Multidimensional Social Interaction and School Refusal Tendency Among Elementary and Junior High School Students: A Three-Year Longitudinal Study with an Exploratory Indirect-Association Analysis

Background/Objectives: School refusal tendency is an important school adaptation issue, but its upstream psychosocial correlates remain insufficiently understood. We examined whether baseline multidimensional social interaction was prospectively associated with later school refusal tendency and whether an exploratory statistical indirect association was observed through psychological and behavioral difficulties. Methods: This three-year longitudinal observational study included 314 students in grades 5–9 aged ≥12 years at baseline in 2020 and followed through 2023. Social interaction was assessed using the Index of Social Interaction (ISI), psychological and behavioral difficulties using the parent-reported Strengths and Difficulties Questionnaire (SDQ), and school refusal tendency using a parent-reported binary item. Adjusted regression and Monte Carlo product-of-coefficients analyses were conducted. Results: School refusal tendency was observed in 84 participants (26.8%). Higher baseline ISI scores were associated with lower follow-up SDQ total difficulties (β = −0.830, 95% CI: −1.033 to −0.627) and lower odds of school refusal tendency (OR = 0.832, 95% CI: 0.741–0.934). The indirect effect through follow-up SDQ total difficulties was significant (−0.120, 95% CI: −0.193 to −0.055). After adjustment for baseline SDQ total difficulties, the indirect effect remained significant (−0.067, 95% CI: −0.124 to −0.020); Firth and parsimonious models yielded similar estimates. Baseline-domain-adjusted analyses retained indirect effects through emotional symptoms, peer problems, and internalizing problems. Conclusions: Higher baseline multidimensional social interaction was associated with lower subsequent psychological and behavioral difficulties and school refusal tendency. Baseline-SDQ-adjusted analyses supported an exploratory indirect association, particularly through internalizing-related difficulties; however, concurrent assessment of follow-up SDQ and school refusal precludes causal interpretation.