Network Analysis: Integrating Social Network Theory, Method, and Application with R

There are many good books on network analysis. Some offer detailed takes on methodology, discuss mathematical foundations, or examine specific analytical tools. Others give a broad theoretical perspective or focus on applications in specific domains such as health or organizations.

Network Analysis: Integrating Social Network Theory, Method, and Application with R stands out because it offers a nuanced and thoughtful take on the relationship between network theory and methods. ISN (the abbreviation of the book title I will opt to use here instead of the more cumbersome NAISNTMAR) may not be the first choice for people who want an easy initial glimpse into network basics. Instead, the book is an excellent resource for researchers looking for a deeper understanding of the way theoretical frameworks and methodological approaches can come together to shed light on a research question. ISN tackles complex themes related to the balance of structure and agency, the primacy of positions or connections, and the interplay of motives, actions, and social environments.

Throughout the text, the authors emphasize an important insight into the nature of network research: many methodological challenges we encounter are in fact theoretical problems in disguise. How we define relationships and construct networks, for instance, depends on our theories and the epistemological assumptions they represent. Theoretical perspectives should similarly inform our understanding of suitable network boundaries, relevant descriptive measures, and appropriate model specifications. Endogeneity is to be treated not as a statistical challenge, but as an issue that requires a better conceptualization of the complex causal relationships between constructs.

When it comes to network methods, ISN covers a lot of ground. It is divided into three sections that discuss network fundamentals, network structures, and network models.

The first section of the book outlines the basics of network research. It starts with a thoughtful grounding chapter describing how researchers in a variety of fields have studied social structure, relationships, and roles. The rest of the chapters in this section discuss the nature and representation of social networks, network data collection, and network visualization. The data collection chapter is a particularly good example of the approach that ISN takes to integrating the conceptual and practical. It goes beyond a simple description of the sources and types of network data and discusses diverging theoretical approaches to conceptualizing structure that lead to different methodological choices. The chapter also outlines critically important (and often omitted) considerations on data quality, missing data, and ethics.

The second section of the book begins with another grounding chapter, discussing the interplay of agency and structure. The chapter outlines a take on human motivations for action and their implications for the way we interact with each other. These theoretical assumptions provide a framework for understanding ego-network research and motivate the choice of specific network measures. The following chapters in this section cover a variety of patterns in network structure, from the dyadic and triadic, to groups and communities. The authors also discuss hierarchy and centrality, roles and structural equivalence, as well as the nature and theory of affiliation networks. The section concludes with an interesting chapter that somewhat deviates from the traditional topics of network analysis books by focusing on the role of culture. Methods discussed in that chapter include dimension reduction approaches, as well as ways to represent structures of meaning through natural language processing and semantic networks.

The third and final section of the book covers network models predicting behavior, structure, and spread. The section opens with a neat metaphor, advising us to think of networks not as pipes transmitting resources, but as watersheds – channels and basins shaped by the flow of water (or in the non-metaphorical case, by social activity). The section covers a variety of ways to examine how these social channels form and how they shape flows of activity and interactions. Methods discussed here include static and dynamic exponential random graph models (ERGM), stochastic actor-oriented models (SAOM), relational event models (REM), as well as models examining network diffusion and behavioral contagion. While the book gives a good sense of mathematical definitions and the intuitions behind each approach, its strength lies in the grounding of each methodological choice in a robust conceptual framework.

The final chapter of the book offers a vision of key new frontiers in network research. It discusses approaches to analyzing large-scale complex networks, concluding that this may be another theoretical problem posing as a methodological challenge. And indeed, many of the questions we want to ask of very large networks are really questions about multiple local subsets of those networks. The concluding chapter of ISN also goes over several other areas of growth in network research, including a look at our cognitive maps of social networks; key network questions at the individual, dyadic, and group levels; ways of expanding existing models of diffusion; relevant approaches to comparing multiple networks, and ways of analyzing multiple different relationships among the same actors.

While the book itself offers a nice integration of theory and methods, its companion website (https://inarwhal.github.io/NetworkAnalysisR-book) demonstrates their applications in practice. The ISN website provides step-by-step instructions for conducting a variety of analyses using the R language for statistical computing. R offers the most versatile analytical toolkit for network scholars available today: it is open source, supports reproducible research, and benefits from a large community of scholars developing cutting-edge tools for it.

The ISN website is a valuable resource, freely available to researchers, and useful in combination with the book or on its own. Much like the book, the website is probably not the first place to go if you are just starting out and have never used R before. It is, however, an excellent guide for researchers with some coding experience who want to understand networks, or expand their analytical toolkit to include new methods. Overall, Network Analysis: Integrating Social Network Theory, Method, and Application with R offers a great mix of theoretical depth and practical advice. The authors guide readers through the complexities of network analysis, from basic concepts to advanced models, continuously highlighting the interplay of theory and method. The additional R tutorials enhance the usefulness of the book, making it a more comprehensive guide to conducting rigorous research. If you are currently planning a network project, you can certainly use ISN to understand what questions you need to ask, what theories to consider, what methods to apply, and how to get it all done in R.