The Minimum Data Set (MDS) is a standardized assessment tool designed to provide a comprehensive biopsychosocial assessment of medical, behavioral, and cognitive status of nursing home residents. This pilot study examined the relationships of three MDS subscales--cognition, depressive symptoms, and behavioral disruptions--to other measures of the same domains (e.g., diagnosed dementia and depression and caregiver ratings on the Revised Memory and Behavior Problems Checklist [RMBPC]). The sample consisted of 135 nursing home residents with a mean age of 84 years. Based on the MDS, there was a high prevalence of cognition-related behaviors but a low prevalence of disruptive and depressed behaviors. The prevalence rates were substantially different according to the RMBPC. In addition, most of the MDS subscales failed to differentiate between residents with and without diagnosed dementia and depression, whereas caregiver ratings on the RMBPC did. The MDS and RMBPC subscales were modestly related but only in residents without dementia. These findings raise questions about the validity of the MDS in measuring nursing home residents' behavior, especially depressive and disruptive behaviors. Thus, caution should be employed in using the MDS as a sole outcome measure for these behaviors, and the use of multiple measures is suggested.