Abstract
The European Commission designs and promotes policy networks in almost all policy fields. Knowledge about networks is necessary for a better understanding of the overall policy-making process in multilevel governance systems like the European Union (EU). Studying EU networks helps to explain how policies emerge, how they are framed and processed, why they take the character they do, and how they might contribute to understanding the course of European integration.
The political system of the EU is characterized by the coexistence and interdependence of formal and informal networks that are part of its multilevel governance. While the interplay of formal and informal actor’s networks has been researched to some extent (Klüver, 2014; Mahoney, 2008), informal policy networks have been paid less attention in EU integration theory (Christiansen and Piattoni, 2004)—one reason being that rules for processes, membership, and responsibilities are often not clearly defined and are not readily detectable from the outside. Informal policy networks may include expert groups, civil society organizations (CSOs) such as lobby groups, social partners, or companies that argue about policy content also discussed in the formal policy-making process.
Researchers used social network analysis (SNA) as an essential technique to investigate such complex networks among actors, be they individuals, small groups, or international organizations (Carrington et al., 2005; Jansen, 2006; Wasserman and Faust, 1994). SNA allows investigating a broad range of professional and personal networks and their quantitative aspects such as group size, the number, and status of contacts.
A common challenge in network analysis is the lack of information on other elements of social life like inter-agency and process-related aspects (Crossley, 2010). Qualitative network analysis (QNA) aims addressing the shortcomings of SNA by applying a micro-perspective instead of a macro-perspective and qualitative instead of quantitative methods, and taking an insider instead of an outsider view. A qualitative approach to networks prompts a focus on personal and social networks, for instance, friendships, family relations, or migration ties. Until now, only a few attempts have been made to apply QNA to policy networks (Baumgarten and Lahusen, 2006; Schiffer and Hauck, 2010; Schindler, 2006), and none of them took EU policy-making as an example.
I contend that QNA enables visualizing where and which formal and informal networks exist in everyday working situations; it also allows to detect whether and why networks are stable and what kind of meaning policy-makers attribute to them. QNA is founded on the same basic assumptions as SNA and has become a more widely used research method during the last decade. It provides a fresh approach to gathering information about qualitative aspects of (policy) networks, which are not available through SNA (Hollstein and Straus, 2006).
In this article, I demonstrate how to employ QNA to investigating supranational EU policy networks. The EU is an important case because participatory democracy is the postulated normative ideal with civil society as a key actor. How civil society is present in policy networks is a question of “throughput legitimacy,” that is “governance processes with the people, analyzed regarding their efficacy, accountability, transparency, inclusiveness, and openness” (Schmidt, 2013: 2). I claim that QNA allows understanding not only quantity and formation but also the quality of networks. I investigated the particular case of (gender) equality CSOs because they are often poorly resourced, and in their strategies to build strong ties with EU institutions, they risk “competition between inequalities” (Verloo, 2006: 211).
The article serves the search for innovative methodological techniques in investigating formal and informal supranational policy networks. QNA is a useful approach to reveal critical network characteristics and to visualize policy networks regarding different aspects like differing formal and informal compositions. I exemplify with two case studies from EU equality policy the appropriateness of QNA as a method of inquiry. First, I briefly discuss SNA and EU policy networks, and summarize findings for the case of equality policy networks. Then, I justify why I chose QNA as an alternative methodology, how I prepared its application, and how I analyzed data. Next, I provide details on practical implications of using QNA in the research process before finally discussing opportunities and challenges when using QNA.
Researching policy networks 1 in the EU
Network analysis became “organizing Babylon” (Straßheim, 2011: 31): definitions are diffuse and often general information and knowledge are seldom well distinguished, and the role of policy learning is ignored. Policy networks can be analyzed in several ways, and each research question requires a suitable methodology. Most of the research hypotheses on (policy) networks center around either transformation (i.e. networks contribute to societal change, organizational deepening, the creation of knowledge, or network society) or problem solving (i.e. networks deliver new and better potentials of problem solving and coordination) (Straßheim, 2011).
SNA has become the key technique of network analysis in sociology and political sciences and was used for a broad variety of research questions, including social movements’ formation, and formal and informal networks inside or between institutions (Carrington et al., 2005; Hollstein and Straus, 2006). SNA is a useful approach to investigate form and structure of a broad variety of networks, not only in politics. It also offers well-developed formula to quantify and visualize various policy networks and to turn them into clear, abstract, and comparable network maps in a way necessitated by research projects.
Research on formal and informal EU policy networks is rich (Beyers et al., 2008; Falkner, 2000; Klüver, 2014; Mahoney, 2008), with SNA as a major method (Scott and Carrington, 2011). Although considerable research has been devoted to describing the emergence, establishment, aims, and influence of EU policy networks, we do not know much about how civil servants perceive their ties in everyday working life, in their “social world” (Hollstein, 2011). So what is often missing in SNA is the possibility to investigate content, meaning, and reasons for interaction in policy networks (Crossley, 2010). Consequently, understanding the relationship between agency and structure—in the EU context the process of policy-making through networks—demands to bring the social world, that is, meaning and individuals back into the research process (Crossley, 2010; Fuhse and Mützel, 2011). Policy networks are also social networks and thereby an interactive “social world” comprising shared meanings, knowledge, norms, identities, and the like, as well as the distribution of resources (Hall, 1987; Strauss, 1973).
Studying EU policy networks from the perspective of meaning and norms was an aspect of studies on transnational social movements (Lang, 2013; Ruzza, 2004), interest groups (Greenwood, 2007), advocacy coalitions (Sabatier, 1998; Sabatier and Jenkins-Smith, 1993), or epistemic communities (Cross, 2011; Haas, 1992). These studies have proven pivotal in understanding how interest representation has functioned in the EU, in particular, the role of civil society in EU policy-making.
The EU itself has promoted and set up policy networks in a range of policy fields, and (gender) equality policies offer a particularly rich case of study. Next to the establishment and creation of new policy networks in different issue areas (Locher, 2007; Montoya, 2008, 2009; Zippel, 2004), research on women’s or gender equality policy networks has focused on participation possibilities (Sperling and Bretherton, 1996), network ties of women’s movements, and their representation on the web (Lang, 2009; Prudovska and Ferree, 2004). Keck and Sikkink (1998) used process tracing for examining transnational advocacy networks (TANs) around violence against women and human trafficking. The analysis of these TANs helped us understand the role of principled ideas, windows of opportunities, and international ties between civil society, states, and international organizations. The power to influence governing actors through networks has become famous through Woodward’s (2004) notion of “velvet triangles” consisting of civil servants, parliamentarians, and academics creating their own rules to advance gender equality policy. Hubert and Stratigaki (2016), furthermore, illustrated how these networks produced a common understanding of what gender equality means and collaborated to push political strategies forward.
Despite the comprehensive research on equality policy networks, a gap remains: studies tend to focus less on the crucial role played by individuals in setting up, maintaining, and changing these networks. Recently, sociological approaches to the EU have returned to such an actor-centered research, because [t]he EU does not do anything by itself; it is people as everyday political agents who make the EU happen. To understand the EU as a distinctive form of social organization and power structure, its influence and the effects of its politics, one has get inside the politics to know who the individuals and groups making up the EU are, where they come from, what kinds of resources and networks they have access to, how they perceive their roles, the institutions in which they work and, more broadly, the social world around them. (Kauppi, 2011: 150–151)
As will be shown in the remainder of this article, QNA offers fruitful methodological tools beyond the ones existing so far that serve the aspiration of bringing actors back in.
Investigating policy networks using QNA
SNA and other EU policy network research employed a macro-perspective focusing on institutions. When we want to understand policy-making from the perspective of the individual, we need to use other methods. For the individual, it makes a difference if they see someone else as belonging to a formal policy network, to a social movement, or to an advocacy coalition. The perception of others also influences self-perception and in effect the overall policy process.
I argue that taking a micro-perspective of individual actors helps explaining better how policy processes and goals develop. If individuals in gender equality policy, for instance, would perceive their policy field as opposed to trade or internal market (Jacquot, 2015), those actors responsible for gender equality will be seen as possible allies. If the same actors would see their position as competing with others within gender equality policy, they think not of allies but probably of rivals. Consequently, the question occurs: “What dominates everyday work and how actors interpret their ties within policy-making?” “What kinds of actors are recognized?” “Whom do they contact how often and what does this tell us about policy-making?”
Trying to answer these questions led me to QNA as a methodological approach specifically designed to investigate networks from a qualitative perspective. QNA established itself next to SNA and utilized qualitative research methods (Hollstein and Straus, 2006). QNA stipulates that networks depend on the actions of individuals and that these are essential to any aspect of networks. Simultaneously, networks may enable as well as constrain individuals in their actions. SNA—when analyzing and visualizing networks between institutions—tends to disregard individuals and their role in influencing the network and anything belonging to it. QNA starts exactly here, with the individual actor, and aims to tease out the subjective interpretation of networks. 2
Literature shows that QNA is a useful technique to shed light on personal perceptions of reality and relationships individuals develop (Hollstein and Straus, 2006). QNA tools allow examining what agency means in a certain context, how policy processes and actors are connected, and which dynamics occur in (social) networks (Hollstein, 2011). Here, QNA aims to analyze everyday communication and actions, their formal and informal content, and the role time constraints may play. Few studies using QNA engage in research about institutions or the relationship between individuals within transnational institutions (Hollstein, 2011). Instead, private networks of individuals (e.g. transnational migration–related family networks, biographical networks) have been the predominant focus of qualitative network research. Although some have studied policy networks (Baumgarten and Lahusen, 2006; Schiffer and Hauck, 2010; Schindler, 2006), there exists no elaborated concept of how to apply QNA to supranational policy networks.
QNA makes use of common qualitative research strategies, like interviewing, observing, and analyzing documents and archival material. Important tools of QNA are network maps. Unstructured maps are “free-style” drawings where interviewees simply receive a piece of paper to develop their perspective on a certain question, for instance, for illustrating private and occupational interactions of individuals. The primary function is to generate narratives on personal systems of relevance and meaning (Hollstein and Pfeffer, 2010). The second category of network maps is structured ones, either standardized or non-standardized. The best-known tool is the “hierarchical mapping technique” by Kahn and Antonucci (1980), also known as the “method of concentric circles.” Interviewees receive a piece of paper with a limited number of concentric circles. The standardized version includes a fixed definition of the circles or sectors of circles, for example, family, job, and friends. Therefore, network maps of different persons are highly comparable. The non-standardized version of concentric circles does not define the meaning of the circles. It is a tool used during interviews simultaneously as a medium of communication and a result of the interview (Hollstein and Pfeffer, 2010). Non-standardized versions of concentric circles are partly comparable on an inter- and intrapersonal level. The network maps produced by using concentric circles offer the chance for interviewees to decide what the final maps look like from a merely visual aspect. 3 They offered the chance to discuss them in the context of the overall EU gender equality policy process. In the following section, I will illustrate how I combined QNA and the method of concentric circles with expert interviews.
Combining the method of concentric circles and expert interviews
I have used QNA in different research projects. I studied how actors cooperated formally and informally in developing the EU policy program “A roadmap for equality between women and men 2006-2010” (Ahrens, 2018). Then, QNA was used in research on equality policy networks among nongovernmental organizations and finally in an ongoing research project studying the interface of CSOs and EU institutions as a basis for supranational participatory democracy. In the latter, I focus on CSOs dedicated to one or more grounds of discrimination covered by Article 19, Lisbon Treaty, 4 and additionally, CSOs committed to social rights in general. I used QNA and its method of concentric circles to explore and visualize policy networks and estimate changes in network collaborations. In the first project, I focused on formal and informal network ties, whereas in the others I looked at the frequency of network contacts. Overall, I investigated EU (gender) equality policy horizontally (among actors) and vertically (within institutions).
For each study, the method of concentric circles was embedded in semi-structured expert interviews. Expert interviews imply a relational characterization of who can be called an expert for what reason; “experts” are an “analytical” construction 5 (Bogner and Menz, 2005). A criterion to be signified as an expert was the actual involvement in EU (gender) equality policy. This criterion was fulfilled, when persons were responsible by definition, so-called gatekeepers, such as gender focal points of the European Commission, members of the European Parliament’s (MEPs) Committee on Women’s Rights and Gender Equality, or representatives from CSOs. I also classified persons mentioned by other interviewees as experts. I used a semi-structured interview guide starting with simple questions about job duties and daily routines aiming to generate a good narrative flow before the method of concentric circles came into play. 6 In the first research project, I used a structured non-standardized network map with concentric circles, while a structured, standardized network map was used in the other research projects. 7
Practical implementation of QNA and expert interviews
Expert interviews on a supranational level are specific in the sense that interviewees are aware that they are “elites” and used to “Euro-speak” (Diez, 2001), to speak in public and also to defend their institution. MEPs as well as Commission civil servants are used to interviews, and CSO representatives and social partners often tend to give interviews as a media strategy. My most likely interviewees were used to interviews, they probably heard questions on a related subject, and they were prepared to answer inconvenient questions. They developed individual narratives during interviews, and these narratives allow (re)constructing institutional logics, norms, and rules.
In my research projects, the method of concentric circles functioned as a tool to break through the typical standard expert interviews. When invited, I did not mention designing network maps as part of the interview. I wanted to use the opportunity to derive original information from the interview instead of letting interviewees prepare a list in advance. The network maps were a surprise element, planned to back up the narrative by transferring it into a product, the network map. Of course, every network map was a product of the moment and might have looked different another day. Still, the intention was to find out what kind of narrative interviewees are telling about EU policy-making and their policy networks. The method of concentric circles placed within an expert interview helps the interviewer as well as the interviewee to focus and narrow down specific information about actor’s constellations. This narrowing down is neither possible with traditional SNA nor with traditional expert interviews.
Since the network map drawing was core to the interview, I prepared them very carefully to ensure a smooth implementation. I used an A4 paper with empty concentric circles (cf. supplemental material online) and custom-labeled stickers to design the network maps. The equipment was chosen for the following reasons:
The equipment guarantees both participants to see the map even if not sitting next to each other, for instance, in case a big table hampers spatial proximity;
Flexible stickers enabled to re-organize the maps in case the interviewee changed their mind on the placement of a certain label;
Customized labels were essential for harmonizing actors’ names across interviews and keeping them also readable for the interviewer.
I allocated the network map creation in the middle of the interview because I expected that interviewees would trust me by then. The placement was crucial: in the beginning, interviewees needed to become familiar with the interview situation; at the end, interviewees might have been too tired. Also, I expected the map to function as a “break” during an interview of often 1 to 2 hours.
In the first research project, interviewees were allowed to freely interpret the ranking of the circles and where to place the labels for actors. Contrary to practices of SNA, I never provided a list of actors to avoid imposing an outsider perspective and I wanted to keep the network map within the reference frame of the interviewee. All interviewees were asked several times if the picture is complete or if any actors from within the institution, from EU institutions, CSOs, or other societal fields were missing.
Analyzing qualitative network maps
As explicated earlier, I embedded creating network maps in expert interviews. The combination fulfilled two functions in generating data with the map: first, the network map as a product, and second, as a means of generating additional and sometimes at first sight inconsistent interview data. For the additional data, researchers need to be aware that interviewees develop narratives or “standard stories” (Tilly, 2002) that can be analyzed regarding similarities, differences, and narrative patterns.
As regards the EU policy process, narratives are usually coherently interwoven with a certain institutional logic and its norms and rules. This also applied to my interviews, but often only until I asked to create the network maps. Unexpectedly, the network maps often supplied a different picture of relevance than the narratives. Without claiming that I systematically planned this as a step in the interview process, the network map creation resulted in a kind of immediate comparison. I sometimes was able to refer back to earlier descriptions that contradicted the network map, and by confronting the interviewee, I gathered additional information about the content or nature of contacts, in other words, the meaning of interaction as seen from point of view of the interviewee. Since the expert interviews were conceptualized with view to their theory-generating function, they concerned individual reasons for actions and their implicit basis of decision-making, in other words, the “subjective dimension” in policy-making (Bogner and Menz, 2005: 38).
The double function of network maps allows for different levels of analysis related to the network maps as products and the context of the interview narrative. Without a sufficient knowledge about the context—in this case, EU policy-making and the role of different actors—data collection and their analysis are hardly feasible. In the following, I will address the two functions separately even though the analysis is undoubtedly a hermeneutic approach in an iterative process that requires contrasting continuously the levels.
Analyzing network maps as output
In a first step, I examined the map of each interviewee regarding the contacts mentioned for the different categories. For instance, in the first research project, I looked at informal and formal contacts, what kind of contacts exist (organizations, individuals), and whether they match with the political system. Comparing the formal and informal network maps provided a better understanding of how interviewees ranked their contacts regarding closeness/distance and who was seen as most important or necessary in the policy process. By inspecting the network maps carefully, it also becomes possible to detect unexpected gaps or unexpected names, an aspect that required matching the network map more intensively with the interview narrative. Network maps in the first and second project often led to actors not previously mentioned in the narrative by diversifying contacts and explaining the differences in the nature of interactions. For example, political parties in the EP translated into a variety of contacts that were connected to the group (party members, rapporteurs, working groups, secretariat).
In a second step, I compared the network maps between interviewees. Here, network maps were analyzed with regard to which actors played a major role across all network maps and which appeared seldom. Furthermore, I examined how actors related to each other, this means, were actors mentioned reciprocally and to the same extent/on the same level or were the maps contradictory in this regard. In the first project, the network maps on formal policy-making from EU institutions showed a striking similarity and mirrored exactly the formal rules and procedures provided in the treaties and organizational rules.
In the ongoing research project, I started from the premise that we should expect no hierarchy between CSOs organized around different grounds of discrimination. Against this background, I compared the network maps, and this step in the analysis revealed that there exist major differences between identity-based CSOs and umbrella CSOs that cover more than one ground of discrimination with the latter contacted more frequently and also closer to EU institutions (Ahrens, 2019 [forthcoming]).
Analyzing network maps in the context of interviews
While the above-mentioned steps relied mainly on the network maps themselves, examining the network maps in the logic of the interview narrative provided particularly fruitful results. The combination of map and narrative allowed for a “thick description” of policy networks, based on the narratives during the production of the network maps and summaries about how interviewees describe each aspect of interaction—content, nature, and meaning. As a matter of course, this becomes easier if interviewers pose questions related to these aspects. The interviews from the different research projects showed, however, that interviewees often anyway (unconsciously) started explaining reasons for contacting others without even being asked directly. Furthermore, they often also mentioned how they perceive the contact; as friendly, complicated, consolidated, important, reciprocal, and so on. Certainly, the reliability of the results can be further consolidated by triangulating with another kind of data, derived, for instance, from a documentary analysis, SNA, twitter analysis, or website analysis.
Opportunities and challenges of QNA
QNA is not the first methodology that researchers come across when thinking of research on EU policy networks, however, as the analysis of the application shows, QNA is worthy thought. QNA is easy to prepare and easy to combine with other qualitative methods, it delivers additional information for analysis, and it covers micro-perspectives on actors that are usually investigated separately. I present the results from using QNA in the next section and discuss the consequences for the research projects and future research.
In the research projects at hand, SNA would not have been sufficient, because it produces formula-based network maps that are then turned into abstract, simplified, and highly comparable maps. In fact, the researcher designs—by using software and a certain calculation—the appearance of the network maps. Unlike SNA, QNA produces visualizations of networks as they appear in interviewees’ heads, so to say the network maps replicate the social world of the research subjects.
QNA opportunities
On the practical side, today’s computer equipment eases the production and adaptation of network maps to various settings. In my interviews, the sheer practice of designing network maps produced the most interesting reactions during interviews. Most of the interviewees screened quite skeptically the equipment before almost always immersing into the practical task. While interviewees tend to produce narratives during semi-structured interviews, concentric circles were a tool to make them focus on specific questions and to transform their most linear reporting structure into a visual overview that was still comprehensive. Some of the interviewees in the ongoing research project even asked permission to copy the network map for planning their political strategies.
Therefore, QNA combined with semi-structured interviews proved particularly fruitful in generating new data and additional aspects of policy-making. Nevertheless, the placement of the network map had to be considered well. In my research projects, it would not have helped to start with the network map, because the main focus of the interview was understanding supranational policy-making and not producing the map. Equally, creating the network map at the end of the interview because interview ended with concluding questions and (re)starting with contacts could have been counterproductive.
QNA allowed me tailoring questions regarding the type of network I wanted to comprehend alongside with the broader research question on policy-making or participatory democracy. The network maps of the first research project included sometimes actors that are “non-actors” such as the interservice consultation. Interviewees, who mentioned it, spoke of the interservice consultation as if it would be an actor. Indeed, it is an automated submission system in the policy process and not connected with any specific individual. If I had only asked about actors in the sense of organizations or individuals, interviewees probably would not have mentioned the interservice consultation. But for the interviewees concerned, the interservice consultation was a necessary, formal contact in the policy process. In other words, it is impossible to “talk” to the interservice consultation, but contacting it is unavoidable for EU civil servants. Their thinking combines actors and policy process; for them, the two are undividable. With SNA or expert interviews alone, I would not have received any information about this contact.
QNA was also useful to collect information about the
Although the original research approach did not directly point to another type of networks, it was possible to gather additional information on the following aspects:
Without a doubt, the different types—actors, ties, the level of governance, policy goals—overlap and are not distinctive. Importantly, interviewees provided additional information they would probably have never revealed in a standard interview situation. This concerned especially whom they are talking to and to whom not. By using QNA, they were motivated to mention those they have in mind; vice versa this means, if an actor was not mentioned, this actor might not be important to the interviewee. When comparing the network maps, this became a surprising finding, as reciprocal recognition was not the norm but limited to some core actors while particularly smaller CSOs were not acknowledged.
Likewise, when comparing the network maps with the formal policy process of, for instance, adopting an EU directive, unmentioned formal actors became apparent and were granted with specific attention in the analysis. Comparing the subjective network maps with the list of formal policy-making actors provided further insights into discrepancies between official and actor-perceived images of the policy process—a finding that explained why policies shift to one or another direction (Ahrens, 2018).
QNA challenges
Despite the illustrated opportunities, QNA also contains challenges linked to qualitative research more generally, and in particular when using structured, non-standardized concentric circles. In the former case, the interviewer cannot ascertain if interviewees forget someone when providing information about whom they talked to during a certain period. In the latter case, it becomes difficult to analyze how they relate actors to each other when the interviewer forgets to ask more detailed questions. Hence, a good knowledge about formal and informal EU policy-making processes is necessary to be able to challenge interviewees on bold and simplified descriptions. In addition, researchers might also face problems with different scenarios of actors mentioned: some interviewees might tend to exaggerate their connections while others might try to downplay them. Interviewees need to be attentive to these possibilities and can try to tease out more details during the rest of the interview. Furthermore, triangulation with additional qualitative or quantitative data derived, for instance, through SNA or website analysis or any other dataset appropriate for the research question at hand may help to compensate these possible pitfalls of QNA.
In my research projects, very few interviewees placed lines and arrows to show distinctive directions of connections among their contacts, but this was not systematic. Interviewees might mention a range of other actors, but without well-prepared questions, the analysis will be hard when aiming to analyze if they are all the same to the interviewee or whether they represent different relations. These aspects must be collected using the other interview questions. Also, time restrictions in interviews sometimes lead to a trade-off between creating network maps and asking questions on additional topics.
The research projects also pointed to the necessity of introducing interviewees better to this way of thinking. Starting with such a network map drawing is not appropriate without putting it into a broader context. By doing this, limitations appear regarding the possibility of asking details about power relations and maybe conflicting policy goals of actors. As an alternative, these nuances can be derived from the overall interview content.
Conclusion
In this article, I demonstrated how QNA and its tool of concentric circles contributed to gathering micro-perspective insider descriptions of EU (gender) equality policy networks. QNA shed light on policy network formations that would not have been available through traditional methods like SNA or classic social movement and policy network research. QNA offered the chance to understand better which everyday working logics and contacts existed and what kind of policy network configurations originate from that. By taking an insider view and a micro-perspective seriously, new aspects of policy networks such as a limited number of actors actively involved in formal/informal policy processes or likewise the role of administrative obligations in the policy process can be disguised more easily and complement thereby other methods of network analysis.
QNA proved to be a useful concept to support interviews on complex questions in multilevel governance of the EU. Nevertheless, limitations exist because the relationships between actors mentioned by interviewees often remain vague if not addressed specifically. Using the maps, I derived a picture of vice versa connections that displayed the actors’ perceptions of EU networks. Because no list of actors and institutions was presented, there existed no insurance that all interviewees kept in mind the same scope of possible actors. It was also not fully possible to compare closeness or distance of actors attributed via the placement of labels. In summary, the method of concentric circles needs to be specifically combined with the questionnaire through a clear definition of circles.
Also, further limits endure which originate in the design of the method. It is almost impossible to produce objective, comparable one-to-one maps because the idea is to leave the definition and introduction of network maps to interviewees. As a consequence, the placement of the map has to be chosen well, and data derived have to be contextualized within the whole interview narrative. Certain circumstances like time pressure or hostile interviewees might further hamper the possibility of employing QNA at all, and it becomes a trade-off between producing a network map and answering further parts of the questionnaire. Future research needs a plan ready, what comes first under which conditions.
On the other hand, such considerations might overburden the method and could lead to more confusion than a clarification of interviewees’ views. Here, it could be useful to combine QNA with SNA to count and measure the overall number of possible actors, and calculate centrality or periphery of certain actors. Thereby, SNA induces a great deal of information about a summarized picture and offers good chances to compare policy fields or political entities like institutions across states. In sum, QNA allows gaining new insights about supranational EU policy-making that would not be available via any other research approach.
Supplementary Material
Supplementary Material, MIO_QNA_Ahrens_online_material_2017 – Qualitative network analysis: A useful tool for investigating policy networks in transnational settings?
Supplementary Material, MIO_QNA_Ahrens_online_material_2017 for Qualitative network analysis: A useful tool for investigating policy networks in transnational settings? by Petra Ahrens in Methodological Innovations
Footnotes
Declaration of conflicting interests
Funding
Supplementary material
ORCID iD
Author Biography
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
