Thank you for providing that direct-link to the Suryanarayanan paper.  Having now read the whole thing, I think it's absolute drivel.  The fact that it got published makes me question the peer-review process at this journal.  The fact that it was funded by a National Science Foundation award (http://www.nsf.gov/awardsearch/showAward?AWD_ID=0924346) makes me despair for our fiscal policy.
 
About half the paper can be boiled down to a call for case studies as an analytical approach.  There is possibly some merit to this argument though not, I believe, in the way the author intends.  Single-factor, properly controlled studies are the gold standard for confirming a causal relationship once you have a plausible hypothesis about what to investigate.  The author is correct that those studies aren't very helpful when you don't know what to test for in the first place.  Case studies can have value in understanding the wider context and then developing the plausible hypotheses for more specific investigation.  Increased use of case studies could be a valuable first step toward better understanding the complex question of bee health.  The author seems to me, however, to be arguing that case studies are ALL that is needed.  Neither research nor experience support that conclusion.
 
The author goes on to describe an improperly-controlled experiment (a fair criticism, though perhaps debatable in that specific case) but then uses that one example to attack the very concept of controls. That over-generalization is a logical fallacy.  The author asserts but provides no evidence for the claim that controlled experiments are a bad idea.
 
Later in the paper, the author has the gall to attack the idea of a 95% confidence interval.  Good grief!  What is he proposing that we do instead?  Make decisions based on guesses without any idea whether the guess is true or not?  Would he treat his own family's medical needs so cavalierly by tolerating similarly low standards for human health care?  Granted, the choice of 95% as opposed to 94 or 96 or even 90 is a "social convention" but is he seriously arguing that we should consider 25?  Or do without?
 
The author's red-herring about Type I vs Type II errors is based solely on the beekeepers perspective and, despite his contant references to the wider community, ignores the fact that the growers' perspective on pesticide value is reversed.  It is a call for the discredited "precautionary principle" without even the honesty of using that phrase.
 
Perhaps I am too harsh but the author's unnecessary use of pejorative language throughout the article did little to encourage the assumption of good faith.  I am unimpressed and do not think that I got value for my tax dollars.

Mike Rossander

             ***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software.  For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html