many entity attributes can be reconceptualized as relations connecting dyads. For example, a nation’s annual volumes of exports and imports are characteristics of its economy. But, the amount of goods and services exported and imported between all pairs of nations represents the structure of trading networks in the global economy. Patents awarded to scientists employed at high-tech firms indicate companies’ research innovations, but patent-citation networks reveal how knowledge flows through industries (Zhang, Kong, Zheng, Wan, Wang, Hu, & Shao, 2016). The number of friends indicates a child’s popularity, but only network analyses of all dyadic friendship choices can uncover important cliques and clusters. Relations reflect emergent dimensions of complex social systems that cannot be captured by simply displaying a variable’s distribution or averaging its members’ attributes. Structural relations potentially influence both individual behaviors and systemic outcomes in ways not reducible to entity characteristics. For example, efforts to control sexually transmitted infections among injection drug users and sex workers require knowledge of both social and geographic distances among street people. Researchers identified 101 “hotspots” of high-risk activities in Winnipeg, Canada, where “the combination of spatial and social entities in network analysis defines the overlap of vulnerable populations in risk space, over and above the person to person links” (Logan, Jolly, & Blanford, 2016). An experiment in a large environmental nongovernmental organization found that “boundary spanners”—individuals who cross internal boundaries, such as departmental or geographic location, via their informal social networks—were more likely to diffuse innovations, although positions in a formal organizational hierarchy mediated this activity (Masuda, Liu, Reddy, Frank, Buford, Fisher, & Montambault, 2018). The strong inference is that exclusively focusing on actor attributes loses many important explanatory insights provided by network perspectives on social behavior.
2.3 Networks
A social network is a structure composed of a set of entities, some of whose members are connected by a set of one or more relations. These two fundamental components are common to most network definitions; for example: “a network contains a set of objects (in mathematical terms, nodes) and a mapping or description of relations between the objects or nodes” (Kadushin, 2012, p. 14). Different types of relations identify different networks, even where observations are restricted to the same set of entities. Thus, the friendship network among a set of office employees very likely differs from their advice-seeking network. Stating that connections exist among members of a network does not require that all members have direct relations with all others; indeed, sometimes very few dyads have direct links. Rather, network analysis considers both present and absent ties and possibly also variation in the intensities or strengths of the relations. A configuration of empirical relations among entities identifies a specific network structure, the pattern or form of that network. Structures can vary dramatically in form, ranging from isolated structures where no actors are connected to saturated structures in which everyone is directly connected. More typically, real networks exhibit intermediate structures in which some entities have more numerous connections than others. A core problem in network analysis is to explain the occurrence of different structures and, at the entity or nodal level, to account for variation in linkages among entities. The parallel empirical task in network research is to detect and represent structures accurately using relational data.
The first researcher credited with using the term social network was John A. Barnes (1954), an anthropologist who studied the connections among people living in a Norwegian island parish. Barnes viewed social interactions as a ‘‘set of points some of which are joined by lines’’ to form a ‘‘total network’’ of relations (Barnes, 1954, p. 43). The informal set of interpersonal relations composed a ‘‘partial network’’ within this totality. Barnes drew on the work of Jacob Moreno (1934), whose hand-drawn sociograms of lines and labeled points displayed children’s likes and dislikes of their classmates. We discuss methods for representing networks visually as graphs and mathematically as matrices in Chapter 4. From anthropology and sociology, network ideas and methods diffused over the past half century to many disciplines, which adapted them to prevailing theories and problems. For historical overviews of the origins and diffusion of network principles, see Freeman (2004, 2011); Knox, Savage, and Harvey (2006); Kadushin (2012); and Scott (2017).
If network analysis were merely a conceptual framework for describing how a set of actors is linked together, it would not have excited so much interest and effort among social researchers. But, as an integrated set of theoretical concepts and analytic methods, social network analysis offers more than accurate representations. It proposes that, because network structures affect actions at both the individual and systemic levels of analysis, network analysis can explain variation in structural relations and their consequences. J. Clyde Mitchell’s (1969, p. 2) definition of social networks emphasized their impacts on outcomes: ‘‘a specific set of linkages among a defined set of persons, with the additional property that the characteristics of these linkages as a whole may be used to interpret the social behavior of the persons involved.’’ The first edition of this book underscored this perspective: ‘‘The structure of relations among actors and the location of individual actors in the network have important behavioral, perceptual, and attitudinal consequences for the individual units and for the system as a whole’’ (Knoke & Kuklinksi, 1982, p. 13). Similarly, Barry Wellman (1999, p. 94) wrote, ‘‘Social network analysts work at describing underlying patterns of social structure, explaining the impact of such patterns on behavior and attitudes.’’
2.4 Research Design Elements
Three elements of network research design shape the measurement and analysis strategies available to researchers: social settings, relational form and content, and level of data analysis. Every network data collection project must involve making explicit choices about these elements before beginning fieldwork. Varying combinations of them generate the wide range of social network investigations published in the research literatures of numerous disciplines.
Social Settings. The first steps in designing a network study are to choose the most relevant social setting and to decide which entities in that setting comprise the network entities. Ordered on a roughly increasing scale of size and complexity, a half-dozen basic units from which samples may be drawn include individual persons, groups (both formal and informal), complex formal organizations, classes and strata, communities, and nation-states. Some two-stage research designs involve a higher-level system within which lower-level entities comprise the actors. Common examples are hierarchical social settings such as corporations with employees, schools with pupils, hospitals with physicians, municipal agencies with civil servants, and universities with colleges with departments with professors.
The earliest and still most common network projects select small-scale social settings—classrooms, offices, factories, gangs, social clubs, schools, villages, artificially created laboratory groups—and treat their individual members as the actors whose relations comprise the networks for investigations. Recent examples include bullying and homophobic teasing among middle school students (Merrin, De La Haye, Espelage, Ewing, Tucker, Hoover, & Green, 2018), helping and gossip networks among employees of a Turkish retail clothing company (Erdogan, Bauer, & Walter, 2015), and the effects of ethnic diversity on the spread of word-of-mouth information in two matched rural Ugandan villages (Larson & Lewis, 2017). Small settings have considerable advantages in sharply delineated membership boundaries, completely identified populations, and usually researcher access by permission from a top authority. However, network analysis concepts and methods are readily applied to larger-scale formations, many of which have porous and fuzzy boundaries, including clandestine networks. Examples include peer network origins of adolescent dating behavior (Kreager, Molloy, Moody, & Feinberg, 2016), criminal organizations in communities of Calabria, Italy (Calderoni, Brunetto, & Piccardi, 2017), and strategic alliances among multinational corporations in the Global Information Sector (Knoke, 2009).
Relational
