Date Approved

4-26-2007

Date Posted

9-19-2013

Degree Type

Open Access Dissertation

Degree Name

Doctor of Education (EdD)

Department or School

Leadership and Counseling

Committee Member

Martha W. Tack, Ph.D., Chair

Committee Member

Elizabeth Broughton, Ed.D.

Committee Member

Patrick Melia, Ph.D.

Committee Member

Yichun Xie, Ph.D.

Abstract

Student retention continues to be a salient issue for administrators and scholars in higher education. For more than 50 years researchers in numerous disciplines, including sociology, psychology, and economic theory, have tried to discover the reasons why students decide to remain in school until graduation. However, retention rates have remained stagnant at about 50%. Serious consequences result when large numbers of students do not graduate (e.g., fiscal appropriations may be reduced). Additionally, an institution’s reputation is created, in part, on its graduation rate and the racial diversity of the student body.

Researchers have recommended finding innovative, interdisciplinary methods to address the “student departure puzzle.” In this pioneering study, a geographic information system (GIS) was innovatively used to develop a habitus retention model based on U.S. Census Bureau socioeconomic and demographic census-tract data. Student addresses were mapped to their census-tract locations to determine whether distance from a university had any relationship to persistence behaviors. In addition, census-tract data were used as a proxy for student and institutional habitus to establish how environmental factors affected retention rates.

This investigation yielded a number of significant findings, especially in regard to females who dropped out of college and students who were still enrolled six years after matriculation. Moreover, habitus and geographic location proved to be important indicators in persistence decisions. The feasibility of using GIS technology for conducting student retention research was confirmed based on the results of this study.

Share

COinS