A few thoughts....

Advanced math is not my strong suit...so I readily admit that sometimes the statistical analysis is over my head when I read some of these studies.

With that said, I'm amazed that people (scientists) think they can take a set of data with an unknown degree of accuracy, analyze it with some math (but comparing it to no data from outside the data set produced by the study), and determine how accurate it is.  I simply don't believe it.

I'm amazed at the lack of real data that is reported..."managed for honey production" could mean anything.  There is rarely any mention of management issues (replacing queens, feeding, spinning out honey, etc).

I think there should be a clear distinction between research talking about "bees" and "beekeeping".  As someone that generally isn't feeding bees (feeding honey when we do feed), I'm amazed at the amount of feeding I see being done...both by beekeepers and researchers.  When I think about "bees", I think of critters that modulate their population based upon environmental conditions...not livestock that is fed to keep them from losing weight.  I understand the need for researchers to look at bees as they exist in the commercial beekeeping industry, but that is separate from bee behaviour (or at least, it needs to be noted that it is how bees behave when being managed in a specific way).

When we did our "No Bee is an Island" experiment (where we looked at the fermentation of bread yeast in a sugar/water medium in the presence of formic acid), the most important step was to prove to ourselves that we could get consistent results with consistent inputs..at first we didn't.  It was simple enough to rectify (we simply used a thermostatically controlled water bath for all samples....it's good to have darkroom equipment around).  I've said before (and I think Jerry disagreed), that bee studies should use two separate control groups (not a larger pool of controls to be averaged....two separately evaluated groups)....if the controls differ very much, the results (or at least the degree of impact of the results) would have to be looked at more closely.

I think Plos One (and other journals) should offer cash rewards for those that show the flaws in papers they publish....to me it is clear that the media and "the people" have become incapable of evaluating science, and we need some incentive to encourage the close reading of research (and to raise the bar on the writers and reviewers...ultimately on the publishers).  Speaking for myself, I've been offered opportunities to write (and publish pending peer review) critiquing published studies...but I don't work somewhere where I get paid to do such things (the study authors and generally the reviewers as well are).  I simply can't afford the time to write the kind of detailed article I would like to write when I have so many other things on my plate that cost me money by not getting them done sooner. 

I have given a few talks titled "The People's Homework" about this very topic....focused mostly on the Harvard (Lu) study, I conclude the talk with a discussion of what it looks like to me when someone cites 3 or 4 studies implicating imidacloprid as the cause of CCD, when one of them is the Lu study.  I don't have time to read every study as carefully as I've read the Lu study (reading the references, taking careful notes, etc)....but I can't possibly believe that someone who cites that study to me as "evidence" has read it carefully (critically)....and therefore hasn't read any of the studies they are citing to me critically.  They don't understand what was done, what was claimed, and what the obvious (never mind not so obvious) flaws are.  When the Lu study is included in a group of studies to show "evidence", the whole group comes into question...and whomever assembled the group has shown themselves to either want to mislead purposefully (a good lie to encourage proper action of others), or just another tool of bad science fueling media sales.

I'm always amazed at the sweeping claims made by some researchers in their study titles and conclusions.

I think that's enough for now :)

deknow

             ***********************************************
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

Guidelines for posting to BEE-L can be found at:
http://honeybeeworld.com/bee-l/guidelines.htm