There is an interesting article in the current issue of the Journal
of the American Medical Association (yeah, go figure) that does
a systematic examination of "Randomized Control Trials" (RCT
aka a between-subjects design with random assignment to
detrmine whether an intervention produces a systematic difference
relative to a placebo group) that have nonsignificant results; 
for the abstract see:

http://jama.ama-assn.org/cgi/content/abstract/303/20/2058?etoc 

Of 72 RCTs published in December 2006 and available through
Medline/PubMed, more than 40% of the articles had "spin" about
nonsignificant results in at least 2 sections of the paper (there is a 
more detailed breakdown in the abstract).  That is, even though
nonsignificant results were obtained, the authors used:

|specific reporting strategies, whatever their motive, to 
|highlight that the experimental treatment is beneficial, 
|despite a statistically nonsignificant difference for the 
|primary outcome, or to distract the reader from statistically 
|nonsignificant results

Back in the day, when I was a young, impressionable researcher,
I worked on a psychiatric research project and my attention was
captured by the conversation of a couple of junior ressearch
psychiatrists while waiting for a research meeting to being.  They
were discussing the profound wisdom of some senior psychiatric
researcher who is alleged to have said "you need to know how to
mine the gold out of the sh*t that's produced in research".  I admit
to not understanding why someone would say this or even think it 
was a good research strategy at the time but, by the looks of this
abstract, some people took the advice to heart.

I wonder if one would find similar results in the psychological literature?

-Mike Palij
New York University
[email protected]
.



---
You are currently subscribed to tips as: [email protected].
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=2788
or send a blank email to 
leave-2788-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Reply via email to