The influence of measurement on interpreting accurate graphing during an asynchronous training.

Abstract

Asynchronous trainings can increase the accuracy of graphing single-case design data. Researchers, however, have measured accuracy using a permanent product when employing asynchronous trainings for graphing. Permanent product measures are insensitive to the graphing process (i.e., adherence to the training), steps of this process yielding intangible outcomes, and steps of this process yielding tangible outcomes that are indiscernible through viewing a final output. A continuous measure of accuracy would account for the graphing process better and further measure the efficacy of the training and the use of an asynchronous training modality overall. Professionals also create graphs at the inception of data collection such that these graphs update automatically with ongoing data entry. A continuous measure of accuracy could further highlight errors during specific steps of the graphing process for which professionals can intervene to mitigate uninterpretable or misinterpreted data likely to affect clinical programming. We replicated the efficacy of enhanced written instructions as an asynchronous training for graphing single-case design data for all seven participants. We evaluated accurate graphing using two permanent product measures from the existing literature and a novel, continuous measure. Both permanent product measures generally overestimated accuracy in the absence of training. We also present data on latency to graph completion, maintenance, and generalization before discussing the implications of measurement on interpreting the outcomes of asynchronous trainings.