As a young scientist I find it increasingly difficult to filter out the good vs not-so-good research publications. Especially given the number of publications that get published every day. Reports on retraction watch like this one, don't help. Very often some aspects of research can be missed or misinterpreted by an inexperienced young researcher.

I was wondering if there is any resource where I could suggest a publication for a quick, short commentary and obtain comments from the community? I am aware of f1000 idea but I'm not particularly fond of the subscription requirements they impose.

Thank you in advance for any suggestions.

  • 1
    You could try ScienceOpen. It's a free platform for researchers with the possibility of open comments, open peer-review and sharing research. Commented May 3, 2017 at 16:21
  • 2
    A journal club with your lab or department is a great way to do this on a paper-by-paper basis to get some other opinions. Obviously you won't be able to cover every single relevant paper in that setting, but it's a start and will help you learn the skills of critically reading literature. I know there are some attempts to do this on a broad basis but at least in my field the participation is far to sparse for you to likely find a write up on the paper you are interested in, and there are so many papers that it isn't really feasible for a small community of experts to review every paper.
    – Bryan Krause
    Commented May 3, 2017 at 16:39
  • If your field was biology, I would also encourage you to post specific questions about a particular paper on Biology.SE (not "please explain this paper for me" but "Scientist, et al. used approach XXYY to study Z, but in other papers I've read they used approach PQRS. The authors didn't mention PQRS. Is XXYY really a valid approach in this context?"). There might be another SE site that is appropriate for your interests if not biology.
    – Bryan Krause
    Commented May 3, 2017 at 16:43
  • 1
    @mjp For this reason my lab has at times had a journal club between 4-5 other labs; this worked out because we were all fairly small groups so the total participation was ~15 people. Our interests overlapped sufficiently that we would typically understand the content, if not the context (such that a key role of the presenter was to help establish context for the group) of the papers interested to others in the group, but diverged sufficiently that we would not typically read the same primary literature. The balance can be tough to achieve but definitely brought different perspectives.
    – Bryan Krause
    Commented May 3, 2017 at 17:06
  • 1
    @BryanKrause that's getting close to what I wish existed on a larger scale. Looking forward to other journals adapting similar features. The trouble with this approach is that many scientist are afraid to attach their scientific name to a paper rant, hence the low throughput.
    – mjp
    Commented May 3, 2017 at 17:18

1 Answer 1


This question was asked a long time ago, and generated some discussion in the comments, but never received an answer. I'll try to answer it to get it off the unanswered queue.

The problem of assessing the quality of some paper is partially solved by the review process. For example, in mathematics we have AMS Math Reviews and zbMATH. These have been discussed on academia.SE before.

It's also meant to be solved partially by the journal quality, but the OP is wise to bear in mind that even published papers have errors.

For sure the best approach is to learn how to read a paper and decide for yourself if it seems correct. I echo the comment suggesting a journal club, which seems like a great way for students to develop this skill together, perhaps with some faculty input. For more isolated researchers, the comment suggesting ScienceOpen seems like a good idea.

Usually assessing the goodness and correctness of a paper is part of the skill set of writing referee reports, and there are tons of resources online about that. Personally, I read the paper carefully and ask myself if the methods being used have enough power to prove the results being stated, and if there are any obvious pitfalls to avoid. Obviously different fields will do this differently. In a data-driven field, it would be good to think about the assumptions the authors are making and the potential for bias in the data. The resources linked above can help guide you in figuring out what questions to ask yourself as you read the paper.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .