This study examined the factor structure, reliability, and validity of the Suicide Probability Scale (SPS) in a sample of 226 (80 male, 146 female) adolescent psychiatric inpatients. Confirmatory factor analyses provided only some support for the original subscales. Exploratory factor analyses revealed some overlap with the original scales, but the factors differed by gender. Internal consistency of the original factors was somewhat better than the factors derived from this sample. A negative change in SPS scores over the course of the inpatient admission was related to risk for readmission due to repeat suicidal behavior.