Rich Ulrich wrote:
>
> - just a note on z-tests -
>
> On 3 Nov 2003 07:37:58 -0800, [EMAIL PROTECTED] (Robert J. MacG.
> Dawson) wrote:
>
> [ ... ]
> >
> > Ninety-nine times out of 100 when people refer to a "Z test" they do
> > not mean the theoretically-justified-but-very-rare test with known
> > variance, but a badly-done t test, in which quantiles of Z are used
> > where those of t should be, that was once taught as "the right thing to
> > do" when n was greater than 30.
>
> Who does that?
> Is it just the old folks went to school 35 years ago?
I try to weed my shelves vigorously, and books that do this sort of
thing are high-risk for weeding. Nonetheless, it took me about 2 minutes
to find, tucked on a top shelf, Mendenhall, Beaver, and Beaver's
"Probability and Statistics", copyright 2003.
Chapter 8 of this book is "Large-Sample Estimation", which introduces
the
z interval for proportions (valid), the z-interval for the mean (with
variance inferred; put forward seriously if n>= 30) and the z interval
for the difference of means (if n1 and n2 are both >=30).
Chapter 9 is "Large-Sample Tests of Hypothesis" with, you guessed it,
the z test for the mean, the z test for difference of means, and the z
test for proportions.
Finally in chapter 10 we reach "Inference from Small Samples". On the
second page it is stated that
"This chapter introduces some equivalent statistical procedures that
can be used when the _sample_size_is_small_" (MB&B's italics.)
This is ambiguous; but in my opinion the stress encourages the reader
to conclude that this is a necessary condition, not just sufficient.
The book (page 367) states that n=30 is the "dividing line between
large and small samples", though it is admitted that this is arbitrary.
I found no reference in the chapter *anywhere* to the fact that it is
legitimate to do a t-test on more than 30 data. The t-table in the
inner cover and appendix reinforces this by giving rows for every value
of nu from 1 to 29 then jumping to infinity.
So, yes, I'd say it's still happening. What really bugs me is that I
taught from a predecessor of this book when I was young and foolish.
-Robert Dawson
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
