Introduction:
Problematic Social Media Use (PSMU) in children has been consistently linked to parenting practices. However, prior research has mainly examined static, group-level associations. Consequently, the long-term within-person pathways remain unclear from a network perspective.
Purpose:
This study employed a longitudinal panel network analysis to disentangle within-level temporal and contemporaneous associations between PSMU symptoms and multidimensional parenting practices, while accounting for between-level variance.
Method:
Participants were 1,639 parents (89.4% mothers) of children aged 6-12 years (46.7% girls; Mage = 9.79 years, SD = 1.57 years) in China, who completed online surveys at three time points (April 2021, September 2021, March 2022). The Social Media Disorder Scale for Parents was used to assess nine symptoms of PSMU: preoccupation, withdrawal, tolerance, persistence, displacement, problem, deception, escape, conflict. The Multidimensional Assessment of Parenting Scale was employed to measure seven domains of parenting practices: proactive parenting, positive reinforcement, warmth, supportive parenting practices, hostility, lax control, physical control. To distinguish dynamic predictive pathways from momentary co-fluctuations, graphical vector autoregression models were used to estimate within‑level temporal and contemporaneous associations.
Results:
In the temporal network, preoccupation and persistence showed positive autoregression (self-reinforcing), deception showed negative autoregression (self-correcting), and physical control and lax control also showed positive autoregression, indicating temporal stability. Most nodes were longitudinally associated. Specifically, Proactive parenting showed the strongest outgoing expected influence, predicting increases in both specific parenting practices and PSMU symptoms, while withdrawal had the greatest ingoing expected influence. In the contemporaneous network, the variables with the strongest influence were hostility, displacement and tolerance. Three parenting practices directly covaried with PSMU symptoms: proactive parenting, hostility, and warmth.
Conclusions:
These findings highlight that specific PSMU symptoms and parenting practices are self-sustaining over time and dynamically co-develop. By identifying key targets such as proactive parenting, this study provides novel insights for interventions aimed at mitigating children's risk for PSMU.