ABA Exam CoachLessons · D. Experimental Design
D.4 · Baseline logic: predict, verify, replicate
How do you prove a treatment worked with just one client and no control group? Single-case designs do it with a three-step logic: predict, verify, replicate. Once that clicks, a lot of design questions get much easier.
Key takeaway
In single-case designs, each person serves as their own control, and behavior is measured again and again in every condition. A stable baseline gives you a prediction. Showing that the untreated behavior would have stayed the same gives verification, for example through a return to baseline or untreated tiers that don't change. Producing the effect again gives replication. In an A-B-A-B design, the second A verifies and the second B replicates.
In single-case designs, each person serves as their own control, and behavior is measured again and again in every condition. A stable baseline gives you a prediction. Showing that the untreated behavior would have stayed the same gives verification, for example through a return to baseline or untreated tiers that don't change. Producing the effect again gives replication. In an A-B-A-B design, the second A verifies and the second B replicates.
Members get the full lesson video, notes and a 5-question test
The video walks through the idea, worked examples, the trap people fall into and an exam-style question. The notes and the checkpoint come with it, and your results feed your Where to focus plan.