What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 122 Stack Exchange communities.

Given the number of questions about the suggested edit system, I think we need some numbers…

On Stack overflow:

  • What is the mean time it takes for a suggested edit to be acted on?
  • What is the length of time before 95% of pending suggested gets acted on?
  • How does the above change according the time of day and day of the week?
  • How many suggested edits get acted on by normal users?
  • How many suggested edits get acted on by 10K users (var the 10K tools if it is tracked)?

Likewise for one of the newer sites…

share|improve this question
1  
Heey, you're not alone in this network! We like our stats too you know! Perhaps it would be better to request it as part of the API, so anyone can figure out whatever they want –  Ivo Flipse Feb 22 '11 at 10:28
    
This will probably find its way into the data dump and the API, but I think it's a bit premature at this point... the team is still hammering out the details. –  Jon Seigel Feb 22 '11 at 18:12
    
@Jon, that way I gust asked for a "add-hock" set of stats, there are two many quesions on meta assuming that we need more people to be able to find edit to aprove. –  Ian Ringrose Feb 23 '11 at 9:13
3  
Maybe you want to propose this here? –  Ben Pilbrow May 10 '11 at 15:46
    
There's a related question here –  M. Tibbits Aug 4 '11 at 11:36
add comment

1 Answer

Data was dumped on ~June 6th, 2011.

I don't have information regarding when or how specific users acted on suggested edits, so I will give some stats on when and how quickly suggested edits were processed (i.e., either accepted or rejected).

None of this data is available on SEDE yet, though it should be at some point.


Stack Overflow

Average time for an edit to be processed: 16.6 minutes
Average of top 95% fastest processed times: 11.9 minutes
Average of bottom 5% slowest processed times: 106.3 minutes

DayOfWeek   EditsProcessed  AvgProcessedTimeInMins
---------   --------------  ----------------------
Sun         4616            23.5
Mon         8634            15.9
Tues        9415            15.5
Wed         9658            14.8
Thurs       9994            14.9
Fri         8924            14.8
Sat         4804            24.5

You didn't ask for this, but it was easy to calculate:

Post Type       Count
----------      -----
Question        38634
Answer          11315
Wiki               96
TagWiki          3393
TagWikiExcerpt   2607

While gathering this data, I found that the processing time varies considerably across sites, so instead of just putting up one random other site, here are the first three stats for all the data sets I have access to. The times are in minutes.

Site           Count  Average  Avg95Pct  Avg5Pct
----           -----  -------  --------  -------
Android          323    209.3     150.5   1314.5
Apple            490    154.7     104.0   1119.0
Cooking          197    187.5     114.3   1608.1
CsTheory         111     71.9      53.0    439.6
DIY               47   1415.9    1084.3   8093.8
Electronics      282    168.6     102.6   1375.6
English          658     40.0      28.7    257.4
GameDev          474     72.4      52.8    448.7
Gaming           741     29.0      20.8    183.8
GIS              210    197.8     135.7   1362.8
Math            1381     14.6       8.3    131.8
Photography      225     75.5      56.5    429.6
Programmers     1180     26.8      17.9    195.7
ServerFault     1807     61.0      43.7    389.0
StackOverflow  56045     16.6      11.9    106.3
Stats            173     63.9      48.8    365.9
SuperUser       2606     17.5      11.5    131.6
TeX              779     16.5       6.9    200.0
Ubuntu          1610     51.8      33.0    411.0
Unix             428     74.2      51.2    506.8
WebApps          362     75.5      52.5    502.6
Webmasters       373    270.4     190.7   1803.1
WordPress        282     75.7      56.8    431.6

Given that the bottom 5% of processed times are massively higher than the averages, I suspect these are the edits that no one really knows what to do with, so they leave them for the next person to make a decision.

The stats for Home Improvement (DIY) are a bit concerning.

share|improve this answer
    
diy's count of edits is so low I wouldn't be that concerned -- not enough data, really –  Jeff Atwood Jul 10 '11 at 1:44
1  
@Jeff: Agreed, but that bottom 5% at 5.6 days means that I think we should be watching that pretty carefully. –  Jon Seigel Jul 10 '11 at 2:45
1  
How many of the "aproves/rejects" come from none 10K users? –  Ian Ringrose Jul 15 '11 at 14:32
    
@Ian: As I mentioned, unfortunately, I was not given data regarding who voted on (and/or approved/rejected) suggested edits. –  Jon Seigel Jul 15 '11 at 22:02
add comment

You must log in to answer this question.

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