Linking administrative data records for the same individuals across services and over time offers a powerful, population-wide resource for child maltreatment research that can be used to identify risk and protective factors and to examine outcomes. Multistage de-identification processes have been developed to protect privacy and maintain confidentiality of the datasets. Lack of information on those not coming to the attention of child protection agencies, and limited information on certain variables, such as individual-level SES and parenting practices, is outweighed by strengths that include large and unbiased samples, objective measures, comprehensive long-term follow-up, continuous data collection, and relatively low expense. Ever emerging methodologies and expanded holdings ensure that research using linked population-wide databases will make important contributions to the study of child maltreatment.