Bill Knight wrote:
> The way this world is ---
...
> SUMMARY:
> * Don't be like a certain social sciences graduate
> * student at our university who, after failing to reject her
> * null hypothesis, nevertheless went on to draw conclusions
> * from her data. (Worse than that, her department
> * had her seminar presented as a star example.)
There's a chapter in J. Utts' mostly wonderful but flawed low-math intro
text "Seeing Through Statistics", in which she does much the same. She
presents a case study based on some of her own work in which she looked at
the question of gender discrimination in pay at her own university, and
fails to reject the null hypothesis [no systemic difference in pay between
male and female faculty]. She heads the example "Important, but not
significant, differences in salaries"; comments (_perhaps_ technically
correctly but misleadingly) that "a statistically naive reader could
conclude that there is no problem" and in closing states:
"As one student suggested, the male faculty who are complaining
about the study's conclusions because the differences are not
^^^^^^^^^^^^^^^^^^^^^^^
statistically significant should donate the "insignificant" amount
of $3,612 to help a student pay fees next year"
from which one might infer that the study (a U.C.Davis technical report,
which I have not read) had said, in effect "well, if hypothesis testing
won't help, w'll just have to draw our conclusions without hypothesis
testing."
This is from the 1996 first edition; if I recall correctly it has
survived more or less unchanged into the second edition. (So, IIRC, has the
highly selective presentation of examples involving ESP -a student
finishing the book could be excused for thinking that experimenters were
unanimous on the existence of this phenomenon! The presentation of the
chi-squared test has improved slightly; in the first edition the 2x2 test
was given along with a "magic" [sic] number for the 5% critical value, and
in the section at the end of the chapter (annoyingly titled "For Those Who
Like Formulas") the rxc formula was given with no indication that the `
"magic" number' was different! I believe that this has been dealt with in
some way.)
Curiously, none of the reviews that I saw for this book thought there
was anything odd about any of this. I concur with the reviewers that this it
is in the main an excellent book; but it has some very worrying bits.
-Robet Dawson
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
