I’ve mentioned in one of my earlier posts, how the western world, particularly the US, are fanatic about seeing numbers/statistics for just about everything that is told.
Sometimes the kind of statistics are too unbelievable. Not because of the data itself — but when you start thinking how the hell in the world, did they derive the kind of deductions they do.
So, what goes into a statistic?
I realize that statistics may be generated based on inputs from real people (surveys) or by juxtaposing/extrapolating arrays of existing data. So what you find next might apply to one or the other or both.
1) Survey base — How many responses (whether human provided or data-backed) made up for the survey?
2) Data Points — Relevance is key. The objective of the survey and the relevance of the data points used to derive this is vital
3) Whose data is it? — It matters who we ask questions to. Of course, assuming contextual relationship is preserved, how does one make sure responses are from a cross section?
4) Which brings us to Who’s doing the survey? What are his motives?
Lastly, if the surveys and statistics we see in the US infact consider all this, they’d be at the CMM 5 of information management. I dont personally think they are.