A Comparative and Multi-level Analysis of Governance Networks: a Pilot Study of Employment Policy

While several empirical studies of governance networks exist (e.g. Marin and Mayntz 1991; Marsh and Rhodes 1992; Heffen et al. 2000) and an array of different theoretical approaches to network governance can be identified (e.g. Rhodes 1997; Scharpf 1997; March and Olsen 1995), the methodological question concerning how to study governance networks has received only scant attention thus far. To some extent, this reflects the general predicament within the qualitatively oriented parts of political science, where the post-positivist critique of epistemology appears to legitimize a general disregard for methodological questions. Hence, hermeneutic and social constructivist approaches have done a great job criticizing the positivist belief that correspondence with an unmediated reality and the application of strict methodological rules can guarantee scientific truth, but the abstract philosophy of science arguments tend to make people forget that we must continue to find ways of tackling the concrete methodological questions about how to produce relevant and plausible research results (Howarth and Torfing 2005). In any case, there are very few studies addressing the ‘nitty-gritty’ of conducting research from a post-positivist stance.