Despite high rates of aggression on female psychiatric inpatient units and research to suggest that risk factors for violence may be different for men and women, violence risk assessment instruments are typically developed and validated through research on male populations. The current study tested a female-specific modification to an existing risk assessment instrument, the Dynamic Appraisal of Situational Aggression: Women’s Version (DASA:WV). The modification involved addition of two factors taken from the Female Additional Manual used in conjunction with the HCR-20 (Covert/Manipulative Behaviour and Low Self-Esteem) and a rating of ward atmosphere to the original DASA. Nursing staff on a high secure female forensic unit rated patients on the DASA:WV at the end of each shift and recorded incidents of verbal aggression and physical aggression against others, objects, and self in the subsequent 24-hr period. Hierarchical (multilevel) logistic regression was employed in the statistical analysis to account for the incorporation of multiple repeated measures for each participant. While the evidence for the predictive validity of the DASA for aggression in a female forensic population was strengthened, the hypothesis that the DASA:WV would significantly improve predictive validity for female patients was not supported.