Assessment & Research

Tutorial: Applying Machine Learning in Behavioral Research

Turgeon et al. (2020) · Perspectives on Behavior Science 2020
★ The Verdict

You can start predicting parent engagement today using the same free machine-learning scripts you use for graphing.

✓ Read this if BCBAs who run parent-training webinars or collect any small data set.
✗ Skip if Clinicians who only use paper forms and have no interest in computers.

01Research in Context

01

What this study did

Turgeon et al. (2020) wrote a how-to guide. They show BCBAs every click needed to run machine-learning tools on small data sets.

The authors use a parent web-training project as the worked example. No new experiment—just the recipe.

02

What they found

The paper does not report new results. It simply proves you can open free software today and predict which parents will finish your online modules.

03

How this fits with other research

Cox et al. (2025) took the next step. They fed live operant data into a Q-learning model and hit 95% accuracy guessing the next response. Their study extends the 2020 tutorial into real-time prediction.

Becraft et al. (2020) offers a sister guide. Instead of machine learning, they walk readers through multilevel meta-analysis of single-case AB data. Both papers lower the tech wall for practitioners.

Li et al. (2018) also teach advanced stats, but they favor Bayesian models. The trio—Turgeon, Becraft, Li—form a toolbox: pick random forest, multilevel, or Bayesian to match your question.

04

Why it matters

You no longer need a data-science partner. Download the free code, swap in your variables, and flag families at risk of dropout before the first Zoom session. One hour of setup can save weeks of chasing no-shows.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Open the tutorial’s GitHub link, run the random-forest script on last quarter’s parent login data, and see who gets a low engagement score.

02At a glance

Intervention
not applicable
Design
methodology paper
Finding
not reported

03Original abstract

Machine-learning algorithms hold promise for revolutionizing how educators and clinicians make decisions. However, researchers in behavior analysis have been slow to adopt this methodology to further develop their understanding of human behavior and improve the application of the science to problems of applied significance. One potential explanation for the scarcity of research is that machine learning is not typically taught as part of training programs in behavior analysis. This tutorial aims to address this barrier by promoting increased research using machine learning in behavior analysis. We present how to apply the random forest, support vector machine, stochastic gradient descent, and k-nearest neighbors algorithms on a small dataset to better identify parents of children with autism who would benefit from a behavior analytic interactive web training. These step-by-step applications should allow researchers to implement machine-learning algorithms with novel research questions and datasets.

Perspectives on Behavior Science, 2020 · doi:10.1007/s40614-020-00270-y