Household composition and school performance

Most children no longer spend their entire childhood in a family with both biological parents, as a sizable proportion of this period is spent in a single-parent family or living with a social parent, in particular a social father. In general, it is believed that children from divorced couples and living in single-parent families tend to perform less well at school than children living in intact two biological parents families, even after controlling for parental characteristics. I compared school performance in the fourth series (or fifth year) in Brazil of children living in six different household arrangements. The highest values for school performance were observed for households with both biological parents, although differences for households with mother and a social father were not large. Values for households with a mother showed intermediate results. The three household compositions without a mother father and a social mother, father (only or with aggregates) and other arrangements without a father or a mother had smaller values for school performance than others. These observed differences in school performance were mostly explained by observable heterogeneity in school inputs, location, household economic and learning resources and household’s interactions. The non-observable differences between households were sizable only for those with father and a social mother and father (only or with aggregates), suggesting mechanisms not captured by the controls of the econometric models.