4
$\begingroup$

I'm wondering if there's a way to see site health diagnostics, like:

  • the number of active users currently on the site (cf. Reddit has this feature), perhaps with further data on the tags to which they contribute most often.
  • the average number of views per question (perhaps with further data of page-views by tag, or by hour/day, even by user perhaps, etc.)
  • the percentage of questions that get answered (per day, per week, per month?)
  • the average number of upvotes for questions and answers
  • the distribution of votes on a post across time (e.g. when were most of the votes cast for those posts with huge numbers of votes)

I want to see if site activity has gone down, or gone up over time, and whether or not the activity of "high quality users" (users who answer many questions with "more advanced tags") have gone up or down.

The reason I ask is because I’m afraid the site is “dying”, and I want to have some numbers to confirm or deny my suspicions.

$\endgroup$
21
  • 7
    $\begingroup$ There is the SEDE data explorer.... (yes, I know that "SEDE data explorer" is redundant; just be thankful that I didn't say "StackExchange SEDE data explorer"). $\endgroup$
    – Xander Henderson Mod
    Feb 19 at 1:13
  • 7
    $\begingroup$ mods and users over 25K rep have acces to Site Analytics. Here's the default view I get today. If you want one or two queries I could provide a screenshot, but I have never gotten any sort of understanding from it! $\endgroup$ Feb 19 at 3:27
  • 2
    $\begingroup$ Why are you afraid that the site is "dying"? It seems very active to me, certainly in comparison with many of the other SE sites $\endgroup$
    – user829347
    Feb 19 at 21:09
  • 1
    $\begingroup$ @OliverHouse there are some high-profile users from whom I remember seeing many posts in the past, but not anymore. I am wondering if this is a defect of my own perception, or is it actually the case that some "heavy-hitters" have "retired". $\endgroup$
    – D.R.
    Feb 19 at 21:22
  • 6
    $\begingroup$ A few of those "heavy hitters" deleted their accounts after encountering consequences for ignoring site policy. That would make the numbers of hi-rep user go down, as well as the upvotes and answers, and "completed" questions. Is that a bad thing though? We have more than sufficient quantity, and it's always been quality that we should seek on SE anyway. $\endgroup$
    – Nij
    Feb 19 at 22:51
  • 9
    $\begingroup$ I think @Nij is talking about the same people as you are, D.R. Only I'd say that, rather than ignoring site policy, they may have become disgusted with site policy, when that policy leads to deletion of serious mathematical content from the site. And, yes, I'd say that's a bad thing. $\endgroup$ Feb 19 at 23:30
  • 3
    $\begingroup$ I would like to widen the scope of Gerry's comment a little bit. We have always had capable users leaving the site for various disagreements about site policies. Early on those leaving the site were disgusted by our laissez-faire attitude to homework questions. True, some other users were against placing any restrictions on the homework questiions at all, feeling that their own answers justified the existence of such questions. The problem has always been A) a search for a compromise, B) making everybody comply with the compromise. $\endgroup$ Feb 20 at 17:26
  • 4
    $\begingroup$ (cont'd) The sad truth has always been that we have too many users unwilling to play along. I should not try and simplify the matter unduly for there is no cover-it-all explanation. Let's just say that mathematicians being idealists of the purest kind, it is not so surprising that some are unable to compromise at all. Stereotypically an aged professor who has gotten used to thinking that whatever they say (in a class) goes, is surprisingly inflexible in their ways. Either disgusted by a homework mill or by not being allowed to answer every question they want. $\endgroup$ Feb 20 at 17:31
  • 4
    $\begingroup$ Those totally disgusted by homework questions getting answered left early. Those wanting to answer everything they can mostly stayed (notable exceptions exist) because the measures against them lacked teeth. With the advent of EoQS that finally changed, but old habits take a while to unlearn, and some refuse. We have not reached a new equilibrium yet. I am not worried about the site losing expertise. The specialists are mostly unaffected because (statistically speaking) the low quality questions often (but not exclusively) deal with lower level math not requiring much expertise. $\endgroup$ Feb 20 at 17:38
  • 3
    $\begingroup$ @OliverHouse EoQS. A rule designed to save the integrity of the site. $\endgroup$ Feb 21 at 10:19
  • 2
    $\begingroup$ This script from SEDE mentions the "status" of the last so-many questions : vote count, closed or open, how many answers etc. One can try 500 for an illustration. The database only updates every Sunday, nevertheless the script still carries some information. It can easily be modified to include details on the last few posts in a specific tag, and (while I haven't done this) I believe one can adjust the script to compare how questions are received now compared to, say a year back. $\endgroup$ Feb 21 at 10:53
  • 3
    $\begingroup$ To add one more thing, it is sad that those user left back then and it is sad that some plan to leave now. This is what you get when diametrically opposite views are held strongly, and people are unwilling to meet the others half-way. $\endgroup$ Feb 21 at 14:06
  • 2
    $\begingroup$ "high quality users" and it's "definition" ... $\endgroup$ Feb 21 at 23:16
  • 1
    $\begingroup$ @JyrkiLahtonen I don't recall any user stating that they left the site due to being "disgusted by our laissez-faire attitude to homework questions". How about some data to back up that dubious claim. Is this yet another example of you making wild guesses about the motivations of other users? $\endgroup$ Mar 3 at 10:31
  • 1
    $\begingroup$ @BillDubuque Just following the precedent you set back in the day. At least I made my comments very impartial, representing both sides. Much more so than yours were back in the day, when you bluntly (without evidence) claimed that every prominent user who left, did so because "there freedom to teach was compromised". $\endgroup$ Mar 3 at 11:22

1 Answer 1

11
$\begingroup$

I feel like I should post an answer, despite not giving any data you specifically asked for, because the Site Analytics I mentioned in comments does in fact show a recent dip in posts (Qs and As) and votes, but not traffic, when you look at the timescale of years. (There is also the minor lie I told that that was the view presented to me, when I have already clicked on a button reducing the data to a weekly sample before taking the screenshot.)

I have overlayed guidelines at roughly each Christmas, and also the date of the EoQS post.

enter image description here

I'm not personally sure that anything actionable can be taken from this (in particular I don't at the moment dare say the dip is because of the EoQS, if that is even preferrable; nor do I know if a few high-profile users would affect this graph), yet I could not leave my comment alone which seems to suggest that there is nothing to see on that page.

We are told not to share the raw data, but I think it should be OK for me to compute some simple statistics if anyone under 25k rep wants (which I will get to in my free time).

$\endgroup$
0

You must log in to answer this question.

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