Many new methods are defined and mobilized in this book to study the large networks outlined in Chapter 1. Their presentation is divided into two chapters, even though there is no rigid divide between them. We present basic network analytic ideas and methods in this chapter: they serve as foundations for the methods we use in Chapters 4–10. Methods used in these chapters are presented fully in Chapter 3. Some of the foundational methods presented in this chapter can be used for large networks. As Chapters 4–6 are explicitly citation networks, the break between this chapter and the next is marked by a sustained consideration of these networks in Chapter 3. Some methods specific to single chapters are provided there. The Chapters 2 and 3 are mostly based on lectures given by Vladimir Batagelj at the ECPR summer school, Ljubljana, 2006–2013.
Nearly all of the network analyses we present were performed in Pajek (Batagelj and Mrvar [PajekConn] ; de Nooy et al. [ESNA], a program suite1 designed to handle large networks very efficiently. Given our extensive use of Pajek, we include Pajek commands, where appropriate, both here and in other chapters. As R was also used extensively, we provide some R commands.2
A network is based on two sets: a set of vertices (nodes), representing the selected units or actors and a set of lines (links) representing ties between units. They determine a graph (defined formally below). If a line is ...