= 9.57 years; 48.5% female), followed annually for three years. Growth mixture modelling identified distinct PSU trajectories, and bidirectional associations with internal (e.g., shyness, depression, anxiety) and external factors (e.g., parental neglect, peer exclusion) were examined. Cross-lagged network analysis further assessed interval-specific variation in predictive associations. Two trajectories emerged: a low-decreasing group (intercept = 1.845, slope = -0.120) and a high-increasing group (intercept = 2.110, slope = 0.348). Early adolescents in the high-increasing group experienced more adverse environments and reported higher levels of risk factors. PSU was primarily predicted by external factors in early intervals, whereas both personal and environmental factors became influential in later intervals. These findings highlight the heterogeneous and dynamic nature of PSU and support early, targeted family-focused interventions.
Hu et al. (Mon,) studied this question.