Should we close questions that do not provide evidence of prior research?

For example, Do stronger muscles lead to a decreased tendon load, and if so, why? seem to have been placed on hold for that reason, while I see many (most) questions being asked without providing any evidence of prior research (and from what I can see they typically do not get closed on the grounds that I prior research). As a result, I'm confused regarding whether one should close questions that do not provide evidence of prior research.

How much should we enforce background research? most upvoted answer seems to indicate that not showing prior research isn't ground for closing a question.

The top 5 most upvoted questions do not show any prior research:

  • 1
    $\begingroup$ In my experience, closing questions for lack of research effort has been rather arbitrary. $\endgroup$
    – canadianer
    Commented May 31, 2019 at 20:54

1 Answer 1


(I intend this answer to be my view as a community member and not as a moderator; I don't speak here for the other moderators nor the community as a whole)

Homework questions

We have an explicit policy regarding homework. This policy is mostly driven by a valid aversion towards "gifting" answers which the community has agreed does a disservice to learners. We extend this policy towards under researched questions more generally, those that ask for a tutorial/explanation of a topic, and those that are very simply answered by other resources. In that way, we may be different than other Stacks that allow for more basic questions as long as they are sufficiently narrowly scoped.

Other questions that need prior research/citation

Another type of question that we perhaps have not fully addressed in Meta is a question like yours, which is based on some sort of claim. Often these questions are in the form of "my textbook says X, but I learned somewhere else Y, who is right?" Other times they just have the "X" part.

In my experience, we always ask for clarification and citation of this sort of question. Questions where this is not provided are closed as unclear. The reason that this sort of question is unclear is that very often the source of the question is rooted in a misunderstanding or incomplete consideration of the source. Often the statement in question is missing an important caveat (i.e., a sentence seems like it is a general statement, but the preceding paragraph makes clear that the author is only talking about mammals).

I feel like there are two reasons these questions need to cite the claim. First, they invite a sort of whack-a-mole where we end up with a bunch of silly questions that cannot be answered by the standards of this Stack because they don't have a basis in scientific interests. "Does toothpaste cure plantar fasciitis?" is never going to get a good answer, because the odds that anyone has actually studied toothpaste in that context are slim.

Second, they invite misleading answers. For example, your question "Do stronger muscles lead to a decreased tendon load, and if so, why?" doesn't really make any sense by itself. The obvious answer is "no" but there may be a more interesting "yes" answer if proper caveats are included. For example, it could be referring to participation of other muscles in a given movement which share the load, it could be referring to approaches to lifting objects like using more of a jerking motion that tend to occur when people are weaker. Without some understanding of the actual claim, it would be wrong and misleading to simply say "no" - and also nearly impossible to back up.

Other communities that get these questions as a larger fraction of their total question load, like MedicalSciences.SE and Psychology and Neuroscience, have more explicit policies that these questions must be referenced with an associated custom close reason. Skeptics.SE, which deals entirely with this sort of question, has even stricter policies on what exactly a suitable "claim" is.

Most upvoted questions

Questions that are "most upvoted" are almost always there because they are on the Hot Network Questions - therefore, the vast majority of those votes come from people who are not active members of this community and instead spark the curiosity of a broader audience. They are not necessarily exemplar questions.

The IF/WHILE/GOTO question is a great example of this: something that is entertaining to the incredible selection bias of StackExchange users towards programmers of some sort.

One way that a question can become a Hot Network Question is simply by attracting multiple answers in a short time. Sometimes this means a question is really good/interesting, but a lot of the time it's pretty random. This is a constant topic of discussion in other Metas across all of StackExchange.

Your most upvoted question as of right now, Where can I find the list of all tendons in the human body? is another good example: you've asked some much better questions, but this one lived on the Hot Network Questions. My most upvoted answer as of right now, Examples of plant families that contain species that are safe for human consumption and species that are poisonous to humans? is a not-so-great answer to a not-so-great and under-researched question, but it lived on the Hot Network Questions.

That said, none of these examples fall into the questions needing prior research categories that I discussed above.

Close is not permanent

Closed questions can be reopened if the reason for closing the question is resolved. In the meantime, closing a question prevents it from attracting poor answers.


Most of your questions on this Stack and others I see you on are well-supported and referenced, and in other cases you typically are responsive to requests for clarification and narrowing of scope, so I thank you for the effort you put in to asking your questions.

  • 1
    $\begingroup$ As an addendum to the most upvoted Q&A discussion, two of our moderators who both write many excellent answers have their most upvoted answer to the same biology.stackexchange.com/questions/56476/… question $\endgroup$
    – Bryan Krause Mod
    Commented May 29, 2019 at 16:13

You must log in to answer this question.

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