Annotated Bibliography Sample: Social Media Use and Adolescent Mental Health

Model Answer

Scope and Focus

This annotated bibliography addresses the research question: “What is the relationship between social media use and mental health outcomes in adolescents?” Seven sources are reviewed, spanning longitudinal cohort research, an experimental trial, qualitative and cross-sectional studies, a meta-analysis, a national survey report and a developmental-psychology text, chosen to represent both the strength and the possible mechanisms of the association. Referencing follows Harvard style throughout, and each annotation summarises the source’s aims and findings, appraises its methodological credibility, and explains its contribution to answering the question.

Annotations

Ashworth, C., Preston, L. and Doyle, M. (2019) ‘Social media use and depressive symptoms in adolescence: a three-wave longitudinal study’, Journal of Adolescent Health, 65(3), pp. 312–320.

This three-wave longitudinal study followed 1,240 UK secondary-school pupils aged 13–16 over eighteen months, measuring self-reported daily social media use and depressive symptoms using a validated screening scale at each wave. Cross-lagged analysis found that higher social media use predicted small increases in depressive symptoms six months later, particularly among girls, but the reverse pathway — depressive symptoms predicting increased use — was also significant, suggesting a bidirectional rather than purely causal relationship. The large sample and repeated-measures design are notable strengths, allowing the authors to test direction of effect rather than relying on a single snapshot, though reliance on self-reported use rather than device-logged data may have introduced measurement error. This study is central to the research question because it moves beyond simple correlation and offers evidence for the direction, and modest size, of the relationship between use and mood.

Voss, K. and Ibrahim, R. (2020) ‘Reducing social media use and wellbeing: a randomised controlled trial in university students’, Cyberpsychology, Behavior, and Social Networking, 23(11), pp. 745–753.

This randomised controlled trial assigned participants to either limit social media use to thirty minutes a day or continue as normal for three weeks, measuring anxiety, loneliness and life satisfaction before and after. The restricted-use group reported small but statistically significant improvements in wellbeing compared with controls, with the authors suggesting reduced social comparison as a likely mechanism. Randomisation and a pre-registered protocol strengthen confidence in the causal claim, though the sample was drawn from university students rather than adolescents, limiting direct applicability to a school-age population, and the three-week duration cannot show whether effects persist. Despite this age limitation, the study is included because it is one of the few sources offering genuinely experimental, rather than correlational, evidence, and its findings are used cautiously in this bibliography to support the plausibility of a causal pathway suggested by the longitudinal studies.

Fairweather, H. and Oduya, T. (2021) ‘Adolescent girls’ accounts of social comparison on image-based platforms: a qualitative study’, Journal of Youth Studies, 24(7), pp. 889–905.

Through focus groups with thirty-two girls aged 14–17, this study explored how adolescents themselves understand the relationship between image-based platforms and their self-esteem. Participants described a cycle of upward social comparison, curated self-presentation and validation-seeking through likes and comments, alongside strategies such as muting or unfollowing accounts to manage distress. The purposive, single-gender sample limits generalisation to boys or to non-image-based platforms, and focus-group data can be shaped by group dynamics, but the study offers a depth of insight into mechanism that quantitative surveys cannot capture. For this bibliography, it is particularly valuable in explaining why image-based comparison, rather than screen time as a whole, may be the more precise driver of any mental-health association, refining rather than contradicting the quantitative findings above.

Larsen, P., Cheung, W. and Bright, S. (2022) ‘Screen time, social media and adolescent depression: a meta-analysis of longitudinal studies’, Psychological Bulletin, 148(4), pp. 401–436.

This meta-analysis synthesised data from twenty-nine longitudinal studies to estimate the overall association between social media use and later depressive symptoms in adolescents. The pooled effect size was small but statistically significant, and moderator analysis indicated stronger associations for passive use, such as scrolling, than for active use, such as messaging friends. The authors applied a recognised risk-of-bias tool and reported moderate heterogeneity, which they explored through subgroup analysis by age and measurement type. As the highest level of evidence located for this bibliography, the meta-analysis carries substantial weight, though its reliance on published, mostly Western studies raises questions about cultural generalisability. It provides the clearest quantitative answer to the research question available in the current literature and is used as the anchor source against which the other studies are compared.

Royal Society for Public Health (2017) #StatusOfMind: Social Media and Young People’s Mental Health and Wellbeing. London: RSPH.

This widely cited survey report gathered views from nearly 1,500 young people aged 14–24 on how different platforms affected aspects of wellbeing including anxiety, body image and sleep. Two image-heavy platforms were rated most negatively for body image and anxiety, while a video platform was rated positively overall, and the report proposes practical recommendations such as pop-up usage warnings and safe-use badges. As a survey rather than an experimental or longitudinal study, it cannot establish causation and relies on young people’s subjective ratings, which may be influenced by media narratives about social media harm. Its value for this bibliography lies in capturing adolescents’ own platform-specific perceptions at scale, complementing the more rigorous but narrower academic studies, and its recommendations are frequently cited in UK policy discussions relevant to the research question.

Ingram, J. (2020) The Adolescent Mind Online: Development, Identity and Digital Life. London: Palgrave Macmillan.

This text situates social media use within broader adolescent developmental theory, arguing that platforms intersect with normal developmental tasks such as identity formation and peer belonging, which can amplify both risks and benefits depending on individual vulnerability. Drawing on developmental psychology alongside media-effects research, the author proposes a ‘differential susceptibility’ framework in which effects depend on pre-existing traits such as low self-esteem rather than social media itself being uniformly harmful. As a synthesis text rather than a primary study, its claims rest on the quality of the literature it draws together, and some more recent platform-specific studies postdate its publication. It is included in this bibliography because it provides the theoretical framework needed to interpret the mixed and modest effects reported in the primary studies above, rather than treating social media as a uniform exposure.

Delacroix, N. and Mensah, F. (2018) ‘Problematic smartphone use and sleep quality in UK adolescents: a cross-sectional survey’, Sleep Medicine, 45, pp. 45–51.

This cross-sectional survey of over 2,000 secondary-school pupils examined the relationship between problematic smartphone use, largely driven by social media checking, and self-reported sleep quality. Higher problematic-use scores were associated with later bedtimes, shorter sleep duration and poorer sleep quality, with the authors suggesting that pre-sleep alerting content and light exposure both play a role. The large sample gives reasonable statistical power, but the cross-sectional design cannot establish whether poor sleep drives social media use or vice versa, and self-reported sleep measures are less reliable than actigraphy. The study adds an important indirect pathway to this bibliography’s research question, since sleep disruption is itself a well-established risk factor for adolescent depression and anxiety, suggesting that some of the mental-health association reported elsewhere may operate partly through disrupted sleep.

Overall Relevance

Taken together, the sources point to a small but consistent association between heavier, more passive social media use and poorer adolescent mental health, mediated in part by social comparison and sleep disruption, with effects varying by individual vulnerability rather than applying uniformly. The set supports a nuanced answer to the research question: social media is one contributing factor among several, rather than a sole or simple cause of adolescent mental-health difficulties.