I'm going to have to agree with most people here that more specifics are needed. After all, what the discussion is about is not 'do people play games' but 'are people part of gaming culture', so if you want to conduct a study to contribute to that discussion, THAT is what you should be measuring. There's nothing wrong with playing farmville, bejeweled or wordfeud, but if you want to know what percentage of people who identify as gamers are women and you measure that by asking 'have you played a game in the last 6 months', your study is poorly operationalized.
Why not do a bunch of case studies? Pick a couple of titles, do surveys amongst the people playing it and see what percentage of them is female. Yeah, it doesn't say anything about gaming as a whole, but since no-one can agree on what that is in the first place I hardly think it matters.
P.S.: this is not me yelling at social scientists for doing agenda studies (like some people here...), I don't even think there's anythign wrong with scientists wanting their studies to do some good (whatever they think that is), I just want it to be good science.
Why not do a bunch of case studies? Pick a couple of titles, do surveys amongst the people playing it and see what percentage of them is female. Yeah, it doesn't say anything about gaming as a whole, but since no-one can agree on what that is in the first place I hardly think it matters.
P.S.: this is not me yelling at social scientists for doing agenda studies (like some people here...), I don't even think there's anythign wrong with scientists wanting their studies to do some good (whatever they think that is), I just want it to be good science.