Do reductions in class size "crowd out" parental investment in education?

We use panel data from the kindergarten and 1st grade waves of the Early Childhood Longitudinal Study--Kindergarten Class to investigate whether increases in schooling inputs in the form of smaller classes "crowd out" parental inputs. We estimate child fixed-effect and fixed-effect-instrumental-variable models to identify the causal effects of class size on three types of parental involvement--parent--child interactions, parent-school interaction, and parent-financed activities for children. Our results suggest that increases in class size are associated with a decrease in parent-child interaction, no change in parent-school interaction, but an increase in parent-financed activities. The magnitude of these effects is between 3% and 7% of a standard deviation. Controlling for parental involvement in test score regressions does not change the achievement effects of class size, suggesting that the benefits of class size reduction are unlikely to be neutralized by adjustments in parental inputs, at least during the first 2 years in school.