Mapping Social Networks over Time and Space

As a novice in social network analysis (SNA), I feel I should be sticking to the basics for now.  I came across this realization while reading Prell (2012), who discusses (pp. 77-78) among other things the potential problems posed by data “reliability,” defined roughly as “the extent to which a measurement will yield the same results time and time again.”  Not yet knowing exactly how to analyze social data, let alone how to account for problems with reliability, I feel comfortable gravitating toward social data and social actor relations that are likely to have a high intrinsic degree of reliability...  

For instance, whether or not two social actors attended elementary school together seems to be a fairly reliable and static measurement, as opposed to whether or not two social actors consider each other to be confidants, which can change over time.  Analysis of the former type of data seems less daunting.

But then I coincidentally stumbled upon (in fact, using StumbleUpon) an article that highlights the limits of analyzing stable ‘snapshot’ data and the powerful possibilities in analyzing relations that shift over time.  The article, “Mapping Our Friendships Over Time and Space: The Future of Social Network Analysis,” argues that while most SNA has focused on relatively static relations, “If new factors could be taken into consideration, specifically changes over time and space, then social network analysis could discover things like emergence or decay of leadership, changes in trust over time, migration and mobility within particular communities online.” 

The concept by itself does not seem entirely new.  After all, the very first datafile introduced in this class—FREEMAN’S EIES DATA—contains two temporally distinguished datasets, which present the “acquaintance information” of the Freeman’s study subjects at the beginning and end of the study.  But what the recent article considers is the possibility of the added temporal or spatial dimension becoming a more standard component of SNA and being “performed on much larger networks, over greater periods of time.”

In fact, one internet technology expert who was quoted in the article observed (presumably with regard to social media data), “This added dimension or set of data points is out there and generally widely available as 'exhaust data', so to harness it and factor it in with the rest of the social graph would be truly valuable.”  How interesting and paradoxical that this so-called “exhaust data” could actually constitute fuel for analytic innovation.  What we need, then, may be more streamlined computational solutions and more user-friendly visualization interfaces to make the integration of the temporal and spatial dimensions into SNA easier and more common.

The temporal element is particularly intriguing.  Using it successfully in SNA would help point out not only the current state of a social network, but also how it got there and, theoretically, where it is headed in the future.  That’s pretty cool.  Combining this with other big data analytics, it seems that predicting the future is getting easier and easier.  That’s pretty scary (but still cool).