top of page
  • Writer's pictureAbigail Blackman

Employee Turnover

Our field is in a staffing crisis as a number of organizations are struggling with workplace instability. Turnover is defined as the separation of an employee from the organization (SHRM, 2015). This installment will cover predictors of turnover and ways to mitigate it. 

Predictors of technician turnover

Kazemi et al. (2015) surveyed behavior technicians and found that support from their supervisors and training were predictors of turnover. Strouse and DiGennaro Reed (2022) provide some other suggestions for reasons for turnover in the I/DD workforce - low pay, work schedules, ineffective hiring practices, and lack of effective training. Cymbal et al. (2022) reviewed BHCOE data to examine predictors of turnover. They found parent satisfaction, supervisor turnover, and behavior technician wage to impact behavior technician turnover. 

Predictors of supervisor turnover

You may be wondering the extent to which supervisor turnover is an issue in our field. I was too! My colleagues and I administered a survey and found that BCBA turnover does occur in our field and several variables contribute to their turnover. These variables include burnout, mentorship, and pay, among others. The overarching theme of the data were that supervisors lack support from upper management and advanced learning opportunities (Blackman et al., under review). See the poster (below) presented at KansABA and ABAI with more details below. 

Ways to mitigate turnover

Strouse and DiGennaro Reed (2022) discuss the importance of organizations looking at their:

  1. staffing models

  2. pay structures

  3. recruitment and hiring practices

  4. training practices

Cymbal et al. (2022) recommends:

  1. targeting behavior technician turnover

  2. enhancing caregiver satisfaction

  3. examining pay structures

  4. focusing on job satisfaction

Wine (2020) recommends:

  1. Calculating turnover rates: Unfortunately, there is no standard calculation in the field (Wine, 2020). Wine discusses the different calculations in the literature to date. I recommend reading the article if you are interested!

  2. Calculating turnover cost: Experts suggest turnover may cost organizations between $5 - 10K dollars per employee. There are a number of variables that determine the cost of turnover for an organization. Behavior Science Technology developed an employee lifetime value calculator that helps organizations determine the time at which they recuperate the costs of hiring and training an employee. Check it out to look at the potential cost of turnover at your organization and when employees begin to be profitable for you.

  3. Conducting a separation assessment: After conducting the separation assessment, Wine (2020) suggests categorizing answers based on good, neutral, and bad turnover. An example of good turnover is the behavior technician leaving to go back to college to become a BCBA. An example of neutral turnover is the behavior technician is moving out of the state. An example of bad turnover is the behavior technician is leaving for a competitor. Below is an example of a flowchart from Wine’s paper that may help organizations adopt processes to address the turnover they are experiencing.

Data from BSTperform

We’ve begun doing analyses to determine whether we can make correlations between fidelity and retention. To date the data suggest that providers should be observed in accordance with the 4-80 rule. Meaning, at least 4 supervisory observations per month with a mean score of 80% procedural fidelity should be the minimal standard. When I say 4 supervisory observations, I do not mean four contacts. Rather, collecting data on the fidelity of 4 different programs that the provider is implementing with clients. This can be done at one time during a single supervisory observation that occurs for 30 minutes or could be broken down across the month into smaller time frames. It does not take long to capture this critical data!  

The data below display the percentage of providers that stayed at their organization 3 months after a months worth of supervisory observations were completed. As you can see, providers that were observed more 4 or more times per month, regardless of their fidelity score, stayed with their organization for at least one subsequent quarter. Providers who were observed less than 4 times per month, but their fidelity scores on the evaluations completed were 80% or higher, were 93% likely to stay with their organization for at least one subsequent quarter. For providers who were observed less than 4 times per month and scores were less than 80%, it was up to chance whether they would stay for a subsequent quarter.

So not only does our large-scale fidelity impact quality, but it also shows that those observed in accordance with the 4-80 rule were 20x more likely to stay at their organization, than their less-frequently observed and poorly-performing peers. Providers performing at 80% or higher were 12x more likely to stay with their organization. And providers observed at least 4 times per month were 17x more likely to stay with their organizations. 

As noted before, the data depicted in the table and below are based on whether the provider stays for one quarter. We extrapolated those data to determine what the annual turnover rate would be for the respective categories. This is helpful to answer the question about how much of a difference is it if 93% or 98% of providers staff for a quarter? As you can see, only 6% of providers who are observed in accordance with the 4-80 rule turnover each year. While, 14% of providers turnover each year if they are observed 4 times per month but their scores are less than 80%. 25% of providers turnover each year if they are observed less than four times per year but their fidelity is high. And 93% of those observed less than 4 times per month with low fidelity scores turnover each year. 

This appears to be the greatest predictor of turnover we've observed. And this is important because we all know that continuity of care impacts quality.


Blackman, A. L., DiGennaro Reed, F. D., Ruby, S. A., & Wine, B. (in prep). Turnover survey.

Cymbal, D. J., Litvak, S., Wilder, D. A., & Burns, G. N. (2022). An examination of variables that predict turnover, staff and caregiver satisfaction in behavior-analytic organizations. Journal of Organizational Behavior Management, 42(1), 36-55.

Kazemi, E., Shapiro, M. & Kavner, A. (2015). Predictors of intention to turnover in behavior technicians working with individuals with autism spectrum disorder. Research in Autism Spectrum Disorders, 17, 106-115.

Soceity for Human Resource Management (2015). How to determine turnover rate.

Strouse, M. C., & DiGennaro Reed, F. D. (2022). Employee turnover and workforce stability. In J. K. Luiselli, R. M. Gardner, F. L. Bird, & H. Maguire (Ed.), Organizational behavior management approaches for intellectual and developmental disabilities (pp. 210-234). Routledge.

Wine, B., Osborne, M. R., & Newcomb, E. T. (2020). On turnover in human services. Behavior Analysis in Practice, 13(2), 492-501.


bottom of page