We estimate a directional distance function to assess the impacts of multiple time-varying parent and child inputs on a cluster of jointly produced child outcomes for children aged 7 to 13 years. The directional distance function specification avoids several well-known empirical problems associated with analysis of household production data, namely, the need to aggregate inputs and outputs, assume separability among inputs and outputs, or estimate reduced form equations. Using a balanced panel of families from the National Longitudinal Survey of Youth-Child Sample for 1996 to 2000, we assess the marginal contributions of home and neighborhood environmental quality and children's time allocations, on their math and reading performance, behavior problems, and body mass index. We also measure productivity growth, technical change, efficiency change, and technical efficiency for production of child outcomes. Our results indicate significant jointness among good and bad child outcomes. Significant improvements in children's good outcomes and reductions in bad outcomes are also associated with a better home and parent perceived neighborhood environment, Head Start participation, and increased family time spent together during meals. Children's productivity growth is found to be highest at age 8 years and diminishes thereafter.