Chapter 4 of 97% of exam

Experimental Design

Domain D covers variables, validity and the single-case designs used to show functional relations. Items ask which design fits a situation and what a data pattern says about experimental control.

Variables and validity

The independent variable is manipulated; the dependent variable is measured. Internal validity concerns whether the manipulation caused the change; external validity concerns how far the result extends. Single-case designs control threats to internal validity by repeated measurement and replication within the individual.

Independent and dependent variables

The intervention is the independent variable; the measured behavior is the dependent variable.

BACB BCBA Test Content Outline (6th ed.) D.1

Internal versus external validity

Confidence that the intervention produced the effect, versus generality to other people, settings and behaviors.

BACB BCBA Test Content Outline (6th ed.) D.2

Threats to internal validity

Uncontrolled factors such as history, maturation and changes in measurement that could explain a change.

BACB BCBA Test Content Outline (6th ed.) D.3

Single-case logic

Each participant serves as their own control, with repeated measurement. Baseline logic has three parts: prediction, verification and replication. Visual analysis of level, trend and variability within and across phases is how these are judged.

Prediction

A stable baseline lets you predict what behavior would do without intervention.

Sidman, Tactics of Scientific Research (1960)

Verification

Showing that behavior would have stayed at baseline levels, for example by withdrawing the intervention.

BACB BCBA Test Content Outline (6th ed.) D.4

Replication

Repeating the effect, for example by reintroducing the intervention, increases confidence in it.

BACB BCBA Test Content Outline (6th ed.) D.4

Strength over group designs

Effects on each individual are visible instead of being averaged across a group.

BACB BCBA Test Content Outline (6th ed.) D.5

Choosing a design

The outline names reversal, multiple-baseline, multielement and changing-criterion designs. The choice depends on whether behavior can reverse and on what question is being asked. Each design shows control in a different way, so match the design to the behavior and the question.

Reversal

Introduce, withdraw and reintroduce the intervention; requires behavior that can return toward baseline.

BACB BCBA Test Content Outline (6th ed.) D.7

Multiple baseline

Stagger the intervention across behaviors, settings or people; suited to behavior that will not reverse. Change in an untreated tier weakens the demonstration.

Baer, Wolf & Risley, JABA 1 (1968)

Multielement

Rapidly alternate conditions and compare their separate data paths.

Barlow & Hayes, JABA 12 (1979)

Changing criterion

Behavior is shown to track a series of stepwise criteria.

Hartmann & Hall, JABA 9 (1976)

Kinds of analysis

Beyond showing that an intervention works, researchers ask which parts matter, which treatment works better, and how much of a variable is needed. Each asks a different question, so name the question before naming the analysis. They are often run after an effective treatment package has been identified.

Component analysis

Isolates which parts of a treatment package are responsible for its effect.

BACB BCBA Test Content Outline (6th ed.) D.8

Comparative analysis

Compares two or more distinct treatments.

BACB BCBA Test Content Outline (6th ed.) D.8

Parametric analysis

Compares different values of one independent variable, such as reinforcer magnitude.

BACB BCBA Test Content Outline (6th ed.) D.8

Keep going: the full BCBA — Board Certified Behavior Analyst guide covers every section of the exam. BCBA Study Guide — 2026 Edition — PDF + EPUB, $19.99 · 14-day refund →

Practice this topic →← All chapters

Studying in order?

Practice stays free. The full BCBA — Board Certified Behavior Analyst study guide is the material itself, taught start to finish — a downloadable PDF + EPUB you keep.

Get the book — $19.99