Reducing Behavior Problems Among Students with Autism Spectrum Disorder: Coaching Teachers in a Mixed-Reality Setting

Background: The growing number of children diagnosed with an Autism Spectrum Disorder (ASD) indicates that school personnel need to be equipped with the appropriate skills to address the far-reaching needs of these students. Children with moderate to severe ASD are at risk for exhibiting behavior problems including aggression and self-injurious behaviors, which can interfere with learning and impact student safety (Munson et al., 2008; Sullivan & Bradshaw, 2012). These behavior problems prove to be challenging for teachers to manage; yet, relatively few teachers have training in the use of evidence-based interventions for educating students while addressing behavior problems in the classroom (Scheuermann et al., 2003; Shyman, 2012).

Objectives: This study aims to fill these gaps, by exploring the extent to which a teacher-tailored coaching approach that utilizes a state-of-the-art mixed-reality simulator was associated with increased teacher capacity to address the diverse behavioral needs of students with ASD. Targeting the practices of teachers serving these students is consistent with the public health approach to prevention (O’Connell et al., 2009) and is likely an effective and cost-efficient way of addressing the complex needs of multiple children with ASD simultaneously.

Methods: Specifically, 19 teachers in two non-public special education settings serving students with ASD were provided with the intervention. Coaches provided data regarding the amount of time they specifically engaged in each coaching activity. Both teachers and coaches provided ratings regarding the acceptability of the intervention. Classroom observations were conducted by an external observer three times (baseline, post-test, and follow-up). The Assessing School Settings: Interactions of Students and Teachers(ASSIST; Rusby et al., 2001) classroom observations were conducted by two research assistants. The ASSIST includes event-based tallies (i.e., counts of specific behaviors) for teacher use of classroom management strategies  and student behavior. In addition, the observer provided responses to Likert-type items regarding teacher classroom management, student behavior, and the engagement of students in meaningful participation.

Results: A repeated measure MANOVA demonstrated a significant measure by time effect. Follow-up repeated measures ANOVAs revealed statistically significant improvements in teacher management over time, as well as some evidence of improvements in student behavior. Specifically, observers tallied increasing levels of teacher use of proactive behavioral expectations F (2, 28) = 6.727, p = .004 and use of approvals F (2, 28) = 8.123, p = .002. Decreases in student non-compliance were also tallied F (2, 28) = 3.584, p = .041. Observer ratings of the classroom environment, using Likert-type scales, also improved with regard to teacher proactive behavior management, F (2, 28) = 6.921, p = .004, teacher monitoring, F (2, 28) = 14.096, p = .000, teacher control, F (2, 28) = 17.109, p = .000, and teacher and student meaningful participation, F (2, 28) = 9.814, p = .001. There were also marginally significant improvement over time in student socially disruptive behaviors approached significance, F (2, 28) = 3.188, p = .057.

Conclusions: Implications for future research and the practice of teachers in special education settings serving students with an ASD are discussed.