Testing the Cinderella effect: Measuring victim injury in child abuse cases

Abstract

Past research finds that stepparents harm and kill their partner's children at higher rates than biological parents do to their own children, a phenomenon called the “Cinderella effect.” Yet one major limitation of these studies is that reporting biases may account for a large share of the effect observed if reporting rates differ based on the victim-offender relationship. As abuse cases with serious injuries are more likely to be disclosed to police than ones with no injuries or minor injuries, using cases where the victim is seriously injured allows us to minimize the impact of differential reporting. Using data from the FBI's National Incident-Based Reporting System (NIBRS) which covers over 500,000 cases of child abuse from 1991 through 2019, we find that, relative to biological parents, unmarried partners, but not stepparents, are significantly more likely to seriously injure the child, partially supporting the Cinderella effect.