Using trials in piloting policy interventions

Roger Jowell, City University and chair of the Working Party and Annette King, Cabinet Office

Slides

Discussants:

Discussants will each have ten minutes to respond to Roger’s presentation and provide a commentary from their own perspective.

Carol Propper, University of Bristol

As an academic I would strongly like to support the use of policy trials but also recognise that the constraints of policy making can mean that such trials change mid-way through the process. The challenge is to work out ways of undertaking data collection and analysis that recognises this.

Slides

Sue Duncan, Government Chief Social Researcher, Prime Minister's Strategy Unit, Cabinet Office

Government researchers are often caught between a rock and a hard place. They recognise the value of methodologically sound policy trials but they also understand the constraints under which policy is made. The process of reconciling these two perspectives is challenging but it
cannot be overlooked if trials are to make a real contribution to policy development.

Slides

Alex Bryson, Policy Studies Institute

Random assignment offers real opportunities for unbiased estimates of policy impacts in a number of spheres. However, they are not a panacea. They are difficult and costly to implement and they don't answer all policy impact questions. It is therefore worth considering the limitations of random assignment from a technical perspective and the value of combining random assignment with non-experimental methods of impact assessment.

Slides