Estimating intergenerational schooling mobility on censored samples: consequences and remedies

In this paper we estimate the impact of parental schooling on child schooling, focus on the problem that children who are still in school constitute censored observations, and evaluate three solutions to it: replacement of observed with expected years of schooling, maximum likelihood approach, and elimination of all schoolaged children. Using intergenerational data from the Wisconsin Longitudinal Study we test how the three correction methods deal with censored observations. The one that treats parental expectations as if they were realizations seems to fix the censoring problem quite well.