Factorial ABAB Design — Factorial Reversal Design
Factorial ABAB Reversal Design · Also known as: factorial reversal design, factorial withdrawal design, multi-factor ABAB design, factorial single-subject reversal
The factorial ABAB design embeds a factorial structure within the classical ABAB reversal framework, enabling a single participant or a small set of participants to experience multiple factor combinations across alternating baseline (A) and treatment (B) phases. By systematically withdrawing and reinstating treatment conditions that vary across two or more factors, the design allows examination of both main effects and interactions at the individual level, providing strong experimental control through within-subject replication.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use the factorial ABAB design when you need to identify which component or combination of components in a multi-factor intervention drives behavior change in an individual participant — and when the target behavior is reversible (i.e., it can safely and ethically return to baseline levels when treatment is withdrawn). It is well-suited to applied behavior analysis, clinical behavioral intervention, special education, and rehabilitation research where mechanistic understanding of multi-component treatments is the goal. Do not use it when the behavior is irreversible after learning (e.g., academic skill acquisition), when withdrawal of treatment poses an ethical risk (e.g., self-injurious behavior at dangerous levels), or when the number of factor combinations is so large that a single-subject ABAB sequence becomes impractically long. In those cases, consider a multiple baseline design or an alternating treatments design.
Strengths & limitations
- Provides strong causal inference through within-subject replication — each A-to-B and B-to-A transition is a built-in experiment.
- Allows examination of factorial main effects and interactions without requiring large samples, making it feasible in clinical and educational settings.
- Continuous measurement across all phases captures the trajectory of change, not just pre-post snapshots.
- Each participant serves as their own control, eliminating most between-subject confounds.
- The factorial structure identifies which treatment components are active, supporting treatment optimization and dismantling research.
- Requires the target behavior to be reversible; behaviours maintained by learning or that carry irreversible health consequences cannot be studied ethically.
- Carryover effects — residual influence of one treatment phase on the next — can threaten the validity of factorial comparisons if withdrawal phases are too short.
- Practical constraints limit the number of factorial conditions that can be tested within a single ABAB sequence; a full factorial with many factors is usually infeasible.
- Findings demonstrate functional relations for the individual participant; generalization to other individuals requires systematic replication across participants and settings.
- Visual analysis, while standard, introduces subjectivity; inter-rater reliability for phase comparisons should be established and reported.
Frequently asked
How is the factorial ABAB design different from a standard ABAB design?
A standard ABAB design tests one treatment condition against a baseline. The factorial ABAB design introduces at least two distinct B-phase conditions that differ on one or more factors, allowing the researcher to compare factor combinations — not just treatment versus no treatment. The reversal logic is identical, but the factorial structure adds the ability to detect main effects and interactions at the individual level.
Can I use the factorial ABAB design with more than one participant?
Yes, and it is advisable. Systematic replication — running the same factorial ABAB sequence with additional participants, ideally counterbalancing the order of B-phase conditions — greatly strengthens external validity. Each participant's data constitute an independent replication, and convergence across replications establishes the generality of the factorial findings.
What if the behavior does not reverse during the A2 phase?
Failure to reverse is informative but limits causal conclusions. It may indicate that the behavior is maintained by factors outside the experiment (e.g., peer reinforcement), that the intervention produced irreversible learning, or that the A2 phase was too short. If reversal is not obtained, the design loses its primary source of experimental control; a multiple baseline design may be a better alternative for behaviours that do not reverse.
How do I report factorial ABAB data?
Standard reporting includes a time-series graph with clearly labelled phase lines for each A and B segment, session-by-session data points, and descriptive statistics (mean and range) per phase. Supplement visual analysis with at least one quantitative non-overlap statistic (e.g., NAP or Tau-U) and report the criterion used to judge phase stability. Make the factorial structure explicit by labelling each B phase with its specific factor combination.
Is the factorial ABAB design ethically acceptable?
It requires careful ethical review. Withdrawing a treatment known or suspected to be effective is only justifiable when the behavior is not dangerous, the withdrawal period is brief and monitored, and the participant or guardian provides informed consent with understanding that treatment will be reinstated. Institutional review boards and behavior analysis ethics codes provide specific guidance; the researcher must document ethical justification for the reversal phases.
Sources
- Kazdin, A. E. (2011). Single-Case Research Designs: Methods for Clinical and Applied Settings (2nd ed.). Oxford University Press. ISBN: 978-0195341881
- Cooper, J. O., Heron, T. E., & Heward, W. L. (2020). Applied Behavior Analysis (3rd ed.). Pearson. ISBN: 978-0134752556
How to cite this page
ScholarGate. (2026, June 3). Factorial ABAB Reversal Design. ScholarGate. https://scholargate.app/en/experimental-design/factorial-abab-design
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- ABA DesignExperimental design↔ compare
- ABAB designExperimental design↔ compare
- Alternating Treatments DesignDisability Studies↔ compare
- Factorial ExperimentExperimental design↔ compare
- Multiple Baseline DesignExperimental design↔ compare
- Single-Subject Experimental DesignExperimental design↔ compare