In this study we examined the ability of a modified Sexual Experiences Survey (SES; Koss, Gidycz, & Wisniewski, 1987) to assess sexual victimization among a local community sample of women (n= 1,014). Women who reported sexual victimization were interviewed regarding the most recent incident. Those who responded negatively to all SES items were asked whether they had ever feared they would be sexually assaulted but were not, and to describe that incident. Independent coders read a subset of transcripts (n= 137) and classified each incident as reflecting: one of the SES items, a form of unwanted sex not included on the SES, or not unwanted sex. Coders viewed nearly all incidents elicited by the SES as reflecting some type of unwanted sex. Respondent-coder agreement for rape and coercion incidents was high, but low for contact and attempted rape incidents. The SES scoring continuum, reflecting objective severity of acts, was only modestly associated with subjective trauma associated with rape, attempted rape, coercion, and contact.