Why are the mod-close rates here on Super User so much higher than on other similar Stack Exchange communities?
For comparison, I pulled the questions for the last (roughly) 11 days to analyze close-vote patterns from 4 different Stack Exchange communities.
For comparison to Super User, I used Ask Ubuntu, Unix & Linux, and TeX - LaTeX since they are the only other non-programming, technology SE sites with at least 50 questions a day. Stack Overflow, of course, would not be a valid comparison as it only requires 3 community-close votes.
For reference, I used the following API query:
https://api.stackexchange.com/2.3/questions?pagesize=100&fromdate=1652400000&todate=1653350400&order=desc&sort=activity&site=tex&filter=!nKzQUR2*et
... with, of course, the
site
adjusted. And, of course, each page pulled in a loop.I limited it to the last 11 days since, beyond that, closed questions start getting Roomba'd. It's likely some of them already were deleted in that period if they were closed quickly.
I limited my analysis to non-duplicate closures, since duplicates are often closed by gold-badge holders, not mods.
I double-checked the data to confirm that every (non-duplicate) question that was closed with fewer than 5 votes was closed by a Mod-vote.
If anyone would like the raw JSON results for validation (and I don't mind additional eyes on the data one bit), I'll try to put it on Github.
The results, honestly, showed what I already suspected. Questions are closed by Mods here on Super User at a much higher rate than the other sites.
Super User | Ask Ubuntu | Unix & Linux | TeX - LaTeX | Total Non-SU | |
---|---|---|---|---|---|
Total Questions (~11 days) | 864 | 961 | 558 | 568 | 2087 |
Questions Closed | 113 | 108 | 43 | 44 | 195 |
↪ Percent of total questions | 13% | 11% | 8% | 8% | 9% |
Non-duplicates Closed | 96 | 55 | 26 | 25 | 106 |
↪ Percent of total questions | 11% | 6% | 5% | 4% | 5% |
↪ Percent of total closures | 85% | 51% | 60% | 57% | 54% |
Non-duplicates Closed with 5 votes (Community-close) | 12 | 47 | 21 | 25 | 93 |
↪ Percent of non-duplicates | 13% | 85% | 81% | 100% | 88% |
Non-duplicates Closed with fewer than 5 votes (Mod-assisted) | 84 | 8 | 5 | 0 | 13 |
↪ Percent of non-duplicates | 88% | 15% | 19% | 0% | 12% |
Non-duplicates Closed with 1 vote (Mod-only) | 45 | 4 | 2 | 0 | 6 |
↪ Percent of non-duplicates | 47% | 7% | 8% | 0% | 6% |
I think the trends are apparent, but to point out the key stat -- 88% of non-duplicate (84:12) closures here were done with moderator involvement. The average for the other sites was 12%, making the SU mod-closures more than 7 times higher than the other sites.
Nearly half (47%) of non-duplicate closures were done with a sole moderator vote. That's almost 8x the rate of the other sites.
It doesn't seem to me that the quality of questions here is so much worse than the other sites that they need to be mod-closed rapidly without reaching community quorum. So what's the reason for this (very large) discrepancy between Super User closures and that of the other communities?