Research Methods 2.0: doing research using virtual communities

Virtual communities offer an unprecedented access to new forms of primary data in many fields of inquiry. In particular, user generated content on sites such as YouTube, Facebook and MySpace provides primary data both about topics of interest to users of those sites as well as the responses of the online community to the posted content. For example, shortly after the launch of the iPhone 3G, there were literally thousands of video clips uploaded to YouTube that offered demos and reviews of the new product (Blythe &Cairns, 2009). Additionally, many people posted comments and ratings about these videos. Such materials could provide a rich resource both in researching the community of people who buy or use the iPhone 3G and in the design of future products. However, both the quality and the quantity of the available data offer substantial obstacles to existing research methods and therefore bring into question the value of research based on this data. Additionally, the very dynamic nature of user generated content sites makes it impossible to be exhaustive and presents something of a moving target to the researcher. In this chapter, we consider how existing quantitative and qualitative methods may be applied to researching the primary data offered by virtual communities. Specifically we focus on analysing YouTube content and we discuss how different methods apply, what their limitations are in this context and how those limitations may be overcome through adaptations of the method or even new methods. The existing methods considered are statistical analysis, content analysis and grounded theory. These methods are not just used to study the virtual community embodied in YouTube but also people's relationships with technology as reflected in the videos and their associated comments. It becomes clear that each method has something to say about understanding YouTube content and could say more if developed for this sort of data. What is also clear, though, is that in some sense these methods miss some of the key sociological phenomena around the videos. In previous work (ibid), we found that the most popular video related to the iPhone 3G was the " Will it Blend? " video where an iPhone 3G is blended to dust in a parody drawing on 1960's gameshows and scientific documentaries amongst other things. The sheer numbers of people viewing this video and commenting on it, more than 2.5 million and 10,000 respectively, shows that this video has …