Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Joseph C. Witt


Independent descriptive and experimental analyses as well as preferences assessments, were conducted. Descriptive analyses were conducted to identify baseline levels of off-task behavior, and to systematically describe the co-variation between off-task behaviors and peer attention, teacher attention, and the instructional task. Data on off-task behavior were analyzed by computing conditional probabilities associated with maintaining stimuli (e.g., peer attention). Intervention development was based on a simple contingency reversal. Experimental analyses were conducted to examine the extent to which off-task behavior was related to task difficulty or to consequences (i.e., peer attention) that were systematically programmed by the experimenter. Hypotheses and intervention development were developed in a manner similar to the interventions used for the descriptive analyses. In addition to developing interventions based upon descriptive and experimental analyses, an intervention for each participant was derived from a reinforcer preference assessment. The procedures were applied to five children between the ages of 6 and 11 years. The results showed that interventions derived from any of the three assessments were effective. For 4 of the 5 students there was little difference between interventions, derived from different assessment methods, and applied to the same child. When a difference in intervention efficacy was present, the intervention derived from the preference assessment generally was more effective than interventions derived from descriptive and experimental analyses. Results are discussed in terms of costs and benefits of various assessment procedures.