ABSTRACT
A network is any set of objects together with the relationships among them. The objects, which often are people, are called "nodes" and the relationships are the "edges" shown as lines or arrows in a network diagram. Types of relationships include attitudes (e.g., likes), behaviors (e.g., works with), flows (e.g., international trade), statistical relationships (e.g., correlates with), and many others. In the first "Quick Start" exercise, the Medici family network in medieval Italy is examined, using the "network" package to show which families were most central. In the second "Quick Start" example, the Marvel comic book here network is examined, using the "linkcom" package to highlight gender roles. Following these illustrations, interactive network analysis of a fictional research team is explored using the "visNetwork" package, to illustrate such concepts as directed vs. undirected networks, visualization of networks, and joining network analysis with cluster analysis of network groups. The "igraph" package is used to illustrate term adjacency networks based on gubernatorial websites during the covid pandemic. The "diagram" and "maps" packages are used to show how networks may be superimposed on geographic maps. The "statnet" and "network" packages are used to illustrate the identification of "network neighborhoods" and to perform cluster analysis. The "sna" package is used for related approaches called clique analysis and K-core analysis, applied to a network of relationships in the U.S. Department of Health and Human Services. Other applications are correlation networks with the "corrr" package and finding community clusters with the "tidygraph" package. A final section of this chapter covers agent-based network modeling with the "SchellingR", "RSiena", and "NetLogoR" packages.
