Interpreting science – understanding, advocacy and policy
[Full report: InterpretingScienceApril2013.pdf]
Interpreting science – implications for public understanding, advocacy and policy formation
A discussion paper
Sir Peter Gluckman KNZM FMedSci FRSNZ FRS
Chief Science Advisor to the Prime Minister
Society uses science in many ways for its benefit – and sometimes individuals and groups within society can misuse science either accidentally or intentionally. However, the proper use of science and technology is essential to our economic, social, and environmental health. As science and technology are being used to address increasingly complex issues and policy makers face the difficult choices on how to reach trade-offs between contesting views and inputs, science has become more important in providing a relatively value-free knowledge base on which the public and policy maker can reflect, integrate their own values and priorities, and make decisions to use or limit technologies or to introduce new policies and programmes or to change current ones.
Thus how society obtains and understands scientific and technical knowledge is critical to a well performing participatory democracy. But because science now deals with very complex matters, many of which have high values content (for example environmental issues, the development of behaviour, the use of genetic modification), how that science is presented and used can have major impacts on societal decision-making and progress.
Clearly those who are active in science communication have major roles to play in allowing the public and the policy maker to understand what we know, what we do not know, and the nature of the inferential gap between what we know and what might be concluded.
Too often a piece of science is misunderstood, misused or overstated – sometimes something is presented as established science when it is not, other times it does not suit advocates to accept the science as established when it is. This paper will give examples of each of these and highlight the questions that should be asked when interpreting evidence. It also explains how scientific conclusions can be established even when all the details may never be resolved or there is still debate over some specifics; classic but diverse examples of this are our understandings of evolution and of earthquakes.
Two matters have given me particular concern. The first has been the increasing trend for the complex nature of science to be ignored or misunderstood. It leads to the – sometimes rhetorically convenient – argument that you can find a scientist to support any given position. This totally misinterprets the scientific endeavour and does grave mis-service to the public interest. Scientific consensus is unlike social consensus – it is not a matter of the loudest voice or compromise; it is a more consultative process by which the expert community examines the currently available evidence and incorporates it into an understanding that integrates what we know and ac-knowledges what we do not know. The very nature of the natural world and scientific observation means that variable results can often be expected and the process of scientific consensus addresses this problem.
The second challenge is that of science being wrongly used as a proxy for a debate over values. This may occur consciously or unconsciously. It is obviously psychologically easier in some situations to say the science is not settled, or not settled enough, than to enter more complex discussions that have strong values components. Much of the climate change debate has used science as a proxy when the real debate, which is valid, is over responsibility between nations and between generations.
The miscommunication and misuse of science in the public domain can engender serious mistrust in the scientific enterprise. The scientific community has to do much to improve its behaviour, but equally society will better served if science is not miscommunicated and not misused in advocacy or policy formation. We do not live in a technocratic society, we live in a democracy and values will always and should be the final arbiter of decisions that are made. Values are formed and moulded by what we know and what we think. We often interpret what we hear or learn in the light of pre-existing biases. Science is the only process we have to gather reliable knowledge about our world, our society and our environment. It should therefore be seen as an essential input into the formation of opinion and values, but it can only do that if its is honestly represented and honestly used, and if society is provided with adequate opportunity to understand the scientific process.
The challenges of the twenty-first century are many – whether we look at the obvious issues of climate change, sustainable economic growth and resource security or the more subtle issues arising from greater urbanisation, changed ways of communicating and the changed nature of our society, science and technology will be essential to navigating a productive and safe path to the next generations. If science is not used and communicated in a way that is appropriate and with high integrity and fidelity we risk sailing into dangerous waters. This paper is intended to assist in better navigation by highlighting how to interpret complex science.
Sir Peter Gluckman
In 2011, I released a discussion paper entitled Towards better use of evidence in policy formation which discussed the interaction between knowledge and policy making. It points out that the process of policy formation is improved if evidence is first incorporated in a value-free manner and only then should the various values-laden domains such as public opinion, fiscal pri-orities, diplomatic concerns and electoral considerations be overlaid upon knowledge. When the science itself is presented in a values-laden way it is compromised and loses its privileged place in policy formation. Conversely, the failure to use evidence properly can lead to decision making which is less likely to produce effective and efficient outcomes.
Beyond the obvious domain of ethics, science is never absolutely value-free. The key values domains to consider are, first, expert judgements over the quality and sufficiency of data and, second, the limits of the data available. There are nearly always inferential gaps between what is known and the decisions that are implied by the knowledge. The gaps and uncertainties must be acknowledged. If that is done with integrity then science advice can be delivered in an effectively value-free way. In Towards better use of evidence in policy formation I argued that science should be presented in a manner that is not based on advocacy but is delivered by ‘honest brokerage’ to the policy maker. It is for the policy maker to overlay the other critical domains of policy formation.
We live in a participatory democracy and ultimately it is for the public and the policy maker to use scientific knowledge – the challenge is how to assist them to use it properly. It is therefore important that citizens and policy makers become aware of the uses and misuses of scientific data. Often it is easy to find an apparent scientific claim or data to support any particular intervention or action, or argue for or against a particular policy change – but is that data sufficiently sound or reliable to act as a basis for decision making? There are many traps in extrapolating from a single study, and this is an important reason why expert but impartial advice is needed in bridging science and policy.
A further issue emerges because the media often inappropriately interprets or profiles a particular single report or claim because it will attract attention and have impact, and it is this over-emphasised and sometimes contrary claim that influences public opinion, even in the face of considerable opposing evidence. A more engaged and higher quality of communication between scientists and society is essential if society is to make better use of scientific knowledge.
This discussion paper is intended to highlight some of the matters that need to be considered when interpreting science.