The Intervention Overload Hypothesis: How Service Diversity Produces a U-Shaped Effect on Recidivism in Juvenile...

Juvenile diversion reduces recidivism on average. However, how many and what combination of services yield the best outcomes remains empirically unaddressed, with contemporary practice operating under an implicit "more is better" logic. Data were drawn from administrative records of the conditional non-prosecution system in Zhejiang Province, China, covering the period from 2014 to 2022. The retrospective cohort comprised 1,612 youth (M age = 15.87, SD = 0.92; 17.6% female; 32.1% non-local household registration) spanning low-, moderate-, and high-risk groups as classified by a standardized risk-need assessment. Using this administrative dataset, the study conducted triangulated causal analyses with risk-stratified moderation. Service diversity exhibited a U-shaped relationship with recidivism, optimal near 4.8 service types. Per-service engagement significantly mediated the relationship between service diversity and recidivism, with engagement declining as service breadth increased. The curvature of the U-shaped relationship was attenuated for higher-risk youth, with inflection point estimates ranging from approximately four service types for low-risk youth to over five for high-risk youth. These findings challenge the prevailing "more is better" logic in diversion programming and point toward a principle of calibrated parsimony: effective diversion requires not the most services, but the right number matched to each youth's capacity to benefit.