Note 17: Universal features vs. contextual interpretation

The last point already made the point that some edge weights representing real-world concepts such as probabilities or friendship do not allow a meaningful interpretation of graph theoretic distances. Such an interpretation is depending on the context, the meaning of the relationships and weights in the real-world. However, hip new "network science" was in part so very hot in contrast to lame, old "social network analysis", because it just applied any kind of measure to all kinds of complex networks to identify structures common to all of them. This was the case for Watts' and Strogatz' seminal paper on Small-Worlds (Watts, 1998) or for Barabási and Albert's paper on Scale-Free Networks (Barabási, 1999).

Notes 8-11: What is the difference between social network analysis and network science?

The following four notes are from my book "Network Analysis Literacy" (Zweig2016).

Summary of the differences between social network analysis and network science. Of course, this is a generalization and will not apply to every single network analytic project from either field.

"Note 8. The first big difference between social and complex network analysis as a part of network science, however, is that the underlying data is not restricted to social systems but comprises all relationships between any kind of entities in any given complex system."

"Note 9. A second important difference between network science and social network analysis is that (in general) the first induces micro-behavior from observed macro-behavior while (in general) the second predicts macro-behavior from hypothesized micro-behavior."

"Note 10. Social network analysis tries to capture many details from the social system of interest. Often, additional parameters of the persons under observation are requested and used for the analysis. The approach is thus a contextual approach that takes the context into account. In network science, the abstraction level is in most cases much higher and individual properties of the entities are much less often taken into account. The approach can be characterized as being largely context-free."

"Note 11. In summary (and a bit bold), social network analysis is a theory-driven, bottom-up approach that carefully models additional social information where available and takes it into account when interpreting the results. Network science follows a data-driven, top-down approach that tries to clean the data from all detail to compare the core structure of different complex networks."

Reference:

 (Zweig2016) Katharina A. Zweig: Network Analysis Literacy, ISBN 978-3-7091-0740-9, Springer Vienna, 2016