Exploring the composition of communication networks of governance – a case study on local environmental policy in Helsinki, Finland

Governance (Rhodes, 1996; Kooiman, 1993, 2003; Pierre, 2000) is one of the most popular new concepts in policy science and administrative science. The literature does not constitute a unified theory, but a single theme runs throughout: policy decisions are referred to as being made by networks of organizations. However, the network governance literature has not built on the previous literature on policy networks and the methods of social network analysis. The structure of the governance network can make a difference in policy‐making, but the structures have often been neglected in governance research. Here, the combination of social network analysis and governance literature is suggested as a possibility for investigating network structures. In particular, we investigate how and why the organizations involved choose their communication partners. The methods of exponential random graph modelling (sometimes referred to as p * models) (Snijders et al. , 2006; Robins et al. , 2007a, 2007b) enable the simultaneous modelling of structural effects and individual variables and their effects on network structure. As of yet, there are few substantial applications. The aim of this paper is to evaluate the composition of a policy network and the effects of network and actor variables on the empirically observed network. The environmental policy of the city of Helsinki, Finland, is used to demonstrate the approach.