> On Mon, 7 Jun 1999, Steve Milliser went:

> I am looking into using Microsoft Excel in my introductory
> level statistics course in the Fall.

Here's my concern.  About a year ago, on one of the stats newsgroups,
someone posted an easily replicable demonstration showing that, for
large numbers, Excel miscalculates standard deviations.  And of
course, if it miscalculates standard deviations, it's going to
miscalculate variances (I tried; it did).  And if it does _that_,
imagine what it might do to an ANOVA.

To my great frustration, I can't seem to dig up a copy of the post.
I'll check my files at home.  I recall that I got the odd behavior on
Excel95 for Windows; I don't know whether it occurs in Excel 98 or in
Mac versions.

--David Epstein

Reply via email to