Introduction World Bank, 2000 Nchinda, 2002 Simon, 2000 Zeleza, 2003 Caldwell, J, Caldwell, P, & Quiggin, 1989 Sitthi-amorn & Somrongthong, 2000 Costello & Zumla, 2000 Zeleza, 2003 RAWOO, 2002 Sitthi-amorn & Somrongthong, 2000 Nchinda, 2002 Sall, 2003 Sitthi-amorn & Somrongthong, 2000 Pang, Lansang, & Haines, 2002 Ramsay, 2002 Sall, 2003 Zeleza, 2003 Jentsch & Pilley, 2003 Lansang & Dennis, 2004 Lansang & Dennis, 2004 Costello & Zumla, 2000 Jentsch & Pilley, 2003 Nchinda, 2002 Sall, 2003 Sitthi-amorn & Somrongthong, 2000 Zeleza, 2003 Costello & Zumla, 2000 Edejer, 1999 processes Green, 2003 Kwesiga, Mbago, & Chimanikire, 2000 Menken, Blanc, & Lloyd, 2002 Mkandawire, 1998 Sall, 2003 Allen, 1986 Sall, 2003 Mkandawire, 1998 Rossi, 2004 Sall, 2003 Allen, 1986 UNDP/World Bank/WHO, 2003 Simon, 2000 Costello & Zumla, 2000 Nchinda, 2002 Nchinda, 2002 Lansang & Dennis, 2004 Nchinda, 2002 Wight, 2005 Methods Table 1 The interview schedule covered leadership of local social science research, training and career paths, ways of strengthening research capacity and barriers to this. The full schedule is available (as Appendix A). Informal conversations were held with five senior and one junior researcher from the University of Dar es Salaam and National Institute for Medical Research, Tanzania, and nine junior researchers from Makerere and the MRC Programme on AIDS in Uganda. A group discussion was held with four of the Ugandans (three men, one woman), following the same schedule used for the in-depth interviews. The interviews were summarised according to analytical themes. There was considerable concordance in accounts, which were not patterned by gender. Divergent opinions are considered in the Discussion. Interviewees were circulated the main report for comments and to confirm that their views were presented accurately. Four provided comments. Findings Severity of the problem Nearly all those interviewed thought that there is a serious shortage of social science research capacity in East Africa, the few really good social scientists being overworked and overwhelmed with requests for collaboration. Most academic health-related social science research in Uganda was said to be run by Northerners, yet Uganda was thought to have stronger capacity than Kenya, with Tanzania coming third. Particular limitations identified were in qualitative research, analysis and writing skills, and health-related specialisms. Interviewees stated that the vast bulk of social science research in East Africa is commissioned by NGOs or government departments, mostly funded from the North; consequently, it is highly applied and determined by external priorities. The few opportunities for academic research were said to come primarily from Northern researchers who win the funding, resulting in unbalanced collaborations. Research processes: inter-collegiate support Wight, 2005 Pfau & Barton, 2004 RAWOO, 2002 Zeleza, 2003 Research processes: consultancies Kwesiga et al., 2000 … in Makerere you can spend your entire time just working on very well paid, short-term consultancy studies for NGOs, … who want something done in three weeks, and will pay you very well … (Senior researcher, previously Uganda) Extremely low university salaries create a powerful incentive for consultancies. A research associate's salary might be $250/month, while consultancies can pay $100–$250/day. In one research institute consultancies augment salaries from around $400/month to about $5000. A head of department explained: ‘… to rely on your salary would never make ends meet at all.’ Furthermore, in contrast to regular salaries, most researchers can avoid declaring consultancy fees for tax (30%). Research commissioners, predominantly government departments or NGOs, usually seek a contract with individuals, or sometimes consultancy firms, but rarely with university departments. Private consultancy firms, often constituted for a particular brief, usually employ university staff to help with the bid and subsequent research. Commissioning bodies are reportedly unwilling to pay overheads to institutions, and when they do, they are generally very low, e.g., 5–20% in Makerere departments, 20% at the University of Dar es Salaam, and a maximum of 15% at a Kampala independent research centre. The senior management at Makerere were said to encourage departments to become consulting firms and demand 30% overheads, but this leads university staff to work independently, undercutting university departments and earning more. There are no research traditions being developed …. We are social scientists but very few are specialists … (Faculty dean) He will leave you. And who loses? This is the person you have trained up to PhD level, and now he is leaving you, and you have no one to teach … (Head of department) Writing consultancy reports provides little incentive to develop analytical skills. Reports generally involve very tight timetables with little opportunity for peers’ critical input, are descriptive and have limited dissemination (sometimes for internal use only). Several researchers said they do not publish from consultancies because they need the funder's permission, but none knew of it being refused. More plausibly, there is rarely time for such writing. Consequently, the CVs of highly experienced researchers often list numerous consultancy reports but very few journal publications, jeopardising their applications for senior jobs. Consultancies is not building the capacity of the person who is doing it. …. [Some] have even refused scholarships to do PhDs because they were busy doing consultancies. Only two interviewees questioned the inevitability that consultancies detract from publications or teaching: ‘… consultancies can … be a source of writing … [and] training.’ (Director large research programme). much Strengthening research capacity and likely barriers Wight, 2005 Five senior interviewees identified the need to develop writing skills, for instance ‘to guide you through … the very complicated processes … and requirements’ to publish in international journals. Suggestions included experienced and in-experienced staff co-authoring, mentoring systems and support networks. The director of a research-facilitating NGO advocated posts dedicated to writing support, but with salaries adequate to prevent appointees taking on consultancies. The potential advantages of institutional research consultancies were explored, and in particular establishing a norm of significant overheads, e.g., 30%. Everyone approved in principle. Overheads could be used for: libraries, computing and internet access; department-initiated research; disseminating reports; training staff and developing writing skills. Institutional consultancies might facilitate a more collective approach to research and assist management by departmental heads. Furthermore, paying overheads might benefit commissioners since they could require reports to be published, at least in an on-line journal. The culture of institutionalising things is not there. Many think the institution is a barrier to them. And … the bureaucracy, you know, many people would prefer to have the money in their own accounts …. (Head of department) Researchers anticipated the frustrations of inefficient institutional administrations, with long delays in finalising contracts or being paid. It was feared that, since some universities do not allow departmental bank accounts, the central administration might appropriate funds raised through departmental consultancies. Furthermore, fees would not only have to be shared with the institution, but would have to be declared for tax. Consequently, the director of a large programme thought: ‘… people will just try to get around it. They will get consultancies privately.’ All the American universities [have] institutional overheads, but tell DfID that [they] have to be factored in [in Kampala] …: “Oh, no!” …. How am I supposed to run the project without institutional overheads? … they have the mentality that they can do it on the cheap. Africa is poor, but it is not cheap! (Director of large programme) It was also argued that some agencies want to commission specific individual researchers, and that individual consultancies incentivise good work produced on time. Some large donors practice their policies of strengthening institutional capacity by only contracting research through institutions, e.g., the Carnegie and Rockefeller Foundations, the World Bank, and the Swedish and Norwegian development agencies. However, some interviewees thought a concordat with all commissioning agencies to pay minimum overheads would be unrealistic, since East Africa is too dependent on donors. Discussion Carlsson & Wohlgemuth, 1996 Sall, 2003 Zeleza, 2003 Zeleza, 2003 Cleland and Watkins (2006, p. 2) Allen, 1986 Rossi, 2004 Sall, 2003 principle Theobald & Nhlema, in press Zeleza, 2003 By and large, initiatives to strengthen research capacity do not address the issue of research consultancies, although in practice they are in competition for researchers’ commitment. This was clear in an academic research centre sponsored by an international NGO where researchers are prohibited from consultancy work. However, as noted above, some large donors further the development of institutional capacity by only contracting research through institutions, not individuals. Wight, 2005 This has only been an exploratory study. Further research needs to clarify: the scale of individual consultancies across East Africa; whether revising commissioning practices would seriously contribute to research capacity; and, hardly represented here, the views of agencies commissioning consultancies. While the underlying causes of poor research capacity require global economic reform, this study also points to the importance of individually contracted research consultancies in perpetuating the problem. Although they greatly augment meagre university salaries, they also seem to divert university staff from academic research and training the next generation of researchers, stunt the institutional capacity of university departments, restrict the sharing of research findings and perpetuate donors’ control of the research agenda. Commissioning bodies committed to strengthening research capacity should consider devising research contracts, and means to improve university administration, that ameliorate rather than exacerbate the problem. Appendix A Supplementary material doi:10.1016/j.socscimed.2007.07.019 Appendix A Supplementary materilas Online supplimentary materials