Features: April 18th, 2008

By Sandra Nutley

The clamour for evidenced based policy can be heard world wide. In the UK policies emerging from departments must have an evidence base. The author describes how a better understanding about social programme implementation, client experiences of those programmes, and the sources and causes of implementation failure are having an impact.

She counters the argument that better research use is unproblematic, rational and linear and that the relationships between research, evidence, policy and practice are straightforward. She describes the agenda for action and calls for a wide-ranging debate and dialogue.

It is a decade since New Labour entered government on the back of its rallying cry of ‘what matters is what works’. Since then evidence-based policy has been much debated and discussed and there has been a range of initiatives to promote evidence use in policy and practice settings. And this is not just a UK affair, debates and initiatives under the broad rubric of evidence-based policy and practice abound in Australia, Europe, New Zealand and the USA too.

So what have we learnt over the past decade about evidence use in policy and practice settings? This is the core question tackled in a newly-published book on Using Evidence(Nutley et al., 2007), which has already produced a great response:‘ Anyone who has ever written or uttered the words “evidence-based policy” should read this outstanding book’ (Carol H. Weiss, Professor Emeritus, Harvard University).

‘This book is a major contribution to the literature’ (Thomas G. Rundall, University of California, Berkeley).‘I am recommending [the book] to knowledge brokers in Canada and I have ordered another seven copies for my staff!’ (David Phipps, York University, Canada).

Research matters—but research, and its uses, are diverse

The phrase ‘evidence-based policy’ can be misleading because it tempts us to think of policy options being shaped by research evidence, and even of research playing a decisive role in the choice between competing options. Such a view emphasizes particular types of evidence: evidence on ‘what is the scale of the problem?’, and evidence on ‘what works?’ in ameliorating such problems. Yet studies of policy repeatedly show that evidence is rarely used in such a direct and instrumental fashion. Instead, research often enters policy debates in much more indirect ways: for example by shaping what is seen as a problem in the first place; by destabilizing current framings of the problem set; or by challenging orthodox approaches to interventions. These more complex and subtle types of evidence use draw on a wide range of research which covers, for instance, basic understanding about the structuring of society and the nature of social problems (including their sources and interrelationships); and a better understanding about social programme implementation, client experiences of those programmes, and the sources and causes of implementation failure.

Even when ‘what works’ evidence is used explicitly, it may be being used tactically or politically, for example, to advance a position, buttress previously taken decisions, or derail customary debates. While such uses may be derided as ‘misuse’ (especially by researchers), they at least admit to the possibility that evidence might have influence rather than irrelevance. They also point to alternative strategies for improving the use of research.

Strategies to improve the use of research

Promote research use by organizations and service delivery systems

The dominant model of research use envisages individual policy-makers and practitioners consciously seeking out and keeping up-to-date with research, and then applying the evidence they thereby glean in their day-to-day work. This means that we know much less about the potential for research to enter policy and practice at the organizational or systems levels—for example through protocols, guidelines and regulatory frameworks, although the evidence we have suggests such uses of research may well be important. We need to move beyond individualized framings of research use and capture what using research might mean within wider organizations and systems, and for groups and communities as well.

Move beyond linear models of researcher/research user interactions

Many of the initiatives developed to promote better research use seem to assume relatively unproblematic, rational and linear, relationships between research, evidence, policy and practice. Much effort has been expended on improving the supply of research: ensuring that the right questions get investigated; supporting rigorous synthesis of existing research studies; and providing for a degree of translation of research findings so that they are accessible both literally and intellectually. All of these have ramped up the availability of concise research-based evidence. However, improving supply alone is clearly insufficient, and attention has been given in parallel to the demand side.

Encouraging the policy and practice demand for research evidence has been a central feature of ‘modernized’ policy processes and ‘effective practice’ initiatives. Departmental spending bids are now expected to be supported by evidence, and the evidence underpinning key policy decisions is increasingly being cited and published. Regulatory frameworks for public service professionals—such as doctors, nurses, teachers and social workers—expect practitioners to be research informed. Staff training and support for evidence use has received much greater attention than hitherto at both policy and service delivery levels. Formal evaluation of these ways of increasing the demand for evidence is rather thin on the ground but nonetheless some positive experiences are reported.

Beyond improving the supply of evidence and stimulating the demand side, other initiatives have sought to bridge these worlds. Strategies used have included strengthening the role of government researchers as knowledge brokers and co-locating these researchers alongside policy-makers. Efforts have also been made to develop research and policy networks in key policy areas that reach out to wider stakeholders such as practitioners and service users. Such initiatives can acknowledge the diversity of knowledge providers for the policy process and can address the issue of whether research needs to be integrated with other forms of evidence such as routine management data and stakeholder consultations. They also promise more interactive engagement with that evidence. Again, however, while such bridging activities seem sensible, there is little systematic evidence on how they affect the uptake and impact of research.

Target a wider range of voices

While action on the supply and demand side of evidence, and attempts to bridge or integrate between these are welcome, there may be more that we can do to promote research use. Research may influence policy indirectly as well as directly, through many non-official routes, such as advocacy networks, that effectively limit the potential for government control of evidence, its interpretation and use.

Outside government, there is already a growing awareness of what might be achieved with evidence through being policy savvy, media aware and advocacy oriented. Some advocacy-oriented agencies (as diverse as the Joseph Rowntree Foundation, the World Wildlife Fund, the Health Foundation and Barnardos) are already showing the way with creative mixed strategies involving vigorous attempts to shape the policy context, the creation of active alliances around key issues, and the development of demonstration projects that signal what can be done on the ground.

Research use strategies that target a multiplicity of voices and agencies offer many opportunities for research to become part of the policy discourse. However, the result is likely to be a far cry from a rational and linear process of instrumental research use. The degree of contestation and debate that inevitably arises is likely to encourage more challenging roles for research: roles that go beyond simply supporting developments within current policy and service paradigms, to roles that question and challenge these paradigms. While clearly demanding in many ways, such open, pluralistic, interactive and informed, policy communities have exciting possibilities and democratizing potential. However, it is a vision that is most challenging for those who would like to see a centrally managed set of strategies for developing research informed policy and practice, as it will involve a good deal of ‘letting go’ of policy and practice debates and the evidence that shapes them.

The agenda for action is extensive, although there is still much that we need to know. And action is important; research and research use matter because wide-ranging yet constructive debate and dialogue that draw on research offer the best chance of enhancing knowledge and practice in public services.

?Reference Nutley, S. M., Walter, I. and Davies, H. T.O. (2007), Using Evidence: How Research Can Inform Public Services (The Policy Press, Bristol).

Sandra Nutley is Professor of Public Management, University of Edinburgh and Director of the Research Unit for Research Utilisation (RURU).