Adaptive Multiple Baseline Design — Adaptive SCED
Adaptive Multiple Baseline Single-Case Experimental Design · Also known as: adaptive MBD, flexible multiple baseline design, adaptive SCED multiple baseline, data-driven multiple baseline design
The Adaptive Multiple Baseline Design is a single-case experimental design that applies the standard multiple baseline logic — staggering intervention onset across two or more tiers (behaviors, settings, or participants) — but allows phase-change decisions to be guided by ongoing data review rather than fixed, pre-specified schedules. This flexibility makes the design more responsive to participant variability while preserving the core replication logic that supports causal inference.
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 an adaptive multiple baseline design when you are studying a behavior, skill, or outcome in one or a small number of participants and want to test whether an intervention causes change, but fixed-duration baselines are impractical or ethically problematic (e.g., withholding treatment for a fixed period regardless of the participant's state). It is particularly well-suited to behavioral intervention research, special education, clinical psychology, and rehabilitation where participant variability is high and individualized timing is important. Do NOT use it when the tiers are not functionally independent — if intervening on one behavior automatically changes another, replication logic fails. Also avoid it when you need group-level generalizations or when the research question requires inferential statistics on population parameters; a randomized controlled trial or group design is more appropriate in those cases.
Strengths & limitations
- Preserves the causal logic of the multiple baseline design — staggered replication across independent tiers supports causal inference without a withdrawal phase.
- Greater ethically practical feasibility than fixed-baseline designs when withholding treatment for a predetermined period is inappropriate.
- Responsive to real participant variability — phase changes are tied to demonstrated stability rather than arbitrary time rules.
- Does not require withdrawal of a beneficial treatment to demonstrate experimental control, unlike reversal (ABAB) designs.
- Can be used with as few as two or three tiers, making it accessible for small-n and practice-based research.
- Requires tiers that are genuinely functionally independent; if behaviors, settings, or participants influence one another, the control logic is compromised.
- Adaptive decision rules must be pre-specified; post-hoc adjustments open the door to researcher bias and inflate the apparent evidence for effectiveness.
- Replication across only two or three tiers provides weaker causal evidence than four or more; confidence in causality increases with the number of replications.
- Does not yield group-level effect estimates or statistically generalizable conclusions; findings speak to the individuals studied.
Frequently asked
How is the adaptive multiple baseline design different from a standard multiple baseline design?
In both designs, the intervention is introduced sequentially across independent tiers and the staggered replication pattern provides evidence of causality. The difference is in timing: a standard multiple baseline design fixes the baseline duration in advance, whereas the adaptive version uses pre-specified data-based decision rules (e.g., stability criteria for trend and variability) to determine when each tier transitions to intervention. This makes the adaptive version more flexible and often more ethically justifiable, but also requires more careful pre-registration of the decision rules.
What counts as a good phase-change decision rule?
A widely used criterion is three to five consecutive data points within a predefined stability band (e.g., no more than 15–20% variability around the mean, no clear trend in the direction of expected intervention effects). The rule should be written down before data collection begins and applied consistently. Some researchers also require a minimum number of baseline sessions regardless of stability to prevent extremely short baselines.
How many tiers do I need for causal inference?
The field generally regards three tiers as the minimum for credible causal inference in a multiple baseline design. Two tiers are sometimes used in pilot work but provide limited replication. Four or more tiers substantially strengthen the argument. The What Works Clearinghouse (WWC) standards for single-case designs specify requirements around minimum data points per phase and minimum number of replications.
Can I combine the adaptive multiple baseline design with statistical analysis?
Yes. Visual analysis is the primary tool, but effect-size statistics developed for single-case data — such as Tau-U, the non-overlap of all pairs (NAP), or the standardized mean difference for single-case designs — can complement visual inspection and are increasingly required by journals and systematic reviewers. These statistics should be chosen and reported transparently alongside visual data.
Is this design appropriate for group research questions?
No. The adaptive multiple baseline design is a single-case (small-n) design; it addresses whether the intervention worked for the specific individuals studied. If your question is about the average effect in a defined population or about statistical generalization across a sample, a randomized controlled trial or group quasi-experiment is the appropriate design.
Sources
- Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Some current dimensions of applied behavior analysis. Journal of Applied Behavior Analysis, 1(1), 91–97. DOI: 10.1901/jaba.1968.1-91 ↗
- Kratochwill, T. R., & Levin, J. R. (Eds.). (2010). Single-Case Intervention Research: Methodological and Statistical Advances. American Psychological Association. ISBN: 978-1433808838
How to cite this page
ScholarGate. (2026, June 3). Adaptive Multiple Baseline Single-Case Experimental Design. ScholarGate. https://scholargate.app/en/experimental-design/adaptive-multiple-baseline-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.
- AB DesignExperimental design↔ compare
- ABA DesignExperimental design↔ compare
- ABAB designExperimental design↔ compare
- Adaptive ExperimentExperimental design↔ compare
- Multiple Baseline DesignExperimental design↔ compare
- Single-Subject Experimental DesignExperimental design↔ compare