Comparing Continuous and Discontinuous Data Collection during Discrete Trial Teaching of Tacting by Children with Autism

The present study compared continuous and discontinuous data collection systems on acquisition, generalization, and maintenance of tacts, and on‐task behavior, during discrete trial teaching with three children with autism. A constant prompt‐delay procedure was used to teach tacts. Performance across data collection systems was compared using continuous measurement collected daily in the presence of teaching, discontinuous measurement of the first trial only, and discontinuous measurement collected weekly in the absence of teaching (i.e., probe data). An adapted alternating treatments design was used. For two of the three participants, tacts were acquired most rapidly in the continuous measurement condition, although the differences were small. For the third participant, no systematic differences were found in skill acquisition across the three data collection systems. For all three participants, minimal to no differences were found across data collection systems with regard to generalization and maintenance of tacting, and on‐task behavior. Additionally, procedural integrity was high and comparable across all data collection procedures. Overall, these results suggest the use of continuous data collection for teaching tacts to children with autism, as it provides the most sensitive measure of change in performance. Copyright © 2016 John Wiley & Sons, Ltd.