It is a common problem that people tend to quote results far more accurately than the data warrant, heedless of the likely errors. Personal calculators have made this much worse; you could not easily quote eight significant figures in the days of four figure logarithms.
Michael Baxter
> Date: Fri, 4 Jan 2013 12:00:10 +0100
> From: William Stanbury <
[log in to unmask]>
> Subject: Re: Comment in Nature on discipline of statistics (Vaux 13 Dec 2012)
>
> For what it's worth, having worked several years with francophone
> statisticians, I've noticed perception diffferences which may be
> socio-cultural or other. A simple arithmetical example will suffice:
> occasionally but regularly whilst doing mental arithmetic to calculate
> answers, in a few seconds I might say:"The answer is circa 140,000." All my
> francophone colleagues would look at me in surprise, get their calculators
> out, take sometimes a minute to cacluate the answer, then answer:"You're
> wrong actually, the answer is 138,989.76."
