This study explored the extent to which differences in social capital among family structures predicted academic improvement in young children using data from the nationally representative Early Childhood Longitudinal Survey, (ECLS) Kindergarten class of 1998-1999, which is administered by the National Center for Educational Statistics (NCES). For all children included in the analyses, parent social and resource capital explained 13% of the variance in reading change scores from kindergarten to fifth grade and 15% of the variance in math change scores from kindergarten to fifth grade. In addition, parent social and resource capital explained 34% of children’s reading change scores in stepparent family structures. In light of these findings, implications for policymakers, parents, and scholars are discussed.