Background: School climate and school identification are important features of the school environment and potential determinants of adolescent internet gaming disorder (IGD). Objective: This novel study investigated their joint effects on IGD and related mediation mechanisms via the interpersonal factor of teacher-student relationship and the individual factors of academic stress and anxiety. Methods: A large-scale cross-sectional study was conducted among adolescent internet gamers of junior, senior, and vocational middle schools in Taizhou City, China, from February to March 2022 (N=5778). Participants self-administered an anonymous, structured questionnaire in classrooms. Adjusted logistic regression and structural equation modeling (SEM) were used for data analysis. Results: Among all participants, the prevalence of IGD was 8% (461/5778). The 4 school climate subscales (student-student relationship subscale: adjusted odds ratio [ORa] 0.88, 95% CI 0.85-0.91; student-staff relations subscale: ORa 0.87, 95% CI 0.84-0.90; academic emphasis subscale: ORa 0.88, 95% CI 0.85-0.91; shared values approach: ORa 0.88, 95% CI 0.85-0.90), the school identification subscale (ORa 0.85, 95% CI 0.83-0.88), and teacher-student relationship (ORa 0.80, 95% CI 0.76-0.84) were significant protective factors against IGD, while academic stress (ORa 1.18, 95% CI 1.14-1.23) and anxiety (ORa 1.16, 95% CI 1.14-1.18) were risk factors of IGD. The SEM showed that the negative associations between school climate and IGD and between school identification and IGD were mediated via (1) three 2-step paths, each involving a single mediator—teacher-student relationship, academic stress, and anxiety, respectively—and (2) two 3-step paths involving 2 mediators—teacher-student relationship and academic stress first, respectively, and then anxiety. The direct effect of school climate on IGD was statistically nonsignificant (ie, full mediation with effect size ranging from 4.2% to 20.4%), while that of school identification was statistically significant (ie, partial mediation with effect size ranging from 4.5% to 38.2%). Conclusions: The relatively high prevalence of IGD among Chinese adolescents may be reduced through school-based interventions to improve school climate and school identification. Such improvements may reduce the levels of risk factors of IGD (poor teacher-student relationship, academic stress, and anxiety) and hence the risk of IGD. Future longitudinal and intervention studies are needed to confirm the findings.