ABA Design Lecture Notes

Overview of ABA Design

• ABA Design refers to a type of experimental design used to assess the effects of an intervention on behavior.
• The sequence involves:
  • A Phase: Baseline
  • B Phase: Intervention
  • A Phase: Return to Baseline
• Aim: To determine if the intervention is responsible for changes in behavior.

Steps in ABA Design

1. Define Target Behavior and Treatment

   • Identify the behavior to change and the intervention to use.

2. Collect Baseline Data (A Phase)

   • Collect until data is stable or for a set number of sessions.

3. Implement Intervention (B Phase)

   • Collect data to see if there is a stable response in the desired direction.

4. Return to Baseline (A Phase)

   • Withdraw treatment to see if behavior returns to baseline levels.

Logic of ABA Design

• If improvement occurs during B and reverts during the second A phase, the treatment is considered effective.

Example: Johnny's Case

• Subject: Johnny, a 7-year-old second grader.
• Issue: Difficulty staying on task.
• Intervention: Verbal praise for on-task behavior.
• Outcome:
  • Baseline: On-task 4/20 observations.
  • Intervention: Increased to 15/20 observations.
  • Withdraw: Returned to baseline behavior.

ABAB Design

• Reintroduces the intervention to confirm findings.
• More common than ABA due to enhanced validation.

Prediction, Verification, and Replication in A-B Design

Prediction

• Anticipate if patterns will remain unchanged if the intervention is ineffective.
• Change expected if intervention is effective.

Verification

• Shows that the intervention causes the change.
• Pattern should revert to baseline after removing the intervention.

Replication

• Reintroducing the intervention should yield similar results if it was effective.

Graph Analysis

• Graph 1: Satisfied prediction, verification, replication.
• Graph 2: Prediction and verification met, replication not satisfied.

Conclusion

• ABA and ABAB designs are critical in validating the effects of interventions on target behaviors.
• Ensure that interventions are responsible for observed changes through systematic analysis of prediction, verification, and replication.