I. Introduction

In a challenging economy, centralized graduate schools are sometimes asked to justify their existence. Our university has asked the graduate school to analyze the differences between centralized and decentralized administration of graduate education. My purpose in this paper is to provide four empirical ways through which universities can estimate the true financial impact of adopting alternative models of graduate education organization. These methods are useful as periodically, and especially in times when university budgets are in crises, recommendations are made to decentralize the administration of graduate education by eliminating the Graduate School and distribute the functions it performs to other academic and administrative units. The motivation behind such recommendations is typically the perceived cost saving that would result from such shift in structure. In a paper on “Value of the Graduate College” Carole Beere¹ wrote: “Often elimination [of the Graduate College] is fiscally motivated; someone believes that the university will save money by eliminating the graduate school. Sometimes elimination is motivated by the belief that there are more efficient models for serving the graduate enterprise.”² Beere’s purpose in that paper was to “document and demonstrate the value of the graduate college and the graduate dean [so that] we can save our institutions from the disruption associated with dissolving and reinstituting the graduate college.”

The savings realized by eliminating the centralized Graduate School appear at first blush to be the entire budget of the Graduate School. The true savings, however, are considerably less, or may even be negative, for there are costs to be incurred by other units in the university which must absorb the functions that had been performed by the Graduate School, and these must be accounted for.

In this paper we complement Beere’s analysis by providing empirical ways through which universities can estimate the net financial gain or loss that would result from alternative models of graduate education organization. I am not aware of any published methodologies to measure these costs.  In this paper I propose and report on four types of analyses, using different methods, for calculating the cost implications of eliminating the Graduate School and distributing its functions to the colleges, programs and other administrative units in the University. The four methods are:

1. Cost Allocation method: This is based on an analysis of the University costs completed by the Business Affairs division and submitted to the Federal government for the purpose of negotiating the indirect cost rates, as well as data from University budget.

2. Functions Analysis method: This method is based on an analysis of the functions performed by the staff of the Graduate School and the resources that would be needed to perform those functions by the colleges, programs and other administrative units of the University.

3. Comparative Data method: Using data from other universities we compare the staffing requirements of universities that have centralized versus decentralized models for administering graduate education.

4. Extrapolation method: In this method we use data from the experience of the College of Business and Economics at our university of providing administrative infrastructure to support the MBA program, to extrapolate and derive an estimate of the needed resources of a decentralized administration.

All four methods use data that are typically available at all universities. Indeed, the data used for illustration in this paper are real data from my university and were produced in response to a request to estimate the cost savings that would result from eliminating the Graduate School and distributing its functions to other units. Collecting the data needed and performing the calculations should not require more than a few hours.

It is my hope that other graduate deans will be able, with appropriate adjustments, to generate cost estimates of alternative graduate school structures for their universities utilizing some of these methods.

< Back          Summary of Findings >