Staff turnover is problematic for behavioral health agencies implementing evidence-based practices (EBPs), which are costly and time-consuming. The current study examined the association between EBP training methods and turnover and explored predictors of turnover for different types of staff. Participants (100 clinicians, 50 supervisors, 50 administrators) were randomized to one of three training conditions for an EBP. Results indicated low annual rates of turnover for clinicians, supervisors, and administrators. However, contrary to hypothesis, no statistically significant differences were found in rates of turnover across training conditions. Partially consistent with prior research, organizational climate was a significant predictor of supervisor and administrator turnover at 24 months, but was not a significant predictor of clinician turnover. Implications and future directions for research are discussed.