Recent studies have shown that quantitative instrumental measurements are more sensitive than clinical rating scales to subclinical dyskinesia and parkinsonism. We therefore hypothesized that an instrumental assessment would be more sensitive to the presence of dyskinetic and parkinsonian movements than the Abnormal Involuntary Movement Scale (AIMS), the Dyskinesia Identification Scale, Condensed User Version (DISCUS), and the Simpson-Angus Scale (SAS). We also hypothesized that the DISCUS, by virtue of its more detailed protocol, would be more sensitive than the AIMS. Using blinded raters, we compared the clinical rating scales with instrumental measurements in 100 patients referred to a movement disorders clinic. We collected demographic data, risk factors for tardive dyskinesia, current medication use, Axis I and III disorders, and an estimate of cognitive functioning using the Mini-Mental Status Examination. There was no significant difference between the AIM and the DISCUS in the identification of dyskinesia. However, an instrumental assessment revealed a significantly greater prevalence of dyskinesia. The Mini-Mental Status Examination was the most prominent predictor of both instrumental and clinical measurements of parkinsonian and dyskinetic movements. It appears that even trained raters, utilizing standard rating scales, may underestimate the prevalence of some motor abnormalities. Instrumental ratings may be helpful to both the clinician and investigator, particularly when abnormal movements are not clinically obvious. The relationship between cognitive impairment and motor abnormalities remains an important area for further research.