An Initial Evaluation of an Assessment Method for the PEAK Relational System Direct Training Module.

Abstract

When considering the best available treatment, it is crucial that assessments yield valid and reliable measures to produce effective treatment options. Currently, this research is limited, giving behavior analysis a platform to evaluate the psychometric properties and content validity of assessment tools used across settings. One major issue that practicing behavior analysts face is conducting the most comprehensive assessment within the time constraints put in place by insurance companies. Utilizing a method of assessment that includes indirect and direct descriptive methods and experimental manipulations could aid in cutting down assessment time, especially if those methods have known correspondence with each other. The purpose of the present study was to assess the components of the Promoting the Emergence of Advanced Knowledge Relational Training System: Direct Training Module (PEAK-DT) for children with autism. More specifically, this study evaluated the correspondence between the PEAK indirect assessment (PEAK-IA) and PEAK preassessment (PEAK-PA) for the Direct Training Module. Comparisons were also made to determine which method offers the best predictive validity of actual performance on the PEAK-DT module. Results indicate that PEAK-IA completed by parents and PEAK-PA share moderate correspondence, with the PEAK-PA offering the strongest predictive validity of direct testing outcomes. Implications for behavior-analytic practice, as well as directions for future research, are discussed.