Accessible Summary: What is known on the subject: Research suggests that the Dynamic Appraisal of Situational Aggression (DASA) is a useful risk assessment instrument to identify individuals who might be at risk of aggression in mental health inpatient units.Although, risk assessment research has typically focused on an individual's risk of aggression, recent research has begun exploring whether the DASA could be used to assess the likelihood that a group of patients would be aggressive. What the paper adds to existing knowledge: While the DASA was useful for assessing whether an individual was likely to be aggressive, the group average score was not a useful indicator for the likelihood of aggression once the individual DASA score was taken into consideration.Unexpectedly, patients who were assessed as high risk on the DASA were more likely to be aggressive on settled units compared to unsettled units, which included other individuals whose risk was elevated. What are the implications for practice: There is not enough evidence to suggest that the group DASA average improves the identification of aggression above the individual DASA score. Introduction: The Dynamic Appraisal of Situational Aggression (DASA) is an inpatient aggression risk assessment instrument. Recently, research explored whether the unit atmosphere, as indicated by a unit's average DASA score, was related to inpatient aggression risk, but failed to control for individual risk. Aim: Investigate whether the DASA unit average score or an interaction between the unit average and an individual patient's DASA score was related to the likelihood that an individual would act aggressively. Method: Cox regression with repeated assessments and recurrent events was used to analyse 11,243 DASA risk assessments of 113 inpatients collected via retrospective file review. Results: The unit DASA average score was not related to aggression towards staff. There was a negative interaction between the individual and the unit DASA average scores when identifying patient‐to‐patient aggression; high‐risk patients engaged in less aggression when the unit average was heightened relative to units with lower DASA average scores. Discussion: It is possible that there were more nursing interventions and/or patients engaged in greater self‐regulation on unsettled units, thus reducing aggression. Implications for Practice: Currently, there is insufficient evidence to suggest that the unit average score should be used to supplement individual DASA scores to identify aggression risk. [ABSTRACT FROM AUTHOR]