BCBA Measurement Common Mistakes
What trips up most candidates on the Measurement, Data Display & Interpretation section.
Using partial interval recording and claiming it measures duration
Partial interval recording overestimates the occurrence of behavior and does not measure duration. If you need to know how long a behavior lasts, use a duration measure. If you need to know how often it occurs, use frequency. Use interval recording only when continuous measurement is impractical.
Misidentifying single-subject designs
A reversal design (ABAB) withdraws and reintroduces the treatment. A multiple baseline introduces the treatment sequentially across different baselines. A multielement rapidly alternates conditions. Know the visual characteristics of each design on a graph.
Concluding a functional relationship from a weak design
An AB design (baseline then treatment) does not demonstrate a functional relationship because there is no replication. At least three demonstrations of effect are needed (e.g., ABAB or three tiers in a multiple baseline).
Ignoring variability in visual analysis
Highly variable data makes it difficult to draw conclusions about treatment effects. If baseline data is unstable, you cannot confidently attribute changes to the intervention. The exam may show you variable data and ask whether a functional relationship has been demonstrated.
Calculating IOA incorrectly
Total count IOA divides the smaller total by the larger total and multiplies by 100. This method can inflate agreement when both observers record different instances. For more precise measurement, use trial-by-trial or interval-by-interval IOA.