25
$\begingroup$

A noticeable pattern I have seen on Math.SE is that some question tags seem to attract more attention than others. I am curious and hope this question is of interest to the meta community; however I apologize if this is off-topic, and for (technically) asking more than one question.

You can view the % unanswered for a particular tag by clicking on "top users" for the tag, then it shows at the top. I looked at a few of interest to me and saw the numbers I expected. Measure theory and probability had significantly higher percentages than elementary set theory and general topology.

My first question is whether there's a way to sort tags by their acceptance rate? If not (and you have some time to kill, like me apparently) any guesses at the best and worst tags are in this respect?

I am also interested in what possible reasons for this there are? To an extent I would guess that some tags have more potential answerers than others (e.g. "propositional calculus" has a higher answer rate than "logic"), but maybe the Math.SE answerer community is skewed towards some fields more than others? I know from my (limited) experience that if you hang around some tags for a while you see the same names popping up very often; this is the case more with some tags than it is with others, are we perhaps (for example) short on PDE experts relative to propositional calculus experts?

$\endgroup$
2
  • 2
    $\begingroup$ The [logic] tag is broader than [propositional-calculus]. More questions are asked, and more difficult questions are asked. It's not surprising that it has more unanswered questions (both in absolute numbers and in relative share). $\endgroup$
    – Asaf Karagila Mod
    May 3, 2016 at 5:49
  • 8
    $\begingroup$ I think we have enough of people qualified to answer PDE questions here however the typical PDE question is often quite specific (i.e. how to apply this particular method to this particular equation) and therefore not so interesting for the general reader. It often takes more time/text to answer such question. More often than not I see good answers in this category getting only 0 or 1 votes so it looks like kind of a thankless task to answer such question (would be nice to see statistics on this). $\endgroup$
    – Winther
    May 3, 2016 at 20:49

1 Answer 1

Reset to default
22
$\begingroup$

The query Unanswered questions by tag ranks the top 50 most frequent tags by the percentage of unanswered questions (defined by SE as a question without upvoted or accepted answers).

These are all-time statistics, not for a period of time. There is a similar query with a stricter definition of unanswered: "no answer". According to it,

As for reasons: obviously, more elementary questions get answered more often. Apart from that, there are fluctuations based on having several very active answerers in some tags, and few or none in others. People were complaining about PDE/analysis situation for years:

$\endgroup$
5
  • 2
    $\begingroup$ Exactly what I was thinking of, thanks. I wonder why there is a lack of analysts. Perhaps there's just a larger volume of questions from undergraduate students in these applied/statistical fields? Though, what was interesting is that the top-5 users in (pde) were all sub-50k reputation, whereas the top-5 users in (algebra-precalculus), (elementary-set-theory), (limits), and (elementary-number-theory) all have 147k+ reputation. Perhaps this is related. $\endgroup$
    – Szmagpie
    May 3, 2016 at 15:46
  • 4
    $\begingroup$ PDEs are hard, y'know. :) $\endgroup$
    – J. M. ain't a mathematician
    May 7, 2016 at 2:14
  • $\begingroup$ Just for reference, the same query on MO: data.stackexchange.com/mathoverflow/query/edit/484734 $\endgroup$
    – roman
    May 11, 2016 at 11:31
  • $\begingroup$ What about Stochastic Analysis, Lévy processes and related fields? The situation is even worse :/ $\endgroup$
    – Pasriv
    May 16, 2016 at 1:07
  • $\begingroup$ Algebraic-geometry occurs twice in the same list, I think it should be differential-geometry $\endgroup$
    – Riemann
    Aug 5 at 9:21

You must log in to answer this question.

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