I would have thought, given the size and visibility of the sites included in the research, that it would be useful to understand how the research was carried out (approach, sample rate, interview criteria, data analysis methodologies, definitions, etc) and under what classifications the resulting data was interpreted.
As far as I can see, none of this information is available. The BBC when reporting on this research quotes Danah Boyd (the research author, a PhD candidate at the University of California-Berkley) as having said that her conclusions were derived from "many" interviews with teenage users of these sites. Why only teenage users I can't help but wonder, as there are plenty of non-teenage users on both.
This hardly inspires confidence in her findings; especially when you consider that in a previous paper entitled "Why Youth (Heart) Social Network Sites: The Role of Networked Publics in Teenage Social Life" (no publication date available, but clearly post 2005) she says: "I have found that race and social class play little role in terms of access beyond the aforementioned disenfranchised population. Poor urban black teens appear to be just as likely to join the site as white teens from wealthier backgrounds."
"aforementioned disenfranchised population" is a reference to those folk (again, teenagers) who do not have a presence on such sites; her categories for such people are a) kids who basically don't do it because their parents say not to and b) politically aware kids who object to Rupert Murdoch owning MySpace (or as she says "conscientious objectors").
In addition, entries like "I have also spent countless hours analyzing the profiles, blogs, and commentary of teenagers throughout the United States." give additional pause; since when did we start taking at face value anything contributed by end users? Unless blog/profile users are the subject of mind control, there is nothing to say that what appears online via such channels is even vaguely true or to be trusted - certainly not as a component part of research that is attempting to identify something as important as the extension of class division to the online space.
Its all far too sloppy for my liking - and really, as an industry we should look a lot harder at such things before assuming they have any veracity at all. Of course, Danah Boyd's findings are only interim at this stage; the full paper is yet to be released and I presume that by the time we do see the final output, it'll have had the necessary academic and investigative rigour applied to it.