The battle over evidence-based approaches to development

11-16-10 Evidence, from iampeas on Flickr (cc)

If you haven't already seen it, it's worth looking at the debate over evidence-based approaches to development assistance that ran over three days on Duncan Green's From Poverty to Power blog.

In the first post, Rosalind Eyben and Chris Roche suggest that evidence based approaches, along with results and best value for money, distract attention away from the politics and power relations involved in development assistance. Does the framing of initiatives as 'laboratories' (Millennium Villages are given as an example) mean that people are seen as subjects of an experiment?

Chris Whitty and Stefan Dercon of DFID respond by saying that evidence is more broadly-defined and carefully than Eyben and Roche appear to suggest: it's also about understanding the social, political, and economic factors that may enable or constrain success of different approaches. Practitioners use a mix of techniques, not just randomised trials, and are aware, on the whole, of the need to be honest about the limits of evidence.

In the final post of the series, Eyben and Roche come back and outline some points of agreement – but it's clearly a topic that will continue to create some discomfort and debate, and rightly so.

The discussion highlights the need to examine what we mean by evidence, and what a complex and contested development landscape means for the way this evidence is interpreted. More engagement between researchers and funders will be one way to move the debate forward. (The points above are a very broad summary of some quite detailed arguments, so I'd recommend reading the original posts to get the full sense of what's being said.)

Both Duncan Green and Chris Whitty will be speaking at the STEPS symposium next week. If you haven't got a ticket, you can still get into the public lecture on 6 February by Anne Glover, the EU's Chief Scientist, on the balance between evidence and the 'real world'.