Evaluation of strategies for building a research culture--an empirical case study at an African university.

Introduction

In a previous publication (Studman 2003a), the first author of this article described a variety of factors involved in the development of a new Office of Research and Development at the University of Botswana. Although in a developing country, which consists mainly of the Kalahari Desert, the University of Botswana has received relatively strong financial government support since its establishment in 1982, and it has experienced dramatic growth in the number of applications for admission from students eligible for tertiary education. Thus, by 2005 there were approximately 15,000 equivalent fulltime students.

The factors that supported the financial well-being of the country and the consequent demand for tertiary education included the combination of a stable society, the discovery of diamonds in 1967, a democratic and peaceful electoral system, and generally benevolent governance with low corruption. However, in recent years, economic pressures, such as the demands on government funding for the civil service, education and other services, have forced the government, to exert more control over expenditure, including restricting the level of support for the university, while still requiring it to accept increasing numbers of students. As a result, between 1997 and 2003, the overall student-staff ratio deteriorated from 12: l to 16:1. In practice, due to staff vacancies, the figure was often around 19:l,

As an institution with a vision for academic excellence (University of Botswana, 2003), the university recognised and acknowledged the principles of research-led teaching (Hattie & Marsh, 1996; Geiger, 1993; Lipset, 1994; Pratt, 1997; Zubrick, 2000), despite its predominantly undergraduate teaching history. The role of research in national development was also recognised (Studman, 2003b). However, in the late 1990s the university also recognised that its research activity was not satisfactory, and so set about improving the situation. It shared the problems of many other predominantly undergraduate institutions as described by Hazelkorn (2002).

Studman (2003a) outlined changes introduced to develop the research culture at the university. An analysis of the strengths, weaknesses, opportunities and threats (SWOT) of a given situation was conducted (SWOT analysis is a commonly used strategy to understand any situation). The key challenge areas identified were: l) no strategic planning or alignment of research with university goals and strategies; 2) poor use of internal funds; 3) an absence of accountability for resources; 4) no management of the quality of outputs; 5) no structure for commercialisation of research; 6) limited postgraduate research; 7) insufficient motivation for some staff; 8) administratively complex research procedures, but no effective research support structure; 9) increasing teaching workloads; 10) insufficient training in research management, methodology, and communication; 11) no database of research capabilities, and few reported research outputs; and 12) lack of funding source information. In addition, some staff preferred private consultancy to research for financial reasons, sometimes at the expense of their teaching responsibilities. Clearly, major changes were required.

After prioritisation, and after assessing the available capability of the Office of Research and Development staff, strategic changes introduced initially included: 1) development of research policy; 2) recovery and utilisation of internal funding through simplified, transparent procedures; 3) introduction of a quality and accountability management programme; 4) introduction of encouragements to undertake research; and 5) training in research proposal writing.

As recommended by Drummond (2003), we developed a plan to evaluate the effectiveness of changes. While an ultimate measure of success in expanding research is an increase in the number of research outputs (i.e., papers, books, presentations, patents), it is too early for the changes described in this paper to be fully realized. Moreover, as Ramsden (1994) has pointed out, many factors may affect research outputs. Other recognised measures such as the Frascati system (an international standard for assessing performance in research and experimental development which was developed by the Organization for Economic Co-operation and Development at a meeting in Frascati, Italy [OECD], 2002) were deemed to be inappropriate to the current level of the university's development. Also, due to the challenges listed later in this paper, complete data are not yet available. Therefore, we opted for a longitudinal study of staff attitudes toward research as a clearer indication of the impact of changes introduced by the research office.

Intellectual challenges exist with the assessment of some of these goals. The perception of quality management in a university context is still a challenging concept, misunderstood by academics and management alike (Houston & Studman, 2001). Internationally the Frascati system has been largely adopted as a measure of research activity and development (OECD, 2002). The Association of Commonwealth Universities has also developed benchmarking procedures for evaluating research offices (Waugaman, 2004; Kirkland & Day, 2005).

Limited support was also provided to enable staff to identify external research funding opportunities. Other desirable changes, such as the development of postgraduate research studies; publicity on research activity; management of consultancies; and policies on intellectual property, ethics and research centres, were developed but delayed for various reasons until 2005.

Development of Research Policy

A research policy was developed and approved in 2002 (University of Botswana, 2002; Studman, 2004). The policy was written in a format that enabled an evaluation of compliance. It was given high priority and developed in harmony with the university's overall strategic goals, as recommended by Drummond (2003).

The policy was designed to be straight-forward and relatively short, It established the basic aims of the university with regard to research, and emphasized those areas where growth was desired. The policy was then circulated by e-mail throughout the university, and went through the normal approval procedures. At the University of Botswana, this was a lengthy process involving several committees, from departmental level to senate and, finally, to the university council. This process typically takes around two years. The research policy was no exception. Therefore, it was necessary to utilize the policy as a working document for decisionmaking even before it could be approved. The policy indicated that the Office of Research and Development would be responsible for implementation, and that the guidelines would be placed in the university handbook. In this way, the practical aspects of policy implementation could be undertaken simply by using a document that could be changed relatively easily, without seeking faculty, senate and council approval. Once the policy was approved, attempts were made to familiarise staff with its content. Few academics can be expected to find time to read a research policy, so we decided to remind staff continually about the conditions and aims of the policy. Electronic media, meetings with faculty boards and faculty executives, individual consultations, and reports to senior management all served the purpose. It was essential to refer to the research policy frequently in discussions with staff so that gradually they became familiar with its terms.

Recovery and Utilisation of Internal Funding through Simplified, Transparent Procedures

The majority of research funds were being allocated to faculties on a per capita basis. Faculties were using their own procedures for approval and allocation of funds. In many cases, these procedures were obscure, poorly advertised, and often excessively bureaucratic and complex. As a result, most faculties were stockpiling research funds in internal accounts. With the deans' agreement, early in 2001 all unused research funds were returned from faculties to a central funding pool. In addition, a review of all existing research projects was initiated, and funds in inactive accounts were also returned to the central funding pool. In this way, almost P3 million (US $800,000), or roughly four years of internal funding, was recovered. The per capita system was abandoned. To meet faculty demands for discretionary research finding, some funds (roughly P 600,000 in total) were then redistributed to faculties based on 1.3 times the total funding each faculty had allocated in the previous year. The message to faculties was clear: use the resources or lose them. After the first two years of operation under this system, the faculty component was calculated according to the number of reported research outputs. Both methods were unpopular with some deans.