A-B Design
Also called: AB design, pre-post comparison
Definition
A two-phase arrangement — baseline followed by treatment — with no withdrawal or staggering. It shows whether behavior changed after treatment began, but because nothing rules out other explanations it does not demonstrate a functional relation. It is common in practice and weak as evidence.
In plain language
Measure before, then measure during treatment. It shows a change but cannot prove the treatment caused it.
Example
Two weeks of baseline data on tantrums, then the plan starts and tantrums fall. Encouraging, but not yet proof.
Not a-b design
The same plan withdrawn and reintroduced with tantrums rising and falling in step. The added phases make it a reversal design.
See also
Written from
- kazdin-2011
Written from Kazdin (2011). No textbook glossary consulted.
This entry has not yet been through clinical review. Treat it as a draft.