Monday, April 26, 2010

The anonymous coversion rate

For #Wikipedia, editors are its most valuable resource. Theory has it that there is a conversion rate whereby anonymous editors become registered and a more visible part of the community. This is the theory and, when it works like it does for the Russian Wikipedia and the English Wikipedia it works beautifully.

When you compare these Russian statistics with those for the Malayalam or the Hindi Wikipedia the absence of anonymous edits is striking. The question is why are there so few anonymous edits.

With so few anonymous edits, there will be no conversion to registered editor. This means that other strategies are needed to grow the community. Having people available to spread the word about "the encyclopaedia that everyone can edit" is one. Making it as usable as humanly possible is another; the usability project is doing what it can. A third is what the Indonesian chapter hopes to achieve by training Wikipedians and make it sustainable by making it a "train the trainer" project as well.

These new statistics from Erik Zachte help to appreciate such diversity. They truly enrich what is already an important resource.


Joseph Reagle said...

Drats, I thought you were going to be able to post statistics about the conversion rate of anonymous users to regular users. Rather, you take that as a continued assumption, and then note rates of poor anonymous contribution in some language Wikipedias. That's fine, but this issue of "anonymous conversion" is very interesting, and taken as a matter of faith at Wikipedia. Of course, answering the question is quite challenging -- which was why I was curious to see your headline. It's almost like that "marijuana as a gateway drug argument." Just because an anonymous user at some point becomes a regular user, that doesn't mean they would not have become a regular user without the gateway capability.

Personally, I think there is merit to the anonymous conversion assumption, but would love to see some compelling arguments.

Sage said...

What I noticed was a striking variability in anonymous-to-registered edits ratios even among successful Wikipedias.

For example, for Japanese, anonymous edits have at times outpaced registered edits. Catalan, on the other hand, has very few anonymous edits, but considering the number of speakers it is doing quite well.

I'm not sure we can tell much about the health of a project just by looking at what portion of edits are anonymous. There are major cultural factors in play; anonymous online activity is a major part of Japanese culture; Catalans, perhaps, take more individual pride in their Wikipedia contributions.

Anonymous said...

thank for share, it is very important . ̄︿ ̄