Temporal Associations between Social Media Use Frequency and Social Media Addiction and Depressive, Anxiety, and Insomnia Symptoms among Hispanic Emerging Adults: Moderation by Identity Exploration and Instability and Negativity

The number of social media users has drastically increased in recent years, and rates of social media use are particularly pronounced among emerging adults and Hispanic groups. Furthermore, emerging adults are at risk for poor mental health and report poor sleep health, and Hispanic groups may experience sleep disparities compared to their non-Hispanic White counterparts. Although many studies have assessed the associations between social media use frequency and social media addiction and depressive, anxiety, and insomnia symptoms, fewer have investigated these associations prospectively. Similarly, few studies have assessed potential moderators of these associations. Thus, the current study investigated the temporal associations between depressive, anxiety, and insomnia symptoms and social media use frequency and social media addiction while also investigating two of the five dimensions of emerging adulthood (identity exploration, instability and negativity) as potential moderators of these associations within the context of Uses and Gratifications Theory and Compensatory Internet Use Theory. Two-hundred and ninety-two Hispanic emerging adults were recruited from Prolific and completed an online survey at two time points three months apart (March 2025 and June 2025) assessing sociodemographics, depressive, anxiety, and insomnia symptoms, identity exploration, instability and negativity, social media use frequency, and social media addiction. Descriptive statistics yielded participant characteristics, and bivariate analyses were conducted for cross-sectional univariate associations of interest. Six structural equation models were performed to test the associations and interactions of interest. Specifically, depressive, anxiety, and insomnia symptoms served as the exogenous variables; identity exploration and instability and negativity as the moderator (also exogenous) variables; and social media use frequency and social media addiction as the endogenous variables. Results showed that across models, depressive and anxiety symptoms at time point one positively predicted social media addiction at time point two. More interestingly, lower levels of instability and negativity demonstrated stronger positive associations between anxiety symptoms and social media addiction (i.e., the positive association between anxiety symptoms and social media addiction was stronger at lower levels of instability and negativity). Exploratory analyses testing the reverse directional paths revealed some support for social media use frequency and social media addiction predicting insomnia symptoms. Additionally, lower levels of identity exploration demonstrated stronger positive associations between social media use frequency and insomnia symptoms (i.e., the positive association between social media use frequency and insomnia symptoms was stronger at lower levels of identity exploration). The clinical implications of these results are discussed, and additional prospective studies are needed to continue assessing temporality of these associations.