(PS8-61) Digital Media Use in Adolescents: Profiles of Use and Predictors

Abstract Text: Nearly all U.S. adolescents report using the internet every day and the proportion of teens who report being online “almost constantly” has almost doubled over the last decade (Pew Research Center, 2024). Despite substantial efforts to understand the relationship between digital media use and a range of psychological and behavioral outcomes, a broad consensus has not been identified in the research. Prior studies use an estimate of hours per day that adolescents engage with screens as a proxy for digital media usage. Recent research advocates for specific assessment to explore beyond the number of screen time hours as a primary predictor of digital media usage (Prinstein et al., 2020). Variation in adolescent digital media use warrants further research. The goal of this study was to identify profiles of digital media use and their relevant predictors. This study reflects one of the first attempts to utilize data from a large nationally representative adolescent survey focused on different digital media categories (watching media, playing video games, and using social media) using latent profile analysis (LPA). Five classes were found: High Users (13.1%), Social Media-Watchers (6.1%), High Gamer-Watchers (6.4%), Moderate Gamer-Watchers (41.5%), and Low Users (32.7%). These classes were assessed for relationships with a selection of predictors (life dissatisfaction, sleep, physical activity, drug use) through multinomial logistic regression. Boredom was associated with increased likelihood of membership in all but the Low User class, and physical activity was associated with decreased likelihood in all classes but the Low User class. The more dissatisfied with life a student is, the greater likelihood they were in the High Gamer-Watcher and Moderate Gamer-Watcher groups over the Low User class. This finding indicates that there may be something unique about gaming that is related to life dissatisfaction. There may also be a protective effect with social media use, which is why the High User and Social Media-Watcher groups were not significantly related to life dissatisfaction. Additionally, gender differences were identified with an increase likelihood in male membership for the High Gamer-Watcher and Moderate Gamer-Watcher classes and females were more likely to be in the High User and Social Media-Watcher classes. This study showed that there are multiple profile patterns of digital media use amongst American adolescents, and that the factors relevant for membership prediction vary depending on the profile of use. It also demonstrates that LPA can be utilized to examine digital media use in a person-centered method. This project aims to support future research focused on the bidirectional effects of adolescent mental health and digital media use.