10

2018 stats on moderation of IPS were given in this post and depict the site's activity smoothed on the whole year - thank you Shog9 for providing us those statistics!

After October 16th 2018, the day when IPS was removed from HNQ, the site's stats significantly evolved, as shown on JAD's meta post about the matter. This post and peufeu's recent one on the same topic (but presenting those stats observed on a bigger period of time) mostly focus on the site traffic, whereas I'm curious to see whether this event also had an impact on IPS' moderation. I imagine those stats could be presented in the following way:

                                         |  Jan 1st - Oct 16th    |   Oct 16th - Dec 31st
---------------------------------------- | ---------- ----------  |  ---------- ----------
                 Action                  | Moderators Community┬╣  |  Moderators Community
---------------------------------------- | ---------- ----------  |  ---------- ----------
Users suspended┬▓                         |          X          Y  |          Z          T    

Would it be possible for the Stack Exchange crew to provide those numbers given in "2018: a year in moderation" splitted before and after October 16th?

| |
10

Here you go:

                                            Jan 1 - Oct 16      Oct 16 - Dec 31
--------------------------------------- -------------------- -------------------- 
Action                                  Moderators Community Moderators Community 
--------------------------------------- ---------- --------- ---------- --------- 
Users suspended                               32         12         7          9  
Users destroyed                              135          0        15          0  
Users deleted                                 10          0         3          0  
Users contacted                               64          0         8          0  
Tasks reviewed: Suggested Edit queue         163       1982        23        114  
Tasks reviewed: Reopen Vote queue             59       1852        15        261  
Tasks reviewed: Low Quality Posts queue       66       6943        17       1433  
Tasks reviewed: Late Answer queue              7        158         1         20  
Tasks reviewed: First Post queue              36       3056        13        246  
Tasks reviewed: Close Votes queue             23       3355        11        650  
Tags merged                                    1          0         2          0  
Tag synonyms proposed                          1          2         1          2  
Tag synonyms created                           1          0         3          1  
Questions unprotected                          7         13         0          0
Questions reopened                            16        165        12         14  
Questions protected                           31        173         0         11  
Questions migrated                             7          2         4          0  
Questions flagged                             33       1173         3        105  
Questions closed                             263        898        88        165  
Question flags handled                       713        493        73         35  
Posts unlocked                                 6         10         3          0  
Posts undeleted                               51         96        14         11  
Posts locked                                  19        253         0         24  
Posts deleted                                652       2218       113        347  
Escalations to the CM team                    11          0         3          0  
Comments undeleted                           213          0        26          0  
Comments flagged                             146      12452        12        978  
Comments deleted                           19751       4009      2134        315  
Comment flags handled                      10569       2029       883        107  
Answers flagged                              384       4848        42        549  
Answer flags handled                        1795       3437       188        403  
All comments on a post moved to chat         104          0         2          0   

Ok, that's boring. Peufeu's post has all those fancy graphs; this big pile o' numbers looks super dull by comparison. So, here's a fancy graph:

That's the entire moderation history of the site, with your October 16th date called out. Enjoy!

| |
  • Thank you so much for those, Shog. – avazula Jan 5 '19 at 9:05
4

I like the premise of the answer by @Noon to look at how moderation stats changed relative to when IPS was removed from HNQ, but given that there are 9.5 months of data before IPS was removed and only 2.5 months of data after, I wanted to take a look at the average moderation workload per month in those time frames.

I took the data that Shog9 provided and calculated the average number of times an action was performed per month both before and after the de-HNQ-ization of IPS. Here are the results:

                                            Jan 1 - Oct 16      Oct 16 - Dec 31
--------------------------------------- -------------------- -------------------- ---------------- ----------------
Action                                  Moderators Community Moderators Community Moderator Change Community Change
--------------------------------------- ---------- --------- ---------- --------- ---------------- ----------------
Users suspended                              3.37      1.26      2.80      3.60          -16.88 %          185.00 %
Users destroyed                             14.21      0.00      6.00      0.00          -57.78 %             NaN %
Users deleted                                1.05      0.00      1.20      0.00           14.00 %             NaN %
Users contacted                              6.74      0.00      3.20      0.00          -52.50 %             NaN %
Tasks reviewed: Suggested Edit queue        17.16    208.63      9.20     45.60          -46.38 %          -78.14 %
Tasks reviewed: Reopen Vote queue            6.21    194.95      6.00    104.40           -3.39 %          -46.45 %
Tasks reviewed: Low Quality Posts queue      6.95    730.84      6.80    573.20           -2.12 %          -21.57 %
Tasks reviewed: Late Answer queue            0.74     16.63      0.40      8.00          -45.71 %          -51.90 %
Tasks reviewed: First Post queue             3.79    321.68      5.20     98.40           37.22 %          -69.41 %
Tasks reviewed: Close Votes queue            2.42    353.16      4.40    260.00           81.74 %          -26.38 %
Tags merged                                  0.11      0.00      0.80      0.00          660.00 %             NaN %
Tag synonyms proposed                        0.11      0.21      0.40      0.80          280.00 %          280.00 %
Tag synonyms created                         0.11      0.00      1.20      0.40         1040.00 %        Infinity %
Questions unprotected                        0.74      1.37      0.00      0.00         -100.00 %         -100.00 %
Questions reopened                           1.68     17.37      4.80      5.60          185.00 %          -67.76 %
Questions protected                          3.26     18.21      0.00      4.40         -100.00 %          -75.84 %
Questions migrated                           0.74      0.21      1.60      0.00          117.14 %         -100.00 %
Questions flagged                            3.47    123.47      1.20     42.00          -65.45 %          -65.98 %
Questions closed                            27.68     94.53     35.20     66.00           27.15 %          -30.18 %
Question flags handled                      75.05     51.89     29.20     14.00          -61.09 %          -73.02 %
Posts unlocked                               0.63      1.05      1.20      0.00           90.00 %         -100.00 %
Posts undeleted                              5.37     10.11      5.60      4.40            4.31 %          -56.46 %
Posts locked                                 2.00     26.63      0.00      9.60         -100.00 %          -63.95 %
Posts deleted                               68.63    233.47     45.20    138.80          -34.14 %          -40.55 %
Escalations to the CM team                   1.16      0.00      1.20      0.00            3.64 %             NaN %
Comments undeleted                          22.42      0.00     10.40      0.00          -53.62 %             NaN %
Comments flagged                            15.37   1310.74      4.80    391.20          -68.77 %          -70.15 %
Comments deleted                          2079.05    422.00    853.60    126.00          -58.94 %          -70.14 %
Comment flags handled                     1112.53    213.58    353.20     42.80          -68.25 %          -79.96 %
Answers flagged                             40.42    510.32     16.80    219.60          -58.44 %          -56.97 %
Answer flags handled                       188.95    361.79     75.20    161.20          -60.20 %          -55.44 %
All comments on a post moved to chat        10.95      0.00      0.80      0.00          -92.69 %             NaN %

By looking at the Moderator Change and Community Change columns, we can see how moderation activities changed around the removal of IPS from HNQ. A few stats that I find interesting:

  • IPS has experienced a 16.88% decline in users suspended by mods, but a 185% increase in users suspended by the community
  • IPS has experienced ~58% decline in answers flagged and ~70% decline in comments flagged
  • IPS has experienced a 92.69% decline in moving comment chains to chat
| |
  • 3
    To be clear, "Community" suspensions are when users either re-create an account deleted while suspended or for spam/abuse or are suspended network-wide. 3 were spammers, 6 were network-wide suspensions... So I wouldn't look too closely at those numbers. :) – Catija Jan 8 '19 at 5:54
  • I wonder what happens if you compute the stats about posts (flagged, closed, deleted, reopened, etc) per post created in that timeframe. That is, are we flagging a lot fewer posts because there are a lot fewer posts or because their quality is higher? Similarly, there's been a 92% decline in moving comment threads to chat; how does that compare to the number of long threads and/or the number of comments overall? – Monica Cellio Jan 11 '19 at 4:46
3

If you are curious about the ratio, I though Shog9 numbers and made this:

                                           Jan 1 - Oct 16      Oct 16 - Dec 31
--------------------------------------- -------------------- -------------------- --------- 
Action                                  Moderators Community Moderators Community Evolution
--------------------------------------- ---------- --------- ---------- --------- ---------
Users suspended                            72,73 %  27,27 %    43,75 %   56,25 %   -28,98 %
Users destroyed                           100,00 %   0,00 %   100,00 %    0,00 %     0,00 %
Users deleted                             100,00 %   0,00 %   100,00 %    0,00 %     0,00 %
Users contacted                           100,00 %   0,00 %   100,00 %    0,00 %     0,00 %
Tasks reviewed: Suggested Edit queue        7,60 %  92,40 %    16,79 %   83,21 %     9,19 %
Tasks reviewed: Reopen Vote queue           3,09 %  96,91 %     5,43 %   94,57 %     2,35 %
Tasks reviewed: Low Quality Posts queue     0,94 %  99,06 %     1,17 %   98,83 %     0,23 %
Tasks reviewed: Late Answer queue           4,24 %  95,76 %     4,76 %   95,24 %     0,52 %
Tasks reviewed: First Post queue            1,16 %  98,84 %     5,02 %   94,98 %     3,86 %
Tasks reviewed: Close Votes queue           0,68 %  99,32 %     1,66 %   98,34 %     0,98 %
Tags merged                               100,00 %   0,00 %   100,00 %    0,00 %     0,00 %
Tag synonyms proposed                      33,33 %  66,67 %    33,33 %   66,67 %     0,00 %
Tag synonyms created                      100,00 %   0,00 %    75,00 %   25,00 %   -25,00 %
Questions unprotected                      35,00 %  65,00 %      NAA      NAA        NAA
Questions reopened                          8,84 %  91,16 %    46,15 %   53,85 %    37,31 %
Questions protected                        15,20 %  84,80 %     0,00 %  100,00 %   -15,20 %
Questions migrated                         77,78 %  22,22 %   100,00 %    0,00 %    22,22 %
Questions flagged                           2,74 %  97,26 %     2,78 %   97,22 %     0,04 %
Questions closed                           22,65 %  77,35 %    34,78 %   65,22 %    12,13 %
Question flags handled                     59,12 %  40,88 %    67,59 %   32,41 %     8,47 %
Posts unlocked                             37,50 %  62,50 %   100,00 %    0,00 %    62,50 %
Posts undeleted                            34,69 %  65,31 %    56,00 %   44,00 %    21,31 %
Posts locked                                6,99 %  93,01 %     0,00 %  100,00 %    -6,99 %
Posts deleted                              22,72 %  77,28 %    24,57 %   75,43 %     1,85 %
Escalations to the CM team                100,00 %   0,00 %   100,00 %    0,00 %     0,00 %
Comments undeleted                        100,00 %   0,00 %   100,00 %    0,00 %     0,00 %
Comments flagged                            1,16 %  98,84 %     1,21 %   98,79 %     0,05 %
Comments deleted                           83,13 %  16,87 %    87,14 %   12,86 %     4,01 %
Comment flags handled                      83,89 %  16,11 %    89,19 %   10,81 %     5,30 %
Answers flagged                             7,34 %  92,66 %     7,11 %   92,89 %    -0,23 %
Answer flags handled                       34,31 %  65,69 %    31,81 %   68,19 %    -2,50 %
All comments on a post moved to chat      100,00 %   0,00 %   100,00 %    0,00 %     0,00 % 

The "Evolution" represent the job done by moderators after HNQ minus the job there were doing before. So if the number is negative, it means mods are, comparatively, working less than before.

| |
  • I like the comparison of moderator work before vs after. I would wonder if there is a better way to classify it than simply subtracting, since there were 9.5 months of data before and only 2.5 after. A better measure might be looking at the amount of work per month or per day and comparing the difference of that before and after. – Rainbacon Jan 7 '19 at 14:48
  • @Rainbacon I agree that this would be interesting. Unfortunately, I only have access to the number that Shog9 give so I can't do a month by months comparison. – Ael Jan 7 '19 at 14:55
  • 1
    you wouldn't need to get specific month by month stats. You could just average it. So for example, in the 9.5 months prior to removal from the HNQ, the moderators suspended 32 users, or an average of 3.3 suspensions per month, while in the 2.5 months after they suspended 7 users or 2.8 suspensions per month. By suspending 0.5 fewer users per month, the monthly workload of suspensions went down by 0.5 / 3.3 or about 15% – Rainbacon Jan 7 '19 at 15:23
  • @Rainbacon You lost me with all your math ^^ I'm not really seeing where you want to go, but I can provide you with a csv file of Shog9 numbers so that you can do your own magic with it. – Ael Jan 7 '19 at 15:55
  • I appreciate the offer, but there's no need. It will only take me a minute or so to copy the data over from the post – Rainbacon Jan 7 '19 at 18:05

You must log in to answer this question.

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