Well, OK, maybe not everything. But perhaps a third of it:

...a third of medical research articles published in major scientific journals and then cited over a thousand times in the literature are later contradicted or have major questions raised over them.

Remember this the next time you hear about a "scientific study," particularly about politically charged issues, such as global warming. As Iain points out, this is an important point:

We acknowledge that most studies published should be viewed as hypothesis-generating, rather than conclusive.

There is a veritable plethora of these "scientific articles may be wrong" articles going around. OK, not a plethora, but today I blogged about one out of the University of Scranton. What gives all of a sudden?

I was at a talk a few weeks ago discussing climate change science. There's a number of studies due out soon (next few months) which refute a lot of the "anti" claims made over the last few years about climate change. Particularly about sampling errors on temperatures indicating accelerating warming.

It looks a little like a pre-emptive strike.

"It looks a little like a pre-emptive strike."

Or the fact that some significant sampling errors were discovered in claims made by those who deny that climate change is occurring.

Or the fact that some significant sampling errors were discovered in claims made by those who deny that climate change is occurring.

The thrust of the talk was more that the sampling errors that those who deny climate change have claimed exist in the data are in fact incorrect and as more data comes available, the more likely there is climate change and that it is occuring faster than data suggests it did in the past.

There's a lot of "noise" in the debate but the person discussing it pointed out that a lot of the "public" data is often a year or more behind what researchers are working on.

86% of all statistics are made up...or was that 92%?