[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Pathological Epidemiology (was "I spent 10 years in radiation data and emerged as the foremost expert ...")
I was struck by Bjorn's comments:
>I was actually searching for some crazy epidemiology. Sometimes it
resembles
>evolutionary theory:
>1. Create variation (multipage questionnaires etc)
>2. Assortment of the information (data, genes etc)
>3. "Natural" selection of the fittest (a large fraction of the variation is
>lost in every new generation/study... and then there is publication bias by
>various media)
>The fittest create the money base for the next study...
>Remember Sir Karl R. Popper in 1962: "...assembled in accordance with the
>collector's interests and points of views".
>My personal initiative only,
>Bjorn Cedervall bcradsafers@hotmail.com
In my former life as a grad student in psychology (Oh so long ago! I never
practiced the craft.) I recall that this was how we were actually trained to
do statistical analysis! Back then, the thinking was that if you couldn't
get a correlation coefficient of at least 0.35 you just weren't trying:
Correlation not strong enough? Just increase the variance (more data points)
until it is! Nowadays Pearson's R isn't much used, but this would
correspond to a relative risk of, what - 1.5 or so?
It is not surprising to me that at least some of the more notorious
purveyors of junk epidemiology seem to have started out in the social
sciences. They are trained to approach research more as providing a support
for their theory-making, than as a Popperian test of their theories. It's
a lawyerly approach to evidence gathering. Select the data to support your
case (theory) and ignore (or suppress) that which tends to contradict it.
The problem with Psychology and the social science in general ( and one
reason why I eventually got out of it) is that there are so many more
theories than facts. Bjorn's comparison to natural selection is very
interesting, but Adam Smith needs to be paid his due as well: The selection
of competing theories in the marketplace of ideas, is based on their revenue
generating power rather than their explanatory power.
Clayton J. Bradt, CHP
Principal Radiophysicist
NYS Dept. of Labor
Radiological Health Unit
voice: (518) 457-1202
fax: (518) 485-7406
e-mail: Clayton.Bradt@labor.state.ny.us
************************************************************************
You are currently subscribed to the Radsafe mailing list. To
unsubscribe, send an e-mail to Majordomo@list.vanderbilt.edu Put the
text "unsubscribe radsafe" (no quote marks) in the body of the e-mail,
with no subject line. You can view the Radsafe archives at
http://www.vanderbilt.edu/radsafe/