A few years ago (many years ago?) someone wrote an article for the
newsletter of the newsletter of the ASA Section on Teaching
Statistics in the Health Sciences in which he described having each
student select a published article "at random" and check for
internal consistency. Round-off errors were NOT counted as
violations. His students found errors in one quarter of all
articles checked. My experience with journal clubs suggests nothing
has changed in the intervening years.
Take some comfort that most errors do not change the article's gross
findings.
> As an example, there was an article in a recent issue of an APA journal
> where the researchers randomly assigned participants to one of six
> conditions in a 2x3 factorial design. The N wouldn't allow equal cell
> sizes, and the reported df exceeded N. Yet the article said the
> researchers ran a two-way fixed-effects ANOVA.
The imbalance could be due to dropouts or spoiled/invalid data.
While the reported df shouldn't exceed the sample size, I don't see
anything in what you've reported here that would preclude a two way
fixed effects ANOVA.
>
> One of my students wrote on her homework, "It is especially hard to know
> when you are doing something wrong when journals allow bad examples of
> research to be published on a regular basis."
That's why you teach your class the way you do, ands that's why it's
important they take it...seriously!
> I'd like to hear what other list members think about this problem and
> whether there are solutions that would not alienate journal editors.
Standard collegial behavior is to contact the author directly and
ask whether s/he feels a correction is appropriate. If the author
feels a correction is in order, s/he will generally do so and
publicly acknowledge you for pointing out the problem. "Dr. Lise
SeShea has kindly pointed out that..."
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
