There's been some friction over our research guidelines and their lack of (simple) definition. See New users and down-voting for lack of initial research There's a lengthy and unwieldy discussion on the definition here: What level of initial research is expected on questions?

We need a quick bit of copy describing our policy on the issue for the FAQ and comments on questions like this.

Ideally what I want:

  • Describes the research requirement clearly
  • Indicates what I should do before I ask a "no research" question
  • Is one or two sentences long, max two wrapped lines or so in length
  • Is simple enough there's no talking around it


Now let's not fret over should be required or not and focus on what's a useful guideline that new users really should know when posting. Required or no this is a very important aspect of a good question. The problem with complex and subjective rules is it's so easy to argue around them. You can always argue around "this IS answerable in a scientific way!" You can't argue around "Yeah I don't really even know what psychology is". Simple rules always win.

3 Answers 3


A good question should contain within itself a clear reason for its being asked. That is what I call the "minimal effort" standard.

Including research (even if there is lack of success with it) provides such a clear reason, and meets the "minimal effort" standard BY DEFINITION.

Research is not ALWAYS necessary, but the "clear reason" (and minimal effort) are.


Questions should be amenable to empirical inquiry and terminologically consistent with prior literature.

  • 1
    not so sure about terminologically consistent; what does that get us? A layman asking is less likely to know the proper current terminology but that doesn't entirely exclude it from being a decent qujestion.
    – Zelda
    Commented Mar 30, 2013 at 16:31
  • Terminological consistency could be as simple as following Wikipedia terminology, or some other clear connection to the literature on the topic. I meant it to discourage users giving imprecise definitions of their questions. Commented Mar 30, 2013 at 16:41

Questions should be answerable and useful. I define useful as making the Internet a better place.

Showing initial research does not need to be a requirement. However, prior research both increases the likelihood that the question is answerable and useful. It also makes an argument for the question being answerable and useful.

Initial research shows that the answer is not already readily available with a quick Google search. Also, initial research can assist in making the question interesting, by defining the scope to what is not already known and in framing the question in terms of the existing scientific literature.

Thus, I think if a question is both answerable and useful, it should not be closed for lack of initial research.

I would like people asking questions to show greater initial research. However, our current rate of questions is relatively low, I'm not sure whether we want to put any impediments in the way of receiving questions in the FAQ. That said, there might be a wording that essentially says that showing initial research is encouraged.

  • 2
    I personally agree with this. I have been closing questions which the community has decided don't meet the "initial research" standards, but I would like to have more questions as long as they are answerable. We're scaring users off now. And I know, others will argue they're not the kind of user we want, but without a community how will we get more users at all? (I will write a longer answer later...)
    – Josh
    Commented Sep 18, 2012 at 2:28
  • 2
    What I see very commonly in these questions is they're laughably broad and could be answered by entire text books. This seems to happen because they haven't even looked slightly into the topic and don't understand enough to ask a remotely answerable question. Hitting them with a bat until they're answerable really doesn't make it much more appropriate IMO, many questions here have been edited into something vastly different than what was asked. While they're usually better, that just doesn't seem right at all. We're not a question-writing service
    – Zelda
    Commented Sep 18, 2012 at 13:20
  • Also, sure, don't make it a requirement, but we really need a guideline then. There needs to be guidance on this issue. These questions are going to continue to be closed whether something's put in the FAQ or not; we may as well help out askers (and not by post-mortem altering their questions entirely)
    – Zelda
    Commented Sep 18, 2012 at 13:22
  • When there's broad questions, why don't we just answer about the current state of research and suggest the user ask a more specific question next time (if we like). We can keep the guideline up, of course, but few people read guidelines. They learn the culture by participating in it. The guidelines are there to help the user get closer to their goal, but it's unrealistic to expect them to to read it before posting. Commented Sep 18, 2012 at 17:44
  • @Xurtio it's something concrete to point them to, and something to guide that culture they're participating; thus we enforce the guideline by craft better questions and implicitly enforcing it before it's even read.
    – Zelda
    Commented Sep 18, 2012 at 18:06
  • I've used the phrase "crap question daycare" to Jeromy in this regard before. I don't think we should feel obligated to nurture questions from users who are not willing to meet us at least halfway. I'm all for closing such questions with a somewhat optimistic moderator comment (or high-rep user comment) to the tune of what Josh already does, explaining that it's not the end of the line. Commented Sep 18, 2012 at 18:17
  • @ChuckSherrington Note that I do this because you, Ben and Artem desperately want me to close them :-) Most of the time I close them I think Boy, I wish we could somehow salvage this...
    – Josh
    Commented Sep 18, 2012 at 21:19
  • @JoshGitlin But both you and I know closing is not the end of the line. Perhaps we need to emphasize that point even more to users. Commented Sep 18, 2012 at 21:22
  • I should ask a new meta question, but today is an extremely busy work day. In the majority of cases, closing is the end of the line for that question. Details to follow when I am not working.
    – Josh
    Commented Sep 18, 2012 at 21:45

You must log in to answer this question.

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