South African science is in decline, measured by its output of scientific publications over the past decade, a study has found.
No country "has lost ground as much as South Africa", which has been overtaken by 12 others in terms of research output since 1980, according to a study of databases of the US Institute for Scientific Information.
While the number of local publications increased during the 1980s and 1990s at a compounded rate of 2.4 per cent a year, from 2,432 in 1984 to 3,592 in 2000, world science expanded by 3.7 per cent.
The survey was conducted by Anastassios Pouris, director of the Institute for Technological Innovation at the University of Pretoria, and published this week in the South African Journal of Science . It shows that South Africa's position has dropped to its lowest in two decades.
The country's world share of publications declined from 0.67 per cent in 1987 to 0.49 per cent in 2000, and it was overtaken by Argentina, Brazil, Czech Republic, China, Greece, Hungary, Mexico, New Zealand, Norway, South Korea, Taiwan and Turkey.
There has also been a decline in citations of South African publications across most disciplines, indicating that research quality has dropped at universities, which produce more than 80 per cent of the country's "visible" research.
Out of six areas, only the social sciences and humanities have improved their research performance. Measuring publications in 25 disciplines, 14 showed decline - this was particularly sharp in clinical medicine, materials science and computer science.
Professor Pouris said that research output at South Africa's two top research universities, Cape Town and the Wi****ersrand, had declined while the universities of Natal, Pretoria and Stellenbosch had improved their performances.
He attributed the decline to lack of funding for research and equipment - the latter could be the reason why only social science and humanities research is improving - and the growing pressure that expanding student numbers is placing on the time of academics.
"We don't have earmarked funds for research from government, so research has become a residual part of the system," he told The THES .
Although about 15 per cent of government funding to universities is supposed to be used for research, this does not generally happen: allocations are distributed by institutions and are often siphoned off for other purposes.
The department of education had changed its funding formula, in recognition of the need to allocate research funds in a more targeted way, Nasima Badha, deputy director-general for higher education, said.
The government is to allocate R70,000 (£5,240) for each scientific publication.