CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 8, September 2012
AFRICA
423
accredited by the Department of Higher Education and Training,
the purpose of the journal must be to disseminate research
results, articles must be peer reviewed, journal contributions
and distribution must be beyond a single institution, the journal
must be published regularly, have an international standard serial
number (ISSN) and have an editorial board. The Department of
Higher Education and Training has also indicated that they plan
to recognise three new journal databases for subsidy purposes
in the future.
A pertinent issue is what constitutes a publication? For
example, how are creative outputs such as artistic performances,
paintings, concerts and novels recognised? Some higher
educational institutions already have internal processes to take
these outputs into account, although they are currently not
recognised by the Department of Higher Education and Training.
However, the Department of Higher Education and Training is at
present wrestling with a procedure to recognise these outputs;
hence this issue will not be discussed in this review. Further
controversies, which are not discussed in this review, are what
constitutes an accredited conference proceeding and the relative
undervaluing of books and chapters in books within the South
African subsidy formula.
Comparisons of the South African system to
encourage research productivity with systems
used in higher educational institutions in
other countries in the world
The system in South Africa is perceived to be unique in that
it is intended to provide direct financial reward to individual
researchers for their outputs.
8,20
Although direct financial benefit
to individuals may be intended, this is often not put into practice
by the higher educational institutions, as will be discussed (see
section titled ‘Distribution of Department of Higher Education
and Training research subsidy within higher educational
institutions in South Africa’).
In comparison, to South Africa’s ‘direct reward system’,
other countries have various different strategies aimed to
stimulate research outputs. For example, in the United States
of America, successful leverage of research support from the
federal government is based on the assessment of an individual’s
research proposal, which incorporates details of his/her research
track record.
2
In the United Kingdom from 2014, obtaining government
funding for research will be based on a system which places
emphasis on an individual researcher’s outputs and the citation of
his/her publications (the Research Excellence Framework, REF).
4
The REF will replace the previous system used in the United
Kingdom in the 1980s, 1990s and 2000s, namely the Research
Assessment Exercise (RAE).
7
The RAE, which was used to
determine the amount of funding provided by the government
to individual institutions, was based upon the credentials of
a limited number of the most prominent researchers in each
academic department within an institution.
7
The South African system is not actually that unique. In
Australia, government funds granted for research are based
on research income, publication counts and higher degrees
completed.
3,5
Importantly, the Australian formula (similar to the
South African NFF) is based solely on quantity, the impacts
of which are discussed below in the section titled ‘Positive
and negative impacts of various approaches used to encourage
research productivity’. As occurs in South Africa, whether
the funding goes to the individual researcher or the institution
appears to differ between the higher degree institutions in
Australia.
3
In Spain, individual researchers are rewarded on the basis
of publications; however a financial bonus is only awarded for
publications in high-impact factor journals.
6
A similar approach
is used in Finland, except the institution rather than the individual
researcher is the recipient of the financial benefits.
1
The merits
of rewarding individuals as opposed to institutions are discussed
under the section titled ‘Financial reward of individuals versus
institution’.
Positive and negative impacts of various
approaches used to encourage research
productivity
In order to encourage research productivity, a measure of
research outputs is mandatory. The common issue that arises is
whether research outputs should be measured purely by counting
(
termed research output audits) or whether an element of
research quality (for example citations or journal impact factor)
should be taken into account. This debate is far from resolved,
however a commentary on the politics of publication, published
in
Nature
in 2003,
9
incited a barrage of correspondence in which
some interesting arguments both for,
6,10,11
and against
12-14
various
measures of research productivity were raised.
Quantity versus quality
A number of criticisms have been raised regarding the potential
negative impact of research output audits on research quality.
3,13
One of the major concerns is an encouragement of researchers
to publish as many papers as possible (salami slicing)
14,22
and to
preferentially choose to publish in those journals which have the
least rigorous review process.
21
Indeed, in South Africa, although there was a 50 to 60%
increase in the number of publications produced by universities
from 1990 to 1994 and 2004 to 2008;
23
only 57% of the
publications that qualified for governmental subsidy in 2007
were published in internationally accredited journals.
23
Although
this percentage has increased to 64% in 2008, 66% in 2009 and
69%
in 2010, this increase is partly attributed to an increase in
the number of South African journals listed in the ISI and IBSS.
24
However, more convincing evidence of an increase in quantity
at the expense of quality is provided by data from a study
conducted on higher educational institutions in Australia.
3,13
In
1995,
the formula by which the Australian government funds
were distributed to higher educational institutions to support
research activities was changed to incorporate publications.
3,13
Consequently, researchers could calculate the financial value
of a publication [A$761 (R6 664.31) to A$1089 (R9 536.71)
between 1995 and 2000].
13
As a consequence of increases in the
total amounts to be distributed this value subsequently increased
to A$3 000 (R26 271.94) in 2002.
13
In her study, Butler
13
allocated all journals from the ISI’s
Science Citation Index into quartiles based on journal impacts
calculated from five-year citation means. She then compared
the Australian Universities’ share of publications within these
