Talk:Universal Code of Conduct/Revised enforcement guidelines/Voting statistics
Add topicVoting anomaly
[edit]The voting graph shows a massive dump of around 1000 votes above-background-rate on day 7 partially stretching through 7.5 and 8 plus day 11 partially stretching through 11.5 and 12. Can someone explain this? I assume the logs show where the votes came from. The graph of the previous vote shows a lesser but identical pattern spiking on days 11.5 & 13.
Also can we get the time graph for the yes/no vote percentage? I'd like to see if the vote spikes differ from the baseline vote. Alsee (talk) 10:27, 15 February 2023 (UTC)
- Hi Alsee. The graph for the yes/no percentage isn't possible, since the voting is anonymous. We cannot see how an individual voted, and the votes in the dump are randomised. Joe Sutherland (Wikimedia Foundation) (talk) 17:24, 15 February 2023 (UTC)
- On the topic of increases in voters - these coerrelate to email messages sent to eligible voters which were sent out on those days. This was done in line with Board Election practices, where you can see similar increases in numbers after email messages are sent. Patrick Earley (WMF) (talk) 17:45, 15 February 2023 (UTC)
- A very interesting letter. Reporting that more than half of the voters supported the guidelines, not that almost half of the voters voted against guidelines. Lol. Iniquity (talk) 00:25, 16 February 2023 (UTC)
- Because the purpose of the emails was to inform people about the second vote, not the first one. In that specific context, it's not important whether supporters won the first vote by a lot or a little. Anyway, the second voting results show a 3:1 ratio in favor of supporters, so your problem is no longer a problem. Adrianmn1110 (talk) 07:17, 16 February 2023 (UTC)
- It's focus shifting and voting control, so the problem hasn't gone away. Iniquity (talk) 08:59, 20 February 2023 (UTC)
- Context matters. The losing side of the first vote was not the subject of the emails. Therefore, shifting focus away from them was fine. As for the alleged voter control, can you prove that not mentioning the exact percentages of the first vote has unfairly influenced the second vote? Adrianmn1110 (talk) 01:55, 21 February 2023 (UTC)
- It is necessary to prove that it did not affect, and not to prove that it did. According to behavioral psychology, any positive reinforcement of a statement will bias the outcome. Iniquity (talk) 10:34, 21 February 2023 (UTC)
- Context matters. The losing side of the first vote was not the subject of the emails. Therefore, shifting focus away from them was fine. As for the alleged voter control, can you prove that not mentioning the exact percentages of the first vote has unfairly influenced the second vote? Adrianmn1110 (talk) 01:55, 21 February 2023 (UTC)
- It's focus shifting and voting control, so the problem hasn't gone away. Iniquity (talk) 08:59, 20 February 2023 (UTC)
- Because the purpose of the emails was to inform people about the second vote, not the first one. In that specific context, it's not important whether supporters won the first vote by a lot or a little. Anyway, the second voting results show a 3:1 ratio in favor of supporters, so your problem is no longer a problem. Adrianmn1110 (talk) 07:17, 16 February 2023 (UTC)
- A very interesting letter. Reporting that more than half of the voters supported the guidelines, not that almost half of the voters voted against guidelines. Lol. Iniquity (talk) 00:25, 16 February 2023 (UTC)
- On the topic of increases in voters - these coerrelate to email messages sent to eligible voters which were sent out on those days. This was done in line with Board Election practices, where you can see similar increases in numbers after email messages are sent. Patrick Earley (WMF) (talk) 17:45, 15 February 2023 (UTC)
┌─────────────────────────────────┘
The burden of proof belongs to the side alleging that a vote was improper. Otherwise, we would constantly overturn elections or referendums just because some people accused them of being unfair. Can you prove that the second vote was unfairly influenced? You talk about positive reinforcement and its potential to create bias, but what about negative reinforcement? I see a lot more criticism than praise of the Enforcement Guidelines. A voter who reads those criticisms could imagine the Guidelines to be much worse than they really are. Isn't that also a kind of bias? Adrianmn1110 (talk) 20:45, 21 February 2023 (UTC)
- Given the way these guidelines have been adopted and how badly the community department is doing that, sorry, no, the burden of proof is on the interviewers. Iniquity (talk) 21:13, 21 February 2023 (UTC)
- No, the burden of proof is still on those who allege an unfairly manipulated vote. They have not provided proof. How do the WMF's actions erase your responsibility to prove your claims? Adrianmn1110 (talk) 01:08, 22 February 2023 (UTC)
- No, of course you are wrong the burden of proof is on the interviewers. Iniquity (talk) 07:11, 22 February 2023 (UTC)
- No, the burden of proof is on people who accuse the second vote of being unfairly influenced. Adrianmn1110 (talk) 01:04, 23 February 2023 (UTC)
- No, of course you are wrong the burden of proof is on the interviewers. Iniquity (talk) 07:11, 22 February 2023 (UTC)
- No, the burden of proof is still on those who allege an unfairly manipulated vote. They have not provided proof. How do the WMF's actions erase your responsibility to prove your claims? Adrianmn1110 (talk) 01:08, 22 February 2023 (UTC)
Emails should be sent before opening the vote to avoid this kind of complaint. However, I don't see any problem with the emails sent during the vote. A l p h a m a Talk 09:18, 25 February 2023 (UTC)