Blog

If we want evidence-accountable policies, stop bashing MPs

Posted by Tracey Brown on 13 April 2015

Evidence in policy making was barely a discussion before the last election. That has changed, and not just on the subjects of climate change or crime, where politicians have made evidence the debating point for a while. But while Whitehall’s discussions about running policy trials and finding out what works are being celebrated, the research community regularly despairs of Westminster, of the low number of scientists there (likely even fewer after 7 May) and of the dreadful stories about how some of the members and committees treat evidence. Today our survey of MPs with Ipsos MORI shows they’re wrong to despair.

Nearly seventy per cent of MPs support or strongly support the use of randomised controlled trials to design and test social policy. Around half think that more trials of policy are inevitable. Few (just 9%) think the cost of RCTs is a barrier.

Should we be surprised by this? Yes and no.

Yes, because RCTs are clearly quite difficult to understand, beyond the initial proposition of testing policy. The survey, and the tentative nature of the views it captures, also shows that MPs in general lack confidence and knowledge about what is involved in trials. Around a third think that ‘randomly choosing’ who gets an intervention is unfair, while two-thirds support the use of pilot studies without control groups - experiments which still depend on groups not getting the intervention yet give less robust results. But that is not a cause for ridicule - it's an opportunity that the research community must leap upon.

And no, MPs’ positive attitudes to policy trials shouldn't actually be such a surprise – Westminster is supposed to hold the executive (Whitehall) to account. ‘Does it work?’ ‘What is the evidence?’ should surely be up there with ‘are you telling the truth?’ and ‘did you do as you promised?’ as questions they can ask of government. We should be fuelling and championing this happy congruity between their constitutional role and our questions about evidence in policy.

What is more, the Ipsos MORI survey – conducted face to face with MPs - also reminds us that ‎MPs have far greater experience of mediating policy issues with the public than many in the research community. They speak in human. While they rated research evidence high among the list of inputs they believe are most important for developing policy, 70% said they have used personal experience and constituents’ concerns to justify policies. Taken together these responses suggest sensitivity both to expertise and evidence, and to the human terms in which evidence and decisions need to be communicated. (The public’s preference for research to be communicated with vivid stories emerged in a study in 2014 by Ipsos MORI and the Royal Statistical Society in 2014.)

So instead of despairing of the lack of scientific research knowledge in Westminster, the research community needs to concentrate on how better to explain the benefits of randomised controls. They are, after all, the closest thing a politician will ever have to a magic wand. And as my colleague Chris Peters explains on the Ask for Evidence website, when voters start asking the MPs returned on 7th May about evidence they will be reminding those MPs that this is more than an esoteric concern of research design: it is an essential part of holding government to account on behalf of the rest of us.