> >From Charles Frost: "Statistics is also calculated guess work. You
> can make statistics give any
> answer you want."
>
> Sorry, this really bugs me. Would you say that engineering can give
> any answer you want?
Statistics are a way of summarizing data.
By choosing which data to summarize, or by tailoring the criteria used
to summarize it, you can make the statistics say anything you want.
In this thread, for example, we have some people doubting that
divining rods work, and others saying they've seen it work, or know
someone who saw it work. But nobody who said, "My friend's second
cousin tried it and it didn't work." Is that because there aren't any
of those, or because the failures don't get publicized, nearly as much
as the stories with positive results?
Andy
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|