Wikipedia talk:Requests for comment/Abortion advocacy movement coverage/CMJ analysis with SoA

CMJ analysis of feedback with subjective strength of argument as a factor
This is my variant analysis of the community feedback on the RFC with weighting by my subjective evaluation of participants' strength of argument. I'm posting it here so as to avoid cluttering up the main talk page and because I don't know that this is a particularly meaningful artifact. Points of interest:


 * You can download my worksheet and play with it to see what it looks like with your own subjective numbers (if you can read OpenOffice Calc .ODS documents) from here.
 * Maximum weight is 1. If an editor is weighted other than at 1.0, that isn't a straight multiplier on his or her numbers; rather, it causes the numbers to draw toward 2, i.e. neutral.  So 0.5 weighting on 4 gives a result of 3, and 0.5 weighting on 1 gives a result of 1.5.  Any weight on 2 gives a result of 2.
 * I weighted the 20% supermajority rows at the mean of assigned weights, so that they wouldn't be swamping actual editors.
 * I gave people weights around 0.5 to 0.8 for not really saying much and apparently mostly just referring to the core RFC arguments, or otherwise not being all that responsive. I gave people weights around 0.1 if I thought they obviously didn't understand the RFC at all.  I gave people weights of 1.0 if they made an effort and said cogent things, even if they were obviously wrong.
 * Yes, I weighted Student7 at -0.2, meaning that I consider his Rethuglican slut-shaming argumentation to be so vile and odious and irrelevant to the maintenance of Wikipedia as to cause his opinions to weigh in the opposite direction from that he desires. If you don't like it, go cry.

This basically causes title option 2 to come out farther ahead. Obviously, this coheres with my evaluation that a decent chunk of the people in favor of 5 didn't give their opinions very strong backup (though certainly the bulk of those supporting 5 did give sensible, policy-grounded reasons). Others may naturally evaluate otherwise. :) The math is interesting, with the field becoming largely much more close because of zeroes turning into 0.2s and 0.4s and whatnot.

So there's that, do with it what you will. :) —chaos5023 (talk) 19:13, 6 November 2012 (UTC)


 * After further experimentation, I made some changes to the underlying methods I was using to calculate the CMJ score, because it seems like it's incredibly sensitive to how the median of the results is handled and is easily driven completely insane by tiny changes to that methodology. The current version reflects the updates.  If you downloaded the old spreadsheet, you should pull the current version. —chaos5023 (talk) 21:44, 6 November 2012 (UTC)