Predictive Model of Intentional and Unintentional Outcomes: An Evaluation Tool for Outcomes-Based Funding in Texas Higher Education

Document Type

Dissertation

Degree Name

Doctor of Education (Ed.D)

Department

Curriculum and Instruction

Date of Award

Spring 2014

Abstract

Designed to incentivize institutions to attain predetermined goals, the accepted business practice of pay for performance continues to gain ground in American public higher education. Texas will soon join states such as Indiana, Ohio, and Tennessee with the implementation of a proposed plan for allocating 10% of undergraduate higher education funding to institutions based on performance on predetermined outcomes. Whether based solely on the public's demands for higher education to operate within the intuitively comfortable business model or the need to allocate increasingly scarce public resources, legislatures are responding with programs that tie funding to institutional performance. Performance funding programs, such as the outcomes-based funding program proposed by Texas, continue to hold the attention of the public and state policy makers. However, they do so largely in the absence of empirical evidence supporting the effectiveness of such funding policies. The primary purpose of this nonexperimental, quantitative study was to attempt the development of a predictive model of change in higher education appropriation allocations. Secondary purposes included an exploration of variations in funding allocations among institutions with different missions, urban-centric locales, enrollment mixes, and enrollment trajectories, and the viability of a multilevel modeling approach. Data from archival public databases such as the Texas Higher Education Coordinating Board's Accountability System and the Integrated Postsecondary Education Data System maintained by the National Center for Education Statistics were collected on 34 general academic institutions (GAIs) over the period 2003-2011. A multilevel conceptual model was proposed. Analysis of variance and multilevel regression were employed to create a model designed to predict changes in institutional funding for GAIs. By creating a model that controls for state- and institution-level variables unrelated to the proposed outcomes-based funding plan, I hoped to establish funding allocation benchmarks to which actual changes in funding subsequent to the implementation of the plan may be compared. The results of the study may provide institutions and policymakers with access to a framework for establishing predictive models designed to assess the effectiveness of performance plans on actual institutional funding.

Advisor

Joyce A. Scott

Subject Categories

Curriculum and Instruction | Education | Higher Education

Link to Full Text

Share

COinS