Robert Dawson wrote:
>
> Jerry Dallal wrote:
>
> > The problem for me
> > with the statement "Z is NEVER a better test for the mean under
> > circumstances they are likely to encounter [in psychology]" is that it
> > reads like an indictment
>
> It is. The last thing students in Intro Stats need is one more red
> herring.
>
If I'm reading you properly ("NEVER"; it is an indictment), then
one of the important rules students are to get from your course is
that there is a critical distinction between the percentiles of the t
distribution and the standard normal distribution when talking about
means of large samples. I don't elevate it to a rule (*) but I tell
my students the direct opposite--the distinction doesn't matter. In
fact, I find it hard to think of an original thing to say about the
issue. This semester, I put the points listed by Mosteller and Tukey
(Data Analysis & Regression, p 5-6) on a set of transparencies, with
attribution, and used them as the basis for discussion about why the
distinction is so *un*important. M&T just about cover it all.
I guess we'll have to agree to disagree on this one. I hope
I'm not misstating your case.
----------------------
(*) My rules are more mundane. "Show me the data!" "In a pre-
test post-test experiment, subjects do *NOT* serve as their own
controls." "<specific oddball situation>, get help! There's no rule
that can save you!"
