I'm dealing with a question on Stack Overflow that's really a "stats review" -- not technically a code review ("is this code OK?"), but "is this statistical approach OK?"

This strikes me as inappropriate for CV as well since it doesn't ask a particular stats question, just "is this OK?", but maybe those rules about specificity are more applicable to SO than CV; I don't see anything about specificity in the on-topic or off-topic lists, although it might fall under "there is no actual problem to be solved". (It's not a stellar question, but it might be improved with editing.)

So my questions are:

  • is a "please review my statistical approach" question on-topic for CV?
  • if not, can anyone recommend a more appropriate venue?
  • 10
    $\begingroup$ +1 This is worth a discussion. As far as the linked SO post goes, though: if it appeared here on CV I would close it with a comment "what is your question?" $\endgroup$
    – whuber Mod
    Apr 9, 2019 at 13:12
  • 4
    $\begingroup$ Another problem I see is that in such cases they often ask the yes / no kind of questions. If questions like this present some solution and ask specific questions about it, they rather receive their answers. Btw, it would be great if someone could post links to examples of such questions so we're not discussing abstract problem. $\endgroup$
    – Tim Mod
    Apr 9, 2019 at 13:17
  • 2
    $\begingroup$ I encounter such questions regularly in the 'first post' queue, I'll try to collect some and link them here. On topic: as Tim stated already, many such questions are posed as yes/no questions, which I actually thought are to be avoided, but couldn't find any reference to it in the help center. I'm quite on the fence about these questions, but I think if properly describing the problem at hand and the approach in question, these questions could provide value to other users even if a short answer suffices. On the other hand, the problems described in such questions tend to be maybe too specific $\endgroup$
    – deemel
    Apr 9, 2019 at 13:47
  • 1
    $\begingroup$ @Tim This recent question seems to fit the bill: stats.stackexchange.com/questions/401122/… $\endgroup$
    – Sycorax Mod
    Apr 9, 2019 at 17:29
  • $\begingroup$ This one asked for a review Can someone help me out and check the results?. It fell more specifically in the category self-study. I believe it is fine although it is not much great value for the site because these reviews and debugging of methods only occur on specific cases and may not be very general to a large public. I answered it because of the effort from the OP. $\endgroup$ Apr 12, 2019 at 12:46
  • $\begingroup$ I tend to downweight the "Yes/No" aspect of many, but not all, of these questions in favor of a more expansive interpretation as "I have this problem, can you help me? BTW, is this idea any good?" that make reference to what's suggested in the OP. It's human nature to try to solve problems, and human nature to wonder if one's proposed solutions are any good - and sometimes they are. Better they ask than that they don't! $\endgroup$
    – jbowman
    Apr 18, 2019 at 19:25


You must log in to answer this question.

Browse other questions tagged .