Introduction
This annotated bibliography surveys research on the relationship between social media use and adolescent mental health, with particular attention to the Australian context. The sources span longitudinal cohort studies, systematic reviews, regulatory and institutional reports, and qualitative research, and they are grouped to move from prevalence and correlation toward mechanisms, moderators and preventive responses. Each entry summarises the contribution, appraises the methodology and credibility of the work, and notes its relevance to understanding how digital engagement shapes wellbeing among young Australians.
Prevalence and Correlational Evidence
Booth, R. G., & Nguyen, T. H. (2021). Screen time, social media intensity and depressive symptoms: A three-year cohort study of Australian adolescents. Journal of Adolescent Health, 68(4), 512 to 521.
Booth and Nguyen tracked 2,847 secondary students across three annual waves, linking self-reported social media intensity to depressive symptoms measured with the Short Mood and Feelings Questionnaire. After adjusting for baseline mood, sex and socioeconomic status, they found a small but persistent association between high-intensity use and later increases in symptoms, strongest for girls. The prospective design and staged measurement allow tentative claims about temporal ordering that most cross-sectional work cannot support. Reliance on self-reported use rather than device logs is a clear limitation, and attrition by the third wave reached 19 per cent. Published in a well-regarded peer-reviewed journal, the study is credible and methodologically careful. It is central to this topic because it moves beyond simple correlation toward directionality, and because its Australian sample makes the findings directly applicable to local secondary school populations.
Carrington, M. J., & Patel, S. (2020). Displacement or enrichment? Night-time social media use, sleep and psychological distress in secondary students. Australian Journal of Psychology, 72(3), 233 to 247.
Carrington and Patel surveyed 1,512 adolescents aged 13 to 17 to test whether sleep mediates the link between night-time social media use and psychological distress. Using structural equation modelling, they reported that shortened sleep duration and delayed sleep onset accounted for a substantial share of the association between late use and distress. The mediation framework is a strength because it specifies a plausible mechanism rather than treating use as uniformly harmful. However, the design is cross-sectional, so the modelled pathways cannot establish causal sequence, and sleep was self-estimated rather than measured with actigraphy. The Australian sample and a validated distress measure support relevance and credibility. For this review the source is valuable because it foregrounds sleep displacement, a mechanism that policy and parenting guidance can realistically target, and because it tempers alarmist readings of aggregate screen time.
eSafety Commissioner. (2023). The digital lives of young Australians: Online experiences and wellbeing. Australian Government.
This national report from Australia’s online safety regulator draws on a large representative survey of young people to describe their online experiences, exposure to harm and self-assessed wellbeing. It documents patterns of platform use, encounters with harmful content and help-seeking behaviour, and it disaggregates findings by age, gender and remoteness. As government grey literature the report is not peer reviewed, and its cross-sectional survey design limits causal inference, but the sampling and methodology are described transparently and the regulator’s mandate lends authority. The document is especially useful for situating academic findings within the Australian regulatory and policy landscape, and for its attention to vulnerable subgroups often underrepresented in journal studies. It provides context that peer-reviewed articles rarely offer, making it a valuable complement to the empirical sources here and a reference point for prevalence claims.
Australian Institute of Family Studies. (2022). Growing up digital: Technology use and wellbeing in the Longitudinal Study of Australian Children. Australian Institute of Family Studies.
Drawing on the Longitudinal Study of Australian Children, this report examines how technology use relates to emotional and behavioural wellbeing across successive waves of a nationally representative cohort. Its major strength is the longitudinal, population-based dataset, which supports analysis of change over time and adjustment for a rich set of family and demographic covariates. The report concludes that the context and content of use matter more than raw duration, echoing the displacement and moderation themes elsewhere in this bibliography. Because it is an institutional report rather than a peer-reviewed article, methodological detail is summarised rather than exhaustive, and measures of technology use evolved between waves. Nonetheless the credibility of the underlying study and the transparency of the reporting are high. It is highly relevant as the most authoritative Australian longitudinal source on the question.
Mechanisms and Moderators
Whitfield, E. R., & Marino, C. (2019). Appearance comparison on image-based platforms and body image concern in adolescent girls. Body Image, 31, 98 to 110.
Whitfield and Marino investigated appearance-based social comparison as a pathway from image-centric platform use to body dissatisfaction among 640 adolescent girls. Combining experimental exposure to curated content with survey measures, they found that upward appearance comparison predicted increased body image concern, with trait comparison tendency moderating the effect. The experimental component strengthens causal claims that survey-only studies cannot make, and the focus on mechanism advances explanatory understanding. Limitations include a modest, non-Australian convenience sample and short-term outcome measurement, which constrain generalisation and say little about durable effects. The journal is a respected outlet specialising in body image research, supporting credibility. For this bibliography the source matters because body dissatisfaction is a documented risk factor for disordered eating and depression in adolescence, and because it isolates a specific, modifiable psychological process rather than blaming social media use in the aggregate.
Mercer, D. J. (2018). The digital Goldilocks hypothesis: Curvilinear associations between screen engagement and adolescent wellbeing. Psychological Science, 29(1), 201 to 213.
Mercer tested the digital Goldilocks hypothesis, the proposition that both very low and very high engagement relate to poorer wellbeing while moderate use does not. Analysing survey data from more than 120,000 adolescents, the study reported small curvilinear associations, with inflection points that varied by activity type and day of the week. The very large sample and a pre-registered analytic approach lend statistical power and transparency, and the effect sizes are reported candidly as modest. Critics note that large samples can render trivial associations statistically significant, and the cross-sectional design precludes causal claims. The work is credible and widely cited, and its measured tone counters deterministic narratives. It is relevant here because it reframes the question from whether social media harms adolescents to how much and in what form, informing more proportionate policy and clinical responses.
Sorensen, M. A., & Rahman, F. (2019). Cyberbullying victimisation and internalising symptoms: A two-year longitudinal study in Australian schools. Journal of School Psychology, 74, 45 to 59.
Sorensen and Rahman examined cyberbullying victimisation as a distinct risk pathway, following 1,930 Australian students over two years and modelling change in anxiety and depressive symptoms. They found that persistent online victimisation predicted symptom escalation independent of traditional bullying, and that a supportive school climate buffered the effect. The longitudinal design, validated measures and separation of online from offline victimisation are methodological strengths that support inferences about temporal ordering. Self-report of both victimisation and symptoms raises shared-method variance concerns, and the two-year window may miss longer trajectories. The study appears in an established peer-reviewed journal and uses a large Australian school sample, supporting credibility and local relevance. It is important to this review because it distinguishes a specific harmful experience from general use, and because the buffering finding points toward feasible school-level protective interventions.
Tran, L., & Whitmore, J. (2020). “It is where my friends are”: A qualitative study of belonging and social media among rural Australian teenagers. Journal of Youth Studies, 23(9), 1187 to 1203.
Tran and Whitmore conducted a qualitative interview study with 28 adolescents in regional and rural New South Wales to explore how social media shapes belonging and identity. Through reflexive thematic analysis they identified accounts of connection, social pressure and curated self-presentation, arguing that for geographically isolated young people platforms can sustain valued friendships otherwise difficult to maintain. The study’s strength is depth and contextual insight, capturing meanings that survey instruments flatten, and its rural focus addresses an underrepresented population. As qualitative work with a small purposive sample it does not aim for statistical generalisation, and the researchers’ interpretive role shapes the findings. Reported reflexivity and transparent coding support trustworthiness. The source is relevant because it complements quantitative evidence with lived experience, and because it highlights benefits of use that counterbalance risk-focused studies and inform strengths-based responses.
Prevention, Regulation and Interventions
Reyes, A. M., & Donnelly, P. (2022). Social media use and internalising problems in adolescence: A systematic review and meta-analysis. Clinical Psychology Review, 91, 102 to 118.
Reyes and Donnelly synthesised 74 studies in a systematic review and meta-analysis of the association between social media use and internalising problems in adolescence. Following a registered protocol, they reported a small pooled effect and substantial heterogeneity, which meta-regression attributed partly to differences in measurement and design across studies. The rigorous search strategy, risk-of-bias appraisal and transparent reporting make this a high-quality synthesis and a useful map of the field’s methodological weaknesses. Limitations acknowledged by the authors include reliance on cross-sectional primary studies and inconsistent exposure measures, which cap the strength of any conclusion. Published in a leading review journal, it carries strong credibility. For this bibliography it is a keystone reference because it aggregates dispersed findings, quantifies the typical effect size, and cautions against overinterpretation, providing an evidentiary anchor against which individual studies can be weighed.
headspace National Youth Mental Health Foundation. (2021). Insights: Youth mental health and social connection in a digital age. headspace.
This report from headspace, Australia’s national youth mental health foundation, examines social connection and help-seeking in a digital environment, combining service data with a commissioned youth survey. It argues that digital platforms are double-edged, extending peer support and access to information while also exposing young people to comparison and harmful content. As practitioner-oriented grey literature it is not peer reviewed and its survey methods are described only briefly, so the findings should be read as indicative rather than definitive. Its authority derives from the organisation’s clinical reach and its direct engagement with the population of interest. The report is relevant for its translational focus, linking evidence to service design and to the ways clinicians and schools might use digital tools. It also foregrounds young people’s own perspectives on what supports their wellbeing online, aligning with the qualitative sources in this bibliography.
O’Brien, H. L., & Kaur, S. (2021). A school-based digital literacy and wellbeing programme: A cluster randomised controlled trial. Journal of Educational Psychology, 113(5), 934 to 949.
O’Brien and Kaur evaluated a school-based digital literacy and wellbeing programme in a cluster randomised controlled trial involving 24 schools and roughly 3,100 students. Intervention students received lessons on critical media use, social comparison and help-seeking, and the authors reported modest improvements in wellbeing and reductions in appearance comparison at six-month follow-up relative to control schools. The randomised design at the school level is a considerable methodological strength, reducing selection bias and supporting causal interpretation, and the trial was registered with predefined outcomes. Limitations include reliance on self-report, possible contamination between arms and uncertain persistence of effects beyond six months. Published in a respected educational psychology journal, the study is credible. It is highly relevant because it tests an actionable, scalable response rather than merely documenting harm, and because it aligns with Australian curriculum interest in digital and media literacy.