Abstract
BACKGROUND: Few studies have investigated the direct and interactive associations between different gender adolescent dissociative experiences (ADEs) and impact factors. This study aimed to analyze the mediating associations of factors influencing ADEs. METHODS: We used a structural equation model (SEM) to explore the direct, interactive, and mediating associations between ADEs and influencing factors by gender stratification. A two-center cross-sectional study was conducted in Hangzhou City, China. The sample comprised 2,400 adolescents who were randomly selected using a multi-phase, stratified, cluster sampling technique. RESULTS: A total of 2,114 participants were included in this study. The ADEs scores in boys were higher than in girls (t = 5.21, p < 0.001). SEM revealed significant direct associations between ADEs in boys and three primary impact factors: mental health parameters (anxiety and sleep/dietary patterns: β = 0.37, p < 0.001), family environment characteristics (conflict and expressiveness: β = 0.16, p < 0.001), and teacher-student conflict (β = 0.19, p = 0.006). Furthermore, mental health factors demonstrated significant mediating associations, explaining 16.7% (β = 0.03, p = 0.006) and 30.5% (β = 0.09, p < 0.001) of the total associations linking family environment and teacher-student conflict to boys ADEs, respectively. Whilst, family environment and teacher-student conflict play a mediating part (β = 0.12, p < 0.001) in the indirect link between mental health and boys ADEs. According to the SEM analysis in girls, among the variables that affected ADEs just through the direct path, the variable of peer interpersonal interaction had a negative relationship (β = -0.06, p = 0.047), but mental health (psychoticism, paranoid ideation, somatic complaints, obsessive–compulsive, hostility) (β = 0.66, p < 0.001) had the most positive relationship with ADEs. The peer interpersonal interaction (β = 0.11, p = 0.002) mediates the relationship between mental health and girls ADEs. CONCLUSIONS: This study indicated that significant gender differences were evident in both the influencing factors and the mediating associations of factors related to ADEs. Evidently, the present cross-sectional SEM—under rigorous measurement, strong model fit, plausible quasi-stationarity, and conservative interpretation—offers a defensible framework for estimating contemporaneous associations in time-varying psychosocial constructs. These findings can guide hypothesis generation and inform the design of longitudinal studies and pilot interventions, while avoiding causal claims. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s40359-026-04171-8.