Beware statistics. Or at least beware how people wield them.

A friend was at work on a public service campaign aimed at meth abuse. Research had shown that young moms represented more addicts than any other demographic, and my friend's objective was to change that. I facetiously suggested that one way of attaining that objective would be to get other groups to increase their usage until it outstripped that of young moms.

As wisecracks often do, this one illustrates a problem. Statistics comparing where you are relative to where someone else is can be meaningless. A better question is, where are you relative to where you wish to be?

Utahans raise a fuss when they learn that their state spends less than any other per student on education. Yet by itself, this statistic isn't necessarily damning. If all other states suddenly dropped their spending to less than Utah's, would Utah's level of spending suddenly be OK? Never mind what other states spend. The real question is whether Utah spends enough, spends it wisely and spends it effectively. While I suspect the answer to all of the above is no, comparing Utah's spending to that of other states doesn't establish as much.

—Steve Cuno
 


Comments

towr
04/08/2010 11:57

While I agree comparative statistics can be meaningless in precisely the way you state, I do think they can be important signals. If one state pends much less on education than other states, then that is good reason to scrutinize why that is.
The same is true for every deviation from norms. If things aren't "normal" it is important to find out whether this is a good thing or a bad thing. "Normal" tends to be a fairly tolerable situation and thus functions as a decent baseline for comparison.

Reply

Your comment will be posted after it is approved.


Leave a Reply