Mindfulness-Based Stress Reduction (MBSR) is an experientially based group intervention empirically supported to reduce psychological symptomology. Although MBSR has shown to be an effective intervention, little is known about which facets of the intervention are important in producing positive outcomes. This study tested several aspects of mindfulness practice (total practice duration, practice frequency and practice quality) with the primary focus being on validating (i.e., predictive and convergent validity) a new measure of mindfulness practice quality (PQ-M). The PQ-M fit a two-factor solution via a Maximum Likelihood Exploratory Factor Analysis (n=99). Using longitudinal multilevel modeling on a smaller subsample (n=19), preliminary support was found for changes in practice quality over the course of the MBSR intervention. Further, change in practice quality was associated with improvements in psychological symptoms. While this study was exploratory, these findings suggest that practice quality is a relevant factor to promote positive outcomes and may guide mindfulness instructors in providing highly tailored interventions.