Image by Raymond Bryson on flickr.

Towards a science policy of ends, not just means

04 September 2013

More information

Paul Harris is Deputy Director of the HC Coombs Policy Forum and a Special Advisor on government and international relations to the Commonwealth Scientific and Industrial Research Organisation (CSIRO), Australia’s national science agency.

You might also like

Related research centres

Recent science policy work by the Federal government has the potential to improve the links between public investment and desired impacts, writes Paul Harris.

It would be fair to say that science policy has not had a particularly prominent role in the Federal election campaign.

But despite the lack of attention, the government has been quietly doing significant science policy work over the past few years which, if further developed, has the potential to improve how we think about the relationships between public investments and desired outcomes in society.

Following on from its wholesale review of the “national innovation system” in 2008, the Labor government released its national, 10-year innovation policy plan Powering Ideas in 2009. The Excellence in Research for Australia (ERA) scheme, which assesses the quality of research in Australian universities, was implemented in 2010.

The Department of Innovation undertook a further review in 2011 entitled Focusing Australia’s Publicly Funded Research, which recommended more coordination and better prioritisation of government investments, along with more attention on the links between researchers and the users of research (particularly in industry) and the “impacts” that flow from public investments.

Since then, we have seen the chief scientist’s Health of Australian Science report, the creation of a new, whole-of-government Australian Research Committee (ARCom), a separate whole-of-government initiative on the role of science in policy, a new National Research Investment Plan, a high-profile review of Australian health and medical research, new national Strategic Research Priorities and further policy work on the design of a new national research impact assessment framework to complement ERA.

Not a bad effort. But the deeper question needs to be asked – why was all this work necessary in the first place? The answer to the question rests on two obvious points – firstly, that science matters. Yes, governments increasingly frame this value in economic terms, through the contribution of science to innovation, productivity and competitiveness, but science also matters in our efforts to achieve diverse environmental, social, health and other objectives, including the advancement of knowledge. Secondly, because governments spend quite a lot of public money on this. In Australia, the federal government alone spends approximately $9bn every year on science, research and innovation programmes.

The problem is that our understanding of the links between these investments and the full range of desired outcomes remains weak, and this is reflected in the practice of science policy. How do we really know that we are investing in the right places and most effective ways to achieve what we want to achieve? The policy work listed above shows that governments are increasingly aware of this issue. It has also been well researched by science policy scholars, with the result that there are new ideas to draw upon.

Existing science investment, prioritisation and evaluation mechanisms tend to rely on input and output measures. The assumption has been that by knowing something about (relatively) easily quantifiable things like the science budget, publications, citations, PhDs and patents, we can assess how effectively we are meeting the full range of desired outcomes. But this is not true. The evidence base for science policy is not always up to the job.

As demonstrated by researchers such as Dan Sarewitz and Barry Bozeman in their ‘public value mapping’ work, and by James Wilsdon, Bryan Wynne and Jack Stilgoe in their Demos report The Public Value of Science, the limitations of current approaches can lead to important questions remaining unanswered, and a narrowing of science policy options and debates.

But simply collecting more input and output metrics is not going to solve the problem. We need to flip science policy around so that it is “outcome-oriented” instead, starting with desired outcomes and letting prioritisation, investment and evaluation decisions flow from there. We need to experiment with science policy, and work to bring good ideas from the academic community into practice.

With its National Research Investment Plan, ARCom and the focus on ‘impact’, the Australian government has started in this direction. State governments in Australia are also grappling with the same problem – for example, the Tasmanian government has been working on a new coordinated approach to science policy, trying to coordinate investment and evaluation across multiple departments and agencies.

In a recent report to the Tasmanian government, I proposed a very practical approach to implementing outcome-oriented science policy. Drawing on the public value frameworks cited above, I show how they can be rolled forward and into whole-of-government decision-making about science.

The key point is that such an approach uses a standard set of questions instead of a standard set of metrics, allowing for both comparison across institutions and a diversity of investments and outcomes. When it comes to science policy, one size doesn’t fit all. Will Grant and I have argued elsewhere that a focus on improving discussions about the impacts of science is a good thing – implementing this kind of outcome-oriented approach allows these discussions to happen further “upstream”.

The questions that logically flow from this – whose job is it to implement an outcome-oriented approach, and how is this linked to other decision-making – are classic policy (and political) issues, not science ones. In science policy, there has been too much science and not enough policy.

With a focus on improving process and evaluation, it is perhaps no wonder that this work is not attracting much high-level attention in the hurly-burly of an election campaign. But doing a better job of linking public investments to desired impacts is crucial for scientists and policy-makers alike. We need to keep taking pragmatic steps towards a science policy of ends, not just means.

This piece was originally published on The Guardian:

Filed under:

Updated:  26 February 2021/Responsible Officer:  Crawford Engagement/Page Contact:  CAP Web Team