Networked Field Studies: Comparative Inquiry and Online Communities

Abstract

In this article, I articulate a methodology for comparative qualitative analysis of online communities, which I refer to as networked field studies. I describe networked field studies as an approach that allows for looking across multiple communities and field sites to build a coherent set of analytical claims about the role of technology and everyday life, drawing on my own research investigating relationships to digital technologies among three countercultural communities. The major aim of this article is to contribute to methodological discussions on comparative qualitative analysis within Internet studies, foregrounding how research on digital technologies can both benefit from and complicate a comparative approach. After a brief summary of the communities studied in the research that has given rise to this methodological approach, I outline key methodological concepts and address the strengths and limitations of networked field studies as a method for analyzing socio-technical practices in everyday life.

With Internet studies maturing into an interdisciplinary area of scholarship, fewer people ask whether or not to study the Internet, and far more people ask how. One strand of the methodological discussions about Internet studies research concerns whether entirely new methods need to be developed for studying online phenomena or whether traditional methods (from long-established fields like sociology, anthropology, and history) can simply be adapted for the web (see Boellstorff, Nardi, Pearce, & Taylor, 2012; Hine, 2015; Hughes, 2012; Markham & Baym, 2008). While there are few who would suggest that Internet studies research divest entirely from traditional research methods, it remains an open question how to develop methods for studying online phenomena in a way that both recognizes what is new and makes relevant connections to existing methods. Studying socio-technical networks, meaning the diverse relationships between humans and non-humans that comprise everyday life, may not demand an entirely new research praxis but different methodological challenges have emerged in the context of social life and digital technologies.

A thorough review of digital media methods is beyond the scope of this article; however, it is worth highlighting a few of the methodological issues tied to studying networks, particularly within qualitative and interpretive research. Early critiques addressed the tendency to impose a strict online/offline binary, ignoring the ways that embodiment matters across different technologies and devices (e.g., Rybas & Gajjala, 2007). Others noted a failure to account for differences in geography and culture in analysis of digital life (e.g., Burrell, 2012), or pointed out limitations of concentrating on a single platform, given that people typically traverse a number of devices and network sites throughout the day (e.g., Lampinen, 2016). As a whole, these critiques underscore the difficulty of studying the digital in a complex and holistic way, recognizing its embeddedness in everyday life (Hine, 2015) and what factors are beholden or agnostic to particular places, people, and practices. The polysemic connotations of the word network helps illustrate these difficulties—humans may arrange themselves in networks of social relations, and they are also increasingly bound up in socio-technical networks, both digital and otherwise. Mindful of the methodological critiques surfacing in Internet studies research, how can we develop methods best suited to articulating the relationships between people and technology? What can comparison between different socio-technical networks reveal in terms of building theory around how groups of people use and play with digital media?

My own research has concentrated on the practices and norms within countercultural groups. I have given the label networked field studies to my approach for comparing social media practices within and across communities. I see networked field studies as a method that lends itself to first asking how groups of people develop practices and norms around a socio-technical network, and then comparing those normative behaviors and attitudes to other communities. In examining communities, networked field studies operates at a meso-level approach in researching human activity, in between the micro-level of individual activities (e.g., ethnomethodology) and the macro-level of societal norms and values. Taking a meso-level approach reflects a more sociological (as opposed to psychological) orientation toward social phenomena, with a focus on collective (rather than individual) norms, ethics, practices, and organization (see Wenger, 2008). Communities are useful to study not because they are simple hierarchies of human relationships, but because they are messy, complex, and often contradictory. Looking at communities allows for a middle ground between studying individual meaning–making and a broader set of societal claims.

The method described here is moreover invested in looking across communities to consider convergences and divergences in relationships to and uses of digital technologies among groups of people. Comparing practices and norms across distinct groups provides a more robust understanding of what role a technology can serve for communities as communities, meaning that this kind of work can address questions of whether technologies support or hinder the work of supporting and sustaining a sense of togetherness. Comparative work comes with challenges, described at length below, and may not be appropriate or feasible in all cases. But this approach does allow for analysis of socio-technical networks, potentially expanding the reach of nodes and edges to encompass the views and practices to multiple groups.

My comparative work has addressed three countercultural communities and their relationships to technology. The first is an online community for people interested in body modification, called Body Modification Ezine (BME) in which I analyzed mechanisms of policing membership and community boundaries (Lingel & boyd, 2013). Within this countercultural community, online tools provide crucial means of social connection but also present obstacles when it comes to maintaining a sense of subversion or otherness in the wake of increased attention and exposure. The second countercultural group is a punk community in New Brunswick, New Jersey, which has a long history of hosting shows in an informal network of residential basements. With collaborators (Lingel, Trammell, Sanchez, & Naaman, 2012), I investigated how members of this community made decisions about how and when to share information about shows with each other while keeping this information secret from the police and other outsiders. The core socio-technical tension in this community is between online platforms as tools for disseminating information and the social mechanisms that arise to control these information flows to meet local needs and ethics. Finally, I analyzed relationships to mainstream social media platforms within Brooklyn’s drag community (Lingel & Golub, 2015), looking at tensions between mainstream and countercultural values. Key tensions in this project surfaced around names and identity work, where the drag community developed a series of tactics and workarounds to navigate platform rules that were experienced as both too strict and too straight. In a monograph (Lingel, 2017), I drew these projects together under a single analytical framework of conceptualizing the capacities of digital technologies to support countercultural identity and community.

My goal in this article is to put what I have learned from this comparative work into conversation with two strands of methods literature: work on comparative qualitative research and methods for studying the Internet. A key argument is that comparative analysis of communities can provide a robust account of socio-technical networks, a thickness to the edges between network nodes. We can see how certain norms and practices may emerge in multiple communities, and whether particular platforms take on similar roles or afford certain functions across groups. Through identifying points of convergence and divergence between communities, we gain added depth about people as well as platforms.

Note that I am not suggesting all Internet studies research should take this approach nor that the method I describe is the only means of gathering thick descriptions of networks. I am advancing two more modest claims, first that qualitative comparison of digital practices across communities is both less common and under-theorized, and second that these studies are useful in developing meso-level theories in socio-technical life. Studying communities as embedded within dynamic arrangements of digital tools and media produces a rich, in-depth, and holistic account of socio-technical networks as shot through with ethics, meaning, and practices. By offering the term networked field studies, I am wary of suggesting what amounts to a methodological neologism by implying that this comparing socio-technical practices across communities is wholly new. In fact, as I will describe in depth below, this methodological approach draws from a range of established techniques and analytical approaches. To be clear, I am less interested in coining a term than (1) contributing to conversations of comparative qualitative methodology and (2) doing so in a way that is specifically attentive to challenges faced in theorizing socio-technical networks.

Unpacking “Networked” and “Field Studies”

Postill (2008) has argued that Internet researchers have been overly reliant on concepts of community and networks, a “paradigmatic dominance [that] blinkers our view of the ongoing adoption of internet technologies by local authorities, companies and residents around the globe” (p. 417). Mindful of critiques like Postill’s, my objective in this section is for precision around methodological terminology, hopefully avoiding the conceptual myopia around concepts of community and networks. The term community is admittedly a fraught one. As Joseph (2006) has pointed out, the label tends to assign easily romanticized connotations of social cohesion and Marxist values of anti-capitalism, belied by realities of infighting and elitism. Loutzenheiser (2007) argued that the lack of flexibility in the term community has to do primarily with the inability to reflect changing group ideology (p. 121), and critiqued scholarship that presents a community as fixed and inert rather than fluid and dynamic. See also Portwood-Stacer (2013) on differences between “subculture,” “movement,” “scene,” “milieu,” and “community” in her discussion of anarchist lifestyle politics (p. 7). Citing Amit (2002), Postill contests the notion that researchers should retain critical use of the term community simply because it remains of cultural value to so many, both currently and historically.

The term community is in fact less central to the present methodological discussion than the following term, networks, but my view is that “community” can be retained as a term with cultural valence to be recognized as meaningful, while also acknowledging its theoretical limitations:

In conceptualizing how networks can be theorized in critical analysis, my approach has parallels with Coleman’s (2011) description of “networked media,]” which she defined as, technologies that are connected to a distributed transmission network such as the Internet or cell towers. In such a case, “networked” speaks to a technical affordance. However, I also use the term to invoke a cultural sense of connectivity with one another. (p. 12)

Like Coleman, I use “networked” to signal an orientation toward technology that acknowledges both the infrastructure sense of networks and the social connections that are represented and enabled by digital media. As well, I follow Hine’s (2015) concepts of the “E3 internet,” meaning that digital technologies are experienced as embodied, embedded, and everyday. Both Hine and Coleman emphasize the embeddedness of online technologies into everyday life, and the ways that these networks comprise people, devices, infrastructure, and the relationships therein. Networked, then, refers both to the topic of digital media and a theoretical approach that views people and technologies as assemblages requiring analytical unpacking. More broadly, I take a social shaping view of technology, meaning that I see technological change and social behavior as mutually implicated and co-constructed. Social shaping occupies a middle ground between technological determinism and social construction of technology (SCOT) as poles of assigning agency to technologies versus people, respectively, in accounts of sociological change (Baym, 2016—for extended discussions of SCOT and technological determinism, see Bijker, Hughes, Pinch, & Douglas, 2012; Pinch & Bijker, 1984).

Note that this conceptualization of technology extends beyond a strict association between “networked” and “digital.” Whereas Boellstorff et al. (2012) described ethnographies of “in-world” communities, meaning communities whose connections are formed primarily through virtual networks, networked field studies takes an approach similar to Hine’s (2015) research on relationships to digital media within a number of social contexts, and Barassi’s (2015) study of how three different activist groups conceptualize digital media as a tool of protest. In contrast to research that concentrates on a single platform, this understanding of networked comprises a media landscape of online and offline tools, as well as related infrastructures and information systems.

My use of the term “field studies” is fairly straightforward, in line with Burrell’s (2009) definition of fieldwork as “the spatial characteristics of a field-based research project, the stage on which the social processes under study take shape” (p. 182). This is not to say that the term is uncontested in social science research. As Hine (2015) has argued, “although we routinely speak of ‘the field site’ in the singular, the object of study in ethnographic tradition has, in practice, rarely been a tightly bound geographic space or cultural unit” (pp. 58-59). Instead, field sites tend to expand and contract over the course of a project and are always connected to other sites, institutions, and communities. Two key influences here are Marcus’ (1995) seminal articulation of multi-sited ethnography and Burrell’s (2012) work on field site as network. Rather than restricting the concept of the field site to a single physical location, Marcus advocated for greater flexibility, incorporating multiple sites into the immersive project of fieldwork. Networked field studies requires similar mobility, not only in terms of investigating separate communities and technologies, but also in accounting for both online and offline practices, and tracing the different platforms, devices, and sites (both physical and digital) that emerge within a community’s socio-technical milieu. This sense of fieldwork has important parallels with Burrell, for whom “network” signals a foregrounding of practices against a dynamic spatial environment, which is moreover highly mediated by technological infrastructures.

Like others who study digital culture and production, I contend that “it is the quotidian experience of media, not the avant-garde or exceptionally expensive, that speaks to what we actually do with media and best forecast the future of mediated worlds” (Coleman, 2011, p. 71—for an early defense of studying everyday rather than exceptional online practices, see Baym, 1994). Although the communities I have studied in developing networked field studies are in many ways exceptional and avant-garde, the platforms and practices that they use are much more mundane, and my overarching research agenda has been to understand the socio-technical field in terms of daily interactions and interests. While my work has tended toward questions of alterity and appropriation, I see networked field studies as agnostic to investigating mainstream versus countercultural identity. The fundamental components of networked field studies as an orientation toward socio-technical phenomena are an expansive and flexible view of networks as both technological and social, and comparative analysis of community practices (see Figure 1 below). My goal in this article is to describe networked field studies in a way that is adaptable to researchers studying a wide range of interests, behaviors, and practices, as manifest in relationships to digital technologies.

Figure 1.

As a method, networked field studies comprise both a conceptual approach to technology and a comparative analysis of multiple field studies. With “networked,” I am signaling a focus on technology and more narrowly, a view of technology as deeply integrated into everyday life and partially constitutive of social relations.

Key Influences: Related Research Models in Media and Internet Studies

In addition to multi-sited ethnography (Marcus, 1995) and field site as network (Burrell, 2009), networked field studies has resonances with case studies (Denzin & Lincoln, 2000; Travers, 2001) and comparative ethnography (Miller & Slater, 2005). Case studies provide an analytical structure of developing theoretical claims (Stake, 2005), in that the case study is structured around the investigation of a coherent, bounded entity, such as a person, space, or incident, where “bounding the case is one of the primary conceptual responsibilities of the researcher” (p. 453). This bounding work of case studies presents a divergence from many of the connotations surrounding ethnographic fieldwork, which (especially in the beginning stages) is less structured and more flexible (Fetterman, 2010). Although structured and involving careful planning, ethnographic fieldwork is less bounded to a single person, site, or object, and it is this approach that differentiates networked field studies from case studies. Miles, Huberman and Saldana (2013) have made a parallel argument for thinking of fieldwork in terms of “sites” over “cases” because doing so “reminds us that a ‘case’ always occurs in a specified social and physical setting” (emphasis in original, p. 27). Similarly in conceptualizing networked field studies, I opt for “study” over site because fieldwork at multiple sites was required for each community I studied.

In terms of theory-building, comparative ethnography offers instruction on developing claims and arguments that emerge between and not just within cases (Miller & Slater, 2005), and given my approach of interviews and participant observations, one might argue that I am simply conducting comparative ethnography of communities that leverage digital technologies. There are some important critiques to offer here. One (somewhat minor) objection is that comparative ethnography has sometimes been viewed as the purview of senior academics after years of fieldwork in distinct sites (Geertz, 1973; Wolcott, 2008). And in fact, comparison has something of a troubled status in ethnography: Wolcott (2008) cautioned that “neophyte” ethnographers should compare as little as possible, worrying that the urge to make compelling arguments through comparison leads to positivist and quantitative evaluations of analytical legitimacy (p. 92). A second and more substantive difference is that comparative ethnography typically takes the form of post-hoc reflection, rather than a more immediate (or even simultaneous) coordination of analysis. More pointedly, I do not see networked field studies as a form of comparative ethnography, largely because the method is open to a broader set of qualitative methods than ethnography, strictly speaking. Although indebted to and inspired by the ethnographic method, particularly in the context of the digital, networked field studies is better thought of as a qualitative method that draws on ethnographic techniques in order to develop theory across different community contexts. Beyond sidestepping debates about what precisely constitutes ethnography, this definition hopefully allows for a broader set of qualitative researchers to engage and play with networked field studies as a method.

Networked Field Studies: Lessons Learned

Turning from methodological influences to grounded descriptions of conducting research, we can now ask, what does networked field studies look like in terms of planning fieldwork? How does comparative analysis unfold across multiple field sides into a coherent set of claims? Having worked on similar themes—alterity and digital technology—in three separate communities, I wanted to build a case for theorizing tensions of community building as reflected in digital technology. Doing so meant conducting additional fieldwork, but it also meant a new process of analysis. After a brief description of fieldwork, I describe this analytical process of comparative theory-building, or developing theoretical claims about the socio-technical networks within and analytical networks between these communities.

Fieldwork

The fieldwork for these projects unfolded over 7 years, in multiple states and countries, with dozens of interviews and hundreds of hours of participant observation. During the summer of 2011, I conducted interviews (18) with members of the online body modification community, in addition to a number of follow-up interviews (3) conducted in 2014 and 2015. Between the fall of 2009 and spring of 2011, I was part of a research team that conducted interviews (14) and a focus group (10) with members of the underground music scene in New Brunswick. With my collaborator Adam Golub, I conducted a series of focus groups with performers (16) in Brooklyn’s drag community and held a workshop with both performers and nightlife goers (approximately 40) about the role of social media in drag culture, all of which took place in the fall of 2013, complemented by a small number of follow-up interviews (3) in the fall of 2015.

In terms of investigating socio-technical practices, flexibility characterized my approach to fieldwork with all three communities. I am influenced here by Marcus’ (1995) and Burrell’s (2009) models for fieldwork, as well as Li’s (2008) description of flexibility as willingness to adjust one’s approach to recruitment while in the field as a way of responding to constraints or circumstances that could not be predicted in advance. Flexibility in fieldwork can be thought of both in terms of embracing a broad definition of digital technology and thinking of acclimating to field sites as iterative rather than immersive. Regarding the former, I approached socio-technical networks as encompassing a wide range of devices, platforms, media, and practices. Instead of confining my work to either online or offline technologies, I see these categories as inter-related and mutually constitutive. And rather than attempting to pre-determine the platforms that I wanted to study, I sought to be flexible in allowing the specific tools and technologies that mattered within each community to emerge during fieldwork. This flexible approach to fieldwork moves past a reductive division between online and offline tools and privileges participants’ perspectives on their local socio-technical landscapes.

As I noted earlier, the division between online and offline as a meaningful way of categorizing online activity becomes increasingly tenuous. Each of the communities I studied involved different kinds of and approaches to accounting for both digital and non-digital technologies during fieldwork. For example, unlike the other two studies, BME’s community took place predominantly online, and the majority of my work was to understand the site’s features as a socio-technical platform. With New Brunswick punks and Brooklyn drag queens, I began research in offline contexts and then worked to understand online practices. In both cases, I went to shows and hung out with performers and lived in the same neighborhoods as many of my participants. I did not participate as fully in the online practices of these communities as I had in the case of BME. Although I spent time reading and analyzing New Brunswick message board posts and Facebook pages of drag queen participants who added me to their online networks, this was very different from having spent years as a member of BME.

As I have described it, a flexible approach to doing fieldwork involves allowing platforms and practices to emerge from the field rather than a predetermined list, in media res rather than a priori. ¹ An example from my fieldwork was the use of multiple Facebook accounts among Brooklyn drag queens. Commonly referred to as “boy pages” and “queen pages,” about half of drag performers I interviewed maintained multiple accounts. Although partly intended to control whom in their diverse social networks knew about their lives as drag performers, through interviews I found that this queen page/boy page tactic also helped queens manage the relational labor (Baym, 2016) of connecting to fans and other performers. Prior to fieldwork, it would have been possible to pre-determine either the meanings of these practices or the tools themselves. Indeed, some media formats that I initially found interesting, like zines, later turned out to be less interesting or crucial to the social phenomena I was studying. Recalling Coleman’s (2011) definition of networked as a complex arrangement of tools and practices, a flexible approach to analyzing socio-technical tactics is a key component to a networked field studies approach.

A common description (or even hallmark) of ethnographic fieldwork is the goal of immersion (Delamont, 2004; see also Boellstorff et al., 2012, pp. 88-89; Fetterman 2010, p. 70). Immersion suggests depth to the point of being surrounded and enveloped; to me it has connotations of slowness or even stickiness, of being pulled deep into a lifeworld over a sustained period of time. But fieldwork can also be more intermittent, where participant observation becomes less about immersion than recursion. As alluded to earlier, there is something of a division in digital ethnographies between “in world” field sites (e.g., Boellstorff et al., 2012; Gray, 2009; Hine, 2015) versus studying phenomena that are more distributed across platforms and devices. Within the latter context, immersion is tied less to a single platform than exploring the network of technologies that matter within a community or group. Developing a thorough accounting of a socio-technical network can benefit from returning to the same community over many months—seeing the same performers, hanging out in the same bars and venues—built up a successive kind of familiarity. This series of snapshots yielded an episodic acclimation to the people, places, and relationships within each field site. The approach to fieldwork I am describing embraces socio-technical networks as a collection of nodes and edges to be understood over time, through fieldwork that may be a series of repeated visits rather than sustained contact.

This iterative, recursive approach to fieldwork has important distinctions from doing “rapid” or “pop up” ethnography (see Hine, 2015)—these labels suggest from a desire to lay claim to the status of ethnography among qualitative methods of observation, when really they constitute interviews with short bursts of participant observation. (See also Jackson’s (2013) arguments about the values of this description.) The episodic nature of non-continuous fieldwork sacrifices continued contact and shared participation in everyday life. Research questions that involve how people develop ideas or discourses among each other may require a different kind of fieldwork than what I have described as networked field studies.

Analysis

A deliberate and thoughtful approach to data analysis is a hallmark of rigorous qualitative work, but this research phase is perhaps particularly crucial in looking at multiple sets of fieldwork data. When qualitative researchers under-describe their analytical process, they may inadvertently support criticisms of qualitative work as “fuzzy” and un-scientific (see useful refutations in Boellstorff et al., 2012, pp. 29-48; Lindlof & Taylor, 2010; Miles et al., 2013). In developing theoretical claims that work across different field sites, coding is a crucial process of binding and cohesion. As such, my goal here is to describe in some depth the analytical work involved in comparing data from different communities and developing a coherent argument about digital technologies in everyday lives of alternative communities.

I had analyzed each community on its own terms prior to working toward a comprehensive framework for online alterity; in developing a comparative analysis, my first step was to survey all the data I had accumulated from prior fieldwork. After assembling transcripts, field notes, media (such as photographs, videos, and screenshots), and ephemera (like flyers and zines), I took notes on themes that emerged, concentrating on high-level issues I knew I wanted to pursue in the book: alterity, community, and relationships to digital media. Rather than an open coding (Corbin & Strauss, 2006) method, my approach at this stage was more like high-level thematic coding (Braun & Clarke, 2006), centered around several central themes, including alterity, community, technology, and space. I then made a plan for additional data sources required to engage these topics thoroughly, including follow-up interviews, textual analysis of social media content, and participant observation. Essentially, I approached the comprehensive review of the data I had gathered as a means of honing research questions and developing a checklist of additional data needed to address those questions. I then made a plan and timeline for completing new interviews and observations.

After conducting additional fieldwork, I recoded all transcripts and media as one (somewhat massive) set of data. This meant setting aside earlier coding structures and reevaluating sources in light of the new comparative project. Arguably the most challenging part of incorporating multiple field sites into a single coding structure is developing a consistently reflexive analysis. The constant comparison method of analysis is instructive here as a model for data reflexivity (Glaser, 1965) as an analytical method “designed to aid . . . in generating a theory which is integrated, consistent, plausible, close to the data, and in a form which is clear enough to be readily operationalized for testing in quantitative research” (pp. 437-438). Glaser’s highly influential method involves four stages: comparing incidents, developing coding categories, bounding theory, and writing theory. Writing in the mid-20th century, Glaser’s account can feel somewhat staid or mechanical, and ultimately beholden to quantitative mechanisms of determining validity (see Charmaz, 2014; Clarke, 2005). My approach to analysis and theory building was more fluid and dynamic, and involved two sets of coding strategies.

I used NVivo software, drawing on an emic/etic approach (Miles et al., 2013; see also Wolcott, 2008 on the emic/etic divide within ethnographic analysis, pp. 141-143). This approach to coding involves a simple, hierarchical structure of high-level themes identified by the researcher (etic), with codes nested below corresponding to how participants refer to these concepts (emic). As an example, I was interested in the relationships between authenticity and alterity. Under the etic code authenticity, I eventually included subcodes like “sellout” and “scene points,” terms that emerged in multiple field sites to refer to practices of assessing the degree to which people are truly ideologically committed versus merely performing membership superficially (see also (Portwood-Stacer, 2013)). Emic/etic analysis provides a dialectical structure that puts concepts into conversation with (or perhaps better, translated through) the terms and perspectives of participants. In terms of comparison across field sites, emic codes demonstrate the manifestations of etic themes. Recalling Stake’s (2005) criticisms of comparisons between cases, a key component of data analysis between field studies involves acknowledging points of divergence as well as convergence between instances, in my case, communities.

In addition, I used an open coding method (Strauss & Corbin, 1990) to develop a set of codes that did not necessarily fall into the scope of etic themes, but were still interesting to me. In combination, the coding methods I have described provide both structure and flexibility for developing observations and theoretical claims. This is also where I tracked the different devices, tools, and platforms within each field study, enabling me to track differences between communities in their orientation to the same technology. For BME members, Facebook presented a source of competition to their niche, interest-oriented site, and was simultaneously viewed as a temptation and a threat. New Brunswick basement punks saw a utility in Facebook for publicizing shows but only up to a point. Trusted practices like texting and face-to-face communication took over as soon as house addresses for shows became involved. Brooklyn drag queens relied on Facebook for self-promotion and fostering solidarity, most queens talked about the pleasure of using social media in their lives as performers and as queer folks seeking community. They nonetheless demanded changes to Facebook’s policies to allow for more fluid mechanisms of verifying identity. Returning to terminology introduced earlier, contrasting these different practices produced a conceptual thickness to the edges between nodes in each socio-technical network.

After completing this new round of coding, I began developing the framework that anchors the analytical component of the book that grew out of comparing these communities. Looking across findings from all three field studies, I drew out three key socio-technical characteristics that supported a sense of community for these countercultural groups: authenticity (the ability of communities to see their values and ethics reflected in the functions and policies of the platforms), legibility (being able adjudicate complaints or tensions within a platform), and flexibility (whether or not a platform encourages tweaking, alteration and appropriation). Developing this framework was not akin to discovering an element out of whole cloth from the data. Rather, it was an iterative process of coming to understand how community members related to a given technology, or what a particular practice meant within a field site. Building the three-part framework described above involved returning to the data repeatedly to make sure that my evolving conceptualization of these components was grounded in the data. For example, I had initially thought of authenticity as locality, meaning a reflection of local practices. Yet locality has connotations of geographic specificity, which fit well in New Brunswick, somewhat for Brooklyn drag queens and not at all for BME as an international collection of body modification enthusiasts. As I continued to think about tensions within each community in terms of their relationships to mainstream social media platforms like Facebook, I realized that these struggles were less about geography per se and more about whether they saw themselves in the policies of the sites they were using. Comparing between networked communities allowed me to refine key theoretical concepts, to interrogate and then attempt to explain characteristics that emerged between platforms in ways that feel more substantive.

Synthesis: Networked Field Studies and Transferable Analysis

I see two key advantages of networked field studies as a method. First, an expansive view of digital technology is definitional of network field studies, which helps to move socio-technical inquiry beyond a simple online/offline binary. Although Internet studies research has long been critical of a tidy division between being on the web versus in real life (boyd, 2001; Burrell, 2012; Hine, 2015; Rybas & Gajjala, 2007), there has nonetheless been a tendency to structure research about online activity in a way that reifies these exact boundaries, whether by concentrating on a single platform or by ignoring offline contexts that intervene in online practices. Thinking about platforms as isolated from other platforms can be just as reductive as insisting on a sharp divide between online and offline life. In contrast, networked field studies lends itself to multiplicity, both in terms of multiple communities and the many different technologies and platforms that matter in the everyday lives of users and communities. Looking at digital technologies this way gets past the reductive division between digital and non-digital media, instead thinking about entwined fabrics of technologies and people.

Second, networked field studies allows for analysis that works across multiple groups. Within the context of studying community-based practices, researchers have tended to look within rather than between communities (e.g., Hodkinson, 2002; Nardi, 2010; (Pearce, 2009)—for an important exception, see Barassi, 2015). Networked field studies provide a framework for thinking about conceptual links that emerge across distinct field sites. This comparative work is a key for building analytical claims that extend beyond a single field site. Note that I am not making an argument about network field studies as yielding more generalizable knowledge, largely because I view generalizability as an inappropriate aim of interpretive work. Instead, I see transferability (Guba and Lincoln, 1982) as a better model for theory building within qualitative work. Where generalizability is about scaling up, transferability is about moving between, tracing connections between communities rather than trying to generalize or making universal claims. Transferability is baked into networked field studies as a method, in that the comparative analysis seeks to identify practices and characteristics that emerge across distinct communities and sites.

Limitations

Every methodology involves tradeoffs and limitations, and as part of a comprehensive account of networked field studies, I want to acknowledge what this method does not do well. In particular, comparative fieldwork presents challenges in precisely the context I have described of studying digital media. Conducting qualitative fieldwork takes time, especially with my emphasis on iterative fieldwork that benefits from multiple visits to a field site. Given the rapid pace of change in digital media, a fundamental challenge of comparing online communities is that investing time in each field site (or set of field sites) can come at the cost of synchronous analysis. While it might be tempting to think that online platforms could afford researchers the ability to conduct fieldwork with multiple communities at the same time, such an approach merits caution. Participants can tell when researchers are genuinely attentive to their community versus always-already recontextualizing their interactions and findings. While overlap is reasonable and perhaps productive, it is crucial to structure a research agenda that allows for sustained attention to a community at a time.

My own approach to mitigating the very real issue of online research feeling dated has been to think of fieldwork as iterative and recursive, a continually updating series of observation and data gathering. In addition, as I noted in the description of my process for data analysis, my first, high-level pass at the entire set of data was intended to yield both high-level themes and a to-do list for additional fieldwork. Conducting follow-up data analysis allowed me to address gaps in my knowledge, but it also allowed me to update my data to produce more current findings.

Less of a limitation and more of a potentially fallibility of networked field studies is that an emphasis on comparison can come at the expense of adequately describing the depth of a field site. While Stake (2005) validated the use of “collective” case studies for “jointly study[ing] a number of cases in order to investigate a phenomenon, population or general condition,” (p. 436), he remained suspicious of comparing between case studies not originally undertaken together. For Stake, comparison necessarily divorces separate cases from their original context, and in doing so “obscures case knowledge that fails to facilitate comparison. Comparative description is the opposite of . . . ‘thick description’” (p. 436). Stake’s concerns center on sacrificing depth and context for shallow connections across distinct instances. This criticism points to a related drawback—in order to provide both thick descriptions within networks and to make substantive connections between networks, monographs seem all but required. Already the gold standard for ethnography, it is difficult to imagine doing justice to comparative field studies in an article. Perhaps there are more experimental solutions in the form of non-traditional means of scholarly communication, like film, theater or multi-moded approaches combining different media formats.

Conclusion

From a conceptual standpoint, the concept of socio-technical networks as the interweaving of bodies, devices, and infrastructure is an increasingly common starting point for Internet studies inquiry. From a methodological standpoint, this concept of network is both appealing and problematic. My goals for this article have been to outline a method for comparative qualitative research on networked communities. I have drawn on a rich set of methodological discourses, both to trace scholarly influences and to identify a niche. Particularly within disciplines that lack a strong tradition of training doctoral students in qualitative methods (Spinuzzi, 2000), there is an unfortunate tendency to collapse distinct methods, techniques, and philosophies under a single label. For example, any project involving participant observation and interviews becomes ethnography, and any coding that uses inductive reasoning is grounded theory. I have sought to be precise in my influences while carving out a niche for comparative socio-technical inquiry.

Ethnographic methods have a rich and growing history within Internet studies, and yet, there is a significant gap in detailed methodological account working across multiple field sties toward a shared analytical inquiry. Comparative qualitative work (whether ethnographic or otherwise) is difficult and time-consuming, but can reveal important insights into practices and phenomena that emerge between communities. My hope is that this article can contribute to continued discussions in Internet and media studies (and indeed broader conversations on qualitative methodology) about investigating socio-technical practices and analytical transferability across field sites and communities.