Abstract: This article reflects on fieldwork experiences with coethnic migrants in London to challenge understandings of insiderness centred in shared ethnicity, as well as the usefulness of the insider-outsider divide in migration research more generally. Drawing on examples from a study of migrants' social relations, it shows how gender, migrant status, and occupational position sometimes shape research encounters in more important ways than shared ethnicity. Furthermore, whilst shared ethnicity is undoubtedly useful in certain respects, participants' ethnicised discourses and practices may also generate feelings of distance in the coethnic researcher. Whilst supporting the "ethnic bias" critique to migration studies (GLICK SCHILLER, ÇAĞLAR & GULDBRANDSEN, 2006), the analysis thus highlights how both ethnic and non-ethnic factors alternate or interact to create perceptions of insiderness or outsiderness in specific research contexts.

Key words: insiderness; ethnicity; ethnic lens; migration research; research relationships

As various scholars note, qualitative studies of migrants tend to focus on particular ethnic groups, and be conducted by researchers who share the same ethnic background as the participants (GANS, 1997, p.887; KUSOW, 2003, p.591; see also the introduction to this thematic section, NOWICKA & RYAN, 2015). Coethnic researchers are commonly seen as "insiders" to the group, benefiting from easier access to participants and their experiences due to their familiarity with participants' language and culture (CARLING, ERDAL & EZZATI, 2014, pp.38, 52). Yet associating insiderness with shared ethnicity reinforces the "ethnic bias" characterising migration studies more generally (GLICK SCHILLER, ÇAĞLAR & GULDBRANDSEN, 2006), overlooking non-ethnic factors that may shape research encounters (CARLING et al., 2014). Furthermore, it may obscure the varied roles ethnicity plays in the research process, which may involve experiences of outsiderness on the part of the researcher. [1]

This article reflects on fieldwork experiences with Romanian migrants in London to address these problems, challenging understandings of insiderness centred in shared ethnicity, and the usefulness of the insider-outsider divide more generally. It looks at two key aspects of the research process, 1. building research relationships, and 2. understanding participants' stories, showing how gender, and migrant status and occupational trajectories respectively, may produce important commonalities or differences of experience that question the relevance of shared ethnicity between the researcher and participants. Furthermore, whilst ethnicity is not entirely absent in either case and undoubtedly useful in certain respects, participants' ethnicised discourses and practices sometimes generate feelings of ambivalence or distance in the researcher. Contrary to assumptions about the advantages of being a coethnic researcher, the analysis thus calls attention to how both ethnic and non-ethnic factors may trigger "moments" of insiderness (see also NOWICKA & CIESLIK, 2014), but also outsiderness in specific research contexts and interactions. [2]

The "insider" status is usually associated with researchers who are seen to be members of the same social group or category as those whom they study (DE ANDRADE, 2000, pp.269-270; NOWICKA & CIESLIK, 2014, p.6). This may be based on achieved or ascribed characteristics (such as occupation, gender, ethnicity), depending on the focus of the research (MERTON, 1972; NOWICKA & CIESLIK, 2014, p.6). In the migration context, being an insider is typically understood in ethnic (ethno-cultural, racial or religious) terms. In other words, the researcher is deemed to be an "insider" when she is a member of the migrant group under study, as opposed to the native population of their destination country (see CARLING et al., 2014, p.38; RYAN, KOFMAN & AARON, 2011). My research on Romanians in London challenges this understanding of insiderness centred in shared ethnicity, looking at the varied ways in which ethnicity may or may not inform research encounters. [3]

Taking for granted the relevance of ethnicity in shaping field relations seems to be a reflection of the "ethnic bias" characterising migration studies more generally (GLICK SCHILLER et al., 2006). As GLICK SCHILLER and her colleagues famously argued, many studies of migration tend to take ethnic groups or communities as the main unit and object of analysis. Even when migrants come from specific places (e.g. a particular village or community of origin), generalisations about the entire migrant population are common (see GLICK SCHILLER, 2008). The "ethnic bias" resulting thereof produces, according to the authors, a partial, if not distorted, picture of migrants' lives and experiences, overlooking non-ethnic forms of attachment, incorporation, or cross-border connections. [4]

GLICK SCHILLER and her colleagues locate the roots of the ethnic bias in the "methodological nationalism" affecting migration research and social sciences more generally. In brief, methodological nationalism implies conflating the boundaries of society with those of the nation-state and taking the nation-state as a natural unit of analysis for the study of social phenomena (WIMMER & GLICK SCHILLER, 2003, p.578). Nationally-defined territories are assumed to contain populations with a shared culture and identity within their boundaries, which entails, in the case of movement from one country to another, portraying migrants, and their ties with natives, mainly in ethno-national terms (GLICK SCHILLER & ÇAĞLAR, 2013, p.496). [5]

The privileged status of ethnicity (or nationality) in migration research is evident not only at the level of research design and analysis, but also in discussions of field relationships and methodological reflections. Attaching the label "insider" to researchers of the same ethnic background as their participants implies that ethnicity has primary and permanent relevance in interactions between researchers and participants. However, many studies of migrants document the heterogeneity within what is typically described as an ethnic group or community (GANGA & SCOTT, 2006; RYAN et al., 2011, p.50), which undoubtedly also affects field relationships (SONG & PARKER, 1995). Ethnicity is one, and sometimes one of the few characteristics shared by the researchers and participants, who may otherwise differ in terms of age, class, gender, occupation, education, migration circumstances, and many other aspects that produce very different experiences and migration trajectories. [6]

Reflecting on these differences, various scholars describe themselves as simultaneously insiders and outsiders to the populations they study (e.g. JIMÉNEZ, 2010; MENJÍVAR, 2000), "partial" insiders (e.g. SHERIF, 2001) or occupying a "hybrid" position (e.g. HALSTEAD, 2001; see also CARLING et al., 2014). That individuals have multiple social statuses (see KUSOW, 2003; MERTON, 1972) may additionally mean that one can be a "multiple insider" (see DEUTSCH, 1981, cited in NOWICKA & CIESLIK, 2014, p.6) or a "multiple outsider", making an understanding of insiderness centred on ethnicity difficult to sustain. For example, studying Salvadorans in the US, MENJÍVAR exposes the marked social differences between her and her participants' background and experiences, despite their common nationality: "I was never an undocumented immigrant, never lived in the neighbourhoods where my informants lived, never held the kinds of jobs they did, and never experienced most of what has shaped their lives" (2000, p.245; see also CARLING et al., 2014, p.51). Although MENJÍVAR (2000, p.247) does not see these differences as precluding "sincere conversations" and understanding of migrants' lives, acknowledging their presence highlights the multiplicity of factors that may impact field encounters beyond common ethnicity or nationality (see also CHERENI, 2014; SHINOZAKI, 2012). [7]

Whilst ethnicity has received disproportionate attention in fieldwork reflections, non-ethnic factors sometimes play a more consequential role (see also CARLING et al., 2014). For example, in their research on young Chinese individuals in Britain, SONG and PARKER (1995, p.251) found that gender differences could outweigh shared ethnic background (see also RYAN et al., 2011, p.52), as also demonstrated by my interviews with migrants I had not met in advance. As BRUBAKER, FEISCHMIDT, FOX and GRANCEA (2006, p.208) note, ethnicity has "intermittent" rather than continuous presence in individuals' everyday experience, "happening" at particular times when participants identify themselves, categorise others, or depict events and experiences in ethnic terms. This observation is equally important for research relationships. Acknowledging the intermittent occurrence of ethnicity requires attention to non-ethnic categories and discourses that may other times prevail (see also MOROŞANU & FOX, 2013). [8]

Apart from downplaying non-ethnic factors, another problem with associating insiderness with coethnic researchers is that it may overlook the varied roles of ethnicity in research encounters. The insider status is generally seen to bring advantages in terms of designing the research, recruiting participants and interacting with them, as well as making sense of their stories and experiences (e.g. DE ANDRADE, 2000). For example, coethnic researchers may face less questioning and suspicion from the targeted participants about their reasons to study them (cp. SONG & PARKER, 1995, p.247; see also SHINOZAKI, 2012, pp.1817-1818; ZINN, 1979, p.210). Although access is not guaranteed, they may inspire more trust in participants, and gather data that might not be readily shared with a perceived outsider (see CARLING et al., 2014, p.52; RYAN et al, 2011, p.51). Furthermore, familiarity with the language and culture of the population studied is often considered crucial for an in-depth, nuanced understanding of migrants' experience, all the more when the focus is on culture or ethnicity (see CARLING et al., 2014, p.38). [9]

There are, however, studies that show how ethnicity does not automatically confer these privileges (e.g. OCHIENG, 2010, p.1727) and may also generate tension. Coethnic researchers, for example, face various expectations from participants, who seek to assess their "insiderness" and ties with the "community" (see CHERENI, 2014; DE ANDRADE, 2000). My research similarly shows that ethnicity can both help and create tension during the research process. Yet this was not due participants questioning my status but their apparent assumptions of our shared views, values, or experiences, which sometimes generated feelings of distance and unfamiliarity on my part. [10]

Drawing on research with Romanian migrants in London, this article thus illustrates the variable relevance and role of ethnicity in the research process, first looking at my interactions with the participants, and second, at making sense of the data they shared. In both cases, I show how gender, and migration status and occupational trajectory respectively, may take precedence in research relationships and understanding participants' stories, which do not necessarily require "ethnic" competence. Furthermore, whilst common ethnic background is beneficial in some respects, it remains an insufficient basis for establishing fruitful research relationships; what is more, ethnicised discourses or practices may also produce a sense of outsiderness in the coethnic researcher. [11]

The analysis thus challenges perceptions of coethnic researchers as insiders to migrants' experiences, and illustrates the difficulty of employing an insider-outsider distinction more generally (see also CHERENI, 2014). Exploring the various ways and contexts in which gender, migrant status, occupational position, and ethnicity alternate or interact to shape the research process does not inevitably reaffirm the privileged status of ethnicity (cp. GLICK SCHILLER & ÇAĞLAR, 2013, p.496); it acknowledges its potential relevance and seeks to understand when and how it might gain or lose prominence in the face of non-ethnic discourses and characteristics (see also MOROŞANU & FOX, 2013). [12]

This article is based on a qualitative study of how Romanian migrants in London built and maintained different social ties locally and transnationally. Like other East Europeans, Romanians have started to become more visible in the UK particularly since Romania's Accession to the European Union (in 2007), which enabled them to travel freely to Britain amongst other member states. During 2008-2010, I conducted in-depth interviews with 20 women and 20 men. Most participants were relatively young, well educated (i.e. three quarters had higher education or were students) and of urban background. [13]

A central concern underpinning my study was the ethnic bias affecting migration research (cf. GLICK SCHILLER et al., 2006) and possible ways to avoid it without neglecting ethnicity altogether. I thus employed several methodological strategies to tackle this concern. Whilst many studies of migration centre on ethnically-marked topics (such as ethnic networks, identities, ethnic entrepreneurship), I started with a general thematic focus, social relations, and aimed to capture both ethnic and non-ethnic dimensions, according to the importance they achieved in different contexts of migrants' daily life. [14]

Although my thematic focus did not privilege ethnicity, the choice of researching Romanians in London may seem surprising when seen against my concern with correcting the ethnic bias. Yet, whilst my participants were Romanian in terms of ethno-national background, I did not automatically assume that they were part of an "ethnic group" (BRUBAKER et al., 2006, pp.11-12, 209-210). My starting point was not an ethnic group, implying entities bounded by solidarity, shared identity, aims and mutual recognition (BRUBAKER, 2002, p.169), but individuals from a particular category (i.e. with shared ethnic background), with the aim to explore the varied contexts in which ethnicity or other factors shape their experiences and the ways in which they narrated them (BRUBAKER, 2002). My sampling strategies reflected this approach. [15]

First, although I focused on Romanians in London, I did not seek to "immerse myself within the life of the city's [...] community" (MARGOLIS, 1994, p.xxi) but reach a more diverse pool of migrants. This meant including "non-ethnic" recruitment routes (e.g. Facebook searches, adverts on local community websites), alongside the traditional "ethnic" ones (e.g. Romanian organisations and online forums). In a multi-diverse city like London, a relatively small-scale population like Romanians had, in principle, multiple opportunities for entering varied social networks in work and non-work contexts. To understand the complexity of migrants' social ties, I considered it essential to also look beyond the visible ethnic landscape, which is the typical entry point and focus of study in many ethnographies of migration, and include migrants whose social ties could have been very different from those of the people who frequented ethnic institutions (see also MOROŞANU, 2013a, 2013b). Recruiting participants in this way required more effort and imagination and unsurprisingly had a lower response rate. For example, conducting Facebook searches with Romanian names and "London" for location is a relatively time-consuming process, often leading to individuals no longer based in London or simply no response. However, devising such strategies can help us move beyond traditional routes and capture a more diverse migrant population. [16]

Another strategy I employed to correct the ethnic bias involved comparing migrants working in high- and lower-skilled occupations in order to uncover different perceptions and experiences of social relations within what is often regarded as an "ethnic group". If other studies emphasise gender or generational divides, my primary interest in migrants' social ties and interactions led me to focus on occupational position as a basis for comparison, since it provides one of the most significant (and few) social sites for entering social relationships, especially during adulthood (see EVE, 2010, p.1235). Furthermore, migrants often invest much of their time and energies in the pursuit of jobs, studies, or careers, which influences the ways in which they reconfigure their social network. A high- or low-skilled occupational position is often associated with a particular residential, social, and legal status. For example, at the time of conducting the interviews, those who were full-time students could also work (part-time), whilst those who came to the UK without an employment contract faced considerable work restrictions (lifted only in 2014). Migrants' different occupational status appeared to influence recruitment. [17]

In searching participants, I have undoubtedly benefited from identifying as a Romanian student who was interested in the social lives of other Romanian migrants in Britain. I circulated an invitation for participation in Romanian and used my institutional email address for contact, to inspire more credibility. However, shared ethnicity did not guarantee unproblematic access to participants (see OCHIENG, 2010, p.1727), especially since I usually approached participants I had not known before, via online messages, adverts or fora. Students or professionals were considerably easier to access. Familiarity with the process and purpose of research, involvement in similar activities during their educational career, or perceived commonalities between us (in terms of occupation) could explain why they seemed more receptive to invitations to participate in the research, beyond our common ethnicity. Those who had lower-skilled jobs proved more difficult to identify and recruit. Limited spare time or precarious status could possibly explain why my study appealed less to this category of migrants. Some additionally worried about their ability to contribute valuable information. For example, one participant who worked as a caretaker initially wrote to me that he was "pleasantly surprised" by my invitation to take part in the study but thought his experience would be rather uninteresting, despite the many years he had spent in London. Whilst my status as a Romanian student may have inspired sufficient trust to participate, I did not immediately succeed in recruiting participants via the different online venues employed. [18]

Building research relationships is a key area where the value of being a coethnic researcher has been emphasised (e.g. OCHIENG, 2010). The type of relationships established with the participants may have a major influence on data collection, the "best" data emerging, according to some, from close, longstanding research relationships, or from those based on common characteristics and experiences, for example, related to ethnic origin (p.1729). As indicated above, being a Romanian certainly contributed to accessing migrants willing to share their stories. Nevertheless, Romanianness was not always the main factor shaping our relationships. This section shows how gender and gender-related considerations took priority in certain contexts. Second, shared Romanianness did not automatically ensure smooth research relationships either. My sustained efforts to create and maintain fruitful research relationships were sometimes faced with moments of tension or emotional distance, which surface more rarely in field accounts, showing how common ethnicity may also produce feelings of outsiderness in the researcher. [19]

Recruiting many participants via online routes often meant I did not know them in advance or someone to recommend them. To create minimal hassle, I offered to meet where participants suggested, as long as the interviews were conducted in public places, such as cafes, pubs or parks. Our encounters, however, sometimes involved spontaneous proposals that revealed the primacy of gender in shaping research relationships. For example, after one interview, a female student participant invited me to her place on a subsequent occasion. I gladly accepted the invitation, which seemed to offer a valuable opportunity to get to know her better, continue our conversations, and gain further insight into her everyday life. A similar situation that involved a male participant, however, generated the opposite reaction. At the end of our interview, which otherwise unfolded without problems, this participant was keen to spend more time socialising over dinner or drinks and offered to drive to an appropriate venue. Contrary to the previous case, I politely declined the proposal. Although the interview did not raise reasons for concern, my reaction in this case was to maintain boundaries and also take what I then saw as basic precautions when conducting research with participants I had not known for more than a few hours. Such spontaneous decisions illustrate the importance of gender-related considerations in relationships with male and female participants. [20]

Gender also informed the organisation of the interview. A telling example is that of purchasing refreshments for the interview. To create a friendly, comfortable interview atmosphere and express my gratitude for participants' willingness to help with the study, I sought to offer them drinks or food, if the interview location permitted. This rarely presented a problem for female participants, after the occasional negotiation. However, in cross-gender interviews, this strategy was less successful. Several male participants firmly declined, insisting on offering me a drink instead. One of the participants even exclaimed in surprise, "I haven't heard this before", signalling that my gesture contravened his gender-related assumptions. Such behaviours may have been enhanced by our common Romanianness and the perceived expectation that our behaviour should match what participants saw as the norm back "home". Another male participant, who arrived accompanied by his wife and daughter, similarly turned down my offer and took the lead in purchasing everyone drinks. His comparatively older age, occupational and family status may have further contributed to a feeling of responsibility in a meeting with a younger female student. Such negotiations over buying drinks could create discomfort at both ends. Not only did I fail in my attempt to offer this minimal sign of recognition for participants' time and willingness to help, but I also felt the interview incurred additional expenses to them. [21]

Being a young female researcher could shape my relationships with participants in other ways too, inspiring, for example, various protective gestures. One family who lived in a distant area in London worried about my journey to the train station at night, and offered me overnight accommodation (which I again declined). This, of course, does not mean that gender was a permanently salient feature during the research process or that my researcher role was endangered. Although cross-gender interactions are sometimes thought to pose difficulties during the research, for example when seeking personal information (SONG & PARKER, 1995), this was not necessarily the case in my study. My questions about social life and relations often led participants to talk about relatively private matters. Female researchers are seen to have easier access to women's stories, (see SONG & PARKER, 1995, p.251; YOW, 1995, p.58), yet the male participants in my sample also proved to talk in a very overt and frank manner about personal aspects of their lives (see also BUCERIUS, 2013). This could involve acknowledging having a limited social circle in London, building or breaking romantic relationships, or worries about finding a suitable partner to set up a family post-migration. The friendly, appreciative approach I sought to adopt during the meetings seemed to encourage both male and female participants to share their life experiences in front of someone they barely knew. [22]

If non-ethnic factors such as gender had variable importance in research encounters, the same applied to ethnicity. Some participants emphasised the feeling of comfort and relief afforded by conversing in their native language (see BRUBAKER et al., 2006, p.241, 255). Being interviewed by a Romanian in Romanian allowed participants not only to narrate their experiences effortlessly and confidently but also effectively (cf. BRUBAKER et al., 2006, p.264; cp. SONG & PARKER, 1995, p.252). "No matter how proficient your English may be, you still feel you're making an extra effort to communicate ...", noted Bianca¹⁾, who missed "that pleasure of language games and spontaneous combinations, things that are more difficult in English". Dan provided a concrete example to illustrate the communication-related frustrations he experienced in interactions with locals. Talking about jokes and humour, he remarked:

"You say Garcea [Romanian comedian] and everyone knows who he is. You see, you started to laugh! Here they have their own, it's just that we don't know them ... that's why Romanians don't have many English friends and don't hang out with them, because when [an English person] makes a joke, you give him a strange look, what did he mean to say?" (See also MOROŞANU, 2013b, p.2167.) [23]

Dan's reference to a Romanian comedian, and subsequent delight at my immediate reaction which indicated that I grasped the message, "you see, you started to laugh!", show how shared ethnicity may become an important source of commonality in research situations. [24]

However, whilst ethnicity could facilitate communication and contribute alongside other factors to creating a comfortable research environment, it could also have more negative implications that are less visible in discussions of insiderness or field relations more generally. In what follows, I show how coethnic researchers may experience "outsiderness" when confronted with the divergent views and values informing some participants' accounts of their status and ethnic belonging. [25]

According to YUVAL-DAVIS (2006), expressions of belonging emerge particularly in times of threat or insecurity and may imply drawing boundaries in an exclusionary way. Belonging, the author notes, can be understood and constructed at several levels: the first concerns individuals' social locations (e.g. gender, class, nation) that constitute "specific positionings" and shape their experiences; the second level refers to individuals' ways of identification and emotional affiliation to various collectives (e.g. ethnic, national, religious); and the third level involves the set of values they share and may deploy to establish the boundaries of such collectives. Determining who is part of the "community" and who is not brings out the importance of one's ethical and political values and attitudes (pp.203-204). A telling example of how these systems of values become active in the construction of ethnic boundaries is the stigmatising discourse some participants voiced towards the Roma. Although many of my interviewees were enthusiastic about London's cosmopolitan profile, and often sought opportunities to befriend people from different ethnic backgrounds, discussions about Romania or Romanians' negative representation abroad led some to express negative views towards the Roma, with whom Romanians have frequently been associated in the (tabloid) press. Some participants' response to negative portrayals of Romanians (and Roma) was thus to distance themselves from the Roma and transfer the stigma to them (MOROŞANU & FOX, 2013). [26]

Frustration with conflations between Romanians and the Roma, and Romanians' attempts to signal to others the difference between ethnic Romanians and the Roma have been documented in the Romanian context (BRUBAKER et al., 2006; VAMANU & VAMANU, 2013). Listening to some migrants adopting similar views provided deeper insight into how those who moved abroad experienced their status and the strategies they developed to cope with stigmatised representations of Romanians, which could involve articulating a stigmatising discourse in turn. For example, noting that Romanians bear a "negative stamp" across Europe, being associated with all sorts of crime, one participant gradually redirected Romanians' stigmatisation towards the Roma, whom he held partly responsible for Romanians' notorious reputation abroad (MOROŞANU & FOX, 2013, p.442). Another one remarked that "what pulls us down is the Roma, who we all know ... I saw them washing car windows here too ... Those who come to work have nothing to do with this", taking distance from the negative discourse around Romanians abroad. [27]

The perceived safe and friendly research environment in the presence of a coethnic researcher may have allowed participants like these to express such views, assuming that I would share or understand their position (JIMÉNEZ, 2010, p.280), as the remark "we all know" might suggest. This echoes JIMÉNEZ's experience of researching Mexican Americans in the US, where his perceived insiderness, due to his light skin and American identity, probably encouraged, according to him, white non-Mexicans to voice both positive and negative views about people with Mexican background. JIMÉNEZ's example is, however, different in that outgroup members expressed negative views. Incidents when migrants themselves articulate negative discourses about other populations are less discussed in empirical (but see FOX, 2013; MOROŞANU & FOX, 2013) or methodological studies. [28]

This relative omission may be partly related to the researcher's perceived role and responsibility. As KUSOW (2003, p.595) notes, participants often expect coethnic researchers to depict the "community" in positive terms. More generally, an important concern amongst qualitative researchers, including those adopting feminist approaches, is to give voice to participants and show empathy towards their perspectives (HOFFMANN, 2007, p.325; OCHIENG, 2010, p.1727), particularly when they are part of a vulnerable or marginalised population. What happens, however, when some members of these vulnerable groups articulate negative views towards other vulnerable populations or present them in an unfavourable light for various reasons, such as redeeming their status? This raises important questions not only about researchers' approach to the data but also about perceptions of insiderness in research relationships. Just as similarities of language or culture may work to bring together the researcher and the researched in some situations, dissimilar views and values reflected in constructions of "Romanianness" may other times disrupt the harmony of field relationships and generate feelings of distance and non-belonging in the researcher. Ethnicity emerges in many different forms during the research process, and it can also be a source of outsiderness. [29]

Apart from building research relationships, shared ethnicity is deemed essential for understanding participants' experiences. My research on Romanians in London focused on the nature and uses of migrants' social relations. In asking questions, I initially refrained from putting ethnicity centre-stage and focusing on "Romanians", unless relevant for or brought up by the participants themselves. For example, I inquired about how migrants built social ties or secured support, instead of asking directly about resorting to "Romanians". Participants' accounts of social relations showed that, whilst ethnicity could inform them in various ways, it was not the main lens through which they made sense of their everyday experiences. Their migrant status and occupational trajectory often proved more relevant than shared ethnicity for understanding many of the stories I listened to. When invoked, ethnicity could even cause feelings of surprise and unfamiliarity in some contexts where I was implicitly or explicitly positioned as an ethnic insider by the participants. [30]

Life in London was for many of the people I interviewed primarily marked by their migrant or non-native status, rather than Romanianness per se. This operated as a main axis of division in various contexts of everyday life, including social interactions, drawing them closer to other migrants from different countries, and away from the native-born (for a detailed discussion, see MOROŞANU, 2013). Referring to the university setting, Bianca, one of the undergraduate students, emphasised the divide perceived to characterise London's otherwise multi-diverse social landscape: "Internationals stick to internationals and home students to home students. They are many, but they're not absorbed". Various non-ethnic factors pertaining to the migrant status contributed to this, including non-native knowledge of the language, unfamiliarity with local norms and institutions, different legal status, or the migration experience more generally. Romanians certainly had Romanian friends too but their social ties extended more widely. The language of ethnicity proved insufficient to capture the cross-ethnic ties developed by those who lived and shared the experience of being non-native in their place of settlement (ibid.). [31]

Stories of building cross-ethnic ties with other migrants (rather than locals) are "classic", as one of my participants remarked, and have been noted in other studies of migration (e.g. on Poles in the UK, see RYAN, 2011). In many respects, understanding their formation does not necessarily require "ethnic" competence. Romanians' stories of cosmopolitan socialisation, their shared concerns and feelings of solidarity with others in similar circumstances, and even work restrictions and feelings of exclusion resonated, to a higher or smaller degree, with a wider category of individuals, including researchers with migrant background or experience more generally. Being an insider to these experiences had more to do with our common migrant than coethnic status per se. [32]

However, whilst being a migrant was a key source of commonality, participants' lives in London were significantly shaped by their occupational status. The concerns, experiences, and everyday interactions of migrants in low-skilled occupations, often found in vulnerable situations due to legal status or precarious employment, differed considerably from those of participants who occupied high-skilled—and implicitly safer and more rewarding—positions. For example, although both categories of migrants could establish cross-ethnic ties, owing much to common migration experiences and preoccupations, these commonalities were particularly based on negative experiences in the low-skilled case, such as work restrictions or other legal barriers (MOROŞANU, 2013). My own student migrant status importantly shaped my degree of insiderness in interview situations, which shifted from close familiarity to outright remoteness from participants' experiences, in a way that further questioned the relevance of shared ethnicity. [33]

Interviewing students (and graduates) generated the fewest surprises. Their stories about integrating and mastering the norms and standards operating in the university environment, communication hurdles between local and international students, the friendships developed with fellow (international) students, or their journeys into the local leisure and partying spaces often generated "moments" of insiderness, since I was able to recognise and relate to participants' experiences. Such similarities stemming from our migrant and occupational status enabled me to recognise clues, probe for further details, test expectations, and thus obtain a deeper insight into participants' lives. [34]

At the other end were participants with lower-skilled occupations, who followed different migration trajectories and faced a different set of challenges, given their more precarious legal or economic situation, despite our common migrant position. These could include difficulties in securing or keeping jobs, coping with low wages, long hours, exploitative employers, or with the degrading status attached to many of the jobs they performed, highlighted particularly by participants with higher qualifications and aspirations. If interviewing students (and professionals to some extent) usually generated a sense of familiarity, I often figured as an outsider to the experiences recounted by lower-skilled workers (see MENJÍVAR, 2000). Talking about the initial difficulties experienced during his casual jobs in construction or agriculture bought via shady "entrepreneurs", one of my participants exclaimed laughing, probably sensing my bewilderment: "yes, that's what happened, that's life, I'm telling you, it's a comedy. You'll see what you have to write about!" His remark suggests that he too positioned me as an outsider to his predicaments (BEST, 2003), and had to decide which aspects to introduce me to and which to avoid, such as his migration route. [35]

Furthermore, in such circumstances, I did not simply become an outsider, but also one who occupied a comparatively advantaged position as a student researcher, at least in some participants' perception, despite my efforts to downplay any differences between us. For example, one woman who worked as a cleaner and had relatively limited prospects in London inquired if I planned to remain in Britain, telling me that I "had better stay here", as I will have "all the opportunities to find a place to work after completing [my] studies". Differences in occupational status and trajectories also surfaced in interviews with migrants who had higher education but experienced downward mobility. Elena, who was university educated but worked as a cleaner and baby sitter for several families in London, provides a clear example of the frustrations and contradictions generated by her downward occupational trajectory:

"What hurts ... is the work you do, you know, what you do in London to live the life you live ... otherwise it's ok, life in London is great, but this rhythm and the work you do don't effectively correspond to [your aspirations]. And then you can't go on, because you do so against your will, your desires, against what you want to become ... Wanting to do something and ending up doing something completely different, at the opposite pole. Because you see, ... you don't do [further] studies to limit yourself to a low status [jos]. For me, it's ok now, I take it lightly, it's not a problem, I haven't yet reached the point where I'm desperate, 'oh, my career', no. But I am aware that I don't want to stay at this level, that I need to go further … ." [36]

Refocusing attention on education level and achievements was a recurrent theme in interviews with migrants working in jobs well below their qualifications, a common trend amongst East European migrants more widely (e.g. PARUTIS, 2014). The incongruity between migrants' qualifications and the jobs they performed was perhaps enhanced in the interview context, when seen against the privileged position they associated with a student researcher. If in the case of students, and professionals to some extent, occupational and migrant status worked to create moments of insiderness, the adversities and disadvantages experienced by lower-skilled workers could have the opposite effect, signalling and augmenting the perceived distance between us, despite our common migrant (and sometimes educational) status. [37]

Whilst my participants often discussed the challenges and opportunities they faced in building or maintaining social relations in non-ethnic terms, this did not mean our common ethnic background was always irrelevant. As in the case of research relationships, ethnicity could be emphasised in particular contexts, generating both privileges and puzzles of understanding migrants' accounts. [38]

The positive facet of ethnicity can be illustrated by examining cross-ethnic ties in more detail. Although these ties were often based on non-ethnic commonalities, ethnicity could become a pretext for, and ingredient of socialisation amongst migrants from different countries (MOROŞANU, 2013). For example, various participants talked about the "cultural exchanges" that marked their cross-ethnic friendships, present in the sharing of "ethnic" gifts, music, food or simply stories from one's homeland. Being familiar with the "ethnic raw materials" (JIMÉNEZ, 2010) they invoked (e.g. dishes, music, traditional artefacts or symbols) certainly helped me get a more tangible, fuller picture of the nature of migrants' socialisation. [39]

Furthermore, discussing various aspects of social interaction, participants sometimes resorted to cultural references or idioms that could remain opaque to non-Romanian researchers (see also GANGA & SCOTT, 2006). These were alluded to in our conversations in a way that indicated that participants assumed my familiarity with them. The example introduced earlier, where Dan talked about interactions with natives and referred to a Romanian comedian, is a case in point. His reference engaged me as an ethnic insider (SONG & PARKER, 1995, p.252-253), who would immediately understand the example and grasp the communication difficulties he experienced. [40]

However, such moments of finding common ground based on shared ethnic background were not pervasive. On various occasions, when Romanianness was invoked, the stories or experiences I came across would, on the contrary, reveal participants' very different lives and life-worlds. Differences in educational, class, or urban/rural background outweighed the relevance of shared ethnicity in understanding migrants' experiences. This could be evident or not for the participants, who sometimes positioned me as an insider, despite my bemusement or unfamiliarity with the references they made. [41]

A telling example is that of Adriana, who grew up in a village in Romania and came to Britain to "build a future" for her young child, left at home in his grandmother's care. In London, Adriana found work as a cleaner for different families, and experienced various forms of exploitation, including remarkably low pay from some employers, deceit, or theft of her belongings. Despite certain shared characteristics, such as gender, age, ethnic background, Adriana's experience abroad was very remote from my own (MENJÍVAR, 2000). Similarly, her life and life-world in Romania evoked images or practices that did not immediately resonate with my experience. Adriana talked about how she could identify Romanian women in London based on their dress, telling me, "you know how our rural women dress like", a remark which did not necessarily evoke closely familiar images, as Adriana assumed. This became more evident when, talking about a group of charity workers visiting Romania, Adriana recalled the joy of her mother who was gifted a scarf by one of the visitors, exclaiming to me, "my mum was crazy about baticuri [scarves], you know, like Romanian women are!" Whilst Adriana identified me as an insider not only in ethnic but also gendered terms, her remark and choice of words, which evoked a head (rather than neck) scarf, caused more surprise than familiarity. If women in rural areas may routinely cover their heads with scarves, this representation of "Romanian women", which my participant presented as typical, did not necessarily coincide with the images I had in mind, based on my twenty-year-long experience of living in a large, university centre in Romania, where women's fashion choices would more frequently involve neck scarves. [42]

Such examples show that being a coethnic is not unequivocally useful when listening to participants' stories. When surfacing in interviews, ethnically-marked references can lead to a more nuanced understanding but also generate unfamiliarity and feelings of "outsiderness" on the part of the researcher. Such feelings could emerge from divergent views and values, as shown in the discussion of field relations, or from differences in social positioning (regarding education, occupation, rural/urban background) which importantly shape the extent to which researchers may relate to participants' experiences and identifications (YUVAL-DAVIS, 2006). [43]

My research on Romanian migrants in London showed that establishing research relationships and making sense of participants' stories did not primarily depend on shared ethnic background. Gender, migration status and occupational trajectories often shaped our interactions in more consequential ways (see also CARLING et al., 2014). Nevertheless, ethnicity was not entirely absent from participants' accounts or research encounters. Its "intermittent" presence (BRUBAKER et al., 2006) yielded both advantages and disadvantages, sometimes in conjunction with non-ethnically marked factors, illustrating the multi-dimensionality of research relationships (KUSOW, 2003, p.593). [44]

The variable presence and role of ethnicity in interviewing coethnic migrants challenges conceptualisations of the insider status centred on shared ethnic background, as well as the insider-outsider distinction more generally. Talking about insiders in ethnic terms can be seen as a reflection of the ethnic bias characterising migration research, which often takes migrant ethnic groups as the main units of analysis and depicts migrants' experiences through an ethnic lens (GLICK SCHILLER et al., 2006). In line with efforts to counter the ethnic bias, I showed how gender, migration status and occupation could eclipse the relevance of ethnicity in particular contexts. Gendered considerations often mediated interactions with participants, gestures, or reactions, from negotiations over purchasing refreshments to those regarding the boundaries of research relationships. Listening to participants' stories further demonstrated how our migrant status and occupational position could elicit important similarities or differences of experience in terms of social relations beyond shared ethnicity per se. [45]

However, the examples presented in the article also suggest that adopting a "non-ethnic lens" to research encounters may have its own limitations, overlooking those contexts and moments when researchers and participants may find common ground in shared ethnicity, or, to the contrary, when ethnicity plays a very different role, generating distance and a sense of outsiderness in the researcher. Being a Romanian student, communicating in Romanian or grasping cultural references could contribute to the development of trusting relationships and a fuller understanding of various events narrated by the participants. But participants' ethnicised discourses or behaviour could at times also generate feelings of surprise or non-belonging. This could be due to my limited familiarity with their practices or experiences, pertaining to our different social locations, or to divergent views and values reflected in the ways in which some migrants constructed ethnic boundaries and belonging (see YUVAL-DAVIS, 2006). Indeed, the examples discussed here not only show that research encounters and insiderness are not inevitably marked by ethnicity but also that when relevant, ethnicity can have the opposite effect, of producing distance and outsiderness. [46]

Best, Amy L. (2003). Doing race in the context of feminist interviewing: Constructing whiteness through talk. Qualitative Inquiry, 9(6), 895-914.

Brubaker, Rogers (2002). Ethnicity without groups. Archives Européennes de Sociologie, 43(2), 163-189.

Brubaker, Rogers; Feischmidt, Margit; Fox, Jon E. & Grancea, Liana (2006). Nationalist politics and everyday ethnicity in a Transylvanian town. Princeton, NJ: Princeton University Press.

Bucerius, Sandra Meike (2013). Becoming a "trusted outsider": Gender, ethnicity, and inequality in ethnographic research. Journal of Contemporary Ethnography, 42(6), 690-721.

Carling, Jørgen; Erdal, Marta Bivand & Ezzati, Rojan (2014). Beyond the insider-outsider divide in migration research. Migration Studies, 2(1), 36-54.

Chereni, Admire (2014). Positionality and collaboration during fieldwork: Insights from research with co-nationals living abroad. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 15(3), Art. 11, http://nbn-resolving.de/urn:nbn:de:0114-fqs1403111 [Accessed: February 20, 2015].

De Andrade, Lelia Lomba (2000). Negotiating from the inside: Constructing racial and ethnic identity in qualitative research. Journal of Contemporary Ethnography, 29(3), 268-290.

Eve, Michael (2010). Integrating via networks: Foreigners and others. Ethnic and Racial Studies, 33(7), 1231-1248.

Fox, Jon E. (2013). The uses of racism: Whitewashing new Europeans in the UK. Ethnic and Racial Studies, 36(11), 1871-1889.

Ganga, Deianira & Scott, Sam (2006). Cultural "insiders" and the issue of positionality in qualitative migration research: Moving "across" and moving "along" researcher-participant divides. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 7(3), Art. 7, http://nbn-resolving.de/urn:nbn:de:0114-fqs060379 [Accessed: February 20, 2015].

Gans, Herbert (1997). Toward a reconciliation of "assimilation" and "pluralism": The interplay of acculturation and ethnic retention. International Migration Review, 31(4), 875-892.

Glick Schiller, Nina (2008). Beyond methodological ethnicity: Local and transnational pathways of immigrant incorporation. Willy Brandt Series of Working Papers in International Migration and Ethnic Relations, 2, Malmö University, https://dspace.mah.se/bitstream/handle/2043/7491/WB%202_08%20MUEP.pdf?sequence=3 [Accessed: February 20, 2015].

Glick Schiller, Nina & Çağlar, Ayşe (2013). Locating migrant pathways of economic emplacement: Thinking beyond the ethnic lens. Ethnicities, 13(4), 494-514.

Glick Schiller, Nina; Çağlar, Ayşe & Guldbrandsen, Thaddeus C. (2006). Beyond the ethnic lens: Locality, globality, and born-again incorporation. American Ethnologist, 33(4), 612-633.

Halstead, Narmala (2001). Ethnographic encounters. Positionings within and outside the insider frame. Social Anthropology, 9(3), 307-321.

Hoffmann, Elizabeth A. (2007). Open-ended interviews, power, and emotional labour. Journal of Contemporary Ethnography, 36(3), 318-346.

Jiménez, Tomás R. (2010) Replenished ethnicity: Mexican Americans, immigration, and identity. Berkeley, CA: University of California Press.

Kusow, Abdi M. (2003). Beyond indigenous authenticity: Reflections on the insider/outsider debate in immigration research. Symbolic Interaction, 26(4), 591-599.

Margolis, Maxine L. (1994). Little Brazil: An ethnography of Brazilian immigrants in New York City. Princeton, NJ: Princeton University Press.

Menjívar, Cecilia (2000). Fragmented ties: Salvadoran immigrant networks in America. Berkeley, CA: University of California Press.

Merton, Robert K. (1972). Insiders and outsiders: A chapter in the sociology of knowledge. American Journal of Sociology, 78(1), 9-47.

Moroşanu, Laura (2013a). Between fragmented ties and "soul friendships": The cross-border social connections of young Romanians in London. Journal of Ethnic and Migration Studies, 39(3), 353-372.

Moroşanu, Laura (2013b). "We all eat the same bread": The roots and limits of cosmopolitan bridging ties developed by Romanians in London. Ethnic and Racial Studies, 36(12), 2160-2181.

Moroşanu, Laura & Fox, Jon E. (2013). "No smoke without fire": Strategies of coping with stigmatised migrant identities. Ethnicities, 13(4), 438-456.

Nowicka, Magdalena & Cieslik, Anna (2014). Beyond methodological nationalism in insider research with migrants. Migration Studies, 2(1), 1-15.

Nowicka, Magdalena & Ryan, Louise (2015). Beyond insiders and outsiders in migration research: Rejecting a priori commonalities. Introduction to the FQS Thematic Section on "Researcher, Migrant, Woman: Methodological Implications of Multiple Positionalities in Migration Studies". Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 16(2), Art. 18, http://nbn-resolving.de/urn:nbn:de:0114-fqs1502183.

Ochieng, Bertha M.N. (2010). "You know what I mean": The ethical and methodological dilemmas and challenges for black researchers interviewing black families. Qualitative Health Research, 20(12), 1725-1735.

Parutis, Violetta (2014). "Economic migrants" or "middling transnationals"? East European migrants' experiences of work in the UK. International Migration, 52(1), 36-55.

Ryan, Louise (2011). Migrants' social networks and weak ties: Accessing resources and constructing relationships post-migration. The Sociological Review, 59(4), 707-724.

Ryan, Louise; Kofman, Eleonore & Aaron, Pauline (2011). Insiders and outsiders: Working with peer researchers in researching Muslim communities. International Journal of Social Research Methodology, 14(1), 49-60.

Sherif, Bahira (2001). The ambiguity of boundaries in the fieldwork experience: Establishing rapport and negotiating insider/outsider status. Qualitative Inquiry, 7(4), 436-447.

Shinozaki, Kyoko (2012). Transnational dynamics in researching migrants: Self-reflexivity and boundary-drawing in fieldwork. Ethnic and Racial Studies, 35(10), 1810-1827.

Song, Miri & Parker, David (1995). Commonality, difference and the dynamics of disclosure in in-depth interviewing. Sociology, 29(2), 241-256.

Vamanu, Alina & Vamanu, Iulian (2013). "Scandalous ethnicity" and "victimized ethnonationalism": Pejorative representations of Roma in the Romanian mainstream media after January 2007. In Jan Kubik & Amy Linch (Eds.), Post-communism from within: Social justice, mobilization, and hegemony (pp.265-292). New York: New York University Press.

Wimmer, Andreas & Glick Schiller, Nina (2003). Methodological nationalism, the social sciences, and the study of migration: An essay in historical epistemology. International Migration Review, 37(3), 576-610.

Yow, Valerie (1995). Ethics and interpersonal relationships in oral history research. The Oral History Review, 22(1), 51-66.

Yuval-Davis, Nira (2006). Belonging and the politics of belonging. Patterns of Prejudice, 40(3), 197-214.

Zinn, Maxine Boca (1979). Field research in minority communities: Ethical, methodological and political observations by an insider. Social Problems, 27(2), 209-219.

Laura MOROŞANU, Ph.D. is a lecturer in sociology at the University of Sussex. Her main research interests are in the areas of migration and ethnicity, with a focus on migrants' social relationships and socialisation practices.

School of Law, Politics and Sociology
Freeman Centre G34
University of Sussex
Brighton, BN1 9QE
United Kingdom

Moroşanu, Laura (2015). Researching Coethnic Migrants: Privileges and Puzzles of "Insiderness" [46 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 16(2), Art. 18,
http://nbn-resolving.de/urn:nbn:de:0114-fqs1502150.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License