1 Introduction

Participant observation and ethnography, at large, are an everyday staple for researchers in migration, ethnic and mobility studies. Even only an overview of the key ethnographies done so far, within national and disciplinary boundaries or across them, would call for a chapter in itself. Nonetheless, there is relatively little of a systematic methodological elaboration around the merits, pitfalls and prospects of ethnography in migration studies (major exceptions including Falzon 2016; Fitzgerald 2006; Glick Schiller 2003; Iosifides 2011). In this chapter, we first sketch out some guidelines on the methodological development of participant observation, on its theoretical underpinnings and on its relevance to this research field. Participant observation should be distinguished from pure observation – that seeks to remove researchers as much as possible from the actions and behaviours they investigate – and from pure participation – that has also been described as “going native” or “becoming the phenomena” (DeWalt and DeWalt 2002; Jorgensen 1989). Rather than choosing one of these two extremes, participant observation aims to find a balance between both. It is important that researchers are aware of their particular place on this “continuum in the degree of observation and participation” and reflect on the impact of this position on the kinds of data collected and the sort of analysis that is possible. As a key ethnographic technique, participant observation is uniquely placed to refine the theoretical understanding of migration as it invites people to adopt the perspective of migrants themselves. Moreover, research based on participant observation allows dynamics of power, agency and politics to be theorized from below.

As human mobility automatically involves multiple locations, the use of participant observation in research on migration and mobility challenges the classical understanding of this method as an in-depth study of a closed locality. In contrast to the classical idea of a “taken-for-granted space in which an ‘other’ culture or society lies waiting to be observed and written” (Gupta and Ferguson 1997, p. 1), the everyday lives of many individuals more often than not transcend the geographical locations in which classical fieldwork took place, challenging ethnographers to include these social spaces in the demarcation of their fieldwork sites (Schrooten 2016). Appadurai (1991, p. 191) has formulated the consequent challenge for ethnographers in the following terms:

As groups migrate, regroup in new locations, reconstruct their histories, and re-configure their ethnic “projects”, the ethno in ethnography takes on a slippery, nonlocalized quality, to which the descriptive practices of anthropology will have to respond. The landscapes of group identity – the ethnoscapes – around the world are no longer familiar anthropological objects, insofar as groups are no longer tightly territorialized, spatially bounded, historically self-conscious, or culturally homogeneous.

In the second part of this chapter, we discuss, at least at a preliminary level, some of the recent methodological developments in ethnographic research at large – and in participant observation more specifically – that have attempted to break away from practices of local, “bounded” and confined ethnography, such as multi-sited ethnography and online ethnography. In doing so, we combine our respective sociological and anthropological backgrounds. This is a very common instance of the relevance of ethnography across disciplinary fields (such as anthropology, sociology, geography, communication studies and history), and of their mutual intersections – all the more so in an inherently interdisciplinary research area such as migration studies.

2 Participant Observation as a Research Method

Participant observation has become the almost identity-giving method for ethnography, although it is certainly not the only one that is being used by ethnographers. Individual or group in-depth interviews, informal conversations, taking fieldnotes, artefact analysis, and many other things may all be part of ethnographic fieldwork. Ethnography, and participant observation as a core part of it, have their own genealogies in any disciplinary realm. The research method is often referred to as “the hallmark of cultural anthropology” (Spradley 2016, p. 3), but is also a common feature of qualitative research in a number of other disciplines. The first anthropologist to write about using participant observation as a research method was Frank Hamilton Cushing, who spent four and a half years as a participant observer with the Zuni Pueblo people around 1879 (DeWalt and DeWalt 2002; Sanjek 1990). Other important early anthropologists who used participant observation were Beatrice Potter Webb (in the 1880s), Bronislaw Malinowski (in the 1920s) and Margaret Mead (in the 1920s). Although other anthropologists had carried out ethnographic fieldwork before him, Malinowski’s (1922) discussion of his participation and observation of the Trobiands still serves as the fundamental description of the method:

Soon after I had established myself in Omarkana Trobriand Islands, I began to take part, in a way, in the village life, to look forward to the important or festive events, to take personal interest in gossip and developments of the village occurrences. (…) As I went on my morning walk through the village, I could see intimate details of family life. (…) I could see the arrangements for the day’s work, people starting on their errands, or groups of men and women busy at some manufacturing tasks. Quarrels, jokes, family scenes, events usually trivial, sometimes dramatic but always significant, form the atmosphere of my daily life, as well as theirs.

When it comes to sociology, the ritual starting tends to be fixed around the twenties at the Chicago School of Urban Sociology. There is a famous quote by Robert Park that nicely captures the spirit of this methodological engagement:

You have been told to go grubbing in the library, thereby accumulating a mass of notes and a liberal coating of grime. You have been told to choose problems wherever you can find musty stacks of routine records based on trivial schedules prepared by tired bureaucrats and filled out by reluctant applicants for aid or fussy do-gooders or indifferent clerks. This is called “getting your hands dirty in real research.” Those who thus counsel you are wise and honorable; the reasons they offer are of great value. But one thing more is needful: first-hand observation. Go and sit in the lounges of the luxury hotels and on the doorsteps of the flophouses; sit in Orchestra Hall and in the Star and Garter Burlesk. In short, gentlemen [sic], go get the seats of your pants dirty in REAL research.Footnote 1

Being “there”, and observing the patterns of everyday life close to the actors engaged in it, are highlighted as a valuable strategy of data collection, whatever the context at stake. What makes this strategy unique is a researcher’s “close observation of and involvement with people in a particular social setting”, thereby relating “the words spoken and the practices observed or experienced to the overall cultural framework within which they occurred” (Watson 2010, p. 205). To put this in just slightly different terms, participant observation is

a method in which a researcher takes part in the daily activities, rituals, interactions, and events of a group of people as one of the means of learning the explicit and tacit aspects of their life routines and their culture. (DeWalt and DeWalt 2002, p. 1)

Whatever the combination between the roles of observer and participant (Whyte 1979; Platt 1983), participant observation is the privileged research tool of ethnography. The latter is first of all a theoretical approach that borrows from such diverse backgrounds as ethnomethodology, symbolic interactionism, phenomenology and grounded theory (Atkinson et al. 2001); second, it can be appreciated as a discipline in itself, cutting across the boundaries between sociology, anthropology and geography, as much as organizational studies, cultural studies and social history, to make the most obvious examples; third, ethnography stands for a textual product – the outcome of empirical data collection – and a way of writing, even a genre in its own right (Emerson et al. 2011; Van Maanen 2011).

What sets participant observation apart from all other ways of doing research, then, is an embodied and extended presence in the social world of those being studied. Social life as it is being lived, rather than only as it is reported by informants (often in ephemeral, artificial or ad-hoc settings), is its fundamental concern. To be sure, this methodological option can be integrated with several others, such as in-depth interviews, life history interviewing, survey research or review of documents and texts, to name a few. What is distinctive of it, anyhow, is a significant degree of researcher participation and involvement in the ordinary life of the social group under study. This can produce insights and findings relevant to a variety of research questions, falling somewhere in a continuum between two ideal-typical stances: a fundamentally naturalistic one, concerning how people live their lives in a given context, influencing and being influenced by the latter in distinct ways; and a more open-ended and interpretative one, regarding what sense they make of their social environments and how a given phenomenon is constructed, negotiated and reproduced by those involved with it, given the relevant external factors. Individuals, social relationships, groups and broader socio-material assemblages are all potentially appropriate units of analysis, to be appreciated in their mutual interactions. The crucial point has however to do with the definition, and then the empirical limitation, of the research field.

Ethnographically speaking, a field may well correspond to one or more specific places. However, it may also amount to a relatively consistent set of social relationships and circumstances, relevant to the group under study, whether produced by proximate or distant forms of interaction. More than “a pregiven entity”, an ethnographic field is “something we construct, both through the practical transactions and activities of data collection and through the literary activities of writing fieldnotes” (Atkinson 2015). Even when the field overlaps with a material environment, it is ethnographically meaningful not only as a physical infrastructure, but also through the ways in which it is “brought into being” by the social actors that co-produce the phenomenon to be studied. In a slightly different understanding, which follows Bourdieu’s metaphorical use of the word, a field stands for any sort of situated, relatively well-bound social arena, defined by a structure of competing pressures, tensions and interests.

In practice, ethnographic fieldwork unfolds along a number of relatively well distinct steps, participant observation being invariably the central and pivotal one. Five research phases can be helpfully sorted out, as follows:

  1. 1.

    Before the field. Ethnographers have typically an exploratory and open-minded stance – which is not, however, an empty-minded one. Some theoretical elaboration, based also on pre-existing research, is necessary from the outset, lest ethnography turns into mere description. Why – a researcher could wonder at first – is an ethnographic option appropriate to her interests? Or at least, what aspects of them could be fruitfully addressed ethnographically? Similar questions are of help in delimiting the field and collecting preliminary information about it. Since the very beginning, however, ethnographers’ attitudes should be flexible enough to fine tune with the social actors’ ways of defining their situation. Instead of fully-fledged hypothesis, “sensitizing concepts” could be sketched and tried out, to be then better developed, refined, or possibly replaced through fieldwork, following a cyclical rather than linear research design.

  2. 2.

    Accessing the field. Gaining respectful, trust-based, hence effective access into the expected research field is critical to all that comes next. This may entail, particularly in community or organizational research, close interaction with informants holding central positions in local hierarchies or networks. Along this critical transition, handbooks invariably emphasize the influence of trust, reputation and sensitivity. They tend to be less emphatic, though, on two equally critical conditions: first, researchers’ skills in managing interpersonal relationships, or their “ability to build mutually supportive relationships with subjects” (Whyte 1979); second, the need for them to figure out and negotiate what “return” their counterparts may expect. That said, how many people should be contacted and followed at first, and then all over fieldwork, is hard to set or predict in advance.

  3. 3.

    Staying in the field. Here comes participant observation as intensive engagement, whatever the degree of participation, the variety of participants, the foci of observation and the underlying driving questions. In practice, observation may involve participants’ accounts as much as their tacit understandings, practices and mutual interactions, with all of their emotional and moral underpinnings. Background settings and all sorts of objects in use are also a major concern for observation. That said, whenever ethnography has a more than descriptive remit, it involves an attempt to infer some insight into the distribution of social resources – knowledge, power, prestige, etc. – that is embedded in the field in question. Ethnographers, warns Atkinson (2015), “are interested in that is told explicitly, and what is withheld, what is regarded as ‘tellable’ and what is treated as ineffable”. Put otherwise, “truth-telling” is not necessarily the key stake of ethnography, as opposed to “the social management of informational and moral states” – how things are said and made plausible to which publics, following which cultural conventions, etc.

There are at least two more issues that emerge quite invariably during participant observation. These have to do, first, with the need for multi-sensorial involvement – what is seen, or heard, being not exhaustive of the sensuous wealth of stimuli and insights an ethnographer can reap from the field; second, with the need for an unusually flexible and open-ended attitude, regarding both one’s own role and position in the field and the limited scope to control the development and temporality of field events. “Learning in (and out of) the field”, as Van Maanen (2010, p. 220) put it, “is uneven, usually unforeseen, and rests more on a logic of discovery and happenstances than a logic of verification and plan. It is the unbearable slowness of ethnography”.

  1. 4.

    Getting out of the field is the following, invariable step – an often contended and not necessarily complete or irreversible one. Against the attendant questions of “Where”, “When”, and “How”, it is once again hard to set criteria in advance, unless there are stronger external constraints. What is clearly important is a degree of theoretical saturation, whereby researchers seem to be seeing, hearing, or feeling the same again and again, with little or no new “evidence” or insight. In practice, exiting the field still entails a variety of relational, emotional and ethical dilemmas to be negotiated on an individual, case-by-case basis.

  2. 5.

    From fieldwork to textwork is the last key transition – one without which ethnography can hardly aim at public scrutiny and scientific relevance. This means to make the most, over time, of the notes ethnographers should take as close as possible to the relevant events or interactions. Fieldnotes are a matter of incremental and open-ended cumulation, albeit increasingly driven by the search for recurrent themes and patterns. As fieldwork itself amounts to much more than words only, visual tools and ICTs have an increasing potential to affect all steps of the ethnographic process – including data assembling and writing. They can hardly replace, though, researchers’ autonomy in deciding which observation items should be included in the fieldnotes, and why; nor in working out a balance between a merely observational style of note-writing and the reflexive collection of their own thoughts, feelings and reactions. In either case, fieldnotes seem bound to make for an irremediably partial and selective account (Fine 1993; Emerson et al. 2011). Whatever the ways of collection, codification and analysis, they are never simply data that reflect what “really” happened out there. Instead, they are affected by the personal circumstances of ethnographers and by fieldwork contingencies, even while being driven by the aim to figure out more general theoretical dimensions, beyond the specific events at stake. Having said that, ethnographers’ claim to achieve a distinctive societal significance (Burawoy 1991) – different from, but not lesser than, statistical significance – is not without its critics (e.g., in sociology, Goldthorpe 2007).

Based on these methodological remarks, we can now approach the specific relevance and implications of ethnography for migration studies.

3 Participant Observation in Migration Studies

Since Malinowski, anthropologists have attempted to understand how individuals move “in and through social systems” (Fortes 1971, p. 2), but active engagement with human movement and mobility in anthropology is a more recent phenomenon (Brettell 2003; Salazar 2013). Much of the research carried out by social and cultural anthropologists during the first half of the twentieth century paid little attention to human mobility as many ethnographers were working with a bounded concept of culture and a static structural-functional theoretical paradigm (Brettell 2013; Gupta and Ferguson 1997). They mostly portrayed societies across the globe as bounded, territorialized, relatively unchanging and homogenous units (Salazar 2010, 2013; Tsing 1993; Vertovec 2010) and cultures “as essentially immobile or as possessing a mobility that is cyclical and repetitive [...] Those with culture are expected to have a regular, delimited occupation of territory. If they move, they must do so cyclically, like transhumant pastoralists or kularing sailors” (Tsing 1993, p. 123).

Since the second half of the twentieth century, however, mobility and mobility-related topics have gained a prominent place in anthropology and in other social sciences. Whereas in migration studies the actual interest is less in movement then in departure and/or arrival (involving issues of uprooting and integration), “mobility” has become a keyword of the social sciences, delineating a novel domain of debates, approaches and methodologies regarding processes of movement in a broader sense (Adey et al. 2014; Cresswell 2006; Salazar and Jayaram 2016; Urry 2007). In the last few years, plenty of ethnographies of mobility have been carried out, focusing on, among many others, migrant trajectories (McKay 2012; Schapendonk and Steel 2014), the everyday lives of migrants (Holmes 2013; Lucht 2013; Smith 2006), families who are divided across borders (Beck and Beck-Gernsheim 2013; Dreby 2010; Hondagneu-Sotelo 2001; Parreñas 2001), migrants’ involvement in transnational politics (Fitzgerald 2004; Ghorashi 2003; Levitt 2001), the experiences of marginalized minorities (Agier 2002; Van Meeteren 2010), and the meaning of “home” and domesticity (Gielis 2011; Giorgi and Fasulo 2013; Levin and Fincher 2010).

Anthropologists were among the first scholars to propose “a transnational perspective for the study of migration” (Glick Schiller et al. 1992), drawing attention to the fact that migrants’ social practices occur almost simultaneously on the territories of more than one national state. This approach challenged previous, rather localized assumptions about identities, and focused on the relationships between places migrated from and to. As such, participant observation has been a driver of conceptual innovation, as the concept of transnationality would have never come to us without it. Criticizing the taken-for-granted equation of society with the nation state, the so-called “methodological nationalism” (Wimmer and Glick Schiller 2003), the transnational approach argues that “national organization as a structuring principle of societal and political action can no longer serve as the orienting reference point for the social scientific observer” (Beck and Sznaider 2006, p. 4).

In the study of transnational migration, ethnographical research distinguishes itself from social survey research, another area of measurement in applied social research, in a number of ways. Firstly, whereas other social scientists generally generate deductively inferred hypotheses that are then verified during research, ethnographers continually question, explore and reformulate systematic explanations of the relations between variables, and even the choice of variables, during research. This enables them to change research questions as new situations that were not expected within the initial set of assumptions present themselves (Glick Schiller 2003). For example, in her PhD research proposal on Brazilian mobility, Schrooten did not refer to the Internet as a possible research site. However, soon after her fieldwork started, she decided to make it one of her central fieldwork locations as Brazilian migrants often mentioned the Internet and other social media as the most important media for keeping in touch with other Brazilians, both inside and outside Brazil. Likewise, the development of ethnography may significantly question one’s previous theoretical assumptions. Boccagni (2011, 2016), for instance, spent a long time tracing the relevance of transnational ties in the everyday lives of Ecuadorians in Italy, only to find out that such ties were far more “fragmented” (Menjivar 2000), and not always so fundamental, as his pre-fieldwork literature review might have suggested.

Secondly, the practice of participant observation, or of ethnographic research more generally, collects data of a different nature than those collected by other methods (DeWalt and DeWalt 2002). Evidence provided on the basis of a survey is emic data, a respondent’s self-presentation and, thus, “front stage behaviour” (Goffman 1999). Yet, people’s behaviour does not always correspond to the opinions they consciously articulate (Giddens 2013; Turner 1991). Participant observation ascertains the typicality of behaviour from on-going observations, over time and within a range of contexts, of what people do, differentiated from what they say that they do (De Munck and Sobo 1998). As such, it is well suited to delve into people’s daily lives well beyond their self-presentations (Boccagni 2012; Glick Schiller 2003).

This point is particularly important in the study of human mobility, as there are many reasons for the discrepancies between self-reported responses and the actual behaviour of migrants. Respondents might deliberately not report certain activities that are considered suspect or illegal. For example, if they obtain financial support from the state, they may report that they don’t receive any financial help from their transnational networks nor send remittances themselves, as they know this could lead to the refusal of the payment of allowances and benefits.

But also, unexpected situations may lead to a difference between the aspirations people have and their actual circumstances. Much has been written, for instance, on “return” as an initially very clear, expectedly short-term aim of migration – one that, however, often tends to be postponed and even to blur away over time. In our research, we met numerous people whose trajectories were very different from what they had expected when they left their country of residence. Many respondents’ narratives show that their (on-going) (im)mobility is often an unintended process and a phase which might end, but could just as well start over depending on circumstances (Withaeckx et al. 2015). For many non-EU immigrants, who initially settled in Southern-European countries, for example, further migration was prompted by the need to seek better opportunities and life circumstances by a subsequent move further north. For many of them, this new migration was unintended, as they had lived in Southern Europe for numerous years and had expected to settle there permanently. Another example is that of transient migrants, who follow an expectedly linear migration trajectory with a specific destination in mind, but for a variety of reasons spend some time in other locations before moving on to their desired destination – as long as they reach it at all. African asylum seekers moving onwards from Southern Europe are a case in point (Belloni 2016). These – and many more – movements illustrate that contemporary processes of human mobility are heterogeneous and varied in terms of purposes, trajectories and durations. Methodologies that continue to work under assumptions of migration as a unidirectional, purposeful and intentional process from one state of fixity (in the place of origin) to another (in the destination) fail to capture much of the complexity of these processes. This makes particularly crucial an ethnographic effort to follow them up over time – as opposed to one-shot techniques of data collection, including in-depth interviews.

Deciding how much to participate or not in the life of people being studied is no easy judgement. Although these topics call for a much broader debate and are not specifically related to participant observation, we want to draw attention to some issues that might play an important role in this judgement. On the one hand, it is important to realise that there are limits to participant observation, not at least so when engaging in certain activities may be illegal, dangerous to the ethnographer, or both. On the other hand, there are occasions during which the researcher faces the decision about whether or not to intervene in a situation; not at least so when they face dilemmas that become difficult ethical issues (DeWalt and DeWalt 2002; Rynkiewich and Spradley 1976; see also the Chap. 15 by Ilse van Liempt and Veronika Bilger in this volume).

4 On the Relationship Between Ethnographers and Their Counterparts

As the instances above suggest, the relationships between ethnographers and their research subjects are particularly critical to participant observation. This is particularly salient in migration studies, and can be appreciated at all the steps of the ethnographic process: while negotiating access to the field, which is very much a matter of gaining trust (or at least respect) of its members, and then of negotiating mutual views and expectations over time; during one’s stay in the field, as the quality of the data collected – hence the validity of an ethnographer’s claims – is affected by informants’ attitudes and willingness to cooperate, no less than by one’s insight and ability to be “in the right place, in the right time”; while leaving the field, during textwork and in publishing research findings, which may raise delicate issues in acknowledging informants’ contribution.

A focus on interpersonal relationships in fieldwork entails revisiting the distinction between insider and outsider research, where the former points to “situations in which the researcher shares membership in a social group with the research participant” (Nowicka and Cieslik 2014, p. 6; see also Carling et al. 2014). At stake is also the variable degree of similarity and proximity between ethnographers and their subjects, as it is constructed, enacted and negotiated over time, primarily (but not exclusively) along ethno-national lines. In practice, there is nothing obvious in the conditions under which an ethnographer acts and is perceived like an insider or an outsider to field members. Most notably, the increasing number of ethnographers with an immigrant background, or with the same ethnic background as their counterparts, is a desirable development in itself. However, it needs not result in automatically better or deeper ethnographic engagement. The very divide between insiders and outsiders is more blurred and context-specific than the distinction between ethnic majorities and minorities would suggest. It is ultimately a matter of boundary-making, where relevant boundaries can involve also gender, age, class, religion and so forth. All of these variables turn out to be more or less salient markers of researchers’ “positionality”, and are subject to more or less intensive forms of “identity management” on a case-by-case basis. Even so, reflecting on the evolving position of ethnographers vis-à-vis field informants is a source of insight in two major respects.

To start with, it stimulates reflection on the weight and consequences of their mutual perceptions and categorizations. In other words, researchers’ positionality mirrors the potential transition from principled differences – those associated with categories such as ethnicity, gender, class, etc. – to the more minute and personal markers of difference that are negotiated between researchers and informants. Carling et al. (2014) develop a fascinating argument around this, by showing how several “markers of status” (Table 12.1) may make field relationships more or less inclusive, symmetric, ultimately sustainable.Footnote 2 Of course, none of these markers is necessarily predictive of researchers’ positionality. Taken together, though, they do provide a map for investigating ethnographers’ attempts to negotiate the “right mix” of proximity and distance vis-à-vis their subjects. They are also telling of the social factors that most likely affect the mutual engagement between researchers and informants; hence, at least indirectly, the outcome of ethnography itself.

Table 12.1 Specific markers of a participant observer that may influence her insider/outsider status (Carling et al. 2014, p. 45)

In the second place, focusing on the insider/outsider identity of ethnographers, and on their interface with ethnographic informants, is instrumental to still another reflexive step: interrogating the identity, roles and purview of field members themselves. The latter can take up a variety of roles over the course of ethnography – “simple” informants, gatekeepers or cognitive mediators, but also, potentially, co-producers of ethnography as a collaborative effort in which (some key) informants parallel the role of researchers themselves (Boccagni 2011). In fact, the underpinnings of field relationships may amount to much more than the traditional and relatively shallow “rapport”. The “essential affinity between observer and observed” (Marcus 2007) can be acknowledged and pave the way for a progressive and empowering approach to fieldwork (Lassiter 2005). If and when such a collaborative approach is enacted, significant issues of authorship – who is the writer of what, on behalf of whom – are also likely to emerge.

Overall, revisiting ethnographers’ field relationships is not simply a background question for purposes of external accountability – i.e. to follow formal protocols, rules of funding agencies, etc. Nor should it be discarded as an ethnographical form of navel-gazing. Instead, the point is that interpersonal relationships, as they are negotiated all over fieldwork, have major epistemological and ethical implications. There is much to be gained from a reflexive stance on one’s evolving position in the field, on its prevalent perceptions among informants, on the influence of the latter on data collection, elaboration and even ownership. Having said this, the emancipatory or politically-oriented potential of ethnography should not be overestimated either. As luring as the labels of “active”, “collaborative” or “participatory” are, they are unlikely to be frequently adopted by immigrant informants – unless for a self-selected minority of them. As much experience shows, providing potential space for their active engagement may well be enough to define a good-enough ethnographic practice. For the bulk of field informants, respectful listening, sensitive acknowledgement and the promise of a more nuanced understanding of their life circumstances are not an irrelevant return for their (mostly limited) involvement.

5 What Next? Multi-sited Ethnography, Online Ethnography, and Beyond

Classical ethnographic research relied on long-term and intensive investigation of one particular place or a local situation, aiming to understand another way of life from the native point of view. Yet, human mobility is a key example of a phenomenon that is irreducible to the scope of a closed, territorially based and fully controllable ethnographic field, as it involves multiple physical, social and symbolic locations, whether simultaneously or over time. One of the basic aspects of a transnational approach is to consider the simultaneity of transnational practices taking place in multiple localities. The development of a transnational approach to migration was interestingly paralleled with the introduction of “multi-sited fieldwork” (Marcus 1995), a new strategy of data collection that encouraged researchers to investigate transnational units of reference in a variety of fields, including migration-related ones. Differing from a merely comparative study of localities, the “multi-sited fieldworker” quite literally follows people and their connections and relationships across space. Much has been written, by now, both on the potential of multi-sited ethnography and on the challenges it typically faces (see also the Chap. 3 by Russell King in this volume). The latter include reconciling breadth and depth of analysis, coping with the huge costs it may entail (hence the need for teamwork and collaborative research designs), and finding strong theoretical grounds to justify site selection (Hannerz 2003; Hage 2005; Falzon 2016; Marcus 2012). After all, the most complex and innovative task for multi-sited ethnography is not only staying somehow in more sites at once (e.g. via Mazzucato’s (2009) “simultaneous matched sampling”). As important and elusive is observing the interpersonal relationships being cultivated between them, and the underlying material and immaterial infrastructures (Boccagni 2016).

Multisited ethnography allows researchers to identify the empirical field as de-territorialised by, for instance, studying migrants’ online communities or the use of new media in transnational relationships (Madianou 2016; Madianou and Miller 2012; Pink et al. 2015; Schrooten 2012). Studies of migration have been particularly important for challenging assumptions about the “degree to which geographically dispersed agents experience a sense of physical and/or psychological proximity through the use of particular communication technologies” (Milne 2010, p. 165). Scientific attention to the variety of ways through which digital media and technologies can be used to create a sense of presence over space and time is related, in turn, to another emerging development in the ethnographic study of migration: digital ethnography, also known as virtual ethnography, and most commonly online ethnography, which extends traditional ethnography to settings where interactions are technologically mediated (see also the Chap. 14 by Koen Leurs and Madhuri Prabhakar in this volume). Like its traditional counterpart, online ethnography generally aims to produce a “thick description” of the behavior in a culture or community. This makes it distinct from other methods such as online interviews, content analysis or web usage mining.

Moving ethnography online requires some adaptations of the method, as many aspects of this online environment are very distinct from those of face-to-face settings. The fact that online communication is often automatically saved and archived, creating permanent accurate records makes researchers able to easily observe and copy these interactions. The enormous amount of available data forces online ethnographers to make explicit choices about the delimitation of their research question, the place and duration of online data collection and the way data will be analysed.

Moreover, the nature of online data is rather different from the data obtained through a face-to-face ethnography. An online ethnography provides mainly textual and visual material (such as the use of pictures, page layout, videos and so on). In these kinds of interactions, body language is absent or replaced by emoticons or text. At the same time, the role of mediated oral communication, with a possibility of visual access to each other is also increasing. Instead of focusing solely on the written word, ethnographers are thus required to also integrate visual aspects of data and to develop a new set of skills and methods of data collection.

Another obvious difference with conventional ethnography is the way researchers can make an entrée into the community they want to study. Although the problem of how to present oneself also exists within traditional ethnography, the challenges involved in obtaining access differ, as ethnographers cannot rely solely upon their physical presence and personal interactional skills (Garcia et al. 2009; Mann and Stewart 2000).

Online ethnographic research has also raised a number of ethical questions. The specificities of this research setting necessitate a re-examination of the institutionalized understandings of research ethics. Although in the emerging literature some concrete guidelines can now be found of how to conduct ethical research online (Bull et al. 2011; Hesse-Biber and Leavy 2010; Schrooten 2016; Wilkinson and Thelwall 2011), an internationally accepted framework for online ethnographic research ethics does not as yet exist.

Overall, multi-sited and online ethnography are different ways of addressing what is arguably the key challenge for the ethnographic study of transnational migration: catching social practices on the move, the associated circulation of a variety of resources, and the interaction between physically proximate, present or visible life environments and their remote, absent or invisible counterparts. The issue, in other words, is to appreciate the influence of migration on geographically distant, but socially interdependent sets of phenomena, and to do so in ways as close as possible to the evolving pathways of migrants themselves. Further connections with the emerging set of mobile methods, as well as with those advanced within “global ethnography”, are arguably necessary to expand further the potential of participant observation for migration studies.