Michael Snow wrote:
Delirium wrote:
Stan Shebs wrote:
Actually, it's sort of interesting that this
is news. While it's
common practice to have peer review for scholarly books, the "peer"
part of the term should be a hint that it's not the publisher doing
the checking! My guess is that our reverence for the printed page
is such that we just assume no one would dare to print without
being certain of its correctness.
In many fields the peer-review of even scholarly books is not all
that high. In the sciences, journal articles hold much more weight
than books, because there's a perception that anybody can get a book
published.
I'm not so sure that the peer review of journal articles is
necessarily that much better. I recently dealt with a situation in
which the submitter of an article was able to specifically request
that his article not be forwarded to one of the logical candidates for
peer-reviewing it, because he anticipated that this person would give
an unfavorable review. Cherry-picking your reviews hardly counts as
rigorous scholarship in my book.
The theory of peer review is nice, but even in academia the execution
is often shoddy and politically skewed. Practices at different
journals vary, of course, so the reputation of the journal needs to be
considered beyond just the question of whether it qualifies as
peer-reviewed.
This past week the CBC has featured the story of Dr. Ranjit Chandra who
faked results about multivitamin benefits for the elderly. He has since
left the country, but seems to be doing well running a non-existent
university in Switzerland, and selling his multivitamins in India. See
http://www.cbc.ca/nl/story/nf-chandra-retraction-20050302.html
http://www.cbc.ca/national/news/chandra/
http://www.sethroberts.net/ (bibliography) and
http://www.sethroberts.net/articles/chandra/1992_Effect_of_Vitamin_and_Trac…
One quotation from the third part of the programme.
The former editor of the British Medical Journal,
Richard Smith, says
there's a good reason for that.
"Anybody who knows about peer review knows that sometimes it will pick
up a fraudulent study, but it's by no means guaranteed to do so,"
Smith says. "And that's largely because it starts from a position of
trust. So if somebody says there were 200 patients, then you assume
there were 200 patients. You don't say, 'Well show me their records
and show me their photographs, I need to see them.' The whole thing is
based on trust."
The pharmaceutical industry in particular has a reputation for making
its studies look good.
It is never safe to say that a peer-reviewed article is ipso-facto valid.
Ec