Is peer-reviewed data "THE DATA"?

[Alric]

And discussions in peer-reviewed articles trump "some guy's blog".

Science does change and data is disputed with more data. However, the latest scientific consensus is what we should use to make decisions.

Can we agree to this and move on?

[Scott1972]

It's not clear to me what you are trying to say.

Being able to reproduce the data is what is important, not a person's or group's data.

For theoretical predictions, it's the prediction conforming to empirical data, and those measurements being reproducible.

[Pipirr]

Blog commentary can be very useful in understanding peer reviewed research. Where things get silly is when some guy with a blog is trying to rewrite science as we know it, and their unpublished, non-peer reviewed blog posts become elevated to a position of privilege.

Take D'Aleo, with his long paper* (http://icecap.us/images/uploads/US_Temperatures_and_Climate_Factors_since_1895.pdf )about problems in the US temperature record: those are grand claims. But are they valid claims, or has he just got it all wrong?

If there really is a problem with the temperature record, then publish*. The people that need to know would notice it. If it is unpublished, I think it is reasonable to wonder why not (http://tamino.wordpress.com/2008/02/03/exclamation-points/). If research has merit, get it through peer review, especially if the claims that are made and the conclusions that are reached are grand. Otherwise, what is there to distinguish it from any other blog post by yet another internet crazy?




*The ability to convert a file into PDF format does not a published paper make.

[Alric]

Right. That is why I started this thread. I see it happen with evolution and climate change where people will start linking to blogs in the face of publication as valid counterpoints.

[Ivor the Engineer]

While I generally agree (and hope) peer-reviewed journal articles have higher validity on average than unpublished work and points of view, I'll just mention Alan Sokal (http://en.wikipedia.org/wiki/Sokal_Affair).

Plus let's not forget other factors such as researcher bias, publication bias and, dare I say it, elitism.

Whether a piece of research is published or unpublished, all that really matters are the method and the results.

[Alric]

Yeah but what can you expect when the publication is on a post-modernist journal. I don't think you could pull the same stunt in Science magazine, at least not for long. And that was precisely the point.

[Ivor the Engineer]

That's what I thought, but I'm sure I've seen positive results for homeopathy in "proper" medical journals, which members of this forum with suitable qualifications have subsequently torn to shreds. (The contents of the articles, not the journals.:)).

It makes me wonder what the peer review process is supposed to be weeding out from publication.

[Pipirr]

{snip}It makes me wonder what the peer review process is supposed to be weeding out from publication.


It isn't just weeding out. The process of peer review itself can benefit the author and the readers by improving the quality of the paper.

Although sometimes, junk does get through.

[Pipirr]

Right. That is why I started this thread. I see it happen with evolution and climate change where people will start linking to blogs in the face of publication as valid counterpoints.

And with climate science, there is so much published research, why ignore it and focus on the unpublished?

Methinks shenanigans.

[mhaze]

And discussions in peer-reviewed articles trump "some guy's blog".

Science does change and data is disputed with more data. However, the latest scientific consensus is what we should use to make decisions.

Can we agree to this and move on?

Data is not the same as scientific method, procedure, conclusion, hypotheses. Data in most cases should or is separate.

the latest scientific consensus is what we should use to make decisions.
Can we agree to this and move on?

No.

[Dr. Imago]

And, on a sidenote, a nice little article here...

http://news.bbc.co.uk/1/hi/sci/tech/290121.stm

If you can't trust the science was conducted with integrity, subsequent peer review means nothing. It is inherently expected within the process that the manuscript being submitted is accurate and the data (and conclusions) are based on events that actually took place. Sometimes this, sadly, just isn't the case.

I'm not suggesting (nor could I possible know) what, if any, science being speculated about here has been fraudulently produced and submitted. But, it (again) underscores the importance of independent validation of all landmark findings.

-Dr. Imago