Objective:Evaluate the measurement properties of the Agitated Behavior Scale (ABS) using rating scale analysis.Samples:Sample 1: 900 observations of 100 individuals with traumatic brain injury; Sample 2: 204 observations of 102 persons with dementia; Sample 3: 241 observations of 6 individuals with anoxia.Results:The calibration indicated that the rating scale was used as intended. The hierarchies of item difficulty were similar across samples. Person and item separation values were within the acceptable range for the TBI sample. Generally, the items work well together, however 3 items misfit the measurement model moderately.Conclusions:Agitation as measured by the ABS is best represented as a unitary construct. Results provide additional support for the reliability and validity of the ABS.