Identifying Cyclical Patterns of Behavior Using Data Smoothing
Author Information
Author(s): Retzlaff Billie J., Craig Andrew R., Owen Todd M., Greer Brian D., O’Donnell Alex, Fisher Wayne W.
Primary Institution: University of Nebraska Medical Center’s Munroe Meyer Institute
Hypothesis
Can data smoothing help identify cyclical patterns in destructive behavior?
Conclusion
Data smoothing can reveal cyclical patterns in destructive behavior that are not apparent through traditional visual analysis.
Supporting Evidence
- Data smoothing revealed cyclical patterns in destructive behavior that were not visible in raw data.
- Both participants showed significant reductions in destructive behavior with treatment, but cyclical patterns persisted.
- Reinforcement-based interventions alone were ineffective for both participants.
Takeaway
This study shows that by averaging behavior data over time, we can see patterns in how often kids act out, which helps us understand their behavior better.
Methodology
The study used a moving-average method to smooth daily rates of destructive behavior for two participants over various time windows.
Limitations
The study did not calculate interobserver agreement for behavior outside of sessions and lacked complete data on medication adherence.
Participant Demographics
Two male participants, aged 8 and 9, diagnosed with various behavioral disorders.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website