Hi
Hi
I know that your study had already started and that you will not change your study.
Part two of your research will be "New Wikipedians at different points in time and how the age of the cohort affects their likelihood of staying with Wikipedia" That is something that interests Wikipedians, because if we would know who will stay more than a few days at Wikipedia, we could care more about them. I asked myself, if we could do a en:Cluster analysis to distinguish new editors. (By the way, if all editors could be clustered, we might find different groups of editors which need different things (maybe a different mediawiki-skin? (that is the software) Or we should change our help-system?). At the moment we distinguish only active and very active editors. (that is a quantitative analysis. I propose a qualitative analysis) I´m sorry that I couldn´t do that on my own.
Although I did some research on my own. I asked myself if there is a connection between the number of reader of the mainpage and of all pages and the number of new editors. My result was, that the mainpage has no influence on the number of new editors. Another question was how much time is between logging in as new user and the first edit. I looked into one day (24 hours, June 8. 2009, a Monday) and counted the numbers by hand. I found interesting, that a lot of new editors took a lot of time (..., more than 15 minutes, more than one hour, some even more than 24 hours) before they start editing.
As a Wikipedian I don´t see those editors. They do a few edits and never come back, but they are the majority of editors. And those people who logged in, but never did one edit at all are a even larger group. (Maybe that is the reason why the foundation start looking into the logg-in process.
Well, thanks for being a research consultant at the Wikimedia Foundation.
Dear Goldzahn,
Thank you for sharing your thoughts with us. I think doing a cluster analysis is indeed a great idea and it could help us in detecting different types of editors as some of the other discussants have observed as well. Do you have a link to your results?
Sorry, but they are written in German. Another point is, that you have to be aware that there are a lot of editors that are logged in via en:Wikipedia:Unified login. In my findings from 267 new editors, coming in via Unified login, 255 never ever edited. At "Account Creation Improvement Project" you will find, that in the German WP there are 579 new user accounts average per day. That means, half of them are Unified login editors which are indeed experienced users from another project (not new editors).
That was me who deleted my link of the list on the frontside of this page. Sorry, I forgot to login.
No problem, and I will ask around to see what the implications of Unified Login are (I assume that it means your userid will be identical across Wikipedia's but let me check).
Maybe bots that are linking articles between different project have an implication on this study. You could detect them via Unified Login, because there you can see which is the home wiki. Or you are able to see if a user has a bot flag, but I don´t know if every bot has a bot flag.
By the way, in your definition of New Wikipedians and Active Editors you don´t say anything about bots and editors with a different home wiki.