Analysis of school-based social skills interventions for children with autism spectrum disorders

A meta-analysis of 55 published research studies reveals programs designed to teach social skills to children with autism are failing to meet their goals.

The study, conducted at Indiana University, found that outcomes for social skills training were poor overall, but programs held in normal classroom settings were more likely to result in positive changes than programs held in other environments.

"The results of the meta-analysis are certainly hard to swallow, but they do shed some light on factors that lead to more beneficial social outcomes for children with autism," said lead researcher Scott Bellini, assistant director of IU's Indiana Resource Center for Autism and assistant professor in the School of Education. "These results underscore the critical need for researchers and practitioners to develop more effective social skills programming."

The reviewed studies included a total of 147 students with an autism spectrum disorder, with students ranging in age from preschool to secondary school. The programs aimed to address skills such as group play, joint attention and language usage, or to improve performance of social behaviors, such as initiating interactions, responding to communication and maintaining interactions. Overall, the programs resulted in little change in the targeted behaviors, and students did not apply the skills outside the programs.

However, students receiving social skills programming in their usual classrooms had substantially more favorable outcomes than students who received services in a pull-out setting. Students in classroom-based programs were more likely to engage the targeted skills during the program, and showed a greater tendency to maintain changes in behaviors and to utilize these skills in other settings.

"This finding has important implications for school-based social skill interventions. Teachers and other school personnel should place a premium on selecting social skill interventions that can be reasonably implemented within naturalistic settings," Bellini said.

The study revealed additional potential improvements in the design and implementation of the programs:

• Increase dosage. The programs in the studies failed to provide sufficient amounts of programming, based on current research recommendations. Implementing services more intensely and frequently may lead to better outcomes, Bellini said.
• Match programs to skill deficits. All but one of the 55 studies failed to distinguish between "performance deficits," which refer to skills that are present but not performed, and "skill acquisition deficits," which refer to the absence of a skill or behavior. Targeting the types of skill deficits exhibited by the participants could lead to more successful programs, Bellini said.
• Ensure proper implementation. Only 14 of the studies in the meta-analysis measured whether the program was implemented as designed. "This makes it extremely difficult to conclude whether a social skills program was ineffective because of an ineffectual strategy or because the strategy was implemented poorly," Bellini said.

The study, "A Meta-Analysis of School-Based Social Skills Interventions for Children With Autism Spectrum Disorders," was published in this month's Journal of Remedial and Special Education .

The Indiana Resource Center for Autism is part of the Indiana Institute on Disability and Community at Indiana University and can be found online at http://www.iidc.indiana.edu/irca

http://www.indiana.edu/