Any teacher and any professional has to rely on other professionals when searching for best practices and evidence based strategies. I have a strong opinion on published research that is not based on observable behaviors. Unfortunately, many authors rely on questionnaires when conducting their research yelding unreliable results. We should adapt strict review standards based on scientific methods. As a teacher, I review many articles in search for best methodology for my students. Here is an example how I screen articles I read:
In this article, the authors analyzed responsiveness of students with autism to peer mediated pivotal response training. Two subjects were 10 years old diagnosed with autism and cognitively on a level of 3 years old typically developing children. Peers trainers were taught to implement complex pivotal response training strategies to these two subjects. Some elements of this complex package included modeling appropriate social behaviors, encouraging conversations, or teaching responsivity to multiple cues. General composition of pivotal response training was inspired by Koegl et al. (1989).
Analysis of the article based on seven dimensions of Applied Behavior Analysis (Baer, Wolf, & Risley, 1968):
Applied: Yes. Children with autism have deficit in simple and in complex social behaviors as documented in applied literature.
Behavioral: Yes. The autorhers choose observable and measurable behavior for this study. They observed the behavior instead of giving questionnaires to third parties. Their dependent variables were complex and were displayed in several areas: time of engagement in maintaining interactions and initiations of play, joint attention, frequency of spoken words and sentences during the intervals of the study, etc.
Analytic: Yes. Functional control was established by detailed description of methods used: a reliable functional relation of peer training with accessible manual and original resources – upon request, demonstrated believability of the study.
Technological: Yes – contingent on review of original resources. The authors refer in their methods to the training manual. The manual is not a part of the published study but is available upon request. Contingent on review of this document, replicability could be determined.
Conceptually Systematic: Yes. This dimension of applied behavior analysis was clearly demonstrated in basic description of Peer training. Some of the features of specific procedures related to basic principles were principles of natural reinforcement, effective establishing operations (varied choices of activities), and modeling.
Effective: Yes. The graphs in figure 1 and 2 demonstrate improvement to a practical degree. The study demonstrates significant effectiveness even with children with very limited verbal and cognitive abilities. Generality: Partially, yes. The strategies were implemented with minimal adult supervision and very effective even after the study was concluded. During the post study, or follow-up maintenance phase, subjects exhibited somehow lower levels of engagement intervals comparing to the study sessions but were still on much higher level than during the baseline. These complex interactions generalized to some degree to untrained peers (Bob – Larry).
Conclusion: this article is a source of inspiring strategy to use in my classroom. It’s effectiveness is demonstrated by reliable and scientific methodology.
Source: Pierce, K. & Schriebman, L. (1995). Increasing Complex Social Behaviors in Children with
Autism: Effects of Peer-Implemented Pivotal Response Training. Journal of Applied Behavior Analysis,28(3),285-295.
Katarina Radi 