Utilization of Service-Learning Pedagogy in Participatory Community Action Research in Homeless Shelters: Randomized Study of Student Outcomes

Date of Award


Degree Name

M.A. in Clinical Psychology


Department of Psychology


Advisor: Roger Reeb


Few studies utilize experimental designs, representing a need for further research in the service-learning research. The present study employed an experimental design in which sections of a course (Engaged Scholarship for Homelessness: A Service-Learning Course) were randomly assigned to (a) direct service-learning (N = 16) or (b) indirect service-learning (N = 13). The students in the direct service-learning worked with guests in a homeless shelter while students in the indirect service-learning worked on the development of a Community Re-Engagement Resource Book, a book containing resources for people experiencing homelessness in the Dayton area. Several civic-related student outcomes were examined, including community service self-efficacy, awareness of privilege and oppression, and stigmatizing attitudes. The study examined two general hypotheses: (a) students in both sections would show improvements on these civicrelated outcomes and (b) the students in the direct service-learning section would show greater improvement on these civic-related outcomes. For each dependent variable, a 2 x 2 ANOVA was conducted with one between-subjects factor (direct service-learning vs. indirect service-learning students) and one within-subjects (repeated measures) factor iv (pre- vs. post-semester measures). Hypotheses were rooted in theory, including Bandura's (Bandura 1977, 1982, 1995) self-efficacy theory and Allport's (1954) Intergroup Contact Theory. In general, the results demonstrated that both groups of students showed significant improvements in community service self-efficacy and awareness of privilege and oppression; qualitative data (written reflections by students throughout the semester) supported this finding. However, the hypothesis that students in direct service-learning would show greater change was generally not supported. Further, the hypotheses regarding changes in social stigma were not supported by quantitative data, but some evidence of decreases in social stigma was evident in qualitative data. Methodological limitations may have played a role in obtaining some unexpected findings. These limitations, as well as directions for future research to address these limitations and also explore other important components of this ongoing program of research, are discussed.


Clinical Psychology, Psychology, service-learning, service learning, civic responsibility, social stigma, community service self-efficacy, experiential learning

Rights Statement

Copyright © 2020, author