The impact of social media on the health of adolescents is an intensely debated topic globally, with conflicting views among researchers, caregivers, and policymakers. Empirical findings remain inconclusive, largely due to methodological and conceptual oversights. To address these oversights, we applied specification curve analysis, a computational method which reduces bias from selective analytical choices, to examine associations between time spent on social media and four related but distinct health outcomes - psychological well-being, mental health, health behaviours, and risk behaviours - in 15-and 16-year-olds in the west of Ireland (N = 2876). While most model specifications indicated a negative association between social media and health outcomes, some associations were positive (e.g., time on social media and vigorous physical activity for boys). Aside from anger management and alcohol consumption, most associations fall below accepted thresholds for reliability or clinical significance, suggesting that most associations may be indistinguishable from statistical noise. Notably, when compared to other relevant factors, time on social media emerged as one of the least influential predictors of adolescent health. While some specific associations between social media and adolescent health are statistically reliable, the overall effects at the population level are small and may not be clinically relevant.