Government Social Research
George Clark, Head of Social Research at the Department for Work and Pensions (DWP) recently retired after 30 years in government including positions with the Department of the Environment, the Manpower Services Commission, the Employment Department and the Department for Transport, Local Government and Transport. DWP employs around 80 social researchers who commission research and evaluation projects worth £18 million on Welfare to Work programmes, pensions, disabled people, lone parents, families and child poverty and housing. He leaves a government social research profession that he says has both grown and changed, and which faces new challenges in terms of the skills of its configuration and the skills of it members, according to an interview he recently gave to the Government Social Research Unit.
It has changed hugely, not least thanks to the availability of computing power. I started as a researcher about six years before the PC was invented, 10 years before they were bought by government and it was another three years before they were networked. They have made a huge difference to our capacity to do things, and to the capacity of outside bodies.
Social researchers used to be much more concerned with the design of data collection instruments. Whilst that is still very important, we are increasingly interested in the quality and speed of analysis. There can be a bit of a trade-off here, but the overall impact is that quality has improved. In 1979 we did small well-designed studies, but didn’t have the technology to fully support what we were doing. We couldn’t easily do analysis with any degree of sophistication - we had to decide what we were doing, tell the computing people what we wanted, and they would do it for us. Now you can go through data repeatedly analysing in different ways, and link your analysis to other studies.
Prior to the rise of the new technologies we had stronger relationships with universities because government social researchers were a small group - a couple of hundred as opposed to 1,000 now. Often researchers operated in quite small units and were more academic than many of the units now are, having moved away from academia towards a focus on professional skills.
Government social research has since developed a slightly different relationship with academia, one of the consequences of which is that researchers are less connected with social theory - the core of the academic approach - and more concerned with analysis of outcomes. The result is a tendency towards straightforward empirical analysis. We’re very good at explaining what data means, but it is some way away from theory.
In the 1970s there were two key insights that people worked from. One was a Marxist insight which was very popular in universities in the 1960s and 1970s, and also classical sociological theories. People tended to come with a strong grounding in one of the two. Now people come from a wider diversity of backgrounds - we are simply looking for people with social research methods knowledge.
Yes. Until quite recently I had warned that we may face a return from evidence-based policy to a policy approach based more on gut feeling, but I no longer see that happening. For some years we have had a political opposition that is as committed as the government to the idea of evidence-based policy.
More of a problem is disappointment over the speed at which we can give somebody a policy ‘answer’. Despite the great investment in the idea of evidence-based policy, research findings don’t give you all the answers - or any easy answers. Research can be uncomfortable if it tends to show that problems are going to be very costly to resolve.
In Departmental Capability Reviews, one of the ‘green lights’ for DWP has been in ‘Use of Evidence and Analysis in Policy Making’. We’ve developed very strong data sources and found ways of linking data, all of which have strengthened our analytical capacity and its contribution to policy. We have a very strong commitment in this Department, and it shows in the kinds of people who become Senior Civil servants here; they tend to come from evidence backgrounds.
The recently published GRSU report ‘Analysis for Policy’ suggests a huge diversity of working arrangements between analysts and policy makers, but that there is broad enthusiasm for ‘bedded out’ teams amongst those involved.
What I have observed is that if you give policy colleagues responsibility for managing and directing social researchers, it results in a much stronger relationship. Until we did that, social researchers saw themselves as having ‘customers’ - a much more detached relationship.
There are some risks to bedding out, with questions around whether it is sustainable over time and the provision of professional direction. Those risks don’t seem to have materialised so far, however. More widely, there’s a merging of policy and analysis, and I think this will be a challenge for government social research over the next 10 years.
Government Social Research is nearing the end of a 20 year period of steady growth and a particular model of social research in government. We’ve built up big central divisions in some Departments, quadrupling the number of social researchers. I don’t think this trend will continue, and we need to explore where we go from here.
There are two things we need to resolve. One is the need to retain and improve professional excellence, and I think there is still a lot to do. We have some very, very good researchers, but they will need to do considerably more in their professional development as opportunities for different kinds of analysis grow, and greater demands are placed upon the discipline.
The other challenge is about finding a home for social researchers in the Civil Service. People like me came to the profession in the 1970s and stayed, but I don’t think this model sustains. In future, researchers will need to make a choice between continuing with lifelong careers in social research, which may mean them moving in and out of government for their development, or taking social research as a starting point for a career in policy making or delivery in the Civil Service. Getting to the top will be about social researchers committing themselves as civil servants, which will require them to learn about other things. That said, an early career in social research should see them well equipped to do that against the requirements of Professional Skills for Government. They may need to think of themselves more as analytical policy professionals or analytical delivery professionals, however, and ‘get their hands dirty’ in one of those ways.
It has made a big difference - it has got impetus, it is well resourced and it is at the centre of things. It has made a big difference in terms of resourcing training and development, giving a more coherent voice to government social research, and taking some of the weight off individual Departments. Prior to GSRU, Heads of Profession had a voluntary arrangement for getting together. This wasn’t particularly effective, whereas GSRU has the scope for greater leverage.
As a Head of Profession, there are now more resources to draw on, and I have felt more empowered in setting out programmes of professional development. We’d like more resources, of course, but what we have is excellent. We’d like more help, say, with annual conferences, but I don’t see any real gaps.
back to top