Interview with Richard Bartholomew – Joint head of the Government Social Research Service

It is important that government policy is supported by the best evidence possible, and this is as true in the area of social science as any other area of research. Richard Bartholomew, Chief Social Researcher in the Department for Education and joint head of the Government Social Research Service, discussed with Socialsciencespace the role of the Government Social Research service, social science more broadly in government, and the problem of the seemingly obvious.

What is the role of the Government Social Research Service?

The Government Social Research Service (GSR) is a professional group of 1,200 members who provide, manage, or commission social research for government departments. There are members within most Whitehall departments and the devolved governments, as well as associate members in other non-departmental public bodies, such as the National Audit Office and the Forestry Commission. GSR members are all social researchers, from a mixture of various academic disciplines including psychologists, sociologists, economists, and geographers, and most join GSR after they’ve done a post-graduate degree or later in their career after they have worked in other fields of research. The role of GSR varies between departments, but the key task is to provide professional research input into each department, influencing the design and conceptualisation of research, the commissioning of research, the management of research, and helping with the publishing, disseminating, and translating of research into something that can be used to inform policy decisions.

What is the value of social science to government policy?

The current government is committed to using research evidence in decision making, although it is obviously not the only factor influencing policy decisions. They are a democratically elected government with views, beliefs and principles. Our role is to contribute to the quality of the overall decision making process by supplying good research-based evidence. Research evidence is key in assessing whether something is likely to provide good value for money, whether it will actually work, or whether it has actually worked. It is however too simplistic to say that you can or should make decisions on the basis of research evidence alone. The skill of being a politician involves building a consensus on a particular issue in order to move forward, and research is one of the factors which contribute to that process.

Is there enough funding for social science research within government?

The funding has been adequate, with considerable expenditure on major evaluations in the last few years. Obviously within the new tighter fiscal environment the case will have to be made for funding research, but it’s right that research has to justify its importance and explain how it will contribute to decision making and evaluation. You can always do more research and get more evidence, but you have to show that there’s an evidence gap, or how that evidence will contribute to improving policy and in what ways.

How much interaction is there between social researchers in GSR and academia?

Our research budgets are used to commission research from academic institutions as well as other organisations depending on the type of research. Some large population surveys require an extensive fieldwork structure, and this is more usually done by research agencies. At other times we need the in-depth knowledge and experience academics have on a particular issue.

We are not just interested in the research we directly commission, but also the work people are doing through different funding mechanisms and through the normal higher education research funding routes. It’s important to know what the state of knowledge is, and if academics are finding things that we’re not aware of. Whilst some academics are very good at communicating their findings, others could improve the clarity of their message and target what they are trying to say.

Do you have any advice for academics who want their research to influence policy?

Sometimes it’s difficult to draw out what the policy implications of research findings are if the researchers don’t understand how policy works. Researchers need to understand their audience, and what the touchstones are, drawing out why the research is relevant, rather than just assuming that it is. People will often respond to social research findings by stating that it’s obvious, or that we know it already. Some things are obvious, or may seem obvious, but only with hindsight, and the key issues have to be drawn out for policy makers. A good example is the Effective Provision of Pre-School Education research, which demonstrated the importance of improving quality in our early years provision, and not just providing childcare. It’s now something that seems very obvious, but was not obvious or widely understood at the time the research was commissioned.

What is the impact of social science, and how should social scientists be demonstrating it?

Generally single studies don’t have a huge impact or greatly influence the course of decision making or expenditure. There are a few studies like that, and the Effective Provision of Pre-School Education was one of those, but for others it is a slower process of influencing opinion and gradually persuading people. It’s not right to base huge decisions on just one study. You’d expect a range of studies to either confirm or refute each other, building up knowledge of what’s likely to be true. There is an argument for replication of studies, and we treat that as normal in medical and natural science research, but it is often more difficult to get funding for replication in the social sciences. In government we attach a lot of importance to review studies, especially systematic reviews, which look at the range of research that has been undertaken on an issue and assessing the quality of the different studies. It’s a slower, more cumulative process than just doing one piece of research and that having an impact or not.

Skip to toolbar