A Bias Radar for Responsible Policy-Making. Foresight-Based Scientific Advice

… a book by Lieve Van Woensel on trustworthy policy advice and on scientific advice beyond the evidence, published by Palgrave.

About the book

Policymakers prepare society for the future and this book provides a practical toolkit for preparing pro-active, future-proof scientific policy advice for them. It explains how to make scientific advisory strategies holistic. It also explains how and where biases, which interfere with the proper functioning of the entire science-policy ecosystem, arise and investigates how emotions and other biases affect the understanding and assessment of scientific evidence. 

This book:

  • Argues that emotions affect the understanding of scientific evidence
  • Examines how foresight thinking can anticipate the impact of scientific developments and emerging technologies on citizens’ lives
  • Develops practical guidance for planning, conducting and communicating robust scientific advice to policymakers

The article below highlights tips and tricks for trustworthy foresight-based policy-advice. It offers a sneak peek of the tools which are explained in the book.

Trustworthy foresight-based policy advice. Trips and tricks

Policymakers prepare society for the future. Lieve Van Woensel has developed a practical toolkit for preparing pro-active, future-proof scientific policy advice. As a sneak peek of her Palgrave book, ‘A Bias Radar for Responsible Policy-making. Foresight-Based Scientific Advice’, she presents four steps, which she argues adds trustworthiness, for preparing foresight-based policy advice. In the book she advocates for explorative foresight, systems thinking, assuring an inter-disciplinary as well as multi-perspective approach, bias awareness and the anticipation of undesirable impacts in policy advising. This paper focuses on four of the guiding components for responsible and trustworthy policy advice which she investigated. These are:

  1. Exploring the topic and its ecosystem: those who are affected by or who can affect the issue;
  2. Exploring possible biases: for the various actors and stakeholders that have been identified, explore their possible or known visions and investigate their possible biases;
  3. Exploring the topic from a wide range of perspectives, i.e. 360 degrees: using a guiding STEEPED scheme for scanning envisioned developments or events and their possible consequences for society, from a wide range of perspectives;
  4. Assessing possible decisions on possible unintended impacts, for instance on other policy areas. By doing so in a systemic way, we can avoid unpleasant surprises such as perverse effects of a policy.

The tools used for the various steps are also valuable facilitation tools for focusing during brainstorming meetings.

Figure 1: Four guiding components for responsible and trustworthy policy advice


Government and parliamentary policy-makers usually need advice on a specific development, problem or trend. They must also make strategic policy decisions to anticipate a more general challenge, threat or trend. Examples of policies for which foresight could be useful are:

  • How can we mitigate, and reverse, plastic pollution?
  • Policies for electric scooters
  • How can we prepare for a rise in the sea level?
  • How can we ensure the access to drinking water and fresh water resilience?

Step 1. Analyse the system, draw the ecosystem

Before dealing with a policy issue, taking a step back can enable you to see the bigger picture.

At the beginning of a foresight exercise for policy it is important to take a step back, to get the broad view, to form an overall picture. This involves first exploring the scope of the topic, as well as outlining the ecosystem with all stakeholders and actors. Systems analysis is a technique that breaks up a system into its component parts for studying how those parts function and interrelate to accomplish the system’s purpose. The book offers practical guidance for sketching the ecosystem. One way to conduct a quick systems exploration is to elaborate the questions: what, who, why, where, when and how. Reiterating these questions can help to gain further depth. This does not only help to detail the topic in focus, but also to produce a rough stakeholders’ analysis, including actors and stakeholders involved or possibly affected.

This phase essentially helps to get the bigger picture of the policy issue at stake, i.e. its entire ecosystem.

Step 2. Bias investigation

By understanding your own and others’ biases, you will become more open-minded.

Everyone in the ecosystem is subject to biases, prejudices or preconceptions. Biases can systematically distort our perceptions of facts, affect how we make up our mind, how we weigh evidence and how we make assessments. They can mislead us and fool us.

The book describes a series of biases and presents these in the form of a ‘bias wheel’, grouping in a systematic way some commonly appearing biases. This wheel is a practical tool for everyone involved in policy-making to check their own and others’ thinking. Lieve Van Woensel argues that bias-awareness helps us to be more open-minded and reflective when dealing with evidence, especially when there can be emotion-based opinions when working on controversial issues, for example genetic engineering, nuclear technologies, chemical use or climate change.

There are dozens of biases, many of which are relevant to dealing with scientific evidence and policy. The book groups biases in the scientific advice context into six categories. The first category is ‘research biases’ (biases which affect the generation of evidence or which influence the availability of evidence). The five other categories are those that affect advisers’ and policymakers’ assessments of evidence and, thus, the decisions they make on its basis. The tool distinguishes ‘culture and value biases’, ‘attention biases’, ‘interest biases’, ‘availability biases’, and ‘association biases’.

Figure 2: A spotlight on ‘association biases’ in the ‘Bias-wheel’

The bias-wheel offers for each of these categories a set of specific biases. To make this clear, the example of ‘association bias’ is explained here a bit more in depth. Association biases occur often, and steer our decision-making. Subdivisions include ‘romantic bias’ and ‘nature bias’. For instance, the ‘romantic bias’, by which one associates wood-stove fires with cosy evenings in Winter rather than with air pollution, or the ‘nature bias’, by which one uses the arguments that something is ‘good’ because it is ‘natural’ or ‘bad’ because it is ‘unnatural’.  This is, for instance, a possible reason behind the resistance to genetically modified food.

Of course, not all biases should be prevented; some are perfectly acceptable. However, advisers and foresighters preparing policy advice should be particularly and acutely aware of their own biases.

A full chapter of the book has been dedicated to biases.

Step 3. Explore the issue from all angles (STEEPED wheel)

Look from all angles, and you will deepen your insight.

When conducting foresight exercises, STOA at the European Parliament applies a STEEPED approach for acquiring insight into policy issues from a wide variety of perspectives. A multi-disciplinary approach, as well as the involvement of representatives from multiple stakeholders, helps to acquire the widest comprehension of the issue in question and to envision its possible intended and unintended impacts (in the case of STOA the studies policy issues are science or technology related).

STEEPED is a variation of STEEP or PEST, which are commonly used by foresighters. It adds ethical and demographical aspects to STEEP.

The STEEPED scheme is a checklist which specifies seven lenses through which we examine the impacts of techno-scientific developments, thereby ensuring all areas of interest or concern are covered:

  • Social aspects include religion, ethnicity, employment situation, financial means, well-being, disabilities, and habits;
  • Technological aspects include purpose of a technology or its application, the accessibility, efficacy, added value, dual use, research & innovation, and challenges;
  • Economic aspects include jobs (creation and losses), value creation, skills dependency, resource dependency, infrastructure dependency and affordability;
  • The environmental aspects include resource efficiency, energy efficiency, water efficiency, recyclability, sustainability, process safety, and product safety;
  • Political and legal aspects include liability, competition, and market regulations;
  • Ethical aspects cover respect for persons, respect for environment, availability justice, collective well-being and individual freedom;
  • Demographic aspects cover age, gender, household status, education level, occupation, and place/region.

It should be noted that not all STEEPED aspects are relevant for every studied policy issue and that overlaps between the seven areas are possible.

Figure 3: A spotlight on the exploration of ‘economic aspects’ using the ‘STEEPED-wheel’

The detailed STEEPED scheme, elaborated in depth in the book, helps policy advisers to zoom out from the details of techno-scientific developments and their related policy questions and study them holistically, ensuring that we investigate their impacts in all areas of concern arranged under seven dimensions.

The STEEPED scheme can also support the systems analysis and the stakeholder analysis, and help to explore possible biases. Hence, the four elements for being trustworthy are interlinked.

Step 4. Cross-impact analysis (doughnut)

Avoid bad surprises by assessing all possible policy options on their imaginable impacts.

Imagine you are investigating a certain policy issue (e.g. plastic pollution) with the purpose to propose policymakers a set of policy options, including their assessment. You have portrayed the ecosystem of the issue, explored the variety of opinions and possible biases throughout the ecosystem. You also collected and synthesised relevant evidence. You investigated possible intended and unintended effects related to the topic.

This should bring you to the position to build a list of possible policy options, i.e. courses of action to tackle the questioned problem/issue.

To allow the policymakers to make a well-reflected choice for the – in their view – most responsible decision, the final policy brief should include a detailed assessment of the selected policy option. For describing these, policy advisors can mainly build upon the previous steps. 

However, to avoid undesirable surprises, an extra assessment should be conducted at this stage, namely a cross-policy impact analysis. An example of a policy leading to adverse effects was the biofuel policy of the EU, which led to a change in land use, causing volatile food prices and threatening global food security.

Lieve Van Woensel argues that such bad surprises can be avoided by adding a systematic cross-policy impact assessment for each of the selected policy options, i.e. an assessment of possible cross-policy impacts for each of the proposed policy options substantially enhances the quality of policy advice.  

To avoid policy actions that will be later regretted, advisers should conduct a cross-policy impact assessment for each of the identified policy options. This involves identifying the existing policies that each policy option may affect and analysing the possible impacts.

Policy advisers can identify such global policies by for instance first scanning the United Nations’ Sustainable Development Goals (SDGs). Next, they can consider the main competences of the policy level at which they are working (e.g., the competences of the European Parliament, or the government or parliament they advise).

Once the relevant associated policies have been identified, another STEEPED-based exploration, focused on imaginable effects of the policy, can add cross-policy impact elements into the assessment for each of the considered policy measures.

The presented doughnut scheme is a guiding tool for preventing undesirable side-effects of policy choices, leading as advisers through STEEPED-based impact analyses of each individual policy option on the identified related policies.

Figure 4: A doughnut for systematic cross-policy analysis of each envisaged policy measure

The idea is to investigate possible related policies, for the topic for which advice has been asked.

Such a cross-policy impact assessment requires that it is first of all clear to which main policy area the issue belongs. Next, the possible associated policies are explored. Finally, a brainstorming should lead to the identification of possible consequences of the measure in the initial policy on each of the associated policies.

Cross-impact analysis is a method of anticipating how present trends may evolve and thus reducing uncertainty about the future. In policy work, it can be used as a sort of stress test to help policy-makers evaluate the adequacy of present policies, by identifying their unintended problematic consequences.


To enhance trustworthiness, policy advisors could apply, in a transparent way, some systematic methodology into policy preparation. Foresighters should be ‘trusted guardians’ of policy-preparation. They can do so, by systematically progressing through these four steps:

  1. Taking a step back to observe the broader context of the policy issue in question and the actors and stakeholders involved;
  2. In addition to the collection and synthesis of available evidence, understanding the opinions and exploring biases of all stakeholders identified. Foresighters should be particularly aware of their own biases. (Tool: the bias-wheel); 
  3. Explore the topic of the project from all perspectives (Tool: STEEPED-wheel);
  4. Assess policy options on their possible impacts on other policies (Tool: Cross-policy doughnut).

In addition, these tools are not just applicable to advisers and foresighters: they can also serve as great facilitators in any brainstorming session.

Lieve Van Woensel, PhD
30 October 2019

Head of Service
Scientific Foresight Unit
European Parliamentary Research Service

Author of ‘A Bias Radar for Responsible Policy-Making. Foresight-Based Scientific Advice’, Palgrave, 131 pages, 15 illustrations. Appearing soon.
Special offer / Get 20% off the printed book or eBook Use the following token on palgrave.com 2WtmrTG5AcwnpPb / Valid Nov 18 – Dec 16, 2019