I posted a high-level summary of the analysis and a number of suggestions and recommendations for the future design of CTAs: Article_feedback/Call_to_action
Topic on Talk:Article feedback
Why are there no completion rates for the Join CTA?
| “ | Given (1) the current positioning of the AFT at the bottom of the page, (2) the fact that it's not actually loaded until users scroll to the bottom of the page and (3) the fact that we know that rating conversions (i.e. the number of people rating an article after reading it) decay very rapidly with the page length, it's highly unlikely that rating may be distracting people from editing the article (people interested in editing would have done so using the main edit tab or section edit links) even if we currently do not have data to test this hypothesis. If this hypothesis holds, though, we can reasonably assume that edits produced via the Edit CTA, however small in volume, are additional to regular edits. | ” |
| —DarTar, Article feedback/Call to action | ||
Emphasis mine.
Why do you think that is reasonable? I would chose another word to describe the quality of that assumption.
| “ | This could be done by sending WikiLove to the top 5 contributors of the article... | ” |
| —DarTar, Article feedback/Call to action | ||
Are you manually going to identify the top 5 contributors to every article? Its impossible to identify the top 5 contributors to an article with software.
Why do you think that is reasonable? I would chose another word to describe the quality of that assumption.
I think that is a reasonable hypothesis based on the 3 arguments above on the conditions under which people can actually see the feedback tool
Its impossible to identify the top 5 contributors to an article with software.
Why are there no completion rates for the Join CTA?
Well, one of the problems is that it is not "highly unlikely" that rating distracts people from editing, see here and here. Therefore the hypothesis is incorrect, and assuming that you have caused an increase in edits by 2.7% is just boasting without data to back up that claim.
What you need to prove that theory is a 2.7% overall increase in edits when the full deployment is finished.
Risker mentioned that 4 FA-level writers stated they will stop participating in the project because of the AFT.
I know WikiSense, it is unable to determine who the top 5 contributors to an article are. Its just a piece of software, it is unable to read and understand English. If person X adds an infobox to an article (s)he adds a lot of bytes, even if (s)he does not fill in all the parameters. If person Y doublechecks all the sources and corrects a couple of important mistakes its possible that (s)he adds (or even removes) just a couple of bytes. According to WikiSense person X would be a top contributor, but person Y would not.
For the record:
- w:en:User:Dr_pda/prosesize provides some additional measurements of the "size" of a page besides the length of its wikitext. Someone could adapt that to provide a table with the data corresponding to each revision of a page and maybe consolidate it in the same way WikiSense does.
Unfortunately we cannot easily infer whether a reader's intention is to edit a page or not as a baseline for comparing the effects of a rating system and I agree that we cannot ultimately settle this question. What we are planning to do is to measure whether AFT significantly affected the edits volume of the pages it was applied to. This is a good suggestion others on this talk page previously made and one that can be easily tackled empirically.