D.6 Critique and interpret data from single-case experimental designs

Understanding data from single-case experimental designs (SCEDs) helps parents and teachers evaluate whether an intervention is effective. Here's how to interpret and critique these designs:

🔑Key Points to Consider When Reviewing SCED Data:🔑

• Trend: Is the behavior improving, worsening, or staying the same over time?

• Level: Did the behavior change immediately after introducing the intervention?

• Variability: How consistent is the behavior during each phase?

Common Designs and How to Analyze Them:

Withdrawal/Reversal Design

What to Look For: Does the behavior change when the intervention is introduced and return to baseline when it’s removed? Example: If a reward system is removed and the behavior worsens, it suggests the reward was effective.

Multiple Baseline Design

What to Look For: Do behaviors change only when the intervention is introduced in each phase? Example: If raising hands improves only after the intervention starts, it supports its effectiveness.

Multiple Survey Design

What It Is: Similar to a Multiple Baseline Design, but baseline data is collected sporadically rather than continuously.

What to Look For: Do behavior changes align with the introduction of the intervention despite less frequent data collection? Example: A teacher records how often a student finishes homework independently on random days instead of every day during the baseline phase.

Multielement Design

What to Look For: Which intervention leads to better outcomes?Example: If verbal praise consistently leads to more focus than stickers, it’s the preferred strategy.

Changing Criterion Design

What to Look For: Does behavior improve step-by-step as the goals change?Example: If a child successfully meets each new goal, the intervention is likely effective.

How to Critique the Data:

• Is the Design Appropriate? Ensure the chosen design matches the goal (e.g., don't use reversal for skills a child can’t unlearn).

• Is There Enough Data? Look for clear patterns over multiple sessions.

• Are There Confounding Variables? Consider whether other factors (like changes in routine) could affect results.