The decision-making process
 


Research knowledge is one input among many in most decision-making situations.

Broadly there are two schools of thought about the role of research in influencing policy developments: one being pessimistic, or cautionary, and the other being more optimistic (Webber 1991; Dunn 1992).

Some analysts have warned that researchers sometimes demonstrate arrogance by suggesting that research knowledge is the only solution to social problems: the authors remind us that many forms of 'ordinary knowledge' are routinely brought to bear on a whole range of complex social problems (Lindblom and Cohen 1979). What is more, many social problems are not caused by a lack of knowledge, but by 'conflicting interests in the social surround' (Huberman and Ben-Peretz 1994-95).

Research interventions take place in this context. It is unwise to see decision-making as a rational linear process that proceeds neatly from problem definition to research, and then to problem solution. Research knowledge can be used by different people and organisations to support their position. The other side of this coin is that research that does not support their interests will either be ignored or actively attacked.

Many explanations of the development of, for example, environmental policies 'have been inclined to view scientific knowledge and policy analysis essentially as instruments to be wielded by interest groups' (Owens and Rayner 1999). However, this has recently begun to change, with a realisation that knowledge, ideas and debate can also exert influence in policy debates (for a useful review of these developments, see Owens and Rayner 1999).

Studies in this more optimistic school of thought tend to adopt a broader view of how research knowledge can be used. An important early contribution was made by Caplan in differentiating between 'instrumental utilization' and 'conceptual utilization' (Caplan 1975). This distinction immediately throws up one of the fundamental differences between the two approaches: rather than simply looking at short-term use of research information, the broader approach studies the uptake and transfer of ideas in helping to frame debates on policy alternatives.

Weiss has found conceptual use to be more common than instrumental use. It may also be more significant, as it can 'gradually bring about major shifts in awareness and reorientation of basic perspectives' (Weiss 1980). The term 'knowledge creep' was coined by Weiss to describe the way in which ideas gradually - or in some cases quickly - spread, enter into use, and sometimes become the conceptual framework of entire policy debates.

 
  Senior scientists have gone so far as to urge the scientific community to forge a 'new social contract' in order to ensure that science makes more effective contributions to solving the big problems of the 21st Century (Lubchenco 1998; Daily and Walker 2000).  
 
Who makes decisions? 		 
  This is reinforced by studies into how policy is formed. Webber, in his review of the effectiveness of research dissemination, states that it is easy to 'focus too narrowly on use (of research) as a behavior attributed to an individual policy maker, while under-emphasizing the importance of dissemination and diffusion activities over which the policy maker exerts little control' (Webber 1991). Research communication activities fit into a systemic process of policy analysis in which many different actors play a part, not simply a small group of 'policy-makers'. This insight offers the possibility of a much greater role for the dissemination of ideas than one that constricts its role simply to connecting individual bits of information with people that can use that information in an instrumental way.  
 
Susceptibility to using research		  
 

Some decision-making situations are, however, more susceptible to research interventions.

First, many policy issues have been found to vary over time in the intensity of interest they receive from policy making communities. This is what is known as an 'issue attention cycle'. In the early stage of the cycle, researchers may be investigating an issue and a few members of a policy community may be interested. These people may interact, building up what has been called an 'epistemic community' or a 'trans-epistemic arena' (Haas 1992; Mayntz and Schimank 1998). Researchers are often active players in building up such communities of interest, in a process that has been termed 'enrollment' (Latour yyy).

Then some event may lead to the issue having a higher profile in policy or public debates. This event may itself be brought about by members of this community - social activists, for example.

Increased profile for the issue then provides the opportunity for researchers to make effective public interventions. This is the finding of a large study by Harvard University that has assessed the effectiveness of research interventions in policy (Clark 1999). These analysts find that assessments are often ineffective at achieving policy change because they produce their policy recommendations too early - that is, before powerful decision-makers are interested or engaged in the debate. Another reason is that when the issue does become high profile, the analysts involved in the issue fail to alter the nature of their analysis, recommendations or style of communication in the light of the expanded set of participants.

Clark and colleagues find three characteristics that research users need to have in order for research to make effective contributions to their decisions: interest, capacity and openness.

  • First, users need to have an interest in the results of research. Clark and colleagues 'suspect that the most effective assessment processes devote a substantial amount of time and energy to negotiating with potential users the particular questions about which those users are most passionately and urgently interested'.

  • Second, in order to be able to take up the research, decision-makers need to have the capacity to do so: civic groups, policy communities and even entire countries can lack capacity and therefore the ability to engage with an issue. This point is echoed in a parallel discussion about the benefits of public sector research (Salter, d'Este et al. 2000).

  • Third, the characteristic of openness is, by their own admission, a somewhat vague and tentative finding of the Harvard team's work. They seem to be saying that any effort that increases the amount of communication between researchers and decision-makers can only be helpful. Decision-makers need to be open to ideas if they are to receive them. This is a consistent finding from other research, and is covered in greater detail in the section on interactions between researchers and decision-makers.

Finally, there is the issue of the timescale in which research can influence decision-making. This is put into perspective if we ask whether researchers should expect to influence decisions more rapidly than activists, politicians or any of the other groups in society who wish to do so. Research knowledge is sometimes taken up rapidly, particularly where all of the above conditions are met. However, like all other influences on decision-making, the impact of research-based knowledge and ideas is more likely to be gradual, diffuse and also difficult to detect, as found in various reviews of the literature on research impact assessment (Kostoff 1995; Martin, Salter et al. 1996; Kostoff 1998; Molas-Gallart, Tang et al. 1999).

But are there features of the research knowledge itself that will increase the likelihood of its being used? This is the subject we now address in "The characteristics of the research knowledge".