Date of this Version
This dissertation involved the application of functional analysis methodology to replacement behaviors (i.e., academic responding). Participants were exposed to the typical school-based functional analysis conditions – (a) teacher attention, (b) peer attention, and (c) escape – in addition to a control condition; yet, replacement behavior (i.e., academic responding) was reinforced across conditions instead of problem behaviors. Two functional analyses were conducted using identical contingencies while measuring condition impact on disruptive behavior, academic engagement, and academic performance (i.e., problems completed, digits correct). Unknown math problems were used during the first functional analysis, and a second functional analysis incorporated antecedent instructional sessions, resulting in the use of known materials. In an extended analysis, both versions of the single-most effective condition from the previous functional analyses (with and without prior instruction) were compared. A multi-element design was used to evaluate the relative effects of functional analysis conditions, with and without instruction on participant performance.
Conditions were implemented with a high degree of integrity, and results demonstrated that the functional analysis with unknown problems produced undifferentiated patterns of responding across participants; however, the functional analysis with known problems resulted in differentiated patterns of responding, allowing for identification of controlling variables for all participants. Additionally, the extended analysis replicated the findings of the functional analyses using a novel mathematics operation for all participants. Results are discussed in terms of the conceptual underpinnings responsible for the findings obtained, as well as the need for future research to refine and extend the functional analysis of replacement behavior.
Advisor: Edward J. Daly, III