How Do People Feel About Citing Arxiv Sources?

Question

[Paul_Muller]

I used a method (testing consistency of shape between 2 histograms) from this e-print (http://arxiv.org/abs/0804.0380) in a manuscript I will be submitting. arXiv is not peer-reviewed, but from what I understand the majority of the information is reliable and the author has over 400 refereed pubs listed on his resume.

Opinions?

Thanks!

FYI is a bioinformatics paper dealing with genome similarity

Answer 1

[David Quigley]

The answer depends on where your manuscript will be submitted. Some journals (I checked Nature and PLoS Biology's author guides) require that papers be "published or submitted to a named publication" to be allowed on the reference list. I may be mistaken, but I don't think "uploaded to a pre-print server" counts. Certainly you should cite the work of anyone whose method you are using, but you may need to consult the an editor to get clarification on whether that particular article can be formally cited, and how to give credit if the answer is no.

[Paul_Muller]

Thank you Laurent, David, and Dan. Certainly, I think the method is sound or I would have found an alternative. I'm submitting to Genome Biology, so I will check with them on the proper way of doing it. It'll be my first "first author" paper. Best, Paul

Answer 2

[Laurent Gautier]

If you used it, cite it ! It seems to me that there should not be any doubt about it. If you do have doubts about citing it, may be you should then have doubts about using it (and have your work rely on it) ;-)

References are here to point readers of your scientific communication to more details about prior work your are using (you know, the "stand on the shoulders of giants" credo). A side-effect is that the authors of that useful work are thanked with an increase in their number of citation (you know, the "publish, get cited, or perish" mantra).

For example, Open Source software is not peer reviewed (hold on... if it is Open Source with a user base, it is peer-scrutinized and peer-improved) but you do cite it, don't you ?

Answer 3

[DG]

Definitely if you used it it should be cited. But my impression of arXiv is that its reputation varies from field to field. Reviewers or the Editor may have an issue with it, but chances are they won't even notice. Like Laurent said, if you used it you should cite it.