6
$\begingroup$

Out of the last 20 questions (most asked more than a day ago), 15 do not have an answer or no upvoted answer (this is the case for the 14 latest). On a slightly longer prespective it also feels like our answer% has dropped a bit (do not have stats though - are there any for moderators on e.g. answer% / month).

Do we have a problem of handling the influx of new questions? I would imagine that it is problematic if people do not get relatively quick answers, especially to basic questions (people asking harder, research-based questions can probably wait a bit longer). If visitors get the impression that questions are not answered, they might get the impression of an inactive community and may not come back.

Maybe this is just a temporary slump, but I was surprised to see this many unanswered questions.

$\endgroup$
5
  • 1
    $\begingroup$ We don't have any really useful statistics for this. And it is not trivial to determine the change of answer rate in general as older answers are more likely to be answered than very new answers, so a simple comparison would be misleading $\endgroup$
    – Mad Scientist Mod
    Commented Nov 7, 2013 at 12:10
  • $\begingroup$ @MadScientist Agree that its not trivial, and I guess this would have to be collected continuously, e.g. looking at answer% of questions that are one week or one month old. $\endgroup$
    – fileunderwater
    Commented Nov 7, 2013 at 12:33
  • 3
    $\begingroup$ Couldn't you take the timestamp of each post and note whether it's answered or not, and make a histogram of the number of unanswered questions as a function of question age (relative to now)? If you make the same type of histogram tracking question age relative to various points in the past, you could watch if there was any drift in the mean, variance, or other moments. I don't know how much metadata we have access to, but that seems like a fairly trivial task for a StackExchange employee with data access privileges at least. $\endgroup$
    – A. Kennard
    Commented Nov 11, 2013 at 0:21
  • $\begingroup$ I guess that's essentially the same suggestion as @fileunderwater. But I don't think it would be that challenging. However, it would completely depend on what data we have access to as concerned site users, or what we are able to ask of SE employees. Anyone know the answers to those two questions? $\endgroup$
    – A. Kennard
    Commented Nov 11, 2013 at 0:44
  • $\begingroup$ @A.Kennard The basic concept is not hard, I played around with it a bit but I didn't get to a fully working self-contained query with my SQL knowledge. If someone would write an appropriate query using the SE Data Explorer I could ask an SE employee to run it for us. Alternatively one could use the Stackexchange API which is also available for beta sites. $\endgroup$
    – Mad Scientist Mod
    Commented Nov 12, 2013 at 6:40

1 Answer 1

Reset to default
3
$\begingroup$

This problem is still relatively actual. It is a two-sided problem: I am not an expert in all fields which are discussed here (and I guess this applies to most people here), so on one day I can answer 5 or 6 questions (happened a few days ago) while on others I can not say anything useful. The other thing is that a lot of useful answers are not voted up, or when they are, they are not accepted. Is there any chance to let moderators do this on old questions with old answers? To accept answers, it would probably be a good idea that the moderators can nominate questions with good answers which haven't been accepted for 3 months (or any other period) for an acception vote. This will only get through, when it is approved by others (like for closing questions). This will give the proper reputation to the people which wrote the answer.

$\endgroup$
2
  • $\begingroup$ I agree that it is unfortunate when answers are not accepted, but I don't see it as a big problem (has mostly to do with new/unexperienced users). Getting votes for good answers is what matters in my mind. However, voting for accept (similar to close votes) after a certain period should be doable, but has to be implemented for all of SE. It would also shift the meaning of "accepted answer" from the subjective perception of the original asker to a community decision. I don't know if this is good or bad. $\endgroup$
    – fileunderwater
    Commented Mar 26, 2014 at 10:44
  • 1
    $\begingroup$ I strongly feel that older unaccepted answers, let's say older than a month , if are good enough to be accepted, should be marked as accepted either by community vote( just like we have for closing questions) or by the moderators. $\endgroup$
    – biogirl
    Commented Mar 26, 2014 at 13:28

You must log in to answer this question.

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