What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 128 Stack Exchange communities.

I see a lot of closed questions that I personally would not have voted to close (assuming I ever manage enough reputation to be able to cast such a vote).

I've mentioned elsewhere that I perceive substantial negativity at SO.

It would be interesting to see statistics about who as individuals and as groups close the most questions and issue the most down votes.

Do such statistics exist?

imo, they would certainly be useful for the purpose of substantiating or disproving my suspicion of excessive negativity here at SO.

It would also be interesting (at least to me) to see SO statistics of most selected answers, up votes versus downvotes, et cetera ~~ i.e., what might be called as positive behaviours weighted against negative behaviours.

Regards,
Gerry (Lowry)

share|improve this question
6  
I just realized that I'd forgotten to go on my end-of-day closing spree today. Thanks for the reminder! –  Aarobot May 6 '10 at 2:37
3  
Is your question really more about why are non-programming questions constantly being closed? –  random May 6 '10 at 2:54
    
If someone figures out how to answer this in detail, can you please do re-opening as well? –  dmckee May 6 '10 at 3:16
    
See also: meta.stackexchange.com/questions/38510/… –  Shog9 May 6 '10 at 3:26
    
@ random: no, I've seen valid programming questions by other members also closed. –  gerryLowry May 6 '10 at 12:13
2  
@gerryLowry: If you would cite some closed questions you think should have stayed open, that would help the discussion. It's hard to discuss a general feeling that fewer questions should be closed. –  David Thornley May 6 '10 at 14:19
add comment

2 Answers 2

After much calculation, I have found a very interesting statistic:

Generally, people with 3,000 or more reputation close more than 90% of all closed questions.

I'm going to search for more info, but that right there blew me away.

share|improve this answer
8  
I'm going to need to check your source on this one. Where did you say you got this info again? –  Aarobot May 6 '10 at 2:38
    
Preposterous. I refuse to believe it. –  Jason Punyon May 6 '10 at 3:16
add comment

This is highly unscientific, but massaging the Google search from S.Mark, a little:

+"username" +"closed as" +site:stackoverflow.com -site:meta.stackoverflow.com

I get:

  • Neil Butterworth: 2160
  • Shog9: 1870
  • George Stocker: 1340
  • Greg Hewgill: 1300
  • gnovice: 1290
  • SilentGhost: 1150
  • dmckee: 1090
  • cletus: 964
  • paxdiablo: 853
  • skaffman: 779
  • S.Lott: 738
  • Mehrdad Afshari: 713
  • Paul Tomblin: 649
  • tvanfosson: 638

For comparison, some other very active or once-active users, all on the first page:

  • Gumbo: 353
  • Andrew Hare: 337
  • JaredPar: 278
  • Konrad Rudolph: 188
  • Darin Dimitrov: 185
  • bobince: 171
  • Jon Skeet: 151
  • VonC: 85

These are relative numbers only, based on fuzzy data, and only include users that I personally recognize (i.e. mainly the ones in my tags). They're also going to be biased toward the most active users and users who have been members the longest.

Nevertheless, the comparison against the "top" users should be a strong indication that activity level and membership duration alone aren't enough to cause that much of a bias... there are definitely some users who close more often than others (and, frankly, I think these users should receive medals of honour).

Just from my own, very anecdotal experience, these are at least credible results. The numbers themselves are nowhere near accurate, but the rankings make sense. I know that Jon Skeet doesn't close many questions.

Also keep in mind that users with the highest close rates are likely to be the users who spend the most time on the site, and therefore contribute the most, so again I must repeat that closing a high number of questions is not a bad thing. These people are committed to the quality of the site, which is virtually the polar opposite of "negativity."


BTW, here's the SQL script that you would be able to use for the Data Dump imported to SQL Server using the SODDI tool, if the data dump actually contained User IDs for close votes (which it doesn't seem to, probably for privacy reasons):

WITH Votes_CTE AS
(
    SELECT
        UserId, COUNT(*) AS VoteCount,
        ROW_NUMBER() OVER (ORDER BY COUNT(*) DESC) AS Seq
    FROM Votes
    WHERE VoteTypeId = 6
    GROUP BY UserId
)
SELECT u.DisplayName, v.VoteCount
FROM Users u
INNER JOIN Votes_CTE v
    ON u.Id = v.UserId
WHERE Seq < 10

Oh well, maybe some day...

share|improve this answer
    
Oh...that Neil. –  Jon Purdy May 6 '10 at 4:06
    
I've closed 371 questions. Wow - I had no idea! –  user27414 May 6 '10 at 5:13
    
-site:meta.stackoverflow.com is really working? I am seeing like this --- Results 1 - 10 of about 2,200 from stackoverflow.com OR meta.stackoverflow.com for "Neil Butterworth" "closed as". (0.17 seconds) , without -site:meta.stackoverflow.com, only 329 results here at the moment for "Neil Butterworth" –  YOU May 6 '10 at 5:31
    
But yes, I realized that my guess is not correct when I see your list. –  YOU May 6 '10 at 5:32
    
@S.Mark: It works, I get 2370 instead of 2160 if I remove the - term. Try it with dmckee, it's easier to notice the difference. –  Aarobot May 6 '10 at 5:49
    
@Aarobot, looks like I dont know how to use google, could be paste your search urls here? let me take a look how those + - operator used –  YOU May 6 '10 at 6:01
4  
@Aarobot Thanks for publishing those stats - I guess it kind of proves a point I've made several times - some high rep users are not really pulling their weight when it comes to maintaining the site. –  nb69307 May 6 '10 at 9:16
    
@S.Mark: It's exactly as shown up top, just replace username with the actual user name. I did notice that sometimes the results can vary - you can usually "fix" them by going to the end of the results, clicking on "show omitted" (if it's there), and then going to the end again. But, nevertheless, as I mentioned before, these aren't intended to be super-accurate, they're just here to give a general sense of the variation in closing habits. –  Aarobot May 6 '10 at 12:07
2  
@Neil: I do think that contributing a lot of great and helpful answers is the most important thing, and users who don't close a ton of questions shouldn't be criticized for it, but closing should still be considered a net positive for the site and encouraged as such. If any action is truly detrimental to Stack Overflow, it's the constant complaining about closed questions and berating of the closers. Not to mention the "this should be CW" comments which imply that CW acts as some sort of immunity... –  Aarobot May 6 '10 at 12:10
    
Hmm, 480 for me. But what surprises me is only 290 for Geoffrey Chetwood. Can that be right? (I suppose since he was most active a year ago, many of the questions he helped close have since been deleted. Still that sounds awfully low.) –  mmyers May 6 '10 at 15:13
    
@mmyers: That makes sense to me, he's been banned for months and months and the Google results are only an heuristic for recent (as in the last few months) closing activity. Besides, IIRC he wasn't known for closing a lot of questions, just trolling... –  Aarobot May 6 '10 at 15:36
    
@mmyers: I don't recall GeoChet being all that much of a closer on SO... Editing and, uh, down-voting were more his thing. –  Shog9 May 7 '10 at 0:51
    
@Aarobot, Thank you. Now working. –  YOU May 7 '10 at 0:59
add comment

You must log in to answer this question.

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