Using numerical data alongside verbal descriptors of uncertainty - terms such as ‘probable’ or ‘unlikely’ - in climate change models reduces the public’s misinterpretation of the guidelines and findings of the Intergovernmental Panel on Climate Change (IPCC). These results, presented this week in Nature Climate Change, could help increase the general public’s understanding of IPCC communications.
The IPCC faces issues with clearly communicating uncertainty in climate change models and predictions in the reports it assembles and disseminates. Although there are aspects of uncertainty inherent in climate change research, there is also the problem of the public’s imperfect understanding of climate-related issues and its misinterpretation of the current scientific consensus on the topic.
David Budescu and colleagues conducted an international study, covering 24 countries and representing 17 languages, comparing the efficiency of various ways of communicating uncertainty about climate models. They found that for a sample statement using the term ‘very likely’, in an attempt to express a greater than 90% likelihood of an event occurring, on average the general public misconceived it as having a likelihood closer to 50%. The authors found that if the same statements actually include the corresponding numerical ranges of likelihood, the public’s interpretation is much closer to the IPCC’s intentions.