Learning to define problems consciously, by evaluating and synthesizing different perspectives, can help scientists consider policy options in a more balanced way. Doing so will help scientists contribute more effectively to societal decision making on topics of importance to science community and to society. Suggestions on problem definition are provided for scientists considering policy issues.

Icône de l'outil pédagogique 1- Pay attention to goals.

Which options appear most favourable depends on one’s goals. An important component of any problem definition is therefore what (and whose) goals are emphasized—and what (and whose) goals are not. This includes goals that are stated explicitly, and goals that are implied or assumed. Paying attention to goals, by identifying and evaluating the goals in different people’s problem definitions (including one’s own), is therefore a key component of developing beneficial policy options and convincing recommendations.

Icône de l'outil pédagogique 2 - Identify and evaluate significant assumptions.

Assumptions in problem definitions can take numerous forms, including assumptions about goals, causes, predictions, and solutions. Such assumptions will be transferred into the analysis and communication of options, and become so embedded that they are difficult to recognize. Although assumptions cannot be avoided entirely, significant assumptions – that is, “those that may significantly affect the conclusions of the analysis” (Morgan and Henrion, 1990, p. 38) – can lead to many of the difficulties discussed earlier. Consequently, identifying and evaluating significant assumptions in problem definitions, and then modifying them or tracking their influence, is an important step toward balanced consideration of options and persuasive communication (Mitroff and Emshoff, 1979). Assumptions embedded in problem definitions will be incorporated into impact assessments. If these assumptions are invalid, inappropriate, or outdated, the impacts are likely to be inaccurate. Such assumptions may also mask sources of uncertainty, leading to overconfidence in predictions and failure to consider alternate outcomes. Moreover, when predictions are inaccurate or overconfident, recommendations are likely to fail to achieve the desired results.

Icône de l'outil pédagogique 3 - Consider different perspectives.

People approach public issues from different perspectives, depending on their knowledge, experience, and values. Scientists can therefore not take for granted that others define problems the same way we do, or even that they view our “problems” as problematic. By considering different perspectives and value systems, we can identify current and potential sources of disagreement about problems and solutions. Incorporating this understanding into how we define problems then produces more societal beneficial, more policy-relevant, and more marketable options. Important perspectives to consider include those of major stakeholders, including the intended audience, and those of people with less evident or more diffuse interests, such as under-represented groups and the general public (Bousset et al., 2005).

For example, analysts suggesting that problem for a given Interest Group (e.g. PE2) could be defined as “area of oil-producing crops – especially oilseed rape and sunflower – are currently growing too slowly in new regions (especially in breeding regions) to improve the biofuel energy balance” (modified from Bardach, 2000, p. 5). Here it is assumed that (1) the goal for PE2 is improving biofuel energy balance, (2) the solution is to develop oil-producing crops, and (3) developing oilseed rape and sunflower in breeding regions will improve biofuel energy balance. Audiences who do not share these assumptions will view the analysis as slanted, irrelevant, or invalid – and may not even perceive the situation as problematic.

Suggesting assessing global performances of cropping systems that contain oil-producing crops in comparison to cropping systems without oil-producing crops and to search for what could occur in term of surfaces at several scales if no specific action is done – with/under various price trends and incites, might be perceived as a more open approach of the problem. “Seemingly inconsequential changes” in framing a choice, such as presenting options in terms of losses rather than gains, can cause “significant shifts of preference” (Tversky and Kahneman, 1981, p. 457). Although results from idealized contexts are not directly transferable to real-world decision-making (e.g., Stewart, 1997), the range of framing effects that have been demonstrated (e.g., Nicholls, 1999) suggests that even subtle aspects of problem definitions can influence which alternative is preferred. Thus, even seemingly minor details in the language used to define a problem can be important (e.g., Stone, 2002).

Régi par la licence Creative Commons Attribution Non-commercial Share Alike 3.0 License

copyright Seamless 2009