Single-blind AB Design — Single-blind AB Single-Subject Experimental Design
Single-blind AB Single-Subject Experimental Design · Also known as: assessor-blind AB design, single-masked AB single-case design, observer-blind AB phase design
The single-blind AB design is a single-subject experimental design that combines the two-phase AB structure — a baseline phase (A) followed by an intervention phase (B) — with assessor or observer masking. The individual collecting or rating outcome data is kept unaware of which phase is being measured, preventing knowledge of treatment status from biasing behavioral observations or ratings. The design improves on the standard AB design by reducing detection bias while retaining the practical and ethical advantages of single-subject methodology.
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 single-blind AB design when you need a single-subject baseline-intervention framework that also controls for observer expectancy bias — particularly appropriate when outcome measurement depends on subjective ratings, behavioral coding, or clinical judgment that could be influenced by knowledge of treatment phase. It suits clinical psychology, applied behavior analysis, rehabilitation, and special education contexts where reversal of the intervention is ethically or practically impossible, but where rigorous, unbiased observation is a priority. Ensure the outcome can be reliably measured by a masked assessor; if the behavior is directly self-reported by the participant (who is necessarily aware of the intervention), assessor masking does not address the primary bias source and a fully blinded group design may be needed instead. Do not use when strong causal evidence is required: without a return-to-baseline phase, history and maturation remain alternative explanations for the observed change.
Strengths & limitations
- Reduces observer expectancy and detection bias by masking the assessor to treatment phase, a clear methodological improvement over the unblinded AB design.
- Ethically feasible when withdrawing a beneficial intervention would be inappropriate, since no reversal phase is required.
- Provides an individualized, time-series picture of behavior change that group-design averages would obscure.
- Practical to implement with modest resources: one practitioner, one masked assessor or scorer, and one participant.
- Compatible with video or audio recording strategies that allow retrospective blinded scoring, increasing flexibility in scheduling.
- Weak internal validity for causal inference: the absence of a return-to-baseline phase means that events coinciding with the phase transition cannot be ruled out as alternative causes of change.
- Assessor blinding may be compromised if recorded sessions contain contextual cues (e.g., references to the treatment) that reveal the current phase to the scorer.
- Results are specific to the individual studied; generalization requires systematic replication across multiple participants, settings, or behaviors.
- Participant blinding is neither attempted nor achievable in most AB designs: the participant is typically aware of the intervention, so participant expectancy bias is not controlled.
Frequently asked
How does single-blind differ from double-blind in a single-subject AB design?
In a single-blind AB design, the outcome assessor or observer is masked to the phase (A vs. B), while the treating practitioner and usually the participant are aware of the intervention. A double-blind AB design would additionally mask the participant, which is rarely feasible because participants typically know whether they are receiving a treatment. The single-blind approach is the practical standard for reducing detection bias in single-case research.
How do I keep the assessor blind in practice?
The most reliable method is to record all sessions (video or audio) and randomize or shuffle segments before providing them to the assessor for scoring, omitting all phase labels and timestamps. Alternatively, a separate data collector can enter raw data without phase information, and a different person conducts analysis. After scoring is complete, phase labels are revealed for analysis. Always document and assess whether masking held.
Does assessor blinding fix the weak internal validity of the AB design?
No. Assessor blinding addresses detection bias — the risk that the observer's expectations distort what is recorded — but it does not address history or maturation threats. If an external event happens to coincide with the phase transition, the blinded assessor's data will faithfully record a change that was caused by the event rather than the intervention. For stronger causal inference, upgrade to an ABA or ABAB design, or a multiple-baseline design.
When is a single-blind AB design preferable to a multiple-baseline design?
A multiple-baseline design is stronger because it provides replication across participants, behaviors, or settings. However, it requires multiple data series and is more resource-intensive. A single-blind AB design is preferable when only one participant, behavior, or setting is available, when time constraints preclude staggered baselines, or as a pilot design to evaluate whether a treatment signal is detectable before committing to a larger study.
Can I use statistical methods with blinded AB data?
Yes. Visual analysis of the time series is primary, but effect-size statistics such as Tau-U, which accounts for baseline trend, can supplement the interpretation. Interrupted time-series regression (piecewise regression) models level and slope changes at the phase boundary. These methods require at least eight to ten data points per phase to be reliable and should be specified before unblinding to prevent selective reporting.
Sources
- Kazdin, A. E. (2011). Single-Case Research Designs: Methods for Clinical and Applied Settings (2nd ed.). Oxford University Press. ISBN: 9780195341881
- Kratochwill, T. R., Hitchcock, J., Horner, R. H., Levin, J. R., Odom, S. L., Rindskopf, D. M., & Shadish, W. R. (2010). Single-case designs technical documentation. What Works Clearinghouse. Retrieved from https://ies.ed.gov/ncee/wwc/Docs/ReferenceResources/wwc_scd.pdf link ↗
How to cite this page
ScholarGate. (2026, June 3). Single-blind AB Single-Subject Experimental Design. ScholarGate. https://scholargate.app/en/experimental-design/single-blind-ab-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
- Multiple Baseline DesignExperimental design↔ compare
- Single-blind Randomized Controlled TrialExperimental design↔ compare
- Single-Subject Experimental DesignExperimental design↔ compare