Report on Proceedings of the Second hesa biennial Research and Innovation Conference

Research Infrastructure: Lessons from the EU

Prof John Wood, Secretary General of the Association of Commonwealth Universities

Prof John Wood, Secretary General of the Association of Commonwealth Universities

NE thing that worried me on the first day of the conference was the very traditional views of research that people expressed. The minister has opened up the discussion about how research should change. We need to empower researchers to own the future. Research is increasingly global, and we can all be more effective if we work together. We need bright people with brilliant ideas, but in addition we need people who can integrate things and bring them together with other disciplines. We should not leave the economy to economists only, or science to the scientists – we need integration between the different disciplines. Perhaps we also need to reconsider the conventional idea of the PhD.
Upcoming issues include the need to train academics to operate in the global environment. However, how would one deal with promotions, and the huge data deluge. There is the issue of publicly funded research being available to everyone. We also have to train and support people to run and manage this environment, and we need to think of the impact of mobile telephony and its democratising reach.
There are four types of infrastructure in the world of research, namely:
• Single physical;

• Dispersed physical;

• Virtual; and

• Massive data sets

All domains of research are included, from humanities to big physics. People are increasingly accessing research sites remotely, no longer having to actually visit specific sites; anybody can access research sites from anywhere in the world. Research infrastructures have links to industry, institutes and universities, education, people, research and innovation. There are huge economic impacts because of the dispersion of intellectual and entrepreneurial activity to smaller businesses. Nelson Mandela Metropolitan University (NMMU) launched a microscopy centre in 2011, which will advance the country’s social and economic development. The Square Kilometre Array is another initiative that will dramatically affect the way science and social structures are managed in South Africa.

The European Strategy Forum for Research Infrastructures was created in 2002 by 15 EU member states and the Commission, originally just as a discussion forum. In 2004, it was decided to create a high-level roadmap to make Europe competitive. In 2005, 12 new member states joined. In October 2006 the Roadmap was published and endorsed by all delegates. In 2008, the Roadmap was updated. The Roadmap looks at issues like legal entities, taxation, employment rights, mobility of staff, in-kind contributions, data management, etc.

The Roadmap looks at research in the following areas (among others):
• A distributed social science data archive, generated from the bottom up, and accessible to anybody in Europe;

• Digital research infrastructure for the arts and humanities;

• European social survey – looking at demographics and health, ageing and retirement in Europe;

• Common Language Resources and Technology Initiative, to look at what words meant historically and today, using semantic web technology. This has a major impact on international relations and legal and economic agreements, so it has a particularly important practical application.

There are many challenges in running these data sets, including technical challenges as well as linguistic challenges. Take-up by the target audience and training for people with no technical background are important issues to address. There were also legal challenges around IP and copyright and different countries’ legal systems.
In the area of biodiversity, a range of different activities take place and have to be consolidated, after which it is analysed and shared in the common exploratory environment. Current technology is democratising research infrastructure.

New developments in data generation and management will result in major developments worldwide. Data is the new oil. It is necessary to think about how data bookkeeping should be done, because of the huge amounts of data generated and storage requirements for it. Researchers should think about whether data is true and what to do with it. Academics and young researchers have to be trained to deal with the huge amounts of new data generated.

Thirty percent of all scientists are trained to do research, but e-science as part of the global research system is a new area. Global collaborations enable engagement on the grand challenges, where data forms part of the infrastructure, a valuable asset. The public should have access to data, and it should inform policy and evidence-based decision making. Politicians should also make decisions based on solid evidence. International trust and interoperability will be essential. A forum should be established to discuss these issues, and to avoid fragmentation of data and resources. Since there are parts of the world where publication of academic papers is well-incentivised, plagiarism is rife throughout the world, which makes data verification a particularly important issue.

3. 
   Discussion
   

Comment (Prof Burton): We should record the intention to establish a national programme to manage our national data. It is something at the forefront of the minds of DVCs: Research, but progress in this area has been slow.
Another question that I want to ask the Minister relates to the Green Paper, where one aspect highlighted is the minister’s attention to differentiation between different types of universities. This will affect the way we manage research and innovation. What is the minister’s view on this?
Comment (Prof Rawlings): I accept the fact that people can work remotely, but could you expand on your statement that the PhD may be past its sell-by date. How do you see a PhD, as an area where people have to make a contribution to knowledge on an individual basis?
Comment (Prof Strangway): I was at MIT for a number of years. I’d like to point out that only part of the activity in the US emanates from universities, but they also have smart procurement and government investment through the space agency and other government departments. Both supply and demand has to be grown to build capacity. It is now established that the US spends 20 times as much on purchasing smart procurement as other governments. You should think about government stimulating supply and demand in order to stimulate new research and innovation.
Comment (Prof Nhlapo): In discussing innovation we should look at both sectoral and regional efforts and collaborations.
Comment (Prof Wood): I am a firm believer in individual endeavour pushing the frontiers of research. One thing we are doing is looking at a European PhD, over and above a national one, probably through a summer school model. We are looking at running a pilot at the University of Botswana for students from Africa on the issue of climate change. There has to be something that shows that somebody has pushed the frontiers.
One of the key recommendations is in terms of public procurement in research, where there are some risks. I had some projects that were complete disasters, but there were also some lessons learnt in those projects. This forms part of the European 2020 Directives, but I know that some countries are reticent to put more money into research.

Response (Minister Pandor): The matter of differentiation was raised by both Minister Asmal and myself, and we were lampooned at the time. I’m glad to see that people are showing more appetite for it now. I don’t think it needs to signal neglect of institutions; we need to ensure that our institutions are capacitated based on their actual needs. I’m thrilled to see that HESA is discussing this issue in a far more disciplined manner.
I think our Science Councils are more active in terms of procurement, as you mentioned. Our policies have identified sectors as challenging areas, and it may be necessary to recast our innovation strategy to identify grand challenges like food security, access to energy, etc. Our policy perspectives may detract from our ability to identify and support research in certain areas, and we are revisiting these issues.
In terms of the National System of Innovation, we identified the need to develop institutionally and also to look at seamless integration rather than pockets of activity all over. We have a difficulty in that we all want to do everything – very few people want to specialise in a specific niche area, which means that our energies are too widely dispersed. I’m hoping that with respect to innovation we can get people to specialise and do in-depth research. We have a tendency where everyone wants to do everything.
I agree that we have to think about regional and sectoral ways of working. I know that some provinces and municipalities and metros want to set up innovation offices. I cautiously look at these issues, and when I see something like that I propose that we should rather co-operate.
After Prof. Wood’s input I thought I would be asked what the Department of Science and Technology was doing in terms of infrastructure in South Africa. The DST is putting a team together to assess what the character of research infrastructure in South Africa is, and what our plan should look like for the next few years. We have had a grant from the NRF which has helped upgrade infrastructure, but we need additional resources.
Comment (Prof Nongxa): The slides presented by Prof Wood mentions huge figures. How do you make the case for that kind of investment in the current financial environment? How will you persuade ordinary people that there is actual value to be gained from that level of investment? Who should make the case for increased expenditure?
Prof Wood mentioned the democratisation of research. Yesterday, we were told of the need to start at the schooling level – how do we start stimulating passion and interest from this level?
Question (Prof Stroebel): It feels like there is a sense of inconsistency in terms of funding, and conflicting priorities for research. At a micro-level we need consistent direct investment in higher education, perhaps through the National Research Fund. We need some consistent signal from bodies like the NRF that research and innovation is guided by a single vision. Could you comment on this?
Question (delegate from National Skills Authority (NSA)): Minister Pandor mentioned the knowledge economy. I wonder to what extent language is used to enhance our understanding of the issues and how we guard against certain concepts becoming elitist. Innovation should also guide and direct how work is distributed in the economy. To what extent have we traced the impact of innovation in the world of work; has there been a reduction of labour in some areas, or have we retained or changed work in other areas as a consequence of innovation?
Question (delegate from North West University): The minister mentioned local knowledge and its role in innovation. As far as I’m aware, the DST has invested in this area. Have we responded appropriately to this investment as the higher education sector? Have we got value for money and responded appropriately? Why does it look as if DST is the only department focusing on this issue? My experience is that other departments have not addressed this aspect to leverage the importance of the issue.
Question (Prof Ijumba, UKZN): How many government departments use local universities for their research and development needs? I think our system is geared towards incentivising and rewarding individual institutional effort, which should be addressed at a system-wide level.
Response (Minister Pandor): I have tried to argue that the rewards system should be looked at, and I get the same reaction I did when I mentioned differentiation. At awards ceremonies, when very few young or black people are rewarded, you ask yourself what it would take to change the landscape. I think we need an appropriate link to national priorities in our research efforts otherwise we will not achieve the change we want.
I don’t know how many Departments use the universities for their research needs, but I know that they do make extensive use of the science councils, because I did a study of that issue. I think the area of Indigenous Knowledge System (IKS) is a difficult area, both in terms of innovation and knowledge production. It is a new area for us, and to some degree it is associated with cultural attributes rather than an area of science. People tend to believe that if you have a department on IKS it must have a hut and bones for divination. I’ve been talking to colleagues in that area to determine how the benefits can be optimised all round. There are some successes, and we will continue to support the sector, but I’m not seeing enough productivity and intellectual activity and scrutiny. I am seeing a reductionist approach to IKS. I think that incredible things can be done in this area. DST linked technology and local community activity, and we met people working in a project that created jobs and products that are used in various guises, creating many jobs. These people were trained by Sasol chemists, and the project is showing real value. I think it is important to generate understanding and build a common set of skills that can be widely used. I hope that other departments will also support IKS.
I was so excited by what was said about language by Prof Wood, that I want to institute a similar programme in South Africa. However, not everything can be done by government. The ANC led the struggle against oppression and apartheid, but you, the people, also have a role to play. People have a misguided notion that the government should do everything, which is a bad approach to have. You will ruin the ANC by enforcing this notion that they are all powerful. You and I have to take some responsibility now that we are free. You have thousands of students that you can tell about the importance of innovation. Get the message out there, and show people what they can do.
In terms of the question about long-term investment and support, I think we tend to look at short-term crises. We need to develop a culture of looking over the medium- to long-term so that our institutions know where they stand in terms of support. In 2010 we made money available for students, and we determined the need to continue the funding. We must think systemically for the long-term. From time to time you should drop an e-mail to the relevant ministers to highlight issues of concern. It is always good to hear from the sector.
We do have programmes at school level, and we want to expand them to reach as many young people as possible. It is a positive area that could develop even more. I think each municipality could have one. I’d like to see young people trained as science communicators. We are thinking of growing our capacity in this regard.
I was told that if we were to successfully build the Meerkat and the SKA, the potential of establishing 1 500 small- and medium-sized firms would be possible. But to do that, we learnt that we would have to spend a few billion Rand. We have to look beyond the short-term and convince our people that the initial investment is worth it. People tend to be short sighted. Even on the SKA I have become a lone ranger, because our scientists are so quiet. We need to be confident and articulate about the importance of infrastructure investment as a prerequisite for economic and social growth. I need that support to convince the finance minister to give funding for these projects.
Comment (Prof Wood): Two years ago I was asked to give a talk on the economic impact of the SKA, which I can share with you. The first thing to realise is that research infrastructure does not have a one-to-one correspondence with economic investment; the impact is clear in a much wider sense. However, the real impact can be measured in terms of the number of children taking up science, and how it affects society. I realise that these issues are not important to somebody who is starving. People do not realise that the initial research into protons now has a major impact in cancer treatment. The most important impact of research is probably in the area of diplomacy, because scientists are often the first point of contact between different countries.
How do you actually take research forward? I think you need to look carefully at the postdoctoral area, which is where people learn to be academics. Scientists have to be able to communicate excitedly in a short space of time. We need to embrace a space where ideas can be discussed and knowledge can be shared.




6 THEME DISCUSSIONS

6.1 Theme 1: An enabling environment

Discussion led by Prof Johan Mouton, Director: Centre for Research on Evaluation, Science and Technology, University of Stellenbosch

Prof Johann Mouton, Director: Centre for Research on Evaluation, Science and Technology, University of Stellenbosch

Report back by session moderator, Prof Tinyiko Maluleke
PROF Mouton highlighted that for those working on research and science policy, the debate around differentiation has been ideologically laden, given the country’s apartheid-era legacy. The subsequent mergers of institutions were similarly ideologically laden. It is therefore refreshing to note that people are moving beyond these issues, instead of focusing on how knowledge production can be improved in a differentiated sector. It will be helpful, therefore, to try and understand differentiation.
Given the different institutional histories, missions and capacities, a high degree of differentiation in terms of key research production dimensions is to be expected in the South African higher education sector. The differentiation constructs and associated indicators presented and discussed are not independent of each other (in statistical terms there are multiple interaction effects).

We still need a proper conceptualisation of the notion of research differentiation. As a first attempt, one should distinguish between the following six types or categories:

• Volume of research production: absolute number of papers in peer-reviewed journals, or normalised output;

• Shape of research production – differences in distribution of output by scientific field;

• Site of publication;

• Research collaboration;

• Research impact; and

• Demographics – differences in output distribution by gender/race, qualification/age.

Proposition 1:
• Following the introduction of a national research subsidy scheme in 1987, research production remained stable until the early 2000s.

• Thereafter production escalated, when the system produced an average of 5 500 article equivalents until 2005, when the new funding system was introduced.

• Over the past six years, output has increased exponentially, despite the fact that the number of academics remained static.
Proposition 2:
• Increases in absolute output over recent years have not affected institutional distribution. Huge differences between the most productive and least productive universities evident 25 years ago have remained mostly unchanged;

• A few universities have managed to improve their positions in the rankings (University of the Western Cape is a good example), but the vast inequalities in knowledge production between the top and bottom have remained.

• Five of the country’s institutions produce 63,9% of the total research output. The next group of seven universities produces 30% of research output, while the remaining 11 institutions produce the remaining 6,1%;

• Figures indicate that stimulating research is a difficult and lengthy process. The top five institutions continue to dominate research production because of their historic background and the benefits of a long period of sustained investment. Third-stream income from research could not be ascertained, and might indicate even greater differences.

The statistics presented on institutional output only referred to absolute output and have not been normalised for the size of institutions. Rankings were presented in terms of research output (normalised for number of permanent staff) and rankings in terms of knowledge output (masters and doctoral graduates) also normalised for size of academic staff. A comparison of the two rankings revealed some interesting shifts in rankings (most notably for NMMU, UNISA and some universities of technology), but the overall difference in normalised output between the top and bottom universities remain huge. In 2009 the highest per capita output was determined as 1,28, despite there being no formal agreement on the norms for knowledge production. In the period 2007 – 2009, a ranking of universities according to average normed knowledge production (including masters and doctorates produced) shows that some institutions move up and down in the rankings. This ranking makes it clear that institutions could adapt their practice in terms of research to change their rankings.
Proposition 3:
• SA universities vary hugely in terms of the shape of their knowledge production. The big differences in scientific field profiles of the different universities is clearly a function of institutional histories (e.g., having a medical school or faculty of theology) and institutional missions (research intensive universities versus teaching-focused universities and former technikons). The disciplines of theology and law are the most prolific producers of research papers. In both fields there are more than 20 journals respectively. These differences mean that the same benchmarks cannot be realistically set for all faculties.
Looking at the shape of knowledge production, it is clear that institutions have different focus areas. UNISA is predominantly focused on the humanities and social sciences, while institutions like Fort Hare University is predominantly focused on the natural sciences.
Proposition 4:
• Distribution of research output by journal index (IS, IBSS and SA) varies hugely. The differences between the universities in terms of this dimension distinguish between South African journals, ISI journals and South African ISI journals. Cape Town University is highest in the rankings at about 70% because it has the best track record for publishing in ISI journals, while other institutions tend to publish in local journals.

• University research output has become significantly more international and collaborative over the past years. Analysis of co-authored ISI papers shows that institutions like UWC, UCT, NWU, UP and Unisa are quite active in this area. Collaborating with top authors in the field multiplies the chances of getting cited exponentially, which is not possible when publishing locally.

Proposition 5:
• The impact of SA’s research production has increased significantly over the past 15 years, mostly because of collaborative publishing in ISI journals. Over the years 2000 to 2010, the number of ISI outputs increased from below 4 000 to close to 8 000. As a country the 8 000 papers produced in the past three years and published in ISI journals have generated exceptional results. South Africa has also featured in the top 1% of highly cited papers. It would be more realistic to look at the impact of publications in particular areas rather than overall research output.
Explaining differences in research production:
• International trends show that demand is stimulated by international rankings;

• National steering instruments include revised funding schemes and an expanded SA presence in ISI and NRF rating systems, which have led to increased research output and increased ISI-production;

• Institutional capacities (Merton and cumulative advantage theory (Matthew effect);

• Institutional histories and structures; and

• Institutional strategies.
Institutional enablers:
• A study for the HEQC done in 2007 showed that a university like Wits has not expanded its number of authors. At UKZN, the number of authors has not increased greatly, but the output has expanded significantly over the years.

• An analysis of the number of staff with PhDs shows that UCT and Stellenbosch University lead the pack. These institutions also lead in terms of the average number of publications in accredited journals per academic staff member. Figures indicate that staff still engaged in completing their PhDs may not have adequate time to generate papers for publication.

We undoubtedly have a highly differentiated university sector in terms of key indicators. Some causes of differences reflect the path dependency of historical factors, missions and structures. Other differences are the results of more recent institutional responses to international and local incentives.