1 The role of evidence in policy: how the mix matters Tim Tenbensel School of Population Health University of Auckland Panel Track: 4) Evidence Based Policy Abstract This paper proposes a way of thinking about the role of evidence in policy derived from a typology of knowledge first developed by Aristotle, and revived by Danish policy theorist Bent Flyvbjerg. I also argue that each of these types of knowledge also has close affinities with particular ‘modes of governance’. Scholars as diverse as Hood, Rhodes and Jessop have each argued that that key challenge of governance is the capacity of the state to mix different modes. As each mode of governance requires a particular type of knowledge, an integrated approach to policymaking and implementation requires that policy actors pay attention to all three types of knowledge. Accordingly, policy actors that are able to draw upon and combine different types of knowledge are more likely to be more effective, and research-based evidence can be seen as a necessary but not sufficient component of robust policy processes. Introduction The term ‘evidence-based policy’ (EBP) became ubiquitous over a very short period of time. One effect of this rapid diffusion has been to shoehorn discussions about policy-relevant knowledge into the relatively narrow confines of the term ‘evidence’. This paper begins by situating ‘evidence’ in the broader context of policy-relevant knowledge. The early part of the paper examines how different normative approaches to policy processes, namely rationalism, pragmatism, and ‘participationism’ have portrayed the role of research-based knowledge and evidence in policy. The differences between these approaches is brought into sharper relief through the introduction of Bent Flyvbjerg’s adaptation of Aristotle’s intellectual virtues – episteme, techne and phronesis. Each of the normative approaches to policymaking emphasises a particular intellectual virtue, or ‘knowledge type’. The middle section of the paper shifts the focus to concepts regarding ‘modes of governance’ (hierarchy, market network, community) that has become prominent in public management literature over the past ten years. As it turns out, these Aristotelian knowledge types also have close affinities with particular ‘modes of governance’ – namely hierarchy, networks and communities. However, compared to the policy process literature, the modes of governance literature is these days less concerned with arguments about which mode is normatively superior, and more concerned with questions regarding the interaction between different modes.