Investigating Policymakers Understanding Of and Preference Over Risk and Uncertainty

In issues like climate change, large scale disease management and terrorism, policy makers are increasingly expected to make science-based decisions.

For this to happen, scientists should keep in mind the fact that policy makers are final users of scientific output which is often characterized by uncertainty and imprecision. Policymakers are individuals with their own biases and preferences for uncertainty, but moreover they are communicators who will use what science produce in order to convince large part of public. When thinking about how to present scientific uncertainty crucial questions are what level of detail it is necessary to produce and in which visual form, if the aim is to improve understanding, while different people can understand differently the same information.

In the presentation, Prof. Bosetti discusses three works that tackle components of these problem and analyze the results.

The first is an experiment done at the COP 21 in Paris with climate negotiators. The experiment aimed at understanding how policymakers perceive scientific information and how they incorporate it in their prior beliefs using also an appropriate format of communication.

The second is yet another experiment done in the same setting and again collecting data from climate negotiators, but it aims at understanding policymakers’ preferences for ambiguity and comparing ambiguity preferences of distinct populations. They find that policy makers are indeed averse to ambiguity but that this is not necessarily due to an irrational behavior, it may rather be driven by intrinsic preferences over subjective probabilities.  

The third work is related to long term emission scenarios, which are typically presented without probabilistic information. Through an expert elicitation they are able to estimate probability distribution over long term GHGs emission ranges and find that indeed experts have in mind a very different distribution than that implied by the users of scenarios. So efforts to quantify and present risk and uncertainty information to policymakers are crucial.