Association for Behavior Analysis International

Task-analysis instruction is widely used for teaching and learning how to visualize data using computer software. Numerous studies demonstrate the efficacy of specific task analyses for teaching students and behavior analysts how to create graphs; however, little is known about the properties of task-analysis instruction that promote skill acquisition. Findings are reported from two component analyses of the instructional content presented in a computer-based tutorial for creating a reversal-design graph in Microsoft Excel. The first study demonstrated the relative effects on learner performance of describing and presenting images of: (a) stimuli to which a learner must respond, (b) the target responses required to create the graph, and (c) stimulus changes in the graph and the software's graphical user interface that are produced by correct responses. The second study further analyzed the relative effects of presenting text and images on graphing performance. The findings informed the development of a checklist of best practices for designing and evaluating graphing instruction, which will be shared.

Behavior analysts visually analyze graphs to interpret data in order to make data-based decisions. Front-line employees, such as the Registered Behavior Technician, are the forefront of behavioral intervention and responsible for its direct implementation. Though front-line employees implement behavioral interventions on a daily basis, they are not often trained to interpret these data. A clinical decision-making model may aid front- line employees in learning how to interpret graphs. Such training will allow front-line employees to evaluate whether the data suggest a learner is struggling with a prescribed intervention, if variations in treatment implementation may be affecting obtained data, and to identify these potential outcomes to provide optimal and individualized therapy. A multiple-baseline-across-participants design was used to evaluate the effectiveness of a clinical decision-making model on the percentage of correct clinical decisions interpreted from line graphs. All of the participants increased their percentage of correct responses after the introduction of the clinical decision- making model. Two of the 8 participants required additional feedback. The implications of these findings are discussed.