This article describes the development and deployment of a framework for measuring parenting capacities in the context of bereavement. Grounded theoretical analysis of interviews with a community sample of 41 bereaved spouses with school-aged children elicited a set of nine bereavement-specific parenting tasks. A corollary coding system (covering all nine parenting tasks) was created to transform interview materials into quantitative data, thus permitting systematic empirical investigation of the parenting capacities of bereaved spouses. Parenting behaviors were coded on a 5-point scale ranging from least child-centered to most child-centered. Sex of surviving parent and circumstances of death proved to be significant mediating variables: mothers were more child-centered than fathers, and parents surviving sudden deaths more child-centered than those surviving anticipated deaths. Lengthy illness was associated with less child-centered parenting. The more child-centered the parenting, the less symptomatic the child as measured by parent report (Child Behavior Checklist) and child self-report (Children's Depression Inventory, Revised Child Manifest Anxiety Scale). Child-centered parenting was associated with more positive and fewer negative perceptions of the surviving parent by the child as measured by the Parent Perception Inventory. Implications of findings are discussed.