Generalization of Social Skills Training on Disruptive Classroom Behavior

Children who exhibit problematic levels of disruptive behavior frequently also present with social skills deficits and poor social relationships. The degree to which children establish and maintain interpersonal relationships is known to predict critical psychological outcomes in adulthood. Thus, social skills training (SST) is a frequently used treatment approach to teach or improve prosocial skills as appropriate replacement behaviors for inappropriate disruptive classroom behavior. However, many skills learned in SST often do not generalize to non-training settings (e.g., classroom) without actively programming for setting generalization. The goal of this study was to evaluate individual generalization procedures implemented by teachers directly in the classroom. The present study used an alternating treatments design to compare social skills training (SST) alone with three teacher-facilitated behavioral strategies to promote generalization. These included: 1) brief direct instruction of social skills with a visual prompt (i.e., positively-stated social skills rules visibly posted in the classroom), 2) verbal prompts, and 3) contingent reinforcement for the demonstration of social skills. Appropriate reinforcers were chosen from results on a preference assessment and teacher interviews. Four second-grade male students, referred for excessive disruptive behavior and poor social relationships, participated in this study. Students were pulled from their classrooms twice weekly to receive SST throughout the study. Students received each generalization component in a rapidly alternating fashion and treatment conditions were counterbalanced among participants. Following the alternating treatments phase, generalization procedures were removed in a withdrawal phase (while SST was ongoing) and the most effective procedure was then re-implemented to verify that behavior change was a function of the treatment condition. Effectiveness of each treatment was determined by visual analysis and standardized mean difference effect sizes using data from direct observations of classroom disruptive behavior. Pre- and posttest ratings of students' conduct problems and social skills were assessed via teacher ratings. Finally, acceptability of each treatment was evaluated by teachers using the Intervention Rating Profile-15 (IRP-15). Contingent reinforcement resulted in the largest decrease in disruptive behavior (d = 3.92) for all participants. Verbal prompting was somewhat effective (d = 1.38), but visual prompting (d = 0.25) had limited effectiveness. Additionally, SST alone was ineffective in producing a behavior change that generalized to the classroom (d = 0.04). As a result of the entire treatment package, conduct problems on the SESBI-R decreased slightly and social skills increased slightly on the SSiS. Teachers rated each procedure as acceptable, with contingent reinforcement most acceptable. Limitations include limited external validity and some variability in baseline conditions. This study demonstrates the importance of implementing similar reinforcement contingencies in the non-training environment.

Department dc:contributor.department

Psychology: School Psychology

Grantor dc:publisher

East Carolina University

Year dc:date.issued

2013

Author dc:creator

• Scott, Emma

Advisor dc:contributor.advisor

• Kazmerski, Jennifer S.

• Psychology
• Behavioral sciences
• Alternating treatments design
• Contingent reinforcement
• Disruptive behavior
• Generalization
• Social skills training

Handle dc:identifier.uri

http://hdl.handle.net/10342/4340

OAI identifier oai:identifier

oai:thescholarship.ecu.edu:10342/4340