The current study bridges literatures on sexual harassment, person-environment systems, and stress and appraisal processes. Conventional wisdom equates severity of sexual harassment with type of harassment. We test this notion empirically and posit a more comprehensive model that examines both person- and situation-level variables. Data came from 13,743 U.S. Armed Forces women responding to survey questions about a significant experience of sexual harassment. Multiple regression results indicate that pervasiveness of sexual harassment relates outcomes better than does type of sexual harassment. Pervasiveness and type interact to predict subjective appraisal of harassment. Additionally, according to multiple-group structural equation models, appraisal mediates relations between pervasiveness and outcomes. Results further suggest that relations among sexual-harassment antecedents and outcomes are consistent, regardless of the type of sexual harassment. These findings highlight the importance of examining both persons and situations when assessing sexual harassment severity.