Watch a 2 minute video clip on what CSOs could do to increase their policy impact & what policymakers could do to increase use of research-based evidence: MP4 (6.15mb) or WMV (1mb)

I am an anthropologist who was trained here in England many years ago when Wilson was Prime Minister. Right now, I have a consortium of thirty-five organisations that includes universities, public and private, NGOs, think-tanks and the government and its research bodies. We are funded by civil sources, Canada and other sources. We identify policy issues and bid competitively for research.

What is your experience of using research-based evidence to influence development policy?

Influencing development policy is usually a long-term process that is based on credibility and trust. In the region I come from, which is Latin America, there has been a long-standing tradition of a lack of trust between academics and government, despite the fact that people move between the two fields very frequently. We found that if we set the agenda in response to the problems that policy-makers want to solve and we involve them from the beginning, the chances of our research influencing policy are increased enormously.

What do you think are the most important tricks of the trade?

As I said before, firstly, set a common agenda. Secondly, look at the timing - policy-makers cannot wait three years for the results of the research. Thirdly, involve the policy-makers throughout the process. Fourthly, ensure that what you are giving them is related to actual programmes and projects.

Do you have any 'good news case studies'?

I hope we do. Let me just mention a few of these. One was a $50 million investment made by the government to produce books for rural children. As I explained in the seminar, we had an anthropologist who received only a small grant of $5000 and she went to six rural schools to find out what had happened to the books. She found that in four out of the six schools, the books had not been distributed. The reason that they had not been distributed was that teachers, the main barriers, had decided that they were very nice books and could not be given to poor individual children because they would spoil them. Instead, the teachers put them away on shelves. In the other two schools where the books were distributed, we interviewed the children. The children could remember the exact pages and page numbers where a little parrot or whatever was mentioned. It is amazing how it is the mindset rather than the thing itself which creates change. We work too much on the hardware and not enough on the software. We need to change the mindsets of teachers in order to be able to use the resources made available.

Do you have any 'horror stories'?

We had a project which dealt with employment, which was a priority. We conducted a study looking at what was more beneficial for a trainee, whether it was university, public or private or on-the-job job training. We found that there was a huge difference in terms of those who were trained on-the-job; they learnt more and were more efficient. We never involved the private sector and we had no constituency to champion the cause. When we did involve the private sector at the end of the study, they said ok, but that training would need to be subsidised. In the event, we were unable to reach an agreement with the government and the private sector regarding on-the-job training that involved some kind of tax relief.

What three things could CSOs do to increase their policy impact?

As I said before, partnership is very important. You have to partner with those who are going to use the study. Maintaining independence is the tricky part. Setting the agenda together so that a common agenda is achieved is important, but you have to make sure that whatever the results are, they are based on good science and they are independent. This is a balancing act because it could well be that the results do not please those who are making the decisions.

A second thing is that it is not a one-to-one relationship; you have to involve other players. We found that focus groups, interest groups and media are key players, especially if you are dealing with something controversial, because it is through them that you can influence policy. For example, with family planning, we had a very conservative group of health ministers who banned family planning for adolescents. Through research, we showed that AIDS was increasing, that the average age of AIDS patients had dropped five years over the past eight to ten years and that it was not a matter of information: adolescents knew what condoms and pills were, they just had no access to them. When we took the results to the minister, he said it was rubbish and he was not prepared to believe it. We had to go through civil society organisations, women's groups and youth groups to demand that the minister was sacked.

What three things could policy-makers do to increase their use of research-based evidence?

First of all, I think that policy-makers need to understand research. In many training activities, many of which are funded by the large development banks, the issue of using research for decision-making is not in the curriculum. I think that it should be. We need to train policy-makers not only to understand research but to know how to request it. I am not saying that every policy-maker needs to be an economist or needs to know econometrics or anthropology. What they have to know is how to ask the right questions, commission the research, respect independence and then use research in their policy. I think that we have looked too much at the supply side and not enough at developing the demand side. Lastly, I would say that we have looked at policy design but we need to look more at policy implementation. Unless we have good science on why policies work and do not work, we will be way off target in terms of what policy-makers need. In Latin America at least, there is a lot of research on policy design but very little research into how policy is implemented.

Back to Talking Heads CSPP interview list