The participants were recruited from public prenatal clinics and postpartum units and by newspaper advertisements and bulletin board postings in four urban locations: Baltimore, Detroit, Springfield, Mass., and Tampa, Fla. Physical and emotional abuse were measured using the ISA and a semistructured interview that also asked about abuse experienced by the women. Factor analysis revealed three factors instead of the two (physical abuse and nonphysical abuse) resulting from the initial factor validity analysis for the ISA. Results demonstrated the need to evaluate critically the extent to which instruments developed to measure spouse or partner abuse are valid and reliable when used with diverse groups that may not have been represented in the samples used for the development and validation work for the initial instrument. Tables and 45 references (Author abstract modified)