Introduction Adolescent risk outcomes related to school issues are widespread, with about 20% parents reporting poor school engagement amongst their youth. Previous literature suggests that adolescents who report strong bonds with their parents are often identified as being less likely to engage in risky behaviours, such as substance use. The current study sought to examine the association between the frequencies of selected family activities and school problems amongst adolescents after adjustments for family connectedness and other characteristics. Methods Data were drawn from the National Longitudinal Survey of Youth, 1997. Of the 8984 youth interviewed, 3855 also had a sibling interviewed who met the selection criteria. School problem outcomes measured were suspension occurrence, poor grades and highest grade completed low for age. Independent variables of interest were self‐reported frequency of family dinner, fun and religious activities in a typical week. Multivariable logistic models were estimated for each outcome, and multivariable linear probability models were estimated adjusting for family fixed effects. Results Adjusting for family connectedness, there were significant associations between certain family activities and adolescent school problem measures. However, these results did not remain significant in models with family fixed effects, suggesting that associations could be driven by family‐level confounders. Discussion This study did not find strong evidence of a protective relationship between family activities and school problems. Therefore, it suggested that programme and policymakers be cautious in overstating the importance of family activities in preventing adolescent risk outcomes until true causal relationships can be determined.