Elsevier

Social Networks

Volume 29, Issue 4, October 2007, Pages 603-608
Social Networks

Book review
Carrington, P.J., Scott, J., Wasserman, S., 2005. Models and Methods in Social Network Analysis. Cambridge: Cambridge University Press.

https://doi.org/10.1016/j.socnet.2007.02.001Get rights and content

Section snippets

The methodological core of social network analysis

The field of social networks is perhaps unusual in being built around a methodological and conceptual core. In many fields, one gets the impression that methodologists are regarded as something of a necessary evil: they are needed to auger troublesome data for hidden meaning and to propitiate hostile reviewers, but one would prefer to keep them far away when not in use. In addition to their reputation for pointing out research flaws at awkward moments, disdain for methodologists sometimes

Data collection, quality and study design

One profound omission in standard network analysis texts has been a serious treatment of data collection, data quality, and study design issues. Therefore, Peter Marsden and Ove Frank's review chapters in CSW on network measurement and sampling (respectively) are especially welcome. Both authors have worked for many years on these topics, and they are able to bring a broad perspective to a domain that is still very much in flux.

Marsden's chapter provides an orderly overview of the methods

Centrality and blockmodeling

Although the field of social network analysis per se is only a few decades old (see Wellman, 1988, Freeman, 2004 for historical antecedents), it can already be said to have several well-established, “classical” traditions (most of which are canonically presented in Wasserman and Faust). Two of these – the study of centrality, and blockmodeling – are treated in this volume, via chapters by Martin Everett and Steven Borgatti (centrality), and Patrick Doreian, and Vladimir Batagelj, and Anuška

Network visualization

If centrality and blockmodeling are classical network analytic approaches, both are predated by visualization. Indeed, Freeman's historical review (2004) goes so far as to make network visualization one of the four requirements for the emergence of modern network analysis. It is thus heartening to see two solid chapters in this area.

Freeman's chapter on visualization in CSW is both conceptually and historically broad, reviewing the primary approaches to network visualization that have been used

Inferential analysis

The principal alternative to the data analytic approach is the explicitly inferential perspective that has increasingly dominated the methodological frontiers of network analysis. The chapters by Stanley Wasserman, Garry Robins, Laura Koehly, Pip Pattison, and Tom Snijders describe progress in one of the fastest-growing areas of inferential network analysis: the explicit use of discrete exponential families to parameterize models for network data.2

Vertex processes: diffusion

While the foregoing has emphasized the measurement and modeling of structure per se, a large and growing body of network literature is concerned with network-embedded vertex processes (i.e., processes involving change in nodal attributes over time). Arguably, theoretical and empirical development has tended to outstrip methodological progress in this area, particularly with recent advances in agent-based modeling of network processes.

One of the more venerable – and methodologically

Network software

The weakest element of this volume, in my opinion, is the chapter by Mark Huisman and Marijtje van Duijn on “Software for Network Analysis”. This is neither a reflection on the authors, who are fine methodologists, nor on the quality of their heroic effort to catalog the large and growing range of software tools available for the analysis of relational data. The chapter is detailed and well written, and it certainly speaks to a frequently voiced need for comparative information on software

An award winning volume

Quibbles aside, Models and Methods in Social Network Analysis is a fine volume, and a welcome update to the literature. It provides a useful summary of developments in a number of areas (particularly the use of discrete exponential family models), and it will be an important reference for those seeking to get up to speed on recent progress. That this volume has recently received the American Sociological Association's Section on Mathematical Sociology book award (rarely given to edited volumes)

References (11)

  • R.S. Burt

    Structural Holes: The Social Structure of Competition

    (1992)
  • P. Doreian et al.

    Generalized Blockmodeling

    (2005)
  • O. Frank et al.

    Markov graphs

    Journal of the American Statistical Association

    (1986)
  • L.C. Freeman

    The Development of Social Network Analysis: A Study in the Sociology of Science

    (2004)
  • P.W. Holland et al.

    An exponential family of probability distributions for directed graphs (with discussion)

    Journal of the American Statistical Association

    (1981)
There are more references available in the full text version of this article.

Cited by (0)

View full text