Methods
The key methodological innovation of METRO lies in novel mix of quantitative Social Network Analysis with qualitative methods such as interviews and Critical Discourse Analysis. In our approach, the qualitative methods allow for contextualising, synthesising and developing the understanding of causal links within the networks of experts in transnational governance. As such our methodological approach allows for research which combines the explanatory power of in-depth qualitative inquiry with a rich exploration of the breadth of connections between actors.
This way we ensure enhanced triangulation of our data, especially data derived from documentation and interview analysis that is applied extensively for the explanation of networks, whilst, on the other hand, we use network data to ensure that we are interviewing all key relevant actors and examine their discourses and ideas.
In more detail, Social Network Analysis has been applied in order to examine the relational ties that link IOs and their actors. The investigation has so far focused on key meetings and events, the participants of which have been mapped in terms of their affiliations and connections. The investigation examines all the meetings that appear as crucial in the development of all four case studies. By creating groupings of actors (sub-nets), we are attempting to explore the degree of shared norms, ideas and common goals, as well as roles played within and between groups (leaders, followers, brokers) and test the extent to which behaviours and beliefs tend to become more similar or diverse within a ‘cohesive subgroup’ over time. The project tries to develop networks of all relevant actors in order to test the degree of centrality, namely locate the actors that appear as central to the network, and those marginal ones – we have so far completed this work in the cases of the SDG4 and Statistical Capacity Development.
As expected, we were not given unlimited access to unravel cohesive and established IOs’ expert communities. More realistically, what METRO has achieved is to explore bridges and brokers in the interconnectedness of multiple policy communities or networks. Crucially to METRO, ‘mediation
takes place on an intellectual plane, involving the “translating” of policy ideas and policy paradigms between policy communities’ (Coleman and Perl 1999; 707). Rather than exhausting the analysis of one specific network of actors, we are looking at the new world vision mediators
create with shared language and norms in order for disparate data agents to come together in the creation of new measurement agendas.
Critical discourse analysis has been applied in all the key documents of each of the four case studies; this is work that is primarily done before commencing with the SNA analysis and interviews. We first familiarise ourselves with the general parameters of the case, before diving into more in-depth analysis of the texts which appear as crucial (e.g., those produced after key events or those that are most referenced). Thus, in interviews we can query actors in relation to specific documents and discourses, as well as inform ourselves about whether there may be other key documents that need analysis. We have conducted at least 60 in-depth, semi-structured interviews in total so far. SNA and interviewing take place in parallel, and largely inform one another. Methodologically, we have been reflexive about our interviewing practice and we have explored some of its dimensions in the chapter ‘Researching Education Elites, 20 years on: sex, lies and…Skype’, (PDF) to be published in the edited collection (N. Piattoeva and C.Addey eds.) The Practice of Method: Intimate Accounts of Researching.