Collecting social network data used to be a tedious, labor-intensive process. In fact, several notable dissertations came out of the researcher’s being at the right place and the right time to be able to observe a social conflagration and gather data on it. Social network data collection is, by nature, more invasive and harder to anonymize (since each respondent must provide names of other people); survey instruments had to be approved by Institutional Review Boards (IRBs), and administration of the surveys was tedious manual labor.
In corporate settings, the situation was not much better. While it was possible to gain access to significant amount of data without conducting a survey (e.g., by tapping into an email server log), the data received that way is not a good indicator of trust relations, or formal and informal structures in the organization. Finally, SNA was (and still occasionally is) viewed as a blunt instrument by HR departments (who might mistake a worker with low betweenness centrality for one who is redundant, and lay him off).
Twitter and other social sites make our lives much easier as researchers. First of all, things that are said on Twitter are considered public data. What this means is that an Institutional Review Board does not have to conduct a full review of the study (if you’re based at a university), and your firm is not in danger of very bad publicity for the mere act of conducting a study.
If you are at a university or a public ...