Evaluating Preschool Programs When Length of Exposure to the Program Varies: A Nonparametric Approach

Nonexperimental data are used to evaluate impacts of a Bolivian preschool program on cognitive, psychosocial, and anthropometric outcomes. Impacts are shown to be highly dependent on age and exposure duration. To minimize the effect of distributional assumptions, program impacts are estimated as nonparametric functions of age and duration. A generalized matching estimator is developed and used to control for nonrandom selectivity into the program and into exposure durations. Comparisons with three groups-children in the feeder area not in the program, children in the program for less than or equal to 1 month, and children living in similar areas without the program-indicate that estimates are robust for significant positive effects of the program on cognitive and psychosocial outcomes with greater than or equal to 7 months' exposure, although the age patterns of effects differ slightly by comparison group. Copyright 2004 President and Fellows of Harvard College and the Massachusetts Institute of Technology.

Behrman, Cheng, and Todd (2004)Urban.Estimation as nonparametric functions of age and duration.Training and unemployment benefits, including a public service component, where local government and other eligible organizations propose public projects and hire ALMP participants to work on them.Statistically significant effects on the likelihood of employment, the likelihood of being employed at least once and on wage levels. Middle aged had biggest impacts. Retraining increases the probability of employment and decreases the wage for females. Program not beneficial for highly educated.Adults. Registered unemployed, having income less than 50% of minimum wage, employed 6 months of last 12 or recent graduate. 45% of sample had university degree.