Globalization,Policy Mobilities,and the Method of Network Ethnography

Over the past three decades, mobilities—the study of the movement of ideas, people, capital, and things—has received significant attention from scholars in sociology and human geography, in what Urry (2003) called the “mobilities turn” and Sheller and Urry (2006) later termed the “new mobilities paradigm.” As Roy (2012, p. 35) put it, this is a shift in emphasis “from the study of mobile policy to the study of the practices through which policy is made mobile.” At the heart of the new mobility studies, notes Merriman (2020, p. 142), lies the argument that “mobility should not simply be reduced to physical movement or displacement. Rather, actual and virtual mobilities, and physical and imagined mobilities, are produced, gather meanings, and are caught up with a complex web of social and political relations, displaying uneven geographies of power.” As a specific body of research, new mobility studies have explored the embodiment and experience of mobility, its governance, and policing. This includes studies on the movement of people under different circumstances, from the homeless to tourists and corporate elites; the policy frameworks and programs that regulate their movement, such as border control and restrictions and control systems for refugees and undocumented migrants; the movement of money across sectors; and the movement of “things,” ranging from waste disposal, to low-cost carriers, to organ transport. In trying to grapple with these complex issues, key figures in the field have brought attention to the theoretical and methodological interdependence of mobility and mooring (Urry, 2003) and fixity and flow (McCann, 2011), warning against “the risk of obfuscating forms of stasis, immobility, and slowness” (Savage et al., 2021, p. 314) in discussing mobility.

With an initial focus on the urban, policy mobility research brings a relational approach to understanding the geography of the city, as “nodes in relational networks linking them to other nodes and across which various material and immaterial objects circulate” (Prince, 2020, p. 186). This emphasis on circulation and anchoring of policy between nodes suggests the existence of a global policy network. A primary aim of mobility-focused research, note Savage et al. (2021, p. 132), is “mapping, understanding and critically assessing the work of global ‘policy elites’ who inhabit positions of power and influence that allow these actors to exert significant influence over policy reforms and processes.” While the notion of global policy “might bring to mind the image of an elite world of the rich and powerful who meet in exclusive spaces where, despite being out of the reach of ordinary citizens, the policies to be foisted upon those ordinary citizens will be decided,” a policy mobility approach “grounds the notion of global policy, encouraging us to think about policy not necessarily as international and so operating in a sphere beyond the national, but as trans local, and so necessarily embedded in different localities that are nevertheless connected up in various and often multiple ways” (Prince, 2020, p. 187). This tension between the national and the global brings attention to the role of the state, sometimes as the vehicle of global policy penetration through commissioners, contractors, and regulators rather than as the party responsible for service delivery, and others fencing policy design against unfettered markets. ¹ Mobilities, notes Lewis (2021, p. 328) “provide a new way of thinking about existing spaces and scales, such as the role of the state in shaping the mobilization of policy, as well as helping to complicate territorial, state-centric accounts of policymaking.” In previous work (Junemann et al., 2015), we have argued for the need to move away from a structured and static community analysis to an attempt to understand mobility and the practices and mechanisms involved in the flows of policy. Policy networks are forms of socio-economic relationship involving affect and exchange, and epistemic-discursive communities joining up multiple, distant “sites” through fluid and evolving forms of financial, discursive, political, technological, and social exchanges. As noted elsewhere (Ball et al., 2017, p. 16), policy mobility research could be informed by the following questions: What spaces do policies travel through on the way from one place to another? Who is active in those spaces and who moves between them? How is space/are spaces reconfigured as policies move through it/them, and how are policies changed as they move? In trying to understand the “complex web of social and political relations” (Merriman, 2020, p. 142) that underpins policy mobility, another important question is, why do some “policy solutions” seem to travel further and faster than others (McCann, 2011; Prince, 2020)?

In trying to contribute to the need for nuanced and adaptive research methods to study the present and futures of education policy, this Special Issue focuses on the uses and extension of the method of network ethnography as a flexible and evolving method that brings ethnographic sensibilities to bear on the portrayal and analysis of the complex and fluid relations that constitute education policy networks (Ball et al., 2017). Peck defines networks as an architecture of relations consisting of “girders,” “pipes,” or of “circulatory systems that connect and interpenetrate” the local and the national (Peck, 2003, p. 229). As a method to further understand the structure, dynamics, and practices within policy networks, network ethnography involves close attention to organizations and actors within a field, to the chains, paths, and connections that join these actors, and to situations and events in which policy ideas are mobilized and assembled. It concerns the work and labor of policy actors, the meaning-making and -reshaping that occur through exchanges in social relations, and the policy discourse shared and enhanced in and by policy networks.

In its call for papers, this Special Issue invited contributors to deploy network ethnography to study education policy and policy mobilities across a range of policy spaces, with varying shapes of policy networks, different relations between the local and global, and particular policy windows for change and reform in education. We provided a series of questions as starting points, or perhaps provocations, for papers that would push the boundaries of the method of network ethnography:

Structure and dynamics of the global education policy field: What policy technologies do key actors use to shape convergent policy practices and discourses in different countries? What policy models are constructed and mobilized to different spaces? How do state actors recognize and respond to the influences of the global education policy field?

Divergences of policy networks in traditionally different policy spaces: How do policy networks work in settings with different cultures and traditions in education policy making? How do we understand convergences and divergences of policy networks in shaping education reforms and practices in different spaces? How should researchers deploy network ethnography differently according to the specifics of the research context?

Politics within policy networks and possibilities of reflexivity and resistance: Although policy networks involve various forms of policy actors, only a reduced group of actors shape the dominant policy discourse. How can we understand the politics within policy networks? Whose voices become dominant and whose are marginalized? Are there possibilities for the marginalized to form actions of resistance and introduce change in policy networks?

The remainder of this introductory paper is structured as follows: First, we offer an overview of the method of network ethnography, with particular attention to the tensions and contradictions experienced by researchers as they “follow” policy. In the final section we offer a succinct account of the papers in this Special Issue and their contribution to the field.

Network ethnography: What do we know so far?

Börzel (1998, p. 253) describes policy networks as “a set of relatively stable relationships which are of non-hierarchical and interdependent nature, linking a variety of actors who share common interests with regard to a policy and who exchange resources to pursue these shared interests acknowledging that cooperation is the best way to achieve common goals.” There are two key elements in social networks: the “nodes” (which can be individuals, organizations, or even subject positions), and the “ties” or links between them. Rather than focusing on “individual attributes,” social network analysis is a method for studying “social relations” (Burt, 1978). However, the “lines” in a network diagram do not always represent the quality of those relations. The challenge, Crow (2004) warns, is to identify what “passes” through networks. That is, schemes, programs, artifacts, techniques, and technologies move through these network relations; indeed, they move at some speed, gaining credibility, support, and funding as they go, mutating and adapting to local conditions at the same time.

As noted elsewhere (Ball et al., 2017), network ethnography shares the fundamental principles of ethnography as a tradition committed to the search for subjectivity and meaning, a suspension of preconceptions and an orientation to discovery (Ball, 1990). This entails that (i) some form of immersion occurs in order to gain a deep understanding of a group's shared culture/values; (ii) studies may be designed and redesigned; (iii) the research is concerned with social processes and meaning; and (iv) data collection and analysis occur simultaneously. There are different sorts of data involved in network ethnography, along with a combination of techniques of data gathering and elicitation. Network ethnography requires deep and extensive internet searches (focused on actors, organizations, events, and their connections). There is a large body of material available online (newsletters, press releases, videos, podcasts, webinars, interviews, speeches and web pages, as well as social media such as Facebook, Twitter, Instagram, LinkedIn, and blogs) that can be identified and analyzed as data in policy research. Network ethnography also involves participating in some of the key occasions where the network participants under consideration come together. As Cook and Ward (2012, p. 139) put it, conferences “continue to be important in creating the conditions under which policy mobility may or may not take place.” Conferences and other events (both face-to-face and online) are moments when both bonding and bridging ties are forged and renewed (Putnam, 2000).

In previous work (Ball et al., 2017; Junemann et al., 2015; Santori et al., 2015), we provide an account of a global education policy network at the level of compulsory schooling and found that many of the same actors and businesses or other organizations recurred. Organizations such as Pearson, McKinsey, Absolute Return for Kids (ARK), Michael and Susan Dell Foundation (MSDF), Omidyar Network, Bridge International Academies, Knowledge Is Power Program (KIPP), Rocketship Education, Kahn Academy, Global Education Management Systems (GEMS), and Centre for Education Innovations, and nodal actors such as Ashish Dhawan, James Tooley, Sharath Jeevan, Jay Kimmelman, Shaheen Mistri, Bob Pattillo, Michael Barber, Ronald Cohen, Irene Pritzker and others appeared time and time again, in different places, activities, and relationships. This level of data saturation became apparent through the process of data collection:

[A]s we systematically followed the links and connections, many of the same actors and organizations recurred; as we looked at events, these organizations and actors came together face to face; as we examined specific initiatives, these organizations and actors were involved; as we explored financial arrangements and exchanges, these organizations and actors were ‘investing’ or being invested in. (Ball et al., 2017, pp. 7–9)

These active actors in the “global education policy field” construct and disseminate particular forms of policy as “best practice” and “policy models” (Peck & Theodore, 2010), including but not exclusively public–private partnerships, charter schools and academies, focus on return on investment, and heavy reliance on standardized assessments and evaluations to measure progress. But what can we learn from existing research about the affordances of network ethnography as a method to study the movement of ideas, money, things, and people? The documented practice of “doing” network ethnography provides valuable insights into the challenges and opportunities of network ethnographic work, which we discuss below.

An unsettling method

Avelar et al. (2022) provide a vivid account of the practice of “doing” network ethnography, reflecting on their identities and positionalities as researchers exploring policy networks through network ethnography: “We note the games we play, the personas we adopt and the lines of ethical morality we draw” (p. 108), as well as the uncertainties and uneasiness they engage with in their methodological practice. One critical aspect of “doing” network ethnography is the implicit or sometimes explicit expectation of reciprocity. Network ethnography-based projects depend entirely on the generosity and openness of participants and organizations who offer their time for interviews and open their doors to give researchers access to their premises, events, and extended networks. Without this access, researchers are restricted to materials available online, which are usually less nuanced and, in many cases, tend to reflect official views. Thus, getting access to organizations and key players results in a pressure of reciprocity, as researchers depend on the goodwill and cooperative attitude of participant organizations. Avelar et al. (2022) refer to the pressure for reciprocity/“loyalty,” providing a grounded account of the complexities around reporting on researched people, as noted below:

These people we study are personable, friendly. It becomes tricky to report on the researched people, because you talk to them and have these sort of natural conversations, and you just think about how far apart are you really? This challenges the part of us that looks for simplistic analysis, or dualistic analysis of progressive and conservative, ‘good’ or ‘evil’, but once you enter these policy networks things are greyer. There's a need to break down the assumptions that we carry as critical policy sociologists. We assume we’re socially progressive, and that many of our participants aren’t. But that's simplistic. (p. 115)

Indeed this “grey zone” represents both a challenge and an opportunity, which future network ethnography research could continue to explore.

A messy business

Another critical aspect of network ethnography relates to decisions about what is included and what is excluded—in other words, to define the boundary around the policy network. One illustration of this tension is in-depth internet searches, usually conducted at the early stages of a project. When conducting these searches on key organizations, we usually look at “about us” tabs to access information about the governance structure of the organization, including board of trustees, advisory boards, etc., as well as “our partners” tabs, which generally include information about associated organizations and businesses, and it is often difficult to make decisions about what nodes are beyond the scope of the network ethnography and hence to draw the boundary around the network.

Relatedly, an essential feature of policy networks is that they are constantly evolving. While we might be able to identify underlaying regularities, patterns of interaction, and recurrent dynamics, the time gaps between the moments we record those connections, interactions, and practices, and the slow journey of those analyses as they transition to publications might mean that some of our claims might already be outdated as governmental agendas and funding priorities shift, policy entrepreneurs move across sectors, philanthropic organizations change priorities and redirect funds, or edu-businesses merge or disappear.

A flexible approach

Unlike other research methods that have rigid, predetermined protocols, network ethnography requires simultaneous planning and improvisation. While usually the research design of a study that uses network ethnography will incorporate a thorough mapping of policy players, including an overview of relevant reports, key organizations/actors in the field, and a list of the main events that bring the network together, as the network ethnography unfolds, new opportunities for data collection will emerge, such as the occasional opportunity to access a partner organization that was not originally included in the research design, or the possibility of attending a network-related event that was not visible to the researcher when the project was conceived. That is, network ethnography requires space for the unknown and the flexibility to navigate those emerging data collection opportunities without diverting from the research aims and objectives.

About the papers in the Special Issue

Drawing on pioneering work in network ethnography studies, a few being mentioned above, the authors in this Special Issue deployed network ethnography in different contexts (Australia, Argentina, the UK, and China) and addressed different kinds of policy actors (venture philanthropies, foundations, international organization program leaders, academics, and school principals) and the different effects of their labor and social relations in shaping policy networks and education reforms. Some of these contexts are less researched by previous literature (e.g., Argentina and China), and some actors have received less, if any, attention (e.g., academics and school principals). Even in areas that have been more often examined in the literature (e.g., venture philanthropies), the papers in this Special Issue offer new insights, such as on territorial expansion through “interrefencing,” where policy ideas are justified by reference to other contexts (Prince, 2020).

The wide engagement of this Special Issue with different contexts and policies (e.g., environmental and sustainability education, artificial intelligence (AI) in education, teacher training programs, and international schools) enables it to establish dialogue and shed light on the dynamics and varied effects of policy networks in shaping education reforms. Contemporary education policymaking is the outcome of “heterarchical” relations and negotiations between bureaucracies, markets, and policy networks (Avelar & Ball, 2019; Ball, 2016). Network ethnography studies, as shown by the articles in this Special Issue, should not only focus on the work of non-state actors in the policy networks but also address the complex relations between state and non-state actors and shed light on “not-so-mobile forces that continue to define education policy and practice” (Savage et al., 2021, p. 307). In what follows, we offer a brief introduction to the articles in this Special Issue, highlighting their key contributions and hopefully providing enough of a hook for the reader to continue to explore the full articles.

With a focus on venture philanthropy in Australia, Rowe (2023) traces the emergence and influence of Social Ventures Australia (SVA), as an example of a pluralistic policy network. SVA is the peak venture philanthropic organization in Australia and an influential node driving large-scale reform in public education. The brainchild of a global consultancy McKinsey & Company, SVA receives philanthropic support from some of the world's largest conglomerates and multinationals, such as Microsoft, Google, and Goldman Sachs. Rowe deploys the notion of philanthrocapitalism as a conceptual tool to examine the workings of SVA. Philanthrocapitalism fundamentally alters the way in which resources and money are redistributed among the population, with an increased amount of resources channeled through intermediaries such as consultancies, private organizations, and not-for-profits, via grants, high-value contracts, or appointments. Drawing upon a wide range of sources (official databases, financial reports, impact statements and interviews with high level nodes), Rowe masterfully demonstrates how SVA has successfully advocated for several intermediaries, such as an evidence broker (Evidence for Learning), a national charity (Schools Plus), and a national education research institute (Australian Education Research Organization), raising important questions about transparency and democracy.

Similar to Rowe, Matovich and Esper (2023) focus their analysis on the Varkey Foundation (VF), a global venture philanthropy organization registered in the UK specializing in teacher education, and map its expansion into Argentina. VF is also part of a large business conglomerate, Varkey Group Ltd., whose subsidiaries include a global chain of international schools, GEMS Education. Interviews with key actors, analysis of official documentation, and systematic newspaper database searches were the main sources of data to evidence how the national government rapidly fostered local reforms in provinces aligned with it politically, creating business opportunities in education for global policy players. Matovich and Esper highlight the important personal connections and affinities in the roll out of VF's teacher training and leadership program across various districts. In their analysis, Matovich and Esper also bring attention to the roles of local non-state actors (a non-profit organization, Edúcere, as an example) in helping VF adapt and “argentinize” their programs. This heterarchical policy network significantly contributed to VF's “quick landing” in Argentina and effectively constructed it as a legitimate actor in local education practices, subsequently spreading their influence in the region.

Gellai (2023) addresses a timely issue in policy and research: AI in education. With a focus on AI-related think tanks, universities and academics, companies, and governmental organizations, Gellai collected available information from websites, Facebook and LinkedIn pages, Twitter posts and replies, blogs, and YouTube channels to develop a comprehensive map of the policy community shaping the AI sector in the UK. This paper discusses several key actors in the policy network and their relations, labor, and effects on AI policymaking in detail. Some key findings are the role of the AI Council—an organization founded by the UK government serving as an interlocutor between industry, academia, and government—in constructing AI as the solution to local education problems and as an effective tool to achieve the government's education export strategy, and the central role played by enterprising academics through setting up “research spin-off” companies and connecting with businesses to draft AI reports. Bound by mutual interests in monetizing on education and educational technology through AI, this paper highlights the need to further understand vested interests in the sector.

Different from the above three papers’ emphasis on the “whos” of policy in network ethnography studies, Stahelin and McKenzie (2023) focus on the “wheres” of policy, which have received less attention in the literature. Drawing on interviews with international policy leaders from across sectors (international organizations, civil society organizations, and academics) participating in the policymaking or implementation of United Nations (UN) environmental and sustainability education programs (ESE programs), Stahelin and McKenzie examine the role of face-to-face meetings in facilitating the policy mobilities of UN ESE programs. This paper shows an interesting tension between global convergence and local adaptations in policy enactments. On the one hand, Stahelin and McKenzie evidence how meetings serve as platforms for policy exchange and consensus-building and how international organizations are dominant in shifting the policy discourse on ESE; on the other, they address the deliberate “in-built ambiguity” embedded in UN documents that aim to encourage localized interpretations and adaptations, and demonstrate how ESE-related meetings have worked to steer UN international organizations in particular directions, putting pressure on national governments to enact policies and modifying the original emphases on regulatory frameworks and guidelines. Face-to-face meetings, the authors conclude, “both fix policy and put it in motion.”

Drawing on a study of international schools in Shanghai, Jin and Chen (2023) explore how external experiences and curricula are imported as a policy tool to rectify and reform the examination-oriented education system, and how these experiences are enacted differently by schools. Against a dominant research focus on the influence of international players in the movement of policy ideas and solutions, this paper brings attention to the strong role of the city government and district governments in shaping the structure and dynamic of policy networks in the international school landscape in Shanghai. This study makes two significant contributions to the field. First, this paper is one of the earliest attempts, if not the first, to experiment with the method of network ethnography in the context of China. Second, the network analysis is the outcome of deep contextualization and key insights from practitioners on the ground, which in turn enabled the authors to identify and subsequently “follow” key nodes in international education in Shanghai.

Taken together, by deploying the method of network ethnography across diverse issues and contexts, the papers in this collection illustrate how “successful policy models are made rather than found” (Peck & Theodore, 2015, p. xx), and also offer critical insights on how “policies morph and mutate along the way, often taking on lives of their own” (Cook & Ward, 2012, p. 142).