http://pubs.acs.org/doi/abs/10.1021/ci700332k

4/12/2009 1:56:45 PM

I knew it was the lemons all along!

4/12/2009 1:57:53 PM

even funnier is the title "The trouble with QSAR (or how I learned to stop worrying and embrace fallacy)

4/12/2009 2:02:24 PM

what am i looking at here?

4/12/2009 2:03:55 PM

It is the abstract of a peer-reviewed article in an ACS journal.

4/12/2009 2:05:09 PM

more like embrace phalluses

4/12/2009 2:07:27 PM

I'm scared that the article is one of those random-generated science articles that makes it past the peer review process

4/12/2009 2:07:53 PM

I get a kick from the fact that it got accepted with a Dr. Strangelove reference in the title. Chemistry is supposed to be serious business.

4/12/2009 2:39:39 PM

this is a cool thread

4/13/2009 10:43:22 PM

i approve

4/13/2009 10:45:26 PM

wat?!?

4/13/2009 10:46:35 PM

I read the article and as a scientist, I think it raises some good points. Beyond the abstract a good highlight is:

Quote :

"...we have largely exchanged the tools of the scientific method in favor of a statistical sledgehammer. Statistical methodologies should be a tool ... but instead have often replaced the craftsman tools of our trades: rational thought, controlled experiments, and personal observation."

The article highlights an issue that I think appears in all areas of science. There are so many publications that abuse statistics and misuse the numbers to support false conclusions. I've seen many cases where researchers performs complex analysis to quantify variance, but then attribute all the variance to whatever causative factor they're advocating with zero attempt to isolate independent sources of variation and deal with the possibility of covariance.

The figure in the abstract was just a tool to illustrate the follies of blinding using statistics without considering covariation and plausible and mechanistic causation.

[Edited on April 13, 2009 at 11:46 PM. Reason : formatting]

4/13/2009 11:45:17 PM

The sources in the abstract indicate that it is actual data. My guess is that it is left out of the paper because of size limitations.

4/14/2009 1:27:38 AM