Pragmatic Multiple Baseline Design — Real-World Single-Case Experimental Design
Pragmatic Multiple Baseline Design · Also known as: PMBD, pragmatic MBD, real-world multiple baseline design, flexible multiple baseline design
The Pragmatic Multiple Baseline Design is a single-case experimental design that staggers intervention introduction across multiple participants, settings, or behaviors in real-world conditions where strict experimental control is impractical. By relaxing some idealized constraints — such as perfectly stable baselines or rigid staggering timelines — it preserves the core logic of the multiple baseline while accommodating clinical, educational, or community realities. It is especially valued when withholding treatment for ethical reasons is untenable and when practitioners need evidence from naturalistic settings.
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 a Pragmatic Multiple Baseline Design when you need causal evidence about an intervention in a naturalistic setting and a randomized controlled trial is infeasible — for example, in clinical practice, special education, or community programs. It is particularly appropriate when withholding treatment from a control group raises ethical concerns, when the sample is too small for group designs, or when individual-level trajectories matter. Avoid it when fewer than two replication tiers are available (a single baseline provides no replication), when outcome data cannot be collected repeatedly and frequently, or when tiers are not functionally independent. If baseline stability is so poor that no meaningful pre-intervention trend can be identified, consider adding more data points or switching to an alternating treatment design.
Strengths & limitations
- Provides causal (internal validity) evidence without requiring a reversal or withdrawal of treatment, which is often unethical.
- Flexible enough to work in real-world clinical, educational, and community settings where strict experimental control is impractical.
- Replication across tiers strengthens the causal inference without inflating sample size requirements.
- Individual-level data reveal heterogeneity in response that group-average designs mask.
- Compatible with small or even single-participant studies where group randomization is impossible.
- Internal validity depends critically on functional independence of tiers; correlated tiers undermine the control logic.
- Relaxing baseline stability standards reduces the clarity of the causal inference compared to the classical multiple baseline.
- Findings from a small number of tiers or participants have limited external generalizability without systematic replication across studies.
- Requires sustained, frequent measurement over an extended period — burdensome in resource-constrained settings.
- Visual analysis is subject to analyst judgment; inter-rater reliability must be established and reported.
Frequently asked
How is a pragmatic multiple baseline different from a standard multiple baseline?
The core logic is identical: stagger intervention introduction across tiers and infer causality from the pattern of change. The pragmatic variant explicitly acknowledges and accommodates real-world constraints — shorter or less stable baselines, flexible staggering timelines, or less frequent measurement — while still requiring the researcher to justify these adaptations and to assess their impact on internal validity.
How many tiers do I need?
A minimum of three tiers is the standard recommendation for adequate replication; two tiers provide weak causal evidence. Adding a fourth or fifth tier strengthens the argument but increases measurement burden. In pragmatic contexts, three tiers with clear staggering and consistent measurement is the realistic target.
What counts as adequate baseline stability in a pragmatic design?
Classic guidelines call for five or more stable data points with no clear trend before introducing the intervention. In pragmatic designs, stability is sometimes impractical. Researchers must document the baseline variability, explain why waiting longer was not feasible, and acknowledge how instability limits the certainty of the causal claim. The goal is transparency, not the pretense of stability.
Can I combine a pragmatic multiple baseline with statistical analysis?
Yes. Visual analysis remains primary, but quantitative effect size measures such as Tau-U, the non-overlap of all pairs (NAP), or multilevel modeling for single-case data can supplement and strengthen the interpretation. Statistical analysis does not replace visual analysis but provides a standardized complement, especially useful for systematic reviews and meta-analyses of single-case studies.
Is this design suitable for a doctoral dissertation?
Yes, with appropriate justification. Many single-case dissertations use multiple baseline designs. You must clearly articulate the pragmatic adaptations you made, explain why the adaptations were necessary, and honestly assess how they affect the strength of your causal inferences. A well-executed pragmatic multiple baseline with three tiers, adequate data density, and documented inter-rater reliability is scientifically defensible.
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 ↗
- Shadish, W. R., & Sullivan, K. J. (2011). Characteristics of single-case designs used to assess intervention effects in 2008. Behavior Research Methods, 43(4), 971–980. DOI: 10.3758/s13428-011-0111-y ↗
How to cite this page
ScholarGate. (2026, June 3). Pragmatic Multiple Baseline Design. ScholarGate. https://scholargate.app/en/experimental-design/pragmatic-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.
- Interrupted Time SeriesCausal inference↔ compare
- Multiple Baseline DesignExperimental design↔ compare
- Randomized Controlled TrialExperimental design↔ compare
- Single-Case Experimental DesignDisability Studies↔ compare