Roughly, there are many ways to "answer" a statistical question: with actual mathematics, with "proofs by pictures" or "proofs by simulation", and with casual explanation.

Is it ever the case that none of those approaches can answer the OP's question satisfactorily? If so, how do you remedy the problem?

Should a third party encourage the asker to select a particular answer as correct if it is a thorough and precise answer to the original question, should the asker be encouraged to retract their question, should the question be closed as "too broad" since the asker has too many missing details? Do you point the asker to the right background material and hope they select your answer (or any answer) two or three years into the future?

  • 5
    $\begingroup$ The key is that OPs decide whether an answer deserves formal acceptance. That's crucial only for who does/doesn't get +15 on an answer. Whether OPs understand the answer is a different dimension. Many may recall what they thought were perverse acceptances, when weaker or even very bad answers were accepted (on Meta too). I'd respect both (a) OPs being reluctant to accept any answer and (b) discreet suggestions that an answer deserves acceptance (preferably from third parties if there are several answers) . $\endgroup$
    – Nick Cox
    Commented Dec 11, 2017 at 16:46

2 Answers 2


Even though I am not an user with skills to answer beyond the asker's experience, I always try to provide the best answer I can.

I try to directly answer the question while being the most didactic as possible, but not only limiting the answer to OP's experience if I feel more is needed I am willing to complement it.

In other words, I believe contributing to the repository of statistical content comes first, helping OP second (like 'votes' being more important than an 'accept').


I believe a well formulated question that falls within the guidelines will be able to be answered in a way that is tailored to the asker's experience.

If the asker is able to genuinely understand and formulate their own question, they will be able to understand a hypothetically well thought-out answer. If the answer is overly simplified, or otherwise not rigorous, that is OK as long as it helps the OP better understand the problem at hand.

In other words, theoretically an answer could be provided that isn't beyond the asker's ability to understand. In practice, that answer may never be provided by the community, but the onus is on the OP to decide.


You must log in to answer this question.

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