In this chapter, we will explore ways of analyzing complex networks with two or more different types of nodes. But first, as usual, we will lead off with a story.
“We have the best government that money can buy,” said Mark Twain.
Every April in Chicago’s Palmer House Hilton, political scientists square off at the Midwest Political Science Association conference (the top venue for discussing special-interest group politics and campaign finance) and debate whether Mark Twain was right, or whether the American system, in fact, is immune to the influence of money.
We entered this fray in 2006, bringing with us a large-scale social network study of campaign finance and its influence on electoral outcome. In this section, we’ll give an overview of this study, and delve into the methods used to derive the results.
But first, let’s play a little game.
Take a look at Figure 5-1. Every node on this chart is a political organization or political action committee (PAC) actively involved in the 2000 congressional and presidential elections. Red and blue nodes, respectively, are Republican and Democratic committees (national and state), green nodes are single-issue groups, purple nodes are industry associations, and yellow are non-profit organizations. The links between PACs are determined by where their money is spent—if PAC-A and PAC-B route donations to the same candidates, they become linked—and the more they have in common, ...