11 May 2010

The secret to good research in #Tanzania? KISS and tell!

No, I am not making a point about building rapport with interviewees. (That would be taking things much too far.) But there is a useful lesson to be learnt from some fascinating recent research by the Centre for Economic Prosperity in Tanzania. They conducted a very simple study of lorry drivers on three routes between Dar es Salaam and the regions, collecting data on how many times they stopped, for what reason, and whether they had to pay any bribes.

The results are interesting. For example, 90% of times a truck is stopped by the police, a bribe is paid. And the average bribe is 1,272 shillings, resulting in a total "bribe-price" of 6,000-8,000 shillings per trip.

But my point is more about the research methods than about their findings.

Think about what they did. They took an important issue - transport costs - and found a simple way of collecting information on it. It also looks like they have focussed their research very carefully in order to collect only the information they really needed. They have then done some simple analysis of the data - no complicated regression models here - and reached some interesting conclusion. I don't think this will have taken them very long, it won't have cost them very much, and yet they have delivered good information on an issue that a lot of people have wondered about.

They Kept It Short and Simple: KISS.

Contrast this to the huge and hugely complicated studies performed regularly all over Tanzania. I've attended REPOA's annual research conference several times and have always been struck by the complicated nature of much of the research that is presented. It seems like research is only considered valuable if it depends on fancy economic modelling and convoluted analysis. The result is that even the researchers don't properly understand what they are saying. More attention is paid to a rigourous methodology than to the clarity (and even the quality) of the end product.

It's not really fair to single out REPOA, who also do a lot of very good research, since the majority of research I've seen in Tanzania has much the same problem. Which is also the case for the second problem: it rarely gets publicised and shared. So much research ends up on a shelf or an obscure website, as if it was only conducted for the benefit of the researcher rather than to add to our overall understanding of society.

Which brings me back to CEP's traffic research. They didn't stop with simple methods and analysis. They took their data and findings, put it together into a simple four page report that has been published and distributed. They've had some press coverage and I came across it online through a post on a facebook page. In other words: KISS and tell.

I wish more research was done in this way.