IMPORTANCE: Increases in screen time have been found to be associated with increases in depressive symptoms. However, longitudinal studies are lacking. OBJECTIVE: To repeatedly measure the association between screen time and depression to test 3 explanatory hypotheses: displacement, upward social comparison, and reinforcing spirals. DESIGN, SETTING, AND PARTICIPANTS: This secondary analysis used data from a randomized clinical trial assessing the 4-year efficacy of a personality-targeted drug and alcohol prevention intervention. This study assessed screen time and depression throughout 4 years, using an annual survey in a sample of adolescents who entered the seventh grade in 31 schools in the Greater Montreal area. Data were collected from September 2012 to September 2018. Analysis began and ended in December 2018. MAIN OUTCOMES AND MEASURES: Independent variables were social media, television, video gaming, and computer use. Symptoms of depression was the outcome, measured using the Brief Symptoms Inventory. Exercise and self-esteem were assessed to test displacement and upward social comparison hypothesis. RESULTS: A total of 3826 adolescents (1798 girls [47%]; mean [SD] age, 12.7 [0.5] years) were included. In general, depression symptoms increased yearly (year 1 mean [SD], 4.29 [5.10] points; year 4 mean [SD], 5.45 [5.93] points). Multilevel models, which included random intercepts at the school and individual level estimated between-person and within-person associations between screen time and depression. Significant between-person associations showed that for every increased hour spent using social media, adolescents showed a 0.64-unit increase in depressive symptoms (95% CI, 0.32-0.51). Similar between-level associations were reported for computer use (0.69; 95% CI, 0.47-0.91). Significant within-person associations revealed that a further 1-hour increase in social media use in a given year was associated with a further 0.41-unit increase in depressive symptoms in that same year. A similar within-person association was found for television (0.18; 95% CI, 0.09-0.27). Significant between-person and within-person associations between screen time and exercise and self-esteem supported upward social comparison and not displacement hypothesis. Furthermore, a significant interaction between the between-person and within-person associations concerning social media and self-esteem supported reinforcing spirals hypothesis. CONCLUSIONS AND RELEVANCE: Time-varying associations between social media, television, and depression were found, which appeared to be more explained by upward social comparison and reinforcing spirals hypotheses than by the displacement hypothesis. Both screen time modes should be taken into account when developing preventive measures and when advising parents.