Webinar series: uncertainty communication - VVSOR - VVSOR

Vereniging voor Statistiek en Operations Research
Recordings available

Webinar series: uncertainty communication

On four Tuesdays in November, speakers have talked about communicating (statistical) uncertainty to the general public in various contexts.


Recordings of the meetings can be found below.

Weather forecasts are inherently uncertain because a weather model is an imperfect approximation of the atmosphere, and because we can’t observe the whole earth system. This results in uncertainty in the location, timing and intensity of weather events, particularly the extreme events that have large impacts on society. The Royal Netherlands Meteorological Institute (KNMI) must quantify and communicate this uncertainty to a variety of users with differing levels of expertise, e.g., the general public or waterboards. We will discuss several case-studies showing the way that KNMI communicates uncertainty, including ensemble forecasts, written text, forecasts for the probability of precipitation, and colour code weather warnings.

Dr. Kirien Whan – Royal Netherlands Meteorological Institute (KNMI)

Every estimate or piece of evidence has inherent uncertainties about it - and for people to make evidence-informed decisions, we need to communicate those uncertainties. So how can we do it without overwhelming people, and does communicating uncertainties undermine people’s trust in the evidence, or in us as communicators? In this talk, Alex will outline some of the Winton Centre’s research into the effects of different kinds of uncertainty communication - on comprehension, perceived trustworthiness and decision-making.

Dr. Alexandra Freeman, Winton Centre for Risk and Evidence Communication - Cambridge University

During the last two years we all experienced a lot of uncertainty about the health risk due to the SARS-CoV-2 virus. Governances and health authorities struggled with how to communicate these uncertainties, sometimes conveying too much certainty about e.g. the transmission route of the virus while experts were still figuring out what the route was. Presumably, the health authorities did not want to frighten the public and aimed to foster the public's trust. Did they do a good job? What do we know about effective uncertainty communication? In order to answer this question, we first have to define what we understand uncertainty regarding health risks is and what we mean with effective. In this presentation we will define uncertain risk, and differentiate between different epistemic or scientific risks, and other uncertainties, e.g. ambiguity, related to health risks, and the public’s perception of uncertainty. Using also our own research about communicating uncertain risks of chemical contaminants in food, we will show that there are discrepancies between scientists and the public and among scientists themselves about the meaning of uncertain risk. This has implications for the effects we aim to attain with uncertain risk communication, e.g. to address concerns and worries, foster trust or increase knowledge.

Prof.dr. Daniëlle Timmermans, professor of Public Health Risk Communication, Department Public and Occupational Health, Amsterdam UMC, Vrije Universiteit
Photo: Tjitske Sluis

In an era of evidence-based medicine, using outcome information (e.g., 10-year survival or risk of experiencing short- and long-term side-effects) to guide medical decisions has become commonplace. Numerous prediction models have been and continue to be developed for this purpose. Originally, this type of information was mainly intended to support healthcare professionals’ decision-making, but increasingly outcome information is being communicated to patients to support shared decision making. Outcome information is also being incorporated into information platforms for patients and patient decision aids. Much of the focus of risk communication research has been on how to communicate probabilities effectively to patients. But what to do with information about the uncertainty associated with these probabilities? During this presentation I will use the context of early-stage breast cancer as an example to zoom in on whether and how uncertainty is discussed by healthcare professionals and patients’ preferences for receiving information about the uncertainty.

Dr. Ellen Engelhardt, Netherlands Cancer Institute - NKI

Gepubliceerd op: October 19, 2022