The South African government’s research funding policy has long been criticised by academics. The policy has three major weaknesses:
it offers incentives for output quantity and productivity rather than research quality;
it promotes the practice of dividing research outcomes between articles, thus diluting the impact of the research; and
it penalises work done in large collaborative projects.
Now South Africa’s Department of Higher Education and Training has approved a revised version of the contentious 2003 policy. The Research Outputs Policy 2015 comes into effect from January 1, 2016. It has been welcomed by some academics. They believe it has the potential to introduce considerable changes in how research output funds are awarded.
But will the apparently “new” policy actually just be more of the same?
The status quo
At the moment, a considerable portion of government funding to South African universities is allocated to an individual institution based on its academics’ annual research outputs. These outputs, collected under the umbrella of “publication units”, include peer-reviewed articles, books, book chapters and conference proceedings.
Academics have complained that university management exploits the policy and uses it as a tool for accumulating funds. Some universities use the number of publication units as a key indicator when measuring academic staff members’ performance.
This drives a dangerous culture of “publish or perish” that makes the quantity of research output far more important to an individual academic’s career track than the quality of their work.
The existing policy also doesn’t encourage collaboration. That’s a big problem for scientists who participate in “big science” projects like the European Organisation for Nuclear Research (SA-CERN) or the Square Kilometre Array (SKA). Hundreds of authors feature on the publications that emerge from these collaborations. But these outputs earn zero subsidies under the government’s existing metric.
The updated research policy introduces some important improvements. For instance, it changes the way that scholarly books are subsidised. These books counted for five publication units under the 2003 policy; they are now worth ten units. The humanities and social sciences will mostly benefit from this.
A particularly positive aspect of the 2015 policy is that it highlights research integrity. This is a long overdue response to the “publish or perish” culture, not only in South Africa but elsewhere in the world. Academic dishonesty and plagiarism increase when quantity is more profitable than quality.
In an important step, universities will be expected to take full ownership when it comes to protecting research integrity. Institutions must establish a Research Integrity Committee to ensure they become compliant with the integrity issues raised by the policy. The practice of dividing research outcomes between articles is also strongly discouraged, for the reason that it undermines the integrity of scholarship.
The policy, in its own words:
… aims to support and encourage scholarship. Institutions and academics must remember the importance of research integrity when submitting their claims and are urged to focus on quality research and not maximum accrual of subsidy funds.
In all, the word “quality” features 12 times in the 2015 policy – a big jump from the 3 mentions it received in the 2003 version.
So, the policy talks the “quality” talk. But does it walk the walk? Sadly not.
Big changes are absent
In October 2013, a committee set up by the higher education and training minister released its list of recommendations about university funding.
How many of the committee’s major recommendations have been implemented in this new policy? Unfortunately almost none.
The committee suggested that a new formula ought to be introduced for calculating accredited publication units. This, it said, should take into account the scientific impact of a publication in terms of citation indexes, journal impact factors and publishers’ rankings. The new policy mentions this recommendation in passing – but as a throwaway line, something that should be discussed some time in the future.
There is also no mention of special rules aimed at incentivising academics’ involvement in international collaborations. For instance, if 15 South African authors are involved in a paper about a big science project, why not split the subsidy between these 15 and their institutions? That they are working with international academics shouldn’t stop local authors or their institutions from being rewarded by the South African government.
Most academics across a range of disciplines choose to disseminate their results through subsidised journal articles. The new policy has made no changes to the funding schemes for such articles, so it’s very likely that academics will feel forced to continue choosing quantity over quality.
The Department of Higher Education and Training’s policy needs a far wider revision. This must be carried out in consultation with all the relevant stakeholders.
The focus must be drifted away from the publication units metric. Instead, it must reward quality and encourage South African academics to get involved in more international collaborations, knowing that they will be subsidised for this sort of “big science”.
Finally, the government policy needs to be aligned with the country’s National Research Foundation metrics. These are based entirely on the quality and impact of research. This will grant South African scholars the full and rounded support they need as they strive for quality, impactful research to support a knowledge-based economy.