Using data visualisation to communicate scientific uncertainty to non-expert decision makers
Communicating the uncertainty and reliability of scientific information and insights to non-expert decision-makers is essential to the application of scientific discoveries in industry, government and business, and is one of the key challenges facing the field of science communication. As decision makers, everyone of us instinctively considers uncertainty in our daily decision making. The future is unknown and information is never complete, it is therefore important to understand how reliable and predictable scientific insights are.
As a simple example, how much money you invest in a new share offer will differ depending on the sources of your advice (a cousin vs a financial advisor).
In the data driven future that is emerging, where non-scientific audiences are demanding insights from the wealth of data being created around us, uncertainty is an essential guide to translating data to insight and applying those insights to the real world. However, evaluating the uncertainty and reliability of data derived information is not intuitive. While statistical and scientific uncertainties are commonly communicated to the statistician or scientist, there are currently no accepted methods or standards for communicating uncertainties to non-expert decision makers.
This research uses the Australian National Cancer Atlas as a case study for developing uncertainty visualisations that communicate to policy-makers and non-expert decision makers. An additional output of this research, is a structured workshop process that guides the creation of targeted uncertainty communication material and visualisations. The National Cancer Atlas uses cutting edge statical methodologies to develop estimates of cancer risk in Australia, at small geographical resolutions. Uncertainty visualisations generated within this project aim to communicate the reliability of estimates so that health policy makers can effectively plan for the future through prioritising future research, addressing inequalities across geographies and demographics, enhancing cancer survivorship, as well as maximising and targeting diagnostic programs.
A copy of the full paper has not yet been submitted.