In 2001, Dance for Parkinson's disease (DfPD(®)) classes for persons with Parkinson's disease and care partners were developed by Brooklyn Parkinson Group and Mark Morris Dance Group. A previous assessment suggested that individuals experience positive benefits from DfPD(®). The current preliminary uncontrolled study investigated the effects of a dance intervention on several motor and quality of life aspects of PD following 16 sessions (8 weeks; 20 h) taught by professional dancers/teachers. A mixed methods design was used to determine the effects of the class. Assessment instruments administered at baseline and post-intervention included the Hoehn and Yahr, UPDRS (part III), Berg Balance Scale, Beck Depression Inventory, and PDQ-39 and individual interviews after the last class. Hoehn and Yahr scores ranged from 1 to 4. UPDRS III total scores and sub scores of gait and tremor improved following the intervention (P < 0.05). During interviews participants reported physical, emotional, and social benefits. Despite the diversity of baseline measures post-class interview results were consistently positive across the sample. Twelve of 14 subjects (mean age 66.2) with idiopathic PD completed the sessions. After 4 years, four participants regularly attended DfPD(®) classes. The low attrition rate and continued attendance suggest notable adherence to the DfPD(®) class. The importance of the results is both clinical and conceptual, highlighting the value of using both quantitative and qualitative data to evaluate the benefits of dance with PD.