7

Thinking about this post: Suggested edit system is causing grief, needs to be rethought

I'm looking for a Data Explorer query to see how many users have received the full 1000 rep limit from accepted edit suggestions. It shouldn't be many because you don't earn them any more after 2000 (with the exception of tag wiki edits), I know I'd personally only made about 20 edits before I hit 2000 rep.

I'm interested in this because I'm concerned about the users who may be submitting low quality edits only to gain rep.

Ideally the results would be sorted by lowest rep first, to see who only has 1001 rep, with all 1000 coming from edits.

Just think, all those people are presumably "established users" who have "been around for a while"! Even if it's just from a few hours of adding tags!

Improve this question
12
  • 3
    I consider myself an extremely conscientious editor (barring an occasional lapse), and I definitely got very close to the 1K cap, but I don't think I actually hit it. And yes, I liked the rep, but I did it mostly because bad writing makes my brain itchy. Still, this would be an interesting statistic to have, just perhaps not for the reasons you're thinking.
    – jscs
    Feb 18 '14 at 6:58
  • Maybe sorting by how long the user has been a member would be better. You have all done valuable services and I didn't intend to discount that.
    – mhlester
    Feb 18 '14 at 7:00
  • Man talk about starting this question off on the wrong foot!!
    – mhlester
    Feb 18 '14 at 7:01
  • 1
    If the edit is adding useless tags just for rep, I think the system is broken. I'm curious how broken.
    – mhlester
    Feb 18 '14 at 7:05
  • 2
    @mhlester I've attempted an edit to put this in terms which involve less mud-flinging. :) I imagine it's more about concern over low quality edits anyway - after all, who cares if someone does 500 great edits all for the rep? They did 500 great edits, and that's fantastic! If the rep system helped motivate them to do that, then the rep system is working!
    – doppelgreener
    Feb 18 '14 at 7:06
  • There is a TargetRepChange in the database, but it seems to only pop up 0 for whatever reason. I wonder if it isn't actually possible other than creating a query to see who has over 500 approved suggested edits under 2000 rep...
    – jmac
    Feb 18 '14 at 7:08
  • @JonathanHobbs, I'd just made an attempt to do the same, but at this point I'll take anything that redirects the question back on the right track. So much egg in my face right now! Eek
    – mhlester
    Feb 18 '14 at 7:08
  • This query returns all zeroes, so I don't think TargetRepChange ever actually returns anything but zero: SELECT TOP 100 SUM(TargetRepChange) AS [Rep Change], TargetUserId AS [User Link] FROM SuggestedEditVotes GROUP BY TargetUserId
    – jmac
    Feb 18 '14 at 7:10
  • 1
    @jmac, or maybe the problem doesn't exist at all!
    – mhlester
    Feb 18 '14 at 7:15
  • @mhlester, there are definitely a lot of people who got 1000 rep from suggested edits I'd wager. Whether it's a problem or not is irrelevant to whether you can query for it. The issue at hand is that you don't seem to be able to in any straightforward way.
    – jmac
    Feb 18 '14 at 7:21
  • @jmac, Sorry that was a not-so-obvious joke. Thanks for trying!
    – mhlester
    Feb 18 '14 at 7:22
  • As much as I'd like to believe that all serial editors have noble intentions, I'd be surprised if the majority of people who've hit that cap haven't submitted a large amount of terrible rep-fuelled edits.
    – OGHaza
    Feb 18 '14 at 7:27
7

Here is a query that gives the top 100 users by number of suggested edits:

SELECT TOP 100
  OwnerUserId AS [User Link],
  COUNT(*) AS [Suggested Edits]
FROM
  SuggestedEdits
WHERE
  ApprovalDate IS NOT NULL
  AND RejectionDate IS NULL
GROUP BY
  OwnerUserId
ORDER BY
  COUNT(*) DESC

There are 86 Users (not including anonymous users) with 500+ suggested edits in the system.

Caveats

  1. Deleted posts are not included, so this number could be higher
  2. My SQL is not grand
  3. This only indicates approved suggested edits and does not indicate whether or not rep was gained from them.
  4. This includes suggested edits to tag wikis and excerpts from users with up to 20k rep.
Improve this answer
7
  • 8
    Hey stop disrespecting mySQL
    – Pekka
    Feb 18 '14 at 12:21
  • @Gilles, if you know how it works, you could improve it rather than just adding another disclaimer (just sayin'). Like I know how to join tables.
    – jmac
    Feb 18 '14 at 14:08
  • That's great! I wonder if it's possible to order by reputation or how long they've been members? It's certainly comforting to see only 86 users though!
    – mhlester
    Feb 18 '14 at 17:28
  • @mhlester, it should be possible. You'd have to do table joins to the users table though. And you should also be able to exclude wiki and excerpt edits with a join to the posttype as well. Both of these would be a bit...complex...for my level of SQL (complete and total novice), though I invite you to fork the linked query and try yourself!
    – jmac
    Feb 18 '14 at 23:32
  • 1
    @Lower, yes. That is exactly it. I assure you it wasn't because I figured I made an error in the query and wanted it to run quick which is why I picked top 100, and that just happened to give the answer when it shockingly ran properly.
    – jmac
    Feb 18 '14 at 23:33
  • @jmac, sadly your SQL is far above mine
    – mhlester
    Feb 18 '14 at 23:38
  • @mhlester, take a look at similar queries, try out ways to combine them, and you will be at my level. Alternatively, ask a separate question on how to combine them, or try in Meta Stack Exchange Chat to find an SQL-related room that may be willing to give you guidance. I know the concept, but it'd take me an hour to try to dig through it. With only 86 users, it shouldn't be too hard to go through manually to collect data if you need to.
    – jmac
    Feb 18 '14 at 23:41

You must log in to answer this question.

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