Comment

John Birch Society Rides Again at CPAC

237
Mad Prophet Ludwig12/22/2009 11:03:13 pm PST

OK, I have been meaning to write something about this for some time. It’s a boring discussion about journal papers, and I will likely try to repost it sometime on a more active thread. However, there are some things I want to point out about scientific papers, peer review and journals in general.

The process is not perfect and I will give examples as I go. However, in some sort of ensemble average, it works well - or perhaps as well as possible.

So first off, not all journals are created equal. There are top flight, journals of note for instance that are hard to get even a very good paper into.

If you are in physics they are PRL, Phys Rev, Nature and Science.

These journals strive to promote papers that they feel are of general import to a large community. The idea is that the paper is there to represent the newest and most important discoveries in a given field. Phys Rev is divided into several sub parts, for instance, phys rev D is particle physics. PRLs (physical review letters) are seen to be important for the whole physics community in principle and are short communications that frequently have a longer paper or two associated with them in Phys Rev or someplace else.

Now getting into these journals does not make a paper perfect. There are dozens of great papers that should have gotten in and dozens that were flawed. In my own career, one of my first papers was actually pointing out flaws in a PRL and the group that wrote that paper later retracted their findings. However, their flaw was not glaring and they reported their results in good faith. They were just wrong. It happens and the main point of science is that the community scrutinizes work. It does not end at peer review.

After this tier of journals, there are more specialized journals. For instance, the British Journal of Fluids is huge and wonderful to be published in if you are a fluid dynamicist. More than once a paper that should have been in a “bigger journal” ended up in one of these journals and later became huge. Also more than once, it was a specialized paper that would be of great interest to a certain community, and that journal was the perfect one for it.

Beneath that tier journals get either more specialized or more obscure.

Occasionally, as in rarely, a paper that later became huge but was way ahead of its time, or just not properly received, could only find its way into one of the more obscure journals. Sometimes These journals are a good place to publish a crazy idea that might be right… and if it turns out to be right, then you got it out there, and if it turned out to be crazy, well not many people read that journal….

The best measure of a paper is how many citations it has received. A paper with 500 citations is one that 500 other papers saw fit to mention as part of their arguments. A paper with over 500 citations is huge.

In fact, a paper with 50 citations is very respectable. That means that 50 other researchers or research groups are using your stuff in their own arguments and presumably as part of their own story with their own results that also in turn back yours up.

Again rarely, a paper might languish for years and then get rediscovered only to later receive the response it deserves. This can happen for many reasons. But also this is rare.

Also it should be mentioned that proceedings of national academies are big deals too. A report from the National Academy or the Royal Society has been past many sets of eyes before publication. These are not trivial things.

So as a lay person looking at a sea of papers. If you see one with 500 citations from a top flight journal, chances are that science in it is very, very sound. If you see a paper with no citations from the Physical Journal of the East Slovenia fire fighters association, quite possibly much less so.

I bring this because if you look at the big papers and the big proceedings, you will find that AGW is supported and that Evolution is real.

On the other hand, the Disco Institute publishes elsewhere.