EFSA‐Member State multilingual online survey on communicating uncertainty to different target audiences

First published in EFSA Supporting Publications
7 May 2018
27 April 2018
Technical Report

During the development of the EFSA guidance on uncertainty analysis, the literature was found to be equivocal about the best methods for communicating scientific uncertainties and there was a lack of empirical evidence on the best approaches for reaching non‐technical audiences. EFSA therefore commissioned a focus group study in 2016 on clear communication of uncertainty information to different target audiences, followed by a larger scale multilingual online survey in 2017 described in this report. The survey was open over a three‐week period, from 23 February until 15 March 2017. As a result, 1,913 people across the EU responded in six languages to questions about expressions of uncertainty: qualitative and quantitative wordings, approximate and precise probabilities, and positive or negative framings. The results showed that similarly large percentages of respondents ranked qualitative and quantitative expressions more helpful for understanding risk, and there was some preference for positively‐framed statements. Nearly all the respondents were positive about receiving uncertainty information. There were moderate differences between the preferences of respondents using different languages and from different professional backgrounds, but due to limitations of the survey method it was not possible to interpret the cause of these. The results are consistent with those of the earlier focus group study. Together they support using positively‐framed, qualitative and quantitative expressions for all audience groups, as in every group some people prefer one and some the other, and nearly all find them useful. These findings together with those in the scientific literature will inform development of an EFSA guidance document on communication of uncertainty in scientific assessments, and updating EFSA's handbook on risk communication.

Question Number