Association for Behavior Analysis International

In order for children to derive the benefits of evidence-based practices, intervention agents must be taught how to implement them accurately and to monitor progress. The extent to which those treatments are subsequently applied with integrity influences their effectiveness when used with students. This symposium will present research on training procedures for teachers and practitioners and the influence that varying levels of treatment integrity may have on student outcomes. The first study consisted of two experiments that compared the effects of training conducted by experimenters to training conducted by peers on teachers' performance of discrete trial training procedures. In the second study, the effects of behavioral skills training were compared to the effects of training delivered via a traditional lecture format on pre-service teachers' performance of eight evidence-based practices. The third study evaluated the effects of using behavioral skills training to teach graduate students to create single-subject graphs. In the fourth study, the effects of errors in treatment integrity on students' acquisition and durability of self-care skills were examined. Together these empirical investigations offer important implications for training of intervention agents and regarding the impact of teaching errors on student outcomes.

Previous studies have examined pyramidal training and consultant-led training in clinical settings (Haberlin, et. al, 2012). This study compared the effects of one-on-one pyramidal-peer training and experimenter training in the context of weeklong summer trainings for teachers of children with developmental disabilities. In Experiment 1, experimenters taught four teachers to implement one form of discrete trial training (DTT), and then peers taught the four teachers to implement a different form of DTT. In Experiment 2, half of the teachers received experimenter training first and the other half received peer training first. All teachers in Experiment 1 successfully trained their peers. However, in Experiment 2, teachers performed more accurately after being trained by the experimenter than after being trained by their peers. Results showed that more experienced teachers preferred experimenter training whereas less experienced teachers preferred peer training. Results have important implications for the use of peer- versus consultant-led trainings for teachers.

In light of high-stakes teacher accountability and the achievement gap between students with disabilities and their typically developing peers, there is a pressing need to identify effective training methods to equip pre-service special education teachers with evidence-based practices. Research has demonstrated the efficacy of behavioral skills training (BST) as an instructional method with adult learners; however, few studies have examined its use in college teaching environments. The purpose of this study was to evaluate the comparative effects of lectures followed by BST sessions versus lectures followed by study sessions on seven undergraduate pre-service special education teachers' performance of eight evidence-based practices. An alternating treatments design was used to evaluate participants' performance during role-play assessments. Evidence-based practice performance was assessed via role-plays with experimenters acting as students. Results demonstrated that, on average, lectures followed by BST sessions produced substantially higher levels of percentage correct performance than lectures followed by study sessions. These results suggest that it may be valuable for pre-service teacher educators to integrate BST into their college teaching practices. Further research is needed to determine the extent to which skills taught through BST are used in the classroom with actual students.

The ability to create single-subject graphs is a crucial skill for behavior analysts and students in the field of behavior analysis or related disciplines. Microsoft Excel is often the chosen medium by which single-subject design graphs are created. Many studies have used task analyses to teach graduate students how to create single-subject design graphs (Deochand, Costello, & Fuqua, 2015; Dixon et al., 2009; Lo & Starling, 2009); however, this type of passive training may result in prompt-dependent performance. That is, students may come to rely on using task analyses to create graphs, and these task analyses are likely to become outdated as new versions of Excel are developed. Active training procedures may circumvent reliance on such prompts. Behavioral skills training (BST) is an evidenced-based, interactive, competency- and performance-based approach (Parsons, Rollyson, & Reid, 2012) that has been used to teach a wide variety of skills (Himle & Wright, 2014; Homlitas, Rosales, & Candel, 2014; Iwata et al., 2000). In the present study, a multiple probe across behaviors design was used to evaluate the effects of BST on three graduate students graphing skills. Results and implications will be discussed, as well as recommendations to improve the behavior analytic training of graduate students.