Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Joseph C. Witt


Variables influencing teacher preference for and actual use of two service delivery options, consultation and referral for evaluation, have previously been studied in isolation using varying methodologies. In this study, several variables including teacher attributions, organizational characteristics, child characteristics, and classroom behavior were studied in a comprehensive format. The relationship between these variables and teacher outcome expectancies, preference for services, and use of consultation versus referral was investigated. Sixty-seven teachers (grades K-8) seeking assistance for a student with behavior problems participated in the study. Teachers completed a demographic questionnaire, and measures of attributions, perceptions of problems, outcome expectancies, and preference for services. Information about organizational variables in the school where the referring teachers worked was also collected. Faculty at participating schools completed measures about school climate and the process of obtaining help with children exhibiting behavior problems in the school. Data were also collected about the referred child's classroom behavior. Information about the child's off-task behavior, and disruptiveness to teacher and peers was collected by a trained observer over three observations. Finally, information was collected about teacher and school referral frequency, final case outcomes, and teacher willingness to attempt interventions in the classroom. Results suggested that variables studied here are differentially important when considering outcome expectancies, preference for services, and actual case outcomes. First, when considering teacher beliefs about expected outcomes of each service delivery option, teacher attributions about his/her control over the problem behavior and the severity of the problem were found to be important. Second, when investigating factors that influence teacher choices for the optimal service in a given case, teacher attributions about the child's ability to control his/her own behavior, severity of the problem, and classroom behavior were found to be most important. Finally, in predicting actual case outcomes, the following variables were found to be significant: teacher willingness to help, and organizational variables, including school climate and staff perceptions of access to and efficiency of consultation services. Future studies are needed in order to further our understanding of conditions under which teachers prefer and use one form of service delivery over another.