Comparing computer-based training and lecture formats to teach visual analysis of baseline-treatment graphs

Accurate analysis of data is vital to the validation of interventions. As such, there has been a recent increase in studies evaluating visual analysis training procedures. However, past investigations have not evaluated direct and indirect visual analysis training methods with matched instructional content that was systematically designed. Furthermore, training has rarely included assessment of generalization and maintenance of visual analysis skills. The purpose of the current dissertation study was to compare the effectiveness and efficiency of (a) computer-based training, (b) lecture formats with and (c) without the opportunity to pause, and (d) a no-training group to teach visual analysis of AB graphs to university students. To make these formats directly comparable, the instructional content was equated by ensuring information and examples were identical across the three training procedures. Eighty-three students were randomly assigned to one of the four groups. Results showed that all three training formats produced increases in accurate responding compared to the no-training group. Visual analysis skills generalized to novel graphs and maintained approximately 2 weeks following all trainings. These results suggest that structured approaches that are carefully designed to train visual analysis are effective and lead to gains that generalize and maintain in the absence of training.