[hyperpape]

I think the net of it is: anonymity encourages some people to be abusive, but enough people are abusive under their real names that you can't say anonymity is the primary factor behind abuse.

Whether curbing anonymity is a necessary component of fighting abuse and/or worth it becomes a further question.

[_red]

It may in fact be self-selecting.

Out of the total corpus of people, there are some who would be abusive if anonymous, however they have enough 'social awareness' to not behave that way when real names are used.

However, there are another subset of this group who lacks the social awareness to curb their abusive behavior even though they are using their real names.

By enforcing a "no anonymity" policy, you filter out the first group effectively but wind up increasing the percentage make-up that the second group has in the overall community.

[ykler]

The study doesn't even address the question of how much real names help. No one uses their real name on Wikipedia. It is just comparing edits from logged in users with edits from people who aren't logged in (sometimes called "IPs" since their IP address is recorded). Also, some abuse from logged in users comes from sock puppet accounts.