Research talk:First edit session/Outline

"Completion"

Re: "with completion of popular articles..." I'd be extremely wary of suggesting that articles on Wikipedia are ever "completed." After all, that's the point – Wikipedia articles are always open for anyone to update/improve; nobody "owns" the article, hence nobody can come in and say it's done and doesn't need any more contributors. I understand the idea you're trying to convey, but I think you should put more stress on the imperfective aspect rather than the perfective one. Something like "with articles becoming longer and more elaborate" or "with articles achieving higher and higher levels of quality," maybe?

And speaking of the completeness thing, jeez, that aspect article needs help. Editing time! :) --Buickmackane 19:54, 6 August 2011 (UTC)

I agree. "Complete" is definitely the wrong term. I've fixed it.  :) Thanks!!! --EpochFail 21:23, 6 August 2011 (UTC)

Correlation versus causation

Regarding this:

This confirms the hypothesis that if a newbie's first experience editing Wikipedia is full of rejection, he or she be unlikely to continue working in the project. It turns out that this effect has existed throughout the history of Wikipedia and has been increasing over time, though it has decreased somewhat in recent years. What's more, while editors who show a high initial investment in the community (by making many edits in their first edit session) are more likely to survive in general, these highly invested new editors suffer even more from having their work rejected than editors who express a lower investment.

I could, from the data as I understand it, write a completely different conclusion:

This confirms the hypothesis that if a newbie's begins by making poor quality edits, including vandalism, that are rejected, he or she be unlikely to respond by subsequently posting quality edits, but rather will stop contributing to the project. It turns out that this effect has existed throughout the history of Wikipedia, and that rejects have been increasing over time, though that has decreased somewhat in recent years. What's more, while editors who show a high initial investment in the community (by making many edits in their first edit session) are more likely to survive in general, such editors, if they post a large number of poor quality or vandalizing edits, are even more likely to stop editing that an editor who has done a lower number of poor edits.

The difference in these two conclusions is that you seem to be assuming that editor's edits that are rejected by being deleted or reverted in their first three edit sessions are good faith edits, and that their rejection is due to excessive expectations for quality, while I'm assuming that these edits, in general, are very poor quality edits, or even vandalism, edits that - if left as is - would degrade the quality of the Wikipedia article.

In short, it's not at all clear that you're controlling for quality of edit. To confirm the hypothesis, you really need to show that for edits of similar quality, whether or not the edits are rejected is correlated with the number of subsequent edits by that newbie. John Broughton 23:53, 7 August 2011 (UTC)

Exhibit A and Exhibit B. Both of these studies were very preliminary and can hardly be considered scientific; the samples were relatively small and the numbers we got out of them probably aren't perfect. But what you learn when you spend a month of your life doing nothing but coding newbie edits (try it some time!) over different years of English Wikipedia's existence is that:

1. Newbie edits have always tended to be less than perfect because... they're newbies! Go back through the contribution histories of the most active veteran Wikipedians and look at their first edits. You'll see shining examples of spam, vandalism (accidental and not), copyvio, self-promotion, and original research.
2. What constitutes "very poor quality" in the minds of Wikipedians has changed dramatically over the years, in a way that challenges one of the basic foundations of Wikipedia, Assume good faith.
3. There are still lots and lots of newbies who make good, even great edits in their first editing session, but whose talk pages are filled with nothing but aggressive template warnings and deletion notifications.

But you're right to bring up this issue and its central importance to our thinking about the current state of the project. It's clear that we need to develop a more rigorous and in-depth study of newbie editing patterns if we really want to convince the community to be more accepting of new editors' initial contributions. --Buickmackane 16:58, 8 August 2011 (UTC)

This is a major caveat of this work and it's the reason why a section was devoted to its discussion. I'd love to find out how much the increase in rejection can be explained by a decrease in the quality of the average new editor's work. We are currently working on a way to test this. --EpochFail 20:58, 9 August 2011 (UTC)