Visual analysis and data-based decisions
Interpret graphs by considering level, trend, variability, overlap, immediacy, and implementation context before changing intervention.
How this shows up in scenario questions
- 1Avoid declaring effect when intervention points overlap baseline.
- 2Notice baseline trend before attributing change to intervention.
- 3Use treatment-integrity data when outcomes worsen.
Common misconceptions
- Making permanent decisions from one or two data points.
- Treating a graph as proof of function.
- Ignoring baseline variability.
Distractor patterns
- Declare the plan effective immediately.
- Stop collecting data.
- Change intervention before checking integrity.
Related terms
Related practice prompts
A BCBA reviews data showing an increase in aggression after a new intervention began. Direct observation shows staff rarely prompt the replacement response and deliver reinforcement inconsistently. What should the BCBA do next?
A line graph shows problem behavior was stable at 8 to 10 responses per session during baseline. After intervention, the first three data points are 9, 7, and 8. What is the best interpretation?
Two observers collect frequency data on repeating questions during requesting help in a early-intervention session. Observer 1 records 14 responses and Observer 2 records 9 responses for the same session. What should the BCBA do with this information? The team wants the next step to be defensible from the current data.