In network analysis a clique is deļ¬ned as a set of nodes in which each is connected to all the others (Scott 1991: 117). It is normally depicted on a diagram as a ācomplete graphā, a set of nodes all connected to each other. In Figure 5.14 the network includes two complete graphs, A and B, each of which constitutes a clique and, in addition, B as a four-point complete graph automatically encompasses two three-point complete graphs. Note that the graph depiction assumes that all links are based upon binary data: nodes are either connected or not connected. However, the links in the inter-city matrix derived above are variable and deļ¬ne a āvalued graphā. Thus in order to carry out a clique analysis a valued graph has to be converted into a binary graph: the values of links have to be dichotomized so that only values above a selected threshold constitute connections (ibid.: 113). Once this conversion is made, clique analysis can proceed.