Tuesday, April 30, 2013

Certainty is Evil: A Musing.

I was preparing a presentation about social media when I came across this discussion of data by Al-Amin on VijanaFM. What a pleasure. Geek/nerds are the best. In years of trying to explain my uneasiness with statistics-worship I have had a hard time finding folks who can see both sides of the coin. 

Anyone who has taken statistics courses will tell you there is a lot more art to measuring things than is publicly admitted. Really good statisticians know it though- that's kind of the fun of the endeavor. There is this idea that numbers are 'hard' and they can define the world in dependable, absolute terms. Well...eh. They are just one of the forms of investigating or defining or 'knowing' what's up. Intuition is a form of intelligence, and there are others but I won't go there.

Today I was telling folks that we have a 73% adult literacy rate because I was using the figure offered by the World Bank in it's database. It is marked as a 2010 statistic, and was probably put together by the Government of the United Republic of Tanzania. I have severe, severe doubts about that figure. Not to blame the WB or anything, I just pick on them because they do try to uphold certain standards of rigor (bless) and carry a certain authoritative clout (blessing revoked). 

Is it accurate in 2013? Hard to believe, after the stunning failure rate of the past Form IV examinations. I know my government too: cutting corners and cooking data is not beyond them. Simple incompetence is not beyond them either. Between the donor-pleasing and face-saving tactics, the mendacity and general lassitude of my public servants* I am not inclined to treat GoT generated "data" with the respect it should command. 

That's story. Although Al Amin's essay focuses on metadata and the story of numbers, I wanted to introduce a further complication: that of positionality. Who is saying what, how did they generate the data, and why are they saying what they are saying? Of course, having studied sociology, I am comfortable with the idea that anything presented as Truth bears examination. Simply studying the history of science (okay, western science) will show you that it's nothing but a journey of discoveries that constantly get challenged and revised. My current delight is the Higgs-Boson affair. So the physics community is pretty sure they've discovered it... and there's already some who are pretty sure that's not the end of the line. 

Knowledge lives, uncertainty is life-affirming. Certainty, though, spells death. Discuss. ;)

*And that's what is horrible about public service. There are some incredibly good people in there. But they are overwhelmed by the awful, awful, awful rest of them. Paying taxes so that these creeps can keep not-showing-up at work is much of the reason we're furious


  1. I think science/geek people are some of the most aware of the limitations of statistics, but probably some deliberately decide to gloss over the uncertainties when they want to score some ideological points.

    So count me as another nerd who know numbers DO lie :-)

  2. Speaking of lying, a cautionary (if not rare) tale from the halls of European academia: http://www.nytimes.com/2013/04/28/magazine/diederik-stapels-audacious-academic-fraud.html?ref=magazine&_r=0

  3. @Dr Bob: yes, indeed. and we should tell the kids about it, lest they defer their own judgements in favor of received "knowledge." ah, but our educational revolution needs to happen.

    @Jetsetter: dah! what an article. yaani, processing. asante.


No biting, spitting, trolling or ugly insults- only pretty ones allowed.

A little birdie told me...

Follow MikocheniReport on Twitter