The Future of Children, Princeton - Brookings: Providing research and analysis to promote effective policies and programs for children.

The term "domestic violence" typically refers to violence between adult intimate partners. The range of conduct included in this term currently varies with the context within which it is used.⁸ Clinical definitions are often broader than legal definitions. For example, one clinical source defines domestic violence as a pattern of assaultive and coercive behaviors, including physical, sexual, and psychological attacks, as well as economic coercion, that adults or adolescents use against their intimate partners.⁹ By contrast, a model code on domestic and family violence limits its definition to acts of physical harm, including involuntary sexual acts, or the threat of physical harm.¹⁰

Several different terms have been used by researchers and others to refer to children in households with domestic violence. Early researchers spoke of these children as either "witnesses" or "observers" of the violence. In the past five years, these terms have been replaced by "exposure" to the violence, which is more inclusive and does not make assumptions about the specific nature of the children's experiences with the violence. Exposure to domestic violence can include watching or hearing the violent events, direct involvement (for example, trying to intervene or calling the police), or experiencing the aftermath (for example, seeing bruises or observing maternal depression).¹¹

No scientifically credible national prevalence data currently exist for child exposure to domestic violence. Estimating prevalence requires clear definitions for what constitutes domestic violence and child exposure to it, as well as ways to verify that the violence and the exposure occurred.⁵ Although the field is in the early stages of formulating common definitions for domestic violence and child exposure to it, studies to date have not used common definitions. Differences in definitions and other research methodologies have resulted in substantial variability in prevalence estimates, and make it difficult to compare data across studies.