Date of Award
Doctor of Philosophy
Jonathan C. Baker, Ph.D.
R. Wayne Fuqua, Ph.D.
Stephanie M. Peterson, Ph.D.
Blair P. Lloyd, Ph.D.
Contingency strength, functional behavior assessment, sequential analysis, structured descriptive assessment, telehealth
The spread of SARS-CoV-2 and acceleration of coronavirus disease cases resulted in a global pandemic and presented unique and unprecedented humanitarian issues. Behavior analysts trained in the area of functional behavior assessment offer important services that may assist families to address behavioral challenges that they face during this difficult time. Despite the growing body of evidence of telehealth functional behavior assessments, practitioners still face several unknowns. In Experiment I, a telehealth functional behavior assessment was conducted with three children with autism spectrum disorder. Assessment methods represented a continuum of experimental control from unstructured conditions (no experimental control), structured descriptive assessment conditions (experimental control of A-B relations), and standardized functional analysis conditions (experimental control of A-B-C relations). Caregivers conducted all face-to-face procedures with synchronous coaching from the remote BCBA. A multielement design was used to compare unstructured observation conditions to structured descriptive assessment conditions. A second multielement design was used to compare standardized functional analysis (Iwata et al., 1982 / 1984) conditions. Timed-event recording was utilized to collect behavioral data. The degree of correspondence between the nonexperimental and experimental methods was assessed by agreement through visual analysis and descriptive statistics. A behavioral function of challenging behavior was identified for 3/3 participants using the standardized functional analysis. A behavioral function was identified for 1/3 participants using a descriptive assessment. There was correspondence between the descriptive assessment and functional analysis for 1/3 participants. Caregivers rated the acceptability of unstructured conditions lower than all structured descriptive assessment conditions and between 75% (3/4) and 100% (4/4) of all functional analysis conditions. Results contribute to the growing body of evidence that supports the utility of telehealth and caregiver-implemented FBA methods. The lack of correspondence between the descriptive assessment and functional analysis is discussed.
Experiment II was an inductive methodological investigation. Behavior sampling methods from Experiment I were expanded to capture and quantify environment-behavior relationships versus behavior-only data across all FBA conditions. Four different dynamic probability calculations were compared across behavior sampling methods and contingency table construction. Estimates were highly correlated within a calculation at t = 1, 2, 5, 10 second intervals and concurrent, lag, and concurrent + lag contingency table constructions. There was a strong positive monotonic relationship between risk difference and Pence et al. (2009). Result interpretation of both absolute and relative degrees of sequential association varied across calculations. Additional research is necessary to improve the precision of analysis methods and sensitivity of assessment practices.
Reuter-Yuill, Lilith Michaele, "Telehealth Functional Behavior Assessment in Practice: A Demonstration of the Clinical Utility of Dynamic Probability Analyses and Estimates of Contingency Strength" (2023). Dissertations. 3968.