Objective  To explore the differences in positive psychological resources among elementary and secondary school students with and without non-suicidal self-injury (NSSI) behaviors.

Methods  Between September and October 2023, the assessment data of 56870 students aged 9 to 18 years in Chengdu, Sichuan Province were collected. The R software package was used to evaluate the network structure and symptom centrality, and the global strength and structural differences of the networks of students with and without NSSI behaviors in elementary, junior high, and senior high schools were compared.

Results  The prevalence of NSSI among elementary and secondary school students stands at 12.21%. Specifically, it peaks in junior high school students, reaching 14.22%, followed by that in senior middle school students at 13.66%, and is lowest in elementary school students at 10.17%. Regarding the network structure in elementary school students, irrespective of the presence or absence of NSSI behavior, subjective support exhibits the strongest network strength, and family functioning demonstrates the highest closeness centrality. Among junior high school students without NSSI behaviors, the strongest network strength was observed in subjective support, whereas for those with NSSI behaviors, self-esteem showed the strongest network strength. In addition, subjective support exhibited the highest closeness centrality in both groups. In senior high school students, self-esteem was the factor with the strongest network strength for students with or without NSSI behaviors. For senior high school students without NSSI, subjective support had the highest closeness centrality, while for those with NSSI, expressive suppression showed the highest closeness centrality. The edge weights differed significantly among elementary school students in family functioning-self-esteem (E = 0.093) and family functioning-objective support (E = 0.087). For junior high school students, significant differences were found in objective support-subjective support (E = 0.131) and family functioning-impulse control (E = 0.109). Among senior high school students, significant differences were observed in impulse control-family function (E = 0.133) and subjective support-utilization of support (E = 0.130) (P < 0.05).

Conclusion  Excessive family involvement in elementary school students and expressive suppression in senior high school students may be important influencing factors for NSSI behaviors. The lack of social support has a greater impact on junior and senior high school students. The strength of the connection among subjective support, objective support, and utilization of support is an important indicator.