Can arXiv papers be considered as reliable sources of information?

[Dadface]

Is it correct to assume that, in general, authors who have papers accepted by arXiv eventually try to get their work published in a journal? If so what should we make of arXiv papers which apparently do not get accepted by any journals? Should we take those papers as seriously as we would take a journal published paper?
As an example, this morning I was scanning through an active thread on the quantum physics forum of PF and reference was made to an arXiv paper from 2005. I couldn't help wondering if the author(s) of this paper were content to just leave it in arXiv or whether they actively tried to get it published elsewhere and failed in their attempts. If the latter is the case it seems to be possible that the referees,who are supposed to be experts, did not agree with the findings of the work. What then should we make of the work?

 

[Orodruin]

Is it correct to assume that, in general, authors who have papers accepted by arXiv eventually try to get their work published in a journal?

In general, but not as a general rule. For example, people will also put thing such as lecture notes and PhD theses on the arXiv. These are typically not published in peer reviewed journals.

Should we take those papers as seriously as we would take a journal published paper?

In general no. You would need to have enough knowledge in the particular field in order to judge the paper. Also note that being peer reviewed is not equivalent to being correct. It just means that the authors managed to convince one or more reviewers to accept the paper.

 

[f95toli]

Is it correct to assume that, in general, authors who have papers accepted by arXiv eventually try to get their work published in a journal? If so what should we make of arXiv papers which apparently do not get accepted by any journals? Should we take those papers as seriously as we would take a journal published paper?

It is important to realize that the arXiv does not really "accept" anything; anyone with a valid e-mail address (e.g. someone@someuniversity.edu) can upload a manuscript to the arXia and all that might happen is that someone takes a very quick look at it (I doubt they have time to do more than to read the title and abstract) to make sure it is not obviously nonsense before it becomes available (which is why it takes less than 24 hours before a manuscript that has been uploaded becomes available). The manuscript is never really reviewed as such.

As an example, this morning I was scanning through an active thread on the quantum physics forum of PF and reference was made to an arXiv paper from 2005. I couldn't help wondering if the author(s) of this paper were content to just leave it in arXiv or whether they actively tried to get it published elsewhere and failed in their attempts. If the latter is the case it seems to be possible that the referees,who are supposed to be experts, did not agree with the findings of the work. What then should we make of the work?

The most common reason by far for getting paper rejected it is that it deemed not suitable or important enough for a journal. It is actually quite rare for referees to claim that something in the manuscript is flat-out wrong. Writing a paper takes a lot of time and effort and you will only submit it once you have convinced yourself and your collaborators that what you have done is correct. This does not mean that all submitted papers are free of errors (far from it), but it does mean that it the errors tend to be less than obvious and the referees are unlikely to spot it; at least in papers from well-known groups.

Once you have worked in a field for a few years you get to know most of the players, and if one of those uploads something to the arXiv I will usually assume that what is in there is correct and will eventually be published (and I might not even bother looking up the published version). This does not necessarily mean that I believe that they've reached the right conclusion, I can still e.g. disagree with their initial assumptions or believe hat they missed something important.

 

[Dale]

If so what should we make of arXiv papers which apparently do not get accepted by any journals? Should we take those papers as seriously as we would take a journal published paper?

Definitely not. Such papers have not undergone peer review. That doesn't mean that they are wrong, but they are definitely less credible.

 

[Dadface]

Thanks for your replies everyone. Personally I don't know what to make of those papers that have been with arXiv for several years. It's easy to imagine that the authors have made several approaches to journals and have been turned down by the referees each time. Just one possibility is that the referees have come to a correct decision.

 

[Orodruin]

Thanks for your replies everyone. Personally I don't know what to make of those papers that have been with arXiv for several years. It's easy to imagine that the authors have made several approaches to journals and have been turned down by the referees each time. Just one possibility is that the referees have come to a correct decision.

I would agree that this is a possibility among many. Unless you know (professionally - not that they are your friends) the authors are reliable, it is probably good to have a larger portion of scepticism (although you should always have some portion of scepticism).

 

[Dr. Courtney]

I've seen the reviewer side of the peer review system too many times to place too much trust in it. In addition to having editors publish papers over my objections as a reviewer, colleagues and I have also had editors refuse to publish our comments pointing out obvious flaws after mistakes make it into print.

For the most part, there is no simple way to determine the quality of a paper from the venue in which it appears. Usually I develop a sense for which authors are careful and which are careless. But I never recommend not bringing your critical thinking skills.

One of my first significant contributions to Dan Kleppner's program at MIT was finding a significant mistake in a Phys Rev Letter that they had published a few years before. The authors, the peer-reviewers, and most of the readers for several years after the paper was published all missed the error.

 

[rootone]

Can arXiv papers be considered as reliable sources of information?
No. but they usually are well thought out proposals that are more digestible than Youtube videos and Facebook memes.

 

[Vanadium 50]

The pros and cons of peer review won't be settled here. However, I think that anyone asking that peer-reviewed papers be error-free is being unrealistic. (One could even argue that by removing the obvious errors, peer-review ensures that remaining errors are subtle and more difficult to find). However, I think the ensemble of peer-reviewed papers is more reliable than the ensemble of all papers, peer-reviewed or not.

There are reasons for not publishing a preprint. I've done it. I had a paper in review hell until it was no longer timely. But it was on the arXiv and everyone who needed to see it has seen it. But this is an exception. Most of the time, papers that never move beyond the arXiv shouldn't.