Scientific studies of child maltreatment victims' memory abilities and court experiences have important legal, psychological, and clinical implications. However, state-of-the-art research on child witnesses is often hindered by methodological challenges. In this paper, we address specific problems investigators may encounter when attempting such research with child victims. Based on our extensive experience, we describe strategies to deal with various methodological and practical issues. Primarily, we focus on methods for: recruiting maltreatment victims, seeking consent for participation in research, ensuring representativeness of samples, interviewing children, and retaining participants longitudinally. Regarding recruitment, legal and mental health agencies have helped us attract research participants. Finding an advocate within the agency is one key to success. Regarding the process of obtaining informed consent, we describe how to ensure participant confidentiality and how to obtain consent on behalf of child maltreatment victims. We further discuss issues concerning the representativeness of maltreatment samples, a topic that is relevant to the generalizability of research findings. In child eyewitness memory studies, it is of interest to ask misleading as well as specific questions. We thus discuss issues of debriefing of child participants to prevent any possible memory taint. Finally, given the importance of studying long-term effects of child abuse and neglect, and of legal involvement, we illustrate methods for tracking maltreated participants over time and obtaining adequate retention rates in prospective longitudinal studies. We hope that the methodological advances made in previous work will help future investigators carry out child witness studies in increasingly effective and successful ways.