In line with the ‘critical perspectives’ suggested in its title, this book is sub-titled ‘Why evidence doesn't influence policy, why it should and how it might’. Although this is slightly tongue in cheek, the various contributions to this edited collection have sought to explore the ways in which research has been suggested to influence policy, the extent to which it really does, how policy gets implemented in practice, the language that policy makers use and why this matters, what constitutes ‘evidence’ in the first place and the contexts into which evidence and policy are introduced. Although the tone has often been somewhat provocative – even slightly cynical at times – this is the result of a firm belief that research can and should influence policy, but that the relationship between the two needs to change in order for this to be a more fruitful partnership. At times, the tone also seems to be the result of the prior experience of the contributors engaging in policy debates at different stages of the process – sometimes with positive and sometimes with less positive results. Overall, we hope that this book provides a challenging overview and something of a ‘warts and all’ approach to the policy process, highlighting potential ways forward as well as some of the problems.

Above all, different chapters repeat key messages about the limitations of traditional rational models of policy formation. While some would argue that the world should function in this way, the evidence (and our personal experience) is that it typically doesn’t. Instead of approaches that are retrospective and summative, research needs to contribute at a much earlier stage in the process to be much more formative and interactive. There are various ways in which this partnership could be more successful than in the past:

• • Policy makers need to seek to learn from research when policy is first being debated, not after it is already formed. Although much research takes place after policy has been developed in order to assess its impact in practice and its future potential, we believe that research and policy can link most fruitfully when ideas are first being discussed and worked up, helping to set the parameters of what may and might not be possible or successful.