What should I believe???

Thanks for that link,Catherine.The internet is just full of so much info it is difficult ,at times ,to filter what is what.

That is a good link Catherine.

I was discussing this issue with my son who is a research scientist.
One additional aspect is the importance of the publication cited in its field. Who are the peers? top scientists or people from obscure universities in Outer Mongolia? If you want to publish a paper, there is probably a journal somewhere that will publish it, if you want your research to have influence it needs to be reported in a journal with credibility in that field. Of course if your research is 'maverick' you may find it difficult to get published in the more mainstream journals.
There is a fairly rough and ready tool that can be used (and misused )to judge this: the impact factor. My son told me that in his field he would not want to be bothered to submit to a journal below a certain impact factor, (what that impact factor is differs from field to field) few would read it , it would have little credibility and add nothing to his CV.
see http://en.wikipedia.org/wiki/Impact_factor
The journals themselves sometimes cite their impact factor,but you have to subscribe to be able to get at the listings.
There is another ranking called eigenfactor which is available freely online. Unfortunately in subjects such as obesity or diabetes many of the journals have fairly low impact overall but at least you can check to see if the journal is in the rankings at all. (if not why not? it may perhaps be a new publication and not yet ranked)
(rankings below for 2006)
endocrinology rankings http://www.eigenfactor.org/results....perpage=100&ordering=perarticle&Submit=Search
nutrition rankings
http://www.eigenfactor.org/results....perpage=100&ordering=perarticle&Submit=Search

I looked at the link and found that there are 4 publications listeed under the same ISSN number.

However, some peer reviews may be relevant at the time and later the paper shows up as much weaker.
I recently read a paper on the benefits of a High carb diet for treatment of diabetes, which seemingly got into the "Lancet".
I wouldn't have given it a "C" as a piece of GCSE coursework. Investigating 2 variables at once isn't valid in my book.
not to mention the conclusions, which were not borne out by the evidence.

looked at the link and found that there are 4 publications listeed under the same ISSN number.

Click to expand...

ie: Obesity, Obesity reseach, that looks like a data entry error! if you look up the link it refers to one publication ( I assume that someone has unsuccesfully tried to recitfy it and ended up compounding the problem with 2 nul entries), needs sorting out!

However, some peer reviews may be relevant at the time and later the paper shows up as much weaker.

Click to expand...

Of course I agree any research can and will be interpreted differently as researchers attempt to replicate old research and go on to investigate different aspects. Thats how knowledge develops.

catherinecherub said:

Everyday there are more and more links to the internet published on this forum. It is difficult to decide if they are valid, especially the research ones.
This link provides a Peer Review Guide on how to decide the validity of the research.
http://www.senseaboutscience.org.uk/PDF ... wGuide.pdf
I, for one, find it mind boggling sometimes and hope this will help others in their decision.

Regards,
Catherine.

Click to expand...

Hi Catherine.
I thought it was a perfectly good link. A clear easy to read guide which can help people who are trying to research on the internet.
A definite way to find valid, scientific research as opposed to just anecdotal 'evidence.'
Thank you.
Ken.

Caherine, I didn't mean to offend.
I just found the wrong data entries jumped out at me.

Having been trained in a strict science school myself, I can be a bit pedantic

Catherine , it wasn't your link Hana was objecting to, it was an obvious data entry error on one of the ranking lists I gave.

Another thing always worth reading is the Disclosures at the end