Longitudinal Associations and Gender-Stratified Patterns in Adolescents’ Emotional and Behavioral Problems—Based on a Cross-Lagged Panel Network Model

Adolescence is a critical period characterized by heightened emotional and behavioral challenges, profoundly impacting young people’s social adaptation and future development. Previous research has predominantly employed traditional latent variable analysis to examine factors associated with emotional and behavioral problems, with less attention to the underlying symptom-level structure and prospective associations of these problems. This study utilized longitudinal data from the China Family Panel Studies (CFPS) collected in 2018 (T1) and 2022 (T2). It combined contemporaneous network analysis with cross-lagged panel network analysis (CLPN), employing the 14-item adolescent problem behavior measure included in the CFPS to examine contemporaneous and prospective symptom-level associations among emotional and behavioral problems in adolescents and to compare gender-stratified networks. The study included 1082 adolescents (mean age at T1 = 12.09 ± 1.54), with boys accounting for 54.5%. In the contemporaneous networks, worry about finishing homework (I10) had the highest expected influence at both waves, although other highly connected nodes differed across networks. In the CLPN, worry about having someone to play with at school (I11) and sadness (I7) had the highest out-expected influence (O-EI) values, whereas difficulty paying attention (E4) and being easily distracted (E6) had the highest in-expected influence (I-EI) values. Permutation tests did not provide evidence of significant between-gender differences. The findings provide a symptom-level perspective on contemporaneous and prospective associations among adolescent emotional and behavioral problems and identify symptom areas for further longitudinal and intervention research.