ABA glossary
Multiple-baseline design
Definition
A single-case design in which baselines run at the same time across two or more behaviors, settings, or people (tiers), and the intervention is introduced to the tiers one at a time, in a staggered sequence.
In plain words
A multiple-baseline design shows that the intervention, not something else, produced the change, without having to take the intervention away. Each tier starts in baseline. The intervention begins in the first tier while the others stay in baseline. If only the treated tier changes, and each untreated tier changes only when it receives the intervention, the pattern is hard to explain by anything other than the intervention.
Example from a session
A BCBA evaluates a social-skills program across three students. All three start baseline at the same time. The program begins with Mia while Leo and Ana stay in baseline; Mia's initiations rise, and the others stay flat. A week later the program begins for Leo, then for Ana, and each student's initiations rise only after their own start date. The RBTs keep baseline conditions unchanged for students who have not started yet.
For BCBA candidates and supervisors
Tiers must be independent (treating one should not change the others) yet similar enough that the same intervention can affect each. Analysts introduce the intervention to the next tier only after the treated tier has changed and the untreated tiers remain stable. Two tiers is the minimum, but three or more give a much more convincing demonstration and are expected by many single-case quality standards. Variations include multiple-baseline across behaviors, settings, or subjects, the multiple-probe design (intermittent baseline measurement for long or reactive baselines), and nonconcurrent multiple baselines, which are weaker because tiers are not measured at the same time. Delayed treatment for later tiers can be an ethical concern.
Often confused with reversal design, changing-criterion design
A multiple-baseline design shows control by staggering the start of the intervention across tiers, with no withdrawal. A reversal design withdraws and reintroduces the intervention on one behavior. A changing-criterion design changes one behavior in graded steps by shifting the criterion.
Exam tip
Look for "staggered" introduction across behaviors, settings, or people. If the behavior is irreversible or withdrawing treatment would be unethical, a multiple-baseline design is usually the best choice.
Can you spot multiple-baseline design in a scenario?
Practice BCBA D.7 with exam-style questions and instant feedback.