Backgrounds: The mental health consequences of social media use remain debated. Drawing on the “rich-get-richer” perspective, this study examines whether social media access interacts with perceived social resources to shape depression risk among Chinese students. Methods: We analyze nationally representative data from the 2020 and 2022 waves of the China Family Panel Studies (CFPS), constructing a two-period unbalanced student panel. High-dimensional fixed effects linear probability models are estimated with province and year fixed effects and province-specific linear trends. Mediation analyses follow the Baron and Kenny framework and are supplemented by Sobel-Goodman and bootstrap tests. Heterogeneity is examined by sex and urban–rural residence. Results: Students with both social media access and high perceived social resources exhibit a significantly lower probability of depression (β ≈ −0.06, p < 0.01). Trust in strangers partially mediates this association, accounting for approximately 5% of the total effect. In contrast, greater perceived entertainment value functions as a suppressor, slightly attenuating the protective relationship. The association is stronger among male and urban students. Robustness analyses using life satisfaction as an alternative outcome yield consistent patterns. Conclusions: Findings support the “rich-get-richer” hypothesis: digital engagement amplifies existing social advantages in mental health outcomes. Rather than exerting uniform effects, social media appears to reinforce underlying social inequalities. Policies should therefore move beyond access expansion toward strengthening trust-building mechanisms and mitigating entertainment-driven risks.
Hu et al. (Thu,) studied this question.