[Air-l] Blogs as source of socio-demographic data

Hello everyone, recently I was surfing Russian facebook-clone vkontakte.ru and decided to count statistics of political preferences. I don't consider my results to be valid, so I've decided to ask about any thoughts, articles etc. on the validity of blogs as a source of socio-demographic data (age, gender, location, political and religious preferences etc.). While I think that other interests such as music, reading, films etc. are quite reliable I can't say the same about socio-demographic data. What do you think? Thanks in advance. Best wishes, Alexander Semenov. MA student Faculty of Sociology Moscow School of Social and Economic Sciences (MSSES) http://www.msses.ru/English/index.html