Which design best supports causal inference between social media use and later depression?

• A. Longitudinal study with measured confounders
• B. Cross-sectional survey with self-report measures
• C. Single-time case series of depressed users
• D. Ecological correlation across countries

Answer: (A)