Leveraging Latinx Adolescents, Photovoice, and Longitudinal Data to Disentangle the Bidirectional Effects of Social Media and Mental Health

PROJECT SUMMARY/ABSTRACT We propose a two-part integrated mixed-methods project to help improve our understanding of the social media landscape for adolescents, especially those with mental illnesses, and further accountability of new technology. Research has documented a concurrent adolescent mental health crisis with mixed effects of the potential role of social media: social media has become an important source of mental health information for the public yet unsafe content is also prevalent such as stigma, cyber-bullying, and hate speech. Efforts designed to study social media have produced only limited success owing to how social media data is dynamic and unique for each user and privately owned/managed in the tech sector; no national/public archive exists. Further, web-scraping methods (e.g., Twitter API) do not yield representative or random samples for study, and web-scraping and survey methods are not yet available for the most popular apps among adolescents of Instagram, Snapchat, and TikTok, or for capturing content in Spanish, limiting both internal and external validity. Extant efforts are thus incongruent with how Latino adolescents use social media more than all other race/ethnic and age groups in the U.S. New approaches are needed that address these methodological problems in a comprehensive fashion that is informed by research and suited to the ecological context surrounding diverse adolescent populations. Responding to this critical need, the proposed project seeks a rigorous test, with long term follow-up, participatory approaches, and both quantitative and qualitative methods. Also, to gain new causal insights and further mental health equity goals, this project focuses on Latinos ages 13-20 to leverage natural counterfactual contexts across their diverse language/cultural social media exposure. Specifically, we will examine the effects of exposure to mental health promoting (Aim 1) and pejorative (Aim 2) content encountered on social media on mental health status, and how this relationship operates via proposed mediators of mental illness stigma, self-perceptions, and help-seeking. Moderation by individual (e.g., age, gender, race), family (e.g., cohesion), and peer (e.g., social support) factors will be assessed, and by mental health status to identify directionality (Aim 3). Achieved in two parts, Part 1 is a prospective cohort design (N=1200) of three biweekly assessments of social media use and its mental health content, potential mediators, modifiers/covariates, and mental health status. Part 2 involves Youth Participatory Action Research (N=50) with SocialsVoice, an adaptation of PhotoVoice, to glean a culturally relevant social media database for community-based analysis. The generated knowledge from the evaluation of data in Parts 1-2 will together inform recommendations for intervention by identifying specific problems and target populations with respect to social media and mental health, cuing how to best leverage the strengths of adolescent ecological contexts to innovatively promote mental health. The proposed project and team to implement it will provide public mental health with critical new knowledge it sorely needs to inform programs, policies, and practices concerning the safety and utility of social media as it relates to population mental health.