Why RBTs leave: what the BACB exit survey shows and what agencies can change

In February and March 2025 the BACB sent a survey to RBTs whose certification had expired in 2024 and who held no other BACB credential¹. It went to 30,018 RBTs across the United States, from all 50 states; the biggest shares of those invited lived in California (15%), Texas (12%) and Florida (10%)¹. In all, 1,386 answered, a 4.6% response rate¹. The question was which reasons influenced the decision to stop working as an RBT, with “select all that apply”¹. The BACB published the options chosen by at least 25% of respondents¹.

Respondents could choose several reasons, so the shares do not add up to 100%¹. Asked in their own words what would bring them back, they pointed above all to better pay, consistent and predictable scheduling and more professional support, and they also asked for ethical and supportive environments, better supervisor training and more respect for the RBT role¹.

• It is not a turnover rate. It asked only people whose certification had expired in 2024, and the newsletter gives no turnover figure¹. Vendor estimates of RBT turnover circulate, but we do not treat any as the rate because none we found traces to a public dataset with a stated method.
• It describes people who stopped being certified, not RBTs who changed employers. A 2026 survey of behavior analytic practitioners found no overlap between what predicted intending to leave a job and intending to leave the field².
• Most people invited did not answer, and we cannot know whether they would have answered the same way¹.
• Respondents chose from a fixed list, and the BACB shows only options chosen by at least 25%¹. It does not explain what “Not applicable” (34%) means¹.
• RBTs who stayed were not asked, so the reasons do not show what separates leavers from stayers.

These summaries follow the authors’ abstracts. The BACB says its findings align with earlier research, which shows pay, workload and organizational culture as central to RBT turnover¹.

• Florida. Nastasi and colleagues interviewed 11 RBTs about burnout and turnover. Four themes came out: challenges establishing and maintaining competency, difficult working conditions, a transient or unheard-of career, and dissatisfaction with pay and benefits. Experiences differed widely across organizations³. Eleven interviews describe experiences; they cannot say how common each one is.
• Supervisor support. Among 81 therapists in ABA schools, perceived supervisor support played a central role in predicting lower burnout and higher therapeutic self-efficacy⁴. In a survey of 92 professionals, supervisor and collegial support seemed to lower burnout and raise job satisfaction, but the relationships were not statistically significant, while excessive work demands went with more burnout and less satisfaction⁵. The direction is consistent and the evidence is small and mixed; the samples were therapists and professionals in ABA, not RBTs specifically.
• What is still unknown. The authors of a 2020 discussion article write that turnover in behavior analysis is still poorly understood and that few studies have examined why employees leave or how to reduce unwanted turnover⁶. None of the studies cited here tested whether a given retention practice lowers RBT turnover.

The survey covers certifications that expired in 2024 and was fielded in early 2025¹. Since January 1, 2026, RBT supervisors must be BCBAs or BCaBAs¹⁰, and the new two-year recertification cycle brings the 12-PDU requirement¹. We found no later BACB survey of RBTs.

The BACB counted 246,109 certified RBTs at the end of 2025¹¹ and 259,577 on October 1, 2026¹². These are certificants, not necessarily people working as RBTs, and they say nothing about how many leave.

In Florida, the minutes of the ABA Task Force meeting of September 28, 2026 list BCBA and non-BCBA turnover and staff tenure among the staffing measures discussed in a presentation on quality and value-based care¹³. That is a topic discussed, not a rule, but it is one more reason to track your own numbers. This article is not legal or HR advice.