The ability to reliably and validly measure client status in Parkinson's disease is essential in the clinical and research settings. Many of the existing measures for PD are not based on a conceptual framework that defines what is being measured, and are not well standardized. This study determined aspects of the reliability and validity of the Bradykinesia Scale (BS) for PD. On two occasions, raters scored videotapes of subjects with and without PD performing the BS. Intra- and inter-rater reliability were high (ICC ≥ .91, p < .01). BS scores discriminated between subjects with and without PD, and between three stages of PD severity ( p ≤ .01). Correlations between BS scores and scores on Unified Parkinson's Disease Rating Scale modules were statistically significant and moderate. The BS is a conceptually-based, reliable, and valid measure for PD, and warrants further standardization and use in this population.