Tablet-based activity schedule for children with autism in mainstream environment

Reviewer: Zachary Alstad

Educational technology is meeting the demands of integrating students with autism into mainstream classrooms in surprising and new ways. Interestingly, there is a precedent of successful iPad-based interventions for students with autism [1]. In this paper, the authors attempt a novel approach by implementing activity schedules as a tablet-based tool to be used in a mainstream classroom for students with autism spectrum disorders (ASD). Several interesting guiding principles are given for the development of tablet-based interventions directed at school-aged children with autism. These include the concurrent development of reading skills, brevity, concrete rather than abstract examples, tangible progress indicators, and the avoidance of audio features (in order to not be disruptive in a classroom setting). Visual cues are also generally preferable to audio cues in this population, as audio information may be problematic both in production and comprehension for students with autism [2]. While overall the authors make several interesting points, issues surrounding subject compliance over the course of the study remain conspicuously unaddressed, as are individual differences with regard to student participation. Students with moderate autism may exhibit vastly different responses to this type of regimented intervention. While the tool is intended to offload some of the time that teachers spend managing ASD student activities, the process of teaching the student the new tool and ensuring persistent use may call into question the practicality of such a system. While the researchers may have demonstrated some effectiveness in this subset of ASD students, broader implementation into more diverse schools (particularly in the US) may not prove to be as effective. This issue is further complicated by focusing only on students with moderate autism. Future research may explore a similar sample that focuses on students diagnosed with a level one diagnosis as well. Furthermore, there are some aspects of the research model that are potentially problematic. A total sample of ten students raises questions with regard to statistical power and generalizability. The authors attempt to account for this through the use of nonparametric measures; however, the Mann-Whitney U or the Wilcoxon (tests for between and within factor differences, respectively) are not robust to both irregularity of variance and problems related to small numbers of subjects in each factor. A power analysis was not performed, and without a power analysis it is impossible to know the number of participants needed to reliably detect an effect of a certain size. Thus, the use of these statistical tools may have been inappropriately implemented. Despite some oversights in the model, this paper provides an interesting solution to the complex problem of facilitating engagement of autistic students situated in mainstream classrooms. To build on this research in the future, inclusion of a more varied population of ASD students and a more deliberate exploration of the weaknesses of these tools would benefit the broader research space. Online Computing Reviews Service