Using the best available evidence to improve public sector decision-making

12 July 2017

Dr Sarah Morton from What Works Scotland (a keynote speaker from our Evidence to Action conference) and Andrew Kibblewhite joined a group of senior public sector managers on 21 June to discuss how to make better use of evidence in decision-making. As Head of the Policy Profession in New Zealand, Andrew opened with a “view from the HoPP” and noted that without evidence, including evidence from evaluation, we have no way of knowing if policies and programmes are working. We need to get better at generating and using data, and to invest in capability to help us look beyond current priorities.

Sarah pointed out the need to shift from a focus on synthesising, spreading and using evidence to one that ensures the best evidence is used for each decision. A key point that resonated with Andrew and the rest of the group was that “knowledge is embedded in relationships”. One of the participants commented that if we don’t hear from different parts of the system, we will miss some of the knowledge that is needed. There is still a need to understand different types and sources of evidence, and how they can be brought together.

In New Zealand, there is a small pool of skilled evaluators (or ‘knowledge brokers’). While numerous agencies now have dedicated teams for research and evaluation we may not be making the most of our collective capability. In answering the question of how we might create more opportunities to connect and join-up the evidence, insights and expertise, the group discussed the need to build an authorising environment. We need to enable and encourage the sharing of what works, and what doesn’t. This requires more “permission” to fail (preferably small and fast) and a culture of sharing the lessons from that failure. There is a tension here within performance management frameworks and risk-averse cultures (often exacerbated by media scrutiny). It was suggested that there is an art to reframing “failures” as being “less valuable than other initiatives”.

Undertaking evaluation as an integral part of policy design and delivery will ensure that we build evidence “as we go”. There is a need for transparent innovation in the public domain, gaining small permissions and preparing the ground for an experimental approach, while thinking about different types of risks and how they can be managed.

This summary is an extract from notes of the meeting made by Simon Olsen of the Policy Project. The full “conversation tracker” can be found here.


Last update: 13 Jul 2017