Prosocial behavior is a relevant indicator of children's socio-emotional development linked to decreased conduct and emotional problems. The present study aimed to identify cross-sectional direct effects of parental involvement on prosocial behavior in three-time assessments (t0, t1, and t2), to identify carryover effects of the study constructs, and to identify the evolution of these effects over time. A sample of 235 Colombian families participated at t0, 220 at t1, and 145 at t2 by completing self-reported questionnaires for prosocial behavior using the Strengths and Difficulties Questionnaire and the Alabama Parenting Questionnaire for parental involvement. Using PLS-SEM path modeling, we found that the contribution of parental involvement to prosocial behavior was significant in the three assessments. Carryover analyses indicated that initial levels of parental involvement and initial levels of prosocial behavior predict later levels. Using multigroup analysis, we tested significant changes in the path coefficients of direct effects, finding nonsignificant results. For carryover effects, we found changes in parental involvement between t0/t1 and t1/t2. Finally, t-test analyses were used to identify changes in the construct's means over time, finding significant changes between parental involvement at t1 and t2. No mean differences were found for prosocial behavior. Results from this study highlight the relevance of parental involvement during childhood for maintaining children's levels of prosocial behavior and reducing the risk of socio-emotional problems. Preventive approaches for these problems should include parents' training on parental involvement from age 3.5 years or earlier.