Research question: In this project we estimate the impact of parental schooling on child's schooling, and focus on the problem that many studies use samples in which part of the children is still in school. Method: We evaluate three solutions to this censoring problem: replacement of observed with expected years of schooling, a maximum likelihood approach, and elimination of all school-aged children. Results: Using intergenerational data we test how the three correction methods deal with the problem that some children have not finished their schooling yet. The method that treats parental expectations as if they were realizations seems to fix this problem quite well. Status: forthcoming in the Journal of Applied Econometrics.