Table 2.
Data charting form including author and year of publication, objectives of the study, method used, country where the study was conducted, depression scale used, number of participants, participant age, results and main social media focus.
| Author and year | Objectives | Methods | Country | N | Ages (years) | Results | Main social media focus |
|---|---|---|---|---|---|---|---|
| Akkin Gurbiiz et al„ 2017 | Evaluate the SNS habits of depressed adolescents and the relationship between depression and disclosure on SNSs | Cross-sectional | Turkey | 53 (cases) and 55 (control students) | 13–18 | The time spent on SNSs increased with depressive symptoms | Frequency of use |
| Banjanin et al., 2015 | Investigate the potential relationship between internet addiction and depression in adolescents | Cross-sectional | Belgrade, Serbia | 336 (65.5% female) | 18 | No relationship between time spent in SNS sites and depression and between depression and SNS activities (i.e.: number of friends) | Problematic use |
| Banyai et al., 2017 | Test the psychometric properties of the BSMAS and assess the prevalence of problematic social media use in Hungarian adolescents | Cross-sectional | Hungary | 6664 (49.06% female) | 15–22 (M = 16.62, SD 0.96) | The class at risk of problematic social media use was more likely to be female, have a higher frequency of use, and have lower selfesteem and higher level of depressive symptoms | Problematic use |
| Barry et al., 2017 | Investigate adolescent and parent reports of adolescent social media use and relation to adolescent psychosocial adjustment | Cross-sectional | USA | 226 (113 parent- adolescent dyads) (51.3% female) | 14–17 (M = 5.27, SD = 1.02) | Number of social media accounts and frequency of checking social media were correlated with depressive symptoms. Parental monitoring of social media was not associated with any of psychosocial adjustment variables | Frequency of use |
| Cole et al., 2016 | Evaluate if cybervictimization is prospectively related to negative self cognitions and depressive symptoms beyond other types of victimization | Longitudinal (2 waves of data collection over a 6-week period) Cross-sectional | USA | 827 (55.1 % female) | 8–13 W = 10.90, SD = 1.18) | Victimization was correlated with negative cognition and depressive symptoms. Cybervictimization predicted depressive symptoms | Cybervictimization |
| Coyne et al., 2014 | Examine association between parent-child use of SNS and feelings of connection and other adolescent outcomes | USA | 491 families | 12–17 (M = 14.4, SD = 1.07), (53% female) | Social networking with parents was associated with increased connection between parents and adolescents. Feelings of connection mediated the relationship between social networking with parents and depression. Adolescent social networking use without parents was associated with depression | Parental involvement | |
| Coyne et al., 2018 | Examine differential patterns of social media use over time and investigate predictors and outcomes of use patterns. | Cross-sectional | USA (Pacific North-west) | 681 families (457 adolescents) (53% female) | 11–14 at baseline (M = 13.5) | Moderate users had higher levels of self-regulation and lower levels of overall media use vs the other 2 classes (peak users and increasers), which had higher levels of depression and physical aggression | Frequency of use |
| Critselis et al., 2014 | Assess the determinants and psychosocial correlates associated with internet addictive behaviours among adolescents | Cross-sectional | Nicosia, Cyprus | 805 | 13–18 | Adolescent BIU was associated with abnormal peer and conduct problems and elevated hyperactivity and emotional symptoms. AIU among adolescents was associated with lower emotional and psychosocial adjustment | Frequency of use |
| Duarte et al., 2018 | Further elucidate which adolescents are at greatest risk for the clinically significant negative mental health outcomes of cyberbullying. | Cross-sectional | USA | 1031 | 13–17 [M = 14.9; SD = 1.39) | Sexual orientation was the only demographic factor correlated with cyberbullying and mental health symptoms. Increased used of SNS correlated with cyberbullying | Cybervictimization |
| Dunlop et al., 2011 | Examine exposure to sources of suicide stories, how knowledge of suicidal behaviour spread among friends and acquaintances, and the relationships between exposure to sources of suicide reports and suicide ideation | Longitudinal | USA | 719 | 14–24 | While friends and family or newspapers remained strong sources of suicide stories, there was considerable exposure to such stories online and especially in SNS. Online discussion forums (but not SNS) were associated with increased suicidal ideation | Suicide contagion |
| Fardouly et al., 2018 | Examine the association between parental control over the child's time spent on social media, number of appearance comparisons, appearance satisfaction, depressive symptoms and life satisfaction. | Cross-sectional | Sydney, Australia | 284 preadolescents (53.2% female) and 1 parent (96.1% mothers) | M = 11.2 (SD = 0.56) | Parental control over preadolescent time spent on social media was not associated with depressive symptoms. Lower frequency of social media appearance comparison was associated with higher preadolescent appearance and life satisfaction, and lower depressive symptoms | Parental involvement |
| Frison & Eggermont, 2015 | Examine relationships among daily stress (i.e., school- and family-related stress), social support-seeking, perceived social support through Facebook and depressed mood among adolescents | Cross-sectional | Flanders, Belgium | 910 (51.9% female) | 13–20 (M = 15.44; SD = 1.71) | Daily stress positively predicted adolescents' seeking of social support through Facebook. When social support was sought on Facebook and subsequently received, it decreased adolescents' depressed mood, but if not received, it increased depressed mood | Social support |
| Frison & Eggermont, 2016 | Provide a deeper understanding of the relationships between different types of Facebook use, perceived online social support, and boys' and girls' depressed mood | Cross-sectional (2-step sampling method) | Flanders, Belgium | 910 (51.9% female) | 13–20 (M = 15.44; SD = 1.71) | Harmful impact of Facebook use occurred among girls who passively use Facebook and among boys who actively use Facebook in a public setting. Girls who actively use Facebook in a public or private setting and subsequently receive online social support, benefit from using Facebook | Characteristics of SNS use |
| Frison & Eggermont, 2017 | Examine relationships between different types of Instagram use (i.e., browsing, posting, and liking) and adolescents' depressed mood. | Longitudinal | Flanders, Belgium | T1 = 671; T2 = 622 at T2, 244 at both time points | 12–19 W= 14.96; SD = 1.29) | Instagram browsing (but not posting or liking) at Time 1 positively predicted adolescents' depressed mood at Time 2. Depressed mood at Time 1 positively predicted Instagram posting (but not browsing and liking) at Time 2 | Characteristics of SNS use |
| Frison et al., 2016 | Address critical gaps in our understanding of online victimization and adolescents' depressive symptoms and life satisfaction | Longitudinal (2-wave panel study; 6-month interval) | Flanders, Belgium | 1621 (48 % female) | 12–19 (M = 14.8; SD = 1.41) | Facebook peer victimization predicted decreases in life satisfaction and vice versa. Depressive symptoms were a risk factor for peer victimization on Facebook. In addition, support from friends was protective from the harmful outcomes of peer victimization on Facebook | Cybervictimization |
| Houghton et al., 2018 | Identify the trajectories of depressive symptoms in adolescents and consider possible associations between trajectory classes and screen use time. Evaluate possible associations between screen use and subsequent depressive symptomatology and vice versa | Prospective cohort (6 waves of data collection) | USA | 1749 (47% female) | 10–17 | Three trajectories of depressive symptoms with differences on screen use (low-stable, high- decreasing, and low-increasing) were identified. Small, positive associations were evident between depressive symptoms and later screen use, and viceversa. Yet, there was no consistent support for a longitudinal association | Frequency of use |
| Isarabhakdi & Pewnil, 2016 | Assess the level of engagement in family and peer activities and Internet use among in-school youth and the effect of engagement in family and friend activities, as well as Internet use on mental well-being | Cross-sectional | Thailand | 1074 | 15–19 | Engagement of family activities improved mental health, and decreased depression and stress among youth. Engagement with peers had a significant effect on mental health and depression, but not on stress. Internet usage had a very low effect on mental well-being | Social support |
| Kircaburun et al., 2018 | Understand how CBP and PSMU are associated with each other and to gender, age, depression, and self esteem among high school students using a structural equation model. | Cross-sectional | Turkey | 1143 students in study 1 [Study 2 with adults, not included] | 14–21 (48% female; M = 16.20, SD = 1.03) | Depression directly predicted PSMU and indirectly predicted cyberbullying perpetration, although the associations were weak | Problematic social media use |
| Lee et al., 2017 | Address weaknesses in the social cognitive model by using an extended version to understand both external and personal antecedents of adolescents' SNSs usage | Cross-sectional | USA | 3753 | 13–21 (M= 14.73) | Depression was positively associated with self-reactive outcome expectation and deficient selfregulation. Positive relationship with father (not mother) is negatively associated with adolescents' dependence on social media for identity formation. In addition to depression, loneliness was included as a psychosocial antecedent factor of high social media usage | Frequency of use |
| Li et al., 2017 | Assess the mediating effects of insomnia on the associations between problematic Internet use, including IA and OSNA, and depression among adolescents | Cross-sectional. | China | 1015 (41.2% female) | 7th—9th graders | IA and OSNA were both associated with depression, with a stronger association for OSNA. Insomnia mediated the associations between IA/OSNA and depression | Problematic use |
| Marengo et al., 2018 | Evaluate the association between social media use, and in particular that of HVSM, with body image concerns and internalizing symptoms in adolescents | Cross-sectional | Northern Italy | 523 (53.5% female) | M = 14.82 (SD = 1.52) | Frequent use of HVSM positively predicted internalizing symptoms and body image concerns, while moderate use was not a significant predictor. Body image concerns mediated this association. Females had higher body image concerns and internalizing problems | Frequency of use |
| Morin-Majora et al., 2016 | Explore the associations between Facebook behaviours (use frequency, network size, self-presentation and peer interaction) and basal levels of cortisol among adolescent boys and girls | Cross-sectional | Montreal, Canada | 94 adolescents (53.1% female) | 12–17 [M = 14.2, SD = 1.7) | There was a positive association between Cortisol systemic output and number of Facebook friends but a negative association with Facebook peer interaction. There were no FB associations with depressive symptoms and HPA axis functioning | Characteristics of SNS use |
| Blomfield-Neira & Barber, 2014 | Investigate whether there was a relationship between adolescents' use of SNSs and their social self-concept, self-esteem, and depressed mood. | Cross-sectional | Western Australia | 1819 students (55% female) | 13–17 [M = 14.6, SD = 1.05) | There was no significant link between social media frequency and depressed mood but social media investment did predict depressed mood. There were differences by gender in the association between having social media and indicators of adjustment | Frequency of use |
| Nesi & Prinstein, 2015 | Examine specific technology-based behaviours (social comparison and interpersonal feedback-seeking) that may interact with offline individual characteristics to predict concurrent depressive symptoms among adolescents | Longitudinal (levels of depressive symptoms at baseline, and 1 year later) | USA | 619 students (M = 14.6; 57 % female) completed both self-report questionnaires | 12–16 [M = 14.6; (57.3% female) | Technology-based social comparison and feedback-seeking were associated with depressive symptoms, with a strong association among females and adolescents low in popularity. Associations were found beyond the effects of frequency of technology use, offline excessive reassurance-seeking and history of depressive symptoms | Social comparisons |
| Niu et al., 2018 | To investigate the association between Chinese adolescents' SNS (Qzone) use and depression, the mediating role of negative social comparison and the moderating role of self-esteem | Cross-sectional | China | 764 (46.8% female) | 12–18 W= 14.23, SD = 1.75) | Negative social comparison mediated the relationship between Qzone use and depression. There were no significant direct effects of Qzone use on depression. Qzone use was less strongly associated with negative social comparison at higher levels of self-esteem | Social comparisons |
| Oberst et al., 2017 | Analyse the link between psychopathological aspects and negative consequences of smartphone use, including role of FOMO and the intensity of social network use | Cross-sectional. | Latin American countries | 1468 (74.3% females) | 16–18 W= 16.59, SD = 0.62) | Depression had a direct effect on CERM. The effect of depression on negative consequences was mediated by FOMO. SNI mediated the association between FOMO and CERM. Being depressed triggered higher SNS involvement in girls | Frequency of use |
| Ophir et al., 2019 | Examine the predictive validity of explicit references to personal distress in adolescents' Facebook postings as well as non-explicit Facebook activity features | Cross-sectional | USA | Study 1: 86 (51.2% female). Study 2: 162 (51.3% female) | Study 1: 13–18 (/W = 15.98, SD = 1.3). Study 2: adolescents (not specified) | While rare, explicit distress references predicted depression among adolescents. There were no additional differences in Facebook activity behaviours that could distinguish between depressive and non-depressive adolescents. Adolescents appeared to publish significantly less verbal content than adults' users of social media | Disclosure of symptoms |
| Pantic et al., 2012 | Investigate the relationship between social networking and depression indicators in adolescent population | Cross-sectional | Pozarevac, Central Serbia | 160 | M = 18.02 (SD = 0.29) | Positive correlation was found between depression and time spent on social networking but not between TV viewing and depression. No statistically significant difference was noted between males and females in TV viewing, social networking, sleep duration and depression | Frequency of use |
| Radovic et al., 2017 | Examine descriptions of social media use among 23 adolescents who were diagnosed with depression to explore how social media use may influence and be influenced by psychological distress | Qualitative study (30–60 min semistructured interviews) | USA | 23 (78.2% female) | 13–20, (M = 16, SD = 2) | Adolescents described both positive (searching for information and social connection) and negative use (risky behaviours, cyberbullying, and making self- denigrating comparisons with others). There were 3 types of use including 'oversharing' (frequent updates or too much personal information), 'stressed posting' (sharing negative updates), and encountering ťriggering posts' | Characteristics of use |
| Rodriguez Puentes & Parra, 2014 | Explore the relationship between the amount of time spent in social networking and the presence of internalizing and externalizing behaviour problems in adolescents | Experimental or quasi-experimental study | Bogota, Colombia | 96 (52.2% female) | 11–15 (M= 11.98, SD = 0.68) | Greater time spent on social networks was associated with externalizing disorders such as aggressive conduct, rule breaking and attention deficits. There was no association with depression | Frequency of use |
| Romer et al., 2013 | Determine the effects of both older and newer media use on academic, social, and mental health outcomes in adolescents and young adults | Cross-sectional | USA | 719 (51% female) | 14–22 | Greater Internet use and video game playing were associated with recent depression. Information users had higher grades, participated in clubs more often, and were lowest in depression. Moderate internet use was best for healthy development | Frequency of use |
| Salmela-Aro et al„ 2017 | Examine the longitudinal paths between excessive internet use, depressive symptoms, school burnout and engagement. Specifically, whether excessive internet use leads to both depressive symptoms and/or school- related burnout, and vice versa | 2 cross-sectional studies; 760 students at Time 1 and 1403 and at Time 2 | Helsinki, Finland | Study 1: 1702 elementary school students; Study 2: 1636 high school students | Study 1: 12–14; Study 2: 16 –18 | Emotional engagement, school burnout and depressive symptoms each made a unique contribution to adolescent excessive internet use. Furthermore, students who burn out at school are at risk for excessive internet use and depressive symptoms | Frequency of use |
| Sampasa-Kanyinga & Hamilton, 2015 | Examine the link between the use of social networking sites and psychological distress, suicidal ideation and suicide attempts, and test the mediating role of cyberbullying victimization on these associations in adolescents | Cross-sectional | Ottawa, Canada | 5126 (48% females) | 11–20 (M = 15.2; SD = 1.9) | Use of social media was associated with psychological distress, suicidal ideation and attempts. Cyberbullying victimization fully mediated the association between SNSs use and psychological distress and suicidal attempts; and partially mediated the association between SNSs use and suicidal ideation | Cybervictimization |
| Sampasa-Kanyinga & Lewis, 2015 | Examine the association between time spent on social media and unmet need for mental health support, self- rated mental health, psychological distress and suicidal ideation in a sample of middle and high school children | Cross-sectional | Ottawa, Canada | 753 (49% female) | M = 15.2 (SD = 0.2) | Those reporting unmet need for mental health support more likely reported using social media for >2 h a day. Use of social media for >2 h a day was associated with fair or poor self-rating of mental health, higher levels of psychological distress, and suicidal ideation | Frequency of use |
| Szwedo et al., 2011 | Determine if youth who experience negative interactions with their mothers as teenagers later prefer online communication, engage in more negative peer interactions on SNS, and have greater likelihood of forming a new friendship with someone they met online | Cross-sectional (Participants drawn from a larger longitudinal study) | USA (sub-urban and urban Southeastern) | 138 (89 had a SNS webpage on Facebook or MySpace; 63 granted access permission) | Time 1: M = 13.23 (SD = 0.66) Time 2: M = 20.53, (SD = 0.97) | Adolescents' depressive symptoms at baseline were positively associated with later preference for online communication. Poor adolescent relationships with mother predicted preference for online communication, likelihood of forming friendships with people met online, and poorer quality of online relationships at an older age | Parental involvement |
| Tseng & Yang, 2015 | Investigate relationships of Internet use, web communication, and sources of social support with adolescent SITBs | Cross-sectional (2-phase sampling design) | Changhua and Nantou counties, Taiwan | 2494 | 13–18 | Web communication in adolescent boys was a risk factor for SITBs. Boys with higher levels of depressive symptoms had lower ability to communicate with others on the Internet due to more impaired functioning. Frequency of use was negatively associated with depression in boys | Suicide contagion |
| Tsitsika, Janikian, et al., 2014 | Explore the prevalence of IAB among adolescents in seven European countries (Greece, Spain, Poland, Germany, Romania, Netherlands, and Iceland) | Cross-sectional | European countries | 13,284 | 14–17 (M = 15.8, SD = 0.7) | The prevalence of DIB was higher among adolescents who spent >2 h per day on SNS. DIB significantly predicted greater emotional and behavioural problems | Problematic use |
| Tsitsika, Tzavela, et al., 2014 | Investigate associations between heavier SNS use, and adolescent competencies and internalizing problems | Cross-sectional | European countries | 10,930 | 14–17 | Heavier SNS use was associated with more offline social competence among older adolescents, but more internalizing problems, and lower academic performance and activities scores, especially among younger adolescents | Frequency of use |
| Twenge et al., 2018 | Determine if the prevalence of depressive symptoms and suicide- related outcomes has increased in U.S. adolescents in recent years and whether these birth cohort trends differ by socio-demographic characteristics and examine possible causes behind trends, primarily focussing on shifts in adolescents' use of leisure time | Cross-sectional | USA | 388,275; YRBSS (N = 118,545) | 13–18 | Adolescents who spent more time on screen activities were more likely to have high depressive symptoms or at least one suicide- related outcome. Social media only had a significant effect on depressive symptoms among those low in in-person social interaction, not among those high in in-person social interaction. Over the same period that depression and suicide outcomes increased, screen activities increased and non-screen activities decreased | Frequency of use |
| Van Rooij et al., 2017 | Explore abandoning a unified approach to problematic 'Internet use' by splitting the concept into more specific application level measurement (gaming, internet use and Social media use) | Cross-sectional | Netherlands | 3945 | 12–15 | PIU was associated with depression and both gaming and social media activities. Specific PIU measures for social media use and gaming differed, with male gender more associated with on and offline gaming. Both problematic social media use and gaming were associated with depression | Problematic use |
| Wang et al., 2018 | Test the mechanisms underlying the association between SNS addiction and depression in adolescents, whether rumination plays a mediating role, and whether self-esteem buffers the mediating effect of rumination | Cross-sectional | China | 365 | 14–18; M = 15.96 (SD = 0.69) | Social Media addiction adolescent depression was positively associated. This association was mediated by rumination. The effect of SNS on adolescent depression was stronger the lower the self-esteem | Problematic use |
| Woods & Scott, 2016 | Explore the association between social Cross-sectional media use (including specific nighttime use and emotional investment in SNS) with sleep quality, anxiety, self-esteem and depression | Cross-sectional | Scotland | 467 | 11–17 | Greater general and nighttime- specific SNS use as well as social media investment were all poorer sleep quality and anxiety and depression. After controlling for depression, anxiety and self-esteem, nighttime-specific SNS use still predicted poor sleep | Frequency of use |
AIU = Addictive internet Use; BIU = Borderline Addictive Internet Use; BSMAS = Bergen Social Media Addiction Scale; BIU = Borderline addictive internet use; CBP = Cyberbullying Perpetration; CERM = Cuestionario de Experiencias Relacionadas con el móvil (Questionnaire of Experiences Related to the cellphone); DIB = Dysfunctional Internet Behaviour; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders (4th edition, Text Revision); FOMO = Fear of Missing Out; HVSM = Highly Visual Social Media; SNI = Intensity of social network use; IA = Internet Addiction; IAB = Internet Addictive Behaviour; OSNA = Online social networking addiction; PSMU = Problematic Social Media Use; RADS-2 = Reynolds Adolescent Depression Scale - Version 2; SITBs = self-injurious thoughts and behaviours; SNS = social networking sites.