Abstract
160 undergraduate students were given a set of 10 photographs or 10 videotaped recordings of women and informed that some of the women were victims of spousal abuse. Half of the participants were asked to select one individual who seemed most likely or least likely to be a battered woman. Participants then sorted another set of photographs or videos of 10 other battered and nonbattered women into the categories according to their guessed status. Analysis showed that certain women were consistently chosen as exemplars of battered or nonbattered women. Categorizations did not correspond with the women's actual status. No reliable differences in category selections or sotrings were found between photographed and videotaped presentations of the battered and nonbattered women.