e361. https:// doi.org/10.1016/j.bodyim.2004.10.002.

Hendrickse, J., Arpan, L. M., Clayton, R. B., & Ridgway, J. L. (2017). Instagram and college women’s body image: Investigating the roles of appearance-related comparisons and intrasexual competition. Computers in Human Behavior, 74, 92e100. https://doi.org/10.1016/j.chb.2017.04.027.

Holsen, I., Kraft, P., & Røysamb, E. (2001). The relationship between body image and depressed mood in adolescence: A 5-year longitudinal panel study. Journal of Health Psychology, 6(6), 613e627.

Jaeger, B., Ruggiero, G. M., Edlund, B., Gomez-Perretta, C., Lang, F., Mohammadkhani, P., et al. (2002). Body dissatisfaction and its interrelations with other risk factors for bulimia nervosa in 12 countries. Psychotherapy and psychosomatics, 71(1), 54e61.

Jamshidian, M., & Jalal, S. (2010). Tests of homoscedasticity, normality, and missing completely at random for incomplete multivariate data. Psychometrika, 75, 649e674.

Jamshidian, M., Jalal, S. J., & Jansen, C. (2014). Missmech: an R package for testing

homoscedasticity, multivariate normality, and missing completely at random (mcar). Journal of Statistical software, 56(6).

Jelenchick, L. A., Eickhoff, J. C., &Moreno, M. A. (2013). “Facebook Depression?” social networking site use and depression in older adolescents. Journal of Adolescent Health, 52(1), 128e130. https://doi.org/10.1016/j.jadohealth.2012.05.008.

Junco, R. (2013). Comparing actual and self-reported measures of Facebook use. Computers in Human Behavior, 29(3), 626e631. https://doi.org/10.1016/ j.chb.2012.11.007.

Kim, J. W., & Chock, T. M. (2015). Body image 2.0: Associations between social grooming on Facebook and body image concerns. Computers in Human Behavior, 48, 331e339. https://doi.org/10.1016/j.chb.2015.01.009.

Lang, C., & Barton, H. (2015). Just untag it. Computers in Human Behavior, 43(C), 147e155. https://doi.org/10.1016/j.chb.2014.10.051.

Lenhart, A. (2015). Pew research center, April 2015. Teen, social media and technology overview. Available http://www.pewinternet.org/2015/04/09/teens-social- media-technology-2015/(April).

Liu, Q.-Q., Zhou, Z.-K., Yang, X.-J., Niu, G.-F., Tian, Y., & Fan, C.-Y. (2017). Upward social comparison on social network sites and depressive symptoms: A moderated mediation model of self-esteem and optimism. Personality and In- dividual Differences, 113, 223e228. https://doi.org/10.1016/j.paid.2017.03.037.

Longobardi, C., Prino, L. E., Fabris,M. A., & Settanni,M. (2017).Muscle dysmorphia and psychopathology: Findings from an Italian sample of male bodybuilders. Psy- chiatry Research, 256, 231e236. https://doi.org/10.1016/j.psychres.2017.06.065.

Manago, A. M., Graham, M. B., Greenfield, P. M., & Salimkhan, G. (2008). Self-pre- sentation and gender on MySpace. Journal of Applied Developmental Psychology, 29(6), 446e458. https://doi.org/10.1016/j.appdev.2008.07.001.

Manago, A. M., Ward, L. M., Lemm, K. M., Reed, L., & Seabrook, R. (2015). Facebook involvement, objectified body consciousness, body shame, and sexual asser- tiveness in college women and men. Sex Roles, 72(1e2), 1e14. https://doi.org/ 10.1007/s11199-014-0441-1.

Marzocchi, G. M., Capron, C., Di Pietro, M., Duran Tauleria, E., Duyme, M., Frigerio, A.,…Th�erond, C. (2004). The use of the strengths and difficulties questionnaire (SDQ) in southern European countries. European Child & Adolescent Psychiatry, 13(Suppl 2), II40eII46. https://doi.org/10.1007/s00787- 004-2007-1.

McElhone, S., Kearney, J. M., Giachetti, I., Zunft, H. J. F., & Martínez, J. A. (1999). Body image perception in relation to recent weight changes and strategies for weight loss in a nationally representative sample in the European Union. Public Health Nutrition, 2(1a), 143e151.

Meier, E. P., & Gray, J. (2014). Facebook photo activity associated with body image disturbance in adolescent girls. Cyberpsychology, Behavior and Social Networking, 17(4), 199e206. https://doi.org/10.1089/cyber.2013.0305.

Moreno, M. A., Jelenchick, L. A., Egan, K. G., Cox, E., Young, H., Gannon, K. E., et al. (2011). Feeling bad on Facebook: Depression disclosures by college students on a social networking site. Depression and Anxiety, 28(6), 447e455. https://doi.org/ 10.1002/da.20805.

Morin, A. J., Maïano, C., Scalas, L. F., Janosz, M., & Litalien, D. (2017). Adolescents’ body image trajectories: A further test of the self-equilibrium hypothesis. Developmental Psychology, 53(8), 1501.

Natsuaki, M. N., Biehl, M. C., & Ge, X. (2009). Trajectories of depressed mood from early adolescence to young adulthood: The effects of pubertal timing and adolescent dating. Journal of Research on Adolescence, 19(1), 47e74.

O’Dea, J. A., & Caputi, P. (2001). Association between socioeconomic status, weight, age and gender, and the body image and weight control practices of 6-to 19- year-old children and adolescents. Health education research, 16(5), 521e532.

O’Keeffe, G. S., Clarke-Pearson, K., & Media, C. (2011). The impact of social media on children, adolescents, and families. Pediatrics, 127(4), 800e804. https://doi.org/ 10.1542/peds.2011-0054.

Paxton, S. J., Neumark-Sztainer, D., Hannan, P. J., & Eisenberg, M. E. (2006). Body dissatisfaction prospectively predicts depressive mood and low self-esteem in adolescent girls and boys. Journal of Clinical Child and Adolescent Psychology, 35(4), 539e549.

Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36(4), 717e731. https://doi.org/10.3758/BF03206553.

Rawana, J. S., & Morgan, A. S. (2014). Trajectories of depressive symptoms from adolescence to young adulthood: The role of self-esteem and body-related predictors. Journal of Youth and Adolescence, 43(4), 597e611. https://doi.org/ 10.1007/s10964-013-9995-4.

Rosenthal, S. R., Buka, S. L., Marshall, B. D. L., Carey, K. B., & Clark, M. A. (2016). Negative experiences on Facebook and depressive symptoms among young adults. The Journal of Adolescent Health: Official Publication of the Society for Adolescent Medicine, 59(5), 510e516. https://doi.org/10.1016/ j.jadohealth.2016.06.023.

RSPH. (2017). Social media and young people’s mental health and wellbeing. Retrieved September 4, 2017, from http://www.rsph.org.uk/our-work/policy/social- media-and-young-people-s-mental-health-and-wellbeing.html.

Sampasa-Kanyinga, H., & Lewis, R. F. (2015). Frequent use of social networking sites is associated with poor psychological functioning among children and adoles- cents. Cyberpsychology, Behavior and Social Networking, 18(7), 380e385. https:// doi.org/10.1089/cyber.2015.0055.

Settanni, M., & Marengo, D. (2015). Sharing feelings online: Studying emotional well-being via automated text analysis of Facebook posts. Frontiers in psychol- ogy, 6.

Simoncic, T. E., Kuhlman, K. R., Vargas, I., Houchins, S., & Lopez-Duran, N. L. (2014).

D. Marengo et al. / Computers in Human Behavior 82 (2018) 63e69 69

Facebook use and depressive symptomatology: Investigating the role of neuroticism and extraversion in youth. Computers in Human Behavior, 40, 1e5. https://doi.org/10.1016/j.chb.2014.07.039.

Smith, A. R., Hames, J. L., & Joiner, T. E. (2013). Status update: Maladaptive Facebook usage predicts increases in body dissatisfaction and bulimic symptoms. Journal of Affective Disorders, 149(1e3), 235e240. https://doi.org/10.1016/ j.jad.2013.01.032.

Statista. (2017a). Number of monthly active Facebook users worldwide as of 3rd quarter 2017 (in millions). Retrieved from: https://www.statista.com/statistics/ 264810/number-of-monthly-active-facebook-users-worldwide/.

Statista. (2017b). Snapchat crowned number 1 by American teens. Retrieved September 4, 2017, from https://www.statista.com/chart/4823/teenagers- favorite-social-networks/.

Statista. (2017c). Active social network penetration in selected countries as of January 2017. Retrieved from: https://www.statista.com/statistics/282846/regular- social-networking-usage-penetration-worldwide-by-country/.

Stice, E., & Bearman, S. K. (2001). Body-image and eating disturbances prospectively predict increases in depressive symptoms in adolescent girls: A growth curve analysis. Developmental Psychology, 37(5), 597e607.

Striegel-Moore, R. H., & Franko, D. L. (2002). Body image issues among girls and women. Body Image: A handbook of Theory, Research, and Clinical Practice, 183e191.

Thompson, J. K., & Cafri, G. (2007). The muscular ideal: Psychological, social, and medical perspectives. Washington, DC: American Psychological Association.

Tiggemann, M., & Miller, J. (2010). The internet and adolescent girls’ weight satis- faction and drive for thinness. Sex Roles, 63(1e2), 79e90. https://doi.org/ 10.1007/s11199-010-9789-z.

Tiggemann, M., & Slater, A. (2013). NetGirls: The internet, Facebook, and body image concern in adolescent girls. The International Journal of Eating Disorders, 46(6),

630e633. https://doi.org/10.1002/eat.22141. Tiggemann, M., & Slater, A. (2014). NetTweens: The Internet and body image con-

cerns in preteenage girls. The Journal of Early Adolescence, 34(5), 606e620. Toma, C. L., & Hancock, J. T. (2010). Looks and lies: the role of physical attractiveness

in online dating self-presentation and deception. Communication Research, 37(3), 335e351. https://doi.org/10.1177/0093650209356437.

Vandenbosch, L., & Eggermont, S. (2012). Understanding sexual objectification: A comprehensive approach toward media exposure and girls’ internalization of beauty ideals, self-objectification, and body surveillance. Journal of Communi- cation, 62(5), 869e887. https://doi.org/10.1111/j.1460-2466.2012.01667.x.

Vogel, E. A., & Rose, J. P. (2016). Self-reflection and interpersonal connection: Making the most of self-presentation on social media. Translational Issues in Psychological Science, 2(3), 294e302. https://doi.org/10.1037/tps0000076.

Walker, M., Thornton, L., De Choudhury, M., Teevan, J., Bulik, C. M., Levinson, C. A., et al. (2015). Facebook use and disordered eating in college-aged women. The Journal of Adolescent Health : Official Publication of the Society for Adolescent Medicine, 57(2), 157e163. https://doi.org/10.1016/j.jadohealth.2015.04.026.

Yang, C.-C. (2016). Instagram use, loneliness, and social comparison Orientation: Interact and browse on social media, but Don’t compare. Cyberpsychology, Behavior and Social Networking, 19(12), 703e708. https://doi.org/10.1089/ cyber.2016.0201.

Zaccagni, L., Masotti, S., Donati, R., Mazzoni, G., & Gualdi-Russo, E. (2014). Body image and weight perceptions in relation to actual measurements by means of a new index and level of physical activity in Italian university students. Journal of Translational Medicine, 12(1), 42.

Zhao, S., Grasmuck, S., & Martin, J. (2008). Identity construction on Facebook: Digital empowerment in anchored relationships. Computers in Human Behavior, 24(5), 1816e1836. https://doi.org/10.1016/j.chb.2008.02.012.