Network modeling of problematic social media use components in college student social media users

While the constitutive features of problematic social media use (PSMU) have been formulated, there has been a lack of studies in the field examining the structure of relationships among PSMU components.

This study employed network analytic methods to investigate the connectivity among PSMU components in a large sample of 1,136 college student social media users (M _age = 19.69, SD = 1.60). Components of PSMU were assessed by the Bergen Social Media Addiction Scale (BSMAS) derived from a components model of addiction. We computed two types of network models, Gaussian graphical models (GGMs) to examine network structure and influential nodes and directed acyclic graphs (DAGs) to identify the probabilistic dependencies among components.

Relapse component consistently emerged as a central node in the GGMs and as a parent node of other components in the DAGs. Relapse and tolerance components exhibited strong mutual connections and were linked to the most vital edges within the networks. Additionally, conflict and mood modification nodes occupied more central positions within the PSMU network for the low-BSMAS-score subgroup compared with the high-BSMAS-score subgroup.

Our findings shed new light on the complex architecture of PSMU and its potential implications for tailored interventions to relieve PSMU.