I’ve heard people quote Viegas’ survey pretty regularly and i find the tendency to generalize quite dangerous. As she notes, this cannot be done because of the snowball methodology. Please note what she states in her caveats:

Respondents to this survey were not selected on a random basis. Announcements for the online survey were posted to mailing lists within MIT as well as on a few high-traffic blogs published by people known to the author of this survey. The viral nature of blogs meant that the links to the survey page quickly spread to many other blogs. Nevertheless, this does not qualify as a random sample of the blogger population and, as such, the results from this survey cannot be generalized to the entire blogging community; instead, these results are representative of the state of affairs in certain portions of the blogging world.

This pilot study is a great first pass at these questions… I really hope folks follow up on them rather than generalizing from them.