Statistics In Africa

I read this article on Africa Check by Georgina Alexander and John Endres- The trouble with statistics about Africa -on how “Bad and incomplete data bedevils African statistics…”

This is so true.

I have had harrowing experiences in two countries, including mine, and in both cases, there were possibilities for immediate correction but the consequences would have been politically and administratively disastrous. The softer option of maintaining the lie was adopted.

Here was a case in point that turned out to be a tragicomedy:

During the skirmishes that normally precede elections in Africa, a University was invaded by the security forces and some students were reported abducted and disappeared. The Minister of Information in that country went public and answered journalists that there had been “zero deaths”. A month later a house under construction by the same Minister was ransacked and petrol bombed at night, and on the charred walls of the burnt-out building was a graffiti stating “Zero destroyed”!

There is light in the tunnel though. Information technology is now making a joke of some of the published data, simply by revealing non-corroborative information. Randomised sampling techniques, mobile phone surveys, satellite imagery – now easily accessible,  are all producing information that are inconsistent with officially published data. 

Nevertheless, some of our Governments still maintain that the only correct data are what they publish. On the other hand one should be wary of ultra-negative data published by some “reputable” NGOs who in their anxiety to show bad governance fail to search for all facts and in the process mislead.

My take is that technology will resolve this soon, and poor data, whether due to deliberate manipulation or weak systems will be an era of the past. Patience is the word.