Note. Longitudinal relations between school-age extracurricular involvement and age 16 internalizing and externalizing problems. Covariates included teacher-reported externalizing and internalizing problems (averaged across ages 7.5, 8.5, and 9.5), study site (Pittsburgh, PA as reference group; rural not in figure), target youth race (not in figure), target youth ethnicity (not in figure), target youth sex, and intervention group (not in figure). All paths were estimated, but only significant predictors with their standardized β values are reported. * p < .05. ** p < .01. *** p < .001.