Opportunities and Challenges of Unplanned Follow-up Interviews: Experiences with Polish Migrants in London

Abstract: Although there is growing interest in qualitative longitudinal research as a way of taking time seriously (ADAM, 2000), this approach still holds many challenges for the social researcher. In this article we use a reflexive approach, drawing on a Goffmanian analysis of self-presentation, to consider our separate but related experience of re-interviewing Polish migrants over intervals of several years. In each case, the repeat interviews were not part of the original research design and were undertaken years later for a range of different reasons. After briefly presenting case studies from our individual interviews, we critically reflect upon some opportunities and challenges of researching change through time. We first consider the ways in which repeat interviews may challenge earlier analyses and findings. We then explore some of the ethical considerations involved in unplanned repeat interviews. Next, we reflect upon dilemmas about self-revelation, particularly in contexts of social media and on-line technologies. Finally, we discuss what we have learned from our different experiences and what implications there are for this kind of ad hoc longitudinal research in migration studies.

Key words: unplanned repeat interviews; reflexivity; presenting the self; migration research

4.1 Louise RYAN on re-interviewing Ewa: Presentation of self in interview encounters

Following the "temporal turn in the social sciences" (THOMSON & McLEOD, 2015, p.244), there is increasing interest in how to "take time seriously" (ADAM, 2000, p.126) in social research. Time cannot be simply included in research as a "neutral medium in which events take place" (ibid.). Instead, we need to understand the complex and subjective dimensions of temporality as a social construct. While a one-off interview encounter may represent a snap shot frozen in time, repeat interviews may help to capture temporal and spatial dynamics that shape individual narratives. Longitudinal research is increasingly valued by funders and holds much promise for the social sciences (THOMSON & McLEOD, 2015). Longitudinal data allows us "to construct a story of a life, which is greater than the sum of its parts (the individual interview) making it possible to gain insight into underlying biographical and social processes at play" (THOMSON, 2007, p.577). [1]

Nonetheless, as we discuss in this article, repeat interviews raise a number of methodological, ethical and epistemological challenges. Interviewing a participant twice in several years does not necessarily address all the shifting nuances and multi-layered complexity of temporality. Constructing a "retrospective story of a life" presents challenges regarding the chronology of writing (HENDERSON, HOLLAND, McGRELLIS, SHARPE & THOMSON, 2012, p.25). The past, present and future interact in complex and shifting ways as we piece together a story told over several interview encounters. There are challenges associated with how to analyse and present data generated from repeat interviews (THOMSON, 2007; THOMSON & HOLLAND, 2003). In addition, repeat interviews raise a number of ethical questions of "confidentiality, privacy and anonymity generated through long term research relationships" (HENDERSON et al., 2012, p.17). [2]

We pay particular attention to the challenge of analysis—how to analyse data gathered from repeat interviews. How may our interpretation of the first interview be challenged or undermined by data generated in a later interview? How is our approach to subsequent interviews shaped by our familiarity with the original data findings? For example, does the first interview frame the questions we ask and topics we wish to explore in all other interviews? Should we present our observations arising from each interview separately or based on the sum of all the interviews? We also consider the ethical issues involved in repeat interviews, as well as the challenge of self-revelation particularly in context of social media. [3]

This article brings together our separate experiences of re-interviewing Polish migrants in London. In all three cases our repeat interviews were not pre-planned as part of the original study design. Given the assumed transience, temporariness and uncertainty of their migration trajectories, in the years immediately following Polish accession to the EU (2004), it would have been difficult to plan a longitudinal study. While much migration research only interviews participants once, there are a few notable exceptions where researchers carried out multiple interviews, either planned or unplanned (inter alia BRADBY, 2002; HOANG & YEOH, 2015; KRINGS, MORIARTY, WICKHAM, BOBEK & SALAMO, 2013; LEUNG, 2015; PILKINGTON, 1998; THOMAS, AGGLETON & ANDERSON, 2010; WHITE, 2011). Drawing on our experiences, we suggest that ad hoc, opportunistic repeat interviews with migrants may be very useful but also pose particular challenges. We use a reflexive approach, paying due attention to the self-presentation and dynamic performances of these interview encounters (GOFFMAN, 1963). [4]

We each present short case studies from our individual research projects and, writing in the first person, reflect on our experiences of ad hoc re-interviewing. In the discussion section, we collectively engage with some risks and challenges associated with re-interviewing including 1. the question of data analysis particularly as part of unplanned longitudinal research, 2. the specific ethical issues associated with this type of research, and 3. self-revelation on the part of the interviewer especially if there has been sustained connection/communication over time. The conclusion considers the possible contribution of this approach to repeat interviewing. But first, we begin with a brief theoretical section to help us critically reflect upon repeat interviews as performative encounters at different moments in time. [5]

Migration not only involves spatial but also temporal movement. Migrants not only move between countries, they also negotiate these environments over time. For example, for Poles who had moved to the UK prior to Poland's EU accession in May 2004, acquiring full rights to live and work in Britain opened up opportunities that were not available to them earlier (TREVENA, 2013). However, capturing these dynamics is challenging for researchers. As McKIE, GREGORY and BOWLBY observe, "the major difficulty is to retain the complexity of time-space at both the levels of the theoretical and the empirical" (2002, p.907). In exploring the social constructedness of time, McKIE et al. highlight the salience of different temporalities—such as the rhythms of life cycle, working patterns or daily routines. Time may be experienced and negotiated differently according to, for example, gender expectations, employment obligations, cultural traditions and spatially specific opportunity structures. Exploring these processes in a research interview means paying attention to the ways in which time is conceptualised and narrated by participants as well as how time, and change over time, are framed by the researcher's questions. [6]

Given the complexity of time, there are questions about how social scientists can adequately study the process and experience of change over time (CORDEN & MILLAR, 2007). ADAM (2000) has suggested timescapes to encompass quantitative time, the connections between space and time, and the multidimensionality of time experienced at different levels. The concept of timescapes provides new ways of thinking about and researching social processes and change drawing on established approaches such as qualitative longitudinal research (NEALE, HENWOOD & HOLLAND, 2012). The theoretical innovation of timescapes involves the combination of micro (biographical), meso (generational) and macro (structural) dimensions of time in order to understand "dynamic relationships between individual and collective lives and broader patterns of social change" (p.5). [7]

Constructions of time underscore research encounters through interlinked performances of storytelling, performativity, memory, re-telling and self-censorship. As Julia BRANNEN notes: "life stories are typically created vertically through time" involving "chronologies or sequences of events that are linked together" and interspersed with evaluation that is "recounted from present time perspectives and with hindsight" (2013, §2.3). Carrying out social research, especially face-to-face encounters with participants, encompasses numerous complex and shifting boundaries (GANGA & SCOTT, 2006; NOWICKA & RYAN, 2015; RYAN & GOLDEN, 2006). Reflexivity can help to reveal the intricate, multi-dimensional positionalities that underpin the research relationship (MAUTHNER & DOUCET, 2003). The necessity of exploring these positionalities takes on even more urgency in the case of repeat interviews "where the multiple temporality of biographical data gathered at intervals over time introduces additional complexities" (HENDERSON et al., 2012, p.17). [8]

In theorising repeated meetings with participants, we draw upon a Goffmanian perspective to explore the interview/re-interview as a dramaturgical encounter involving multiple, interactive presentations of self through a process of "information control" and readable signs (GOFFMAN, 1963). According to GOFFMAN, an interaction may be defined as "the reciprocal influence of individuals upon one another's actions when in one another's immediate physical presence" (1971, p.26). Using the metaphor of the stage, he considers the "performance" of an individual in relation to an audience of observers. Such "performance" is also part of interview interactions and we return to this discussion in our analysis. [9]

As presented in the following data sections, the interview encounter is a process of individual "identity work" and mutual identity construction. It is not always easy or indeed possible to predict how one will be placed by the participants (RYAN, 2015a). Far from being stable, these identities may be re-formed and re-shaped throughout the encounter as various verbal and non-verbal clues are used to piece together a sense of the other actor. Na'amah RAZON and Karen ROSS (2012) refer to the fluidity of identities throughout the research encounter: a dance in which both parties attempt to size up each other. This "research dance" (LIN, 2014) takes on added complexity when relations are sustained or re-activated over time such that researchers and interviewees develop a different kind of relationship and find out more information about each other. Before presenting our research findings, we summarise our methods. [10]

Although we three worked independently, our studies share a common approach of unplanned re-interviewing Polish migrants living in London after a considerable interval of time—in some cases eight years after the first interview encounter. Our discussion draws on data collected between 2004 and 2014. [11]

Louise RYAN conducted a study on Polish migration in 2006-07 for which 30 participants were interviewed (in addition there were three focus groups and several key informant interviews—thus over 50 people took part in that initial study). The results of that study were analysed and published in several academic papers (RYAN, SALES, TILKI & SIARA, 2008). In 2014, Louise carried out a project involving Polish migrants who had been resident in the UK for approximately ten years (RYAN, 2015b). As part of that study 20 participants were interviewed of whom 9 had been interviewed previously. The later follow up study had not been initially planned and Louise had not maintained contact with participants during the intervening years. The key aims of the later study were to understand how migration plans, especially initial expressions of temporariness and uncertainty, may develop over time into longer term settlement. In particular Louise wished to understand what factors may enable or hinder processes of embedding into London society (ibid.). [12]

Magdalena LOPEZ RODRIGUEZ's analysis is based on repeated interviews conducted for her PhD research for which she interviewed 40 Polish mothers living in the UK, almost all of them in London (LOPEZ RODRIGUEZ, 2010). There were two waves of semi-structured, in-depth interviews (2005/2006 and 2008/2009). In the second wave, 10 out of the initial 40 participants were re-interviewed. Her study focused on how "capital" influences the Polish mothers' strategies with respect to their children's education throughout the migratory process. The repeated interviews focused specifically on life changes and identity transformations but they were not originally planned as part of the research design. However, observing how mothers' experiences change through the life course, Magdalena decided to undertake some follow-up interviews. For the second set of interviews those mothers whose children transferred, or would soon transfer, to secondary schools were selected. Most of the re-interviewed mothers had stayed in touch with Magdalena, following the first interview, and thus had some kind of interaction with her between the two interviews. [13]

Paulina TREVENA carried out a case study of young Polish graduates living in London and working in low-skilled jobs for the purposes of her PhD (TREVENA, 2013). The first wave of interviews was carried out in 2004 with a view to exploring how working below qualifications impacts on the migrants' identity, feelings of well-being and plans for the future, yet the interview itself was of a much broader nature and did not focus on work issues in particular. Receiving additional funding facilitated a second wave of interviews in 2007-2009; the rationale behind re-interviewing was to see whether, how and why original study participants had progressed in the British labour market. Of the original 21 study participants, 14 were re-interviewed after three plus years (with seven new interviewees, the study consisted of 42 interviews). Paulina had maintained close contacts with several (but not all) of the participants following the first interview. [14]

Thus, between us, we generated a large body of data about these participants over a considerable period of time. Faced with this massive corpus of data, we decided to adopt a case study approach. As NEALE et al. observe, in longitudinal research case studies are a good way to "condense cumulative data into a meaningful narrative that carries interpretation and analysis" (2012, p.8). The individual case study can locate subjectivity in a cultural and historical context (THOMSON, 2007). Hence, these case studies are not written as "static case profiles" but rather as "dynamic case histories" (NEALE et al., 2012, p.8). For the purposes of this article, in order to allow for an in-depth, reflexive approach, we focus on three participants each of whom was interviewed twice. We have selected these particular case studies because they each depict interesting and thought provoking issues that are pertinent to the discussion in this article. While each is unique in its own way, they also reflect wider patterns found across each of our individual data sets. To explore our individual reflexivity, we switch to the first person in the case studies. [15]

4.1 Louise RYAN on re-interviewing Ewa: Presentation of self in interview encounters

I interviewed Ewa¹⁾, aged 30, for the first time in the summer of 2006. Having completed a Master's degree in Poland, Ewa "wasn't too happy" with her career prospects there. Like many Polish young people in the early 2000s, she decided to come to London. She arrived in 2002, with quite short term plans—"my plan was to stay 3 months, up to 6 months". However, she quickly decided to extend her stay because "I really felt so good here in London that I didn't want to go back to the sad reality in Poland, struggling". [16]

Throughout the interview Ewa constructed "England" as "a country of lots of opportunities" where "I have better possibilities". Nonetheless, she also expressed frustration: "I find it difficult, I had to go down in my career, to the bottom, very bottom". Having started off volunteering for a charity, at the time of the first interview she was working as a personal assistant (PA) in another charity. Although she described this as a "proper" job it was quite clear that this was not her ideal career—"But I am a sociologist, that is what I want to do". There was tension between the impression Ewa sought to convey and the image of herself given off by her PA job. [17]

She related an anecdote about being undermined in the workplace. A former manager regularly diminished Ewa in front of colleagues by making negative remarks about her lack of English fluency: "ok, let's start the meeting, who will take minutes, not Ewa, because she can't understand a word we are saying". [18]

Because of nationality, migrant status, language/accent, an actor may become stigmatised, labelled and stereotyped; resulting in discrimination and loss of status (GOFFMAN, 1963). However, rather than "remain passive in the face of potential meanings" generated against them (GOFFMAN, 1961, p.104), actors seek consistency with their former frame of reference and sense of self (RYAN, 2011; TREVENA, 2010a). Thus, by asserting that "I am a sociologist", Ewa drew on a frame to maintain a sense of continuity with her self-identity, in the fluid and unpredictable migratory context. [19]

In the summer of 2014 I re-interviewed Ewa after a gap of eight years. Second interviews will always to some extent be framed in relation to the previous encounter (HENDERSON et al., 2012). One of the key aims of a repeat interview is to assess what has changed since the first interview. It is all too easy therefore to construct a linear narrative along a clear diachronic axis. As THOMSON and HOLLAND (2003) warn, the research design of repeat interviews leads to a focus on "development" over time. I am aware that I was looking for change but of course the complex interplay between continuity and change both temporal and spatial form key dynamics in repeat interviews. [20]

Re-reading the new transcript, I was struck by how I had immediately framed the second interview in relation to the first: "when I last met you in the summer of 2006 ... you were unsure how you wanted your career to develop and now you're married and have children". So I set the tone by measuring development since the last interview not just in terms of passing time (since 2006) but also in terms of key life events (marriage and motherhood). Ewa then took up the narrative and went back to explain how events unfolded during the interim period. [21]

She spoke at length, without interruption from me, explaining what had happened since we met. Interestingly, she did not use dates to frame her narrative but rather highlighted key life events—such as changing jobs, getting promoted, getting married and having children. In terms of temporality, it was sometimes difficult for me to understand when specific events had occurred and I occasionally sought clarification about what year a particular thing happened. Thus I was attempting to impose "clock time" (see MELUCCU cited in McKIE et al., 2002, p.904) on her deeply personal temporal narrative of life cycle and significant events. [22]

Her account of the period between our interviews was framed by challenges encountered and overcome in developing her career. She explained how she had "been searching for possibilities but it's not easy". It is noteworthy that she repeated the word "possibilities", which had been such a strong feature of the first interview. She had encountered obstacles as people were sceptical of "only a Polish degree". Despite this she was determined to build up her career—"I was really so desperate to succeed". According to THOMSON, the longitudinal study is not simply a "chronological description of change but an exploration of how the individual is involved in creating that change, in a specific temporal and social context" (2007, p.578). While working full time as a PA, Ewa undertook several self-funded training courses. Eventually, she got a "lucky break" and was offered a job more fitting to her qualifications and ambitions. Building on that job, she moved to another organisation where she is "heading a team" and "in a senior position". [23]

However, while there was remarkable consistency and clarity in how she recalled and presented the past years, there were also some potential inconsistencies. Throughout the first interview, Ewa repeatedly presented herself as quite independent: "I want to be my own person". She emphasised self-reliance: "I don't have family or children, I can easily make it by myself". She used the evocative phrase: "I am quite an ivy person, put me somewhere and I will put down roots and grow". By contrast, in the second interview, she highlighted the role of family life and parenthood in facilitating settlement: "I think it is impossible to really settle without starting a family". Thus perhaps one could argue the "ivy person" needs a fence, or some support, to grow alongside. There is no right or wrong interpretation here. Rather, her observations reflect changing socio-temporal perspectives in the context of key life changing events. We return to the role of changing life circumstances in the next two case studies and later in the discussion section. [24]

I interviewed Nina for the first time in 2005, when her son, Bruno, was 7 years old. The second interview took place in 2008 when Bruno was 10 and about to transfer to secondary school. At the time of our first interview he was being assessed by various specialists who were trying to identify his social and learning difficulties. Three years later, he was "statemented"²⁾ as having Asperger's syndrome. When I met Nina for the first time she planned to rent a studio flat in order to be within the catchment area of a school, rated by Ofsted³⁾ as "outstanding". Albeit insecure, her and her partner's jobs were well remunerated and at the time it seemed a viable strategy. She lived near two other comprehensive schools that were not considered an option for her son's secondary education. Yet, when I met her for a follow-up interview three years later, both she and her partner were unemployed and renting a studio in a "better area" was beyond their means. Consequently, they had applied and been given a place for Bruno at one of the previously rejected local comprehensive schools. In interviewing, time, even in RICOEUR's less linear conceptualisation⁴⁾, often goes unquestioned (FATHI, 2014; GASPANI, 2014). However in my interview, temporality and futurity were substantially ruptured by the context of the situation; there was a noticeable shift in Nina's narration of the pre-planned linearity of events. Although I insisted on structuring the presence by her past and kept a horizontal course of events, she lived in "here and now" and contradicted retrospective successive temporality, which I imposed. [25]

Interviewing mothers for the second time, my strategy usually involved re-reading excerpts of the first interview and applying a "recall technique", whereby a fact, event or situation from a previous interview is brought back into the respondent's mind by the interviewer. Here, Nina's past mind-set was brought back to her and meant to be reflected upon by her in order to trigger further insights. It was only when I started analysing this interview with a methodological mind-set that I realised to what extent this "recall technique" influenced Nina's discourse. Using this technique with her, for example, I unwittingly framed the interview through "negative" recall. In the beginning of the interview I mentioned that I remembered she had been facing certain "problems" and, in particular, I explicitly problematised the process of statementing Bruno:

"How has it been? I remember 3 years ago you had quite a lot of concerns because you were not sure about Bruno‘s abilities, literacy etc.. I remember from our last interview...I remember you mentioned to me that there was an issue of whether Bruno would be statemented or not". [26]

BRANNEN (2013) points out that interview questions can act as hooks upon which people choose to hang their stories. Following my "negative" recall Nina spoke for approximately 15 minutes about the challenging process of her son being statemented and about the slow pace of referrals, mainly listing problems she faced and describing how she was searching for solutions. However, now reflecting on this, I am aware that were it not for my negative recall, the interview could possibly have focused on a different fragment of Nina's reality, possibly more corroborative and affirmative and subsequently may have followed a different trajectory. [27]

Regarding recall, another issue to consider is the ethical dimension. When conducting Nina's interview I asked her: "3 years ago I asked you about aspirations for your son ... what would you say today?" Nina, appearing slightly vexed, answered: "I don't think I remember what I said". At this point, I explained to Nina that the focus of my research was on possible changes in interviewees' lives so she would feel free to break away from the linearity of her narrative and not worry that she was giving me the "wrong answer". Prompting my interviewees to position themselves in relation to their previous statements has at times triggered consternation or surprise and resulted in interviewees' feeling a sense of intimidation as they felt almost obliged to recall exactly the same facts and feelings which surfaced in the first interview. [28]

Finally, throughout the second interview the mothers had been reflecting on their own lives—"recalling life". As GOFFMAN (1961) observed, in social interaction, just as in dramaturgy, there is a front stage where the individuals perform positive aspects of the self and convey the desired impressions by changing or fixing setting, appearance or manner, as if "rejecting an image of the self as abnormal" (p.50). Such performativity of success is constructed in BRANNEN's words by "adding positive gloss on an unfortunate life" (2013, p.5). I noted it was somewhat uncomfortable for participants to discuss lack of progress and development in their own biographies and they were consistently recalling "stepping up" rather than "stepping down" and always positioned themselves in a superior position in relation to their circumstances at the first interview. This was apparent in the case of Nina, despite the clear challenges she had encountered during this period. As a researcher, I recaptured those facts and remembrances that were most illustrative for the purpose of my constructed and, consequently, re-constructed discourse. Inadvertently, I recalled particular events and issues, and hence mothers might have scaffolded their new biographical narratives around discussion points that I found critical for my study. Needless to say, there could have been other significant and pressing aspects of their life which were not mentioned or explored. [29]

I initially interviewed Łucja in summer 2004. Łucja, who holds an MSc degree in Biology, was deeply disappointed by lack of career opportunities in Poland. Therefore, after a couple of years of working in Poland she decided to leave for London in order to learn English. Arriving in London in 2000, Łucja first worked as a nanny, then in coffee shops and restaurants, and when I first met her she was working as a sales assistant in a clothes shop. Although Łucja made it clear that she could not stand her "stupefying" job or co-workers, she presented it as a matter of lifestyle choice and seemed to be "ok" about it:

"As for today, working in a shop is a matter of my choice. I think of myself that I am aware of what I'm doing and what sort of work I have and I do not get involved emotionally. (...) My shitty job, excuse the expression, is the price I have to pay for other things". [30]

Łucja, who used to have high educational ambitions (she was even considering doing a PhD in Poland) explained that she had "re-evaluated" her earlier views on professional status:

"When I look at these people [who are in higher positions], I feel sorry for them, what do they need it all for? They're killing themselves just to be able to write ‘manager' or something on their visit cards (...). Meanwhile, I just do 5 hours and have the whole afternoon to myself and this is great. (...) The work I'm doing now is absolutely stress-free (...). I have no responsibility whatsoever, and don't want to have any". [31]

Though in the interview she mentioned thinking of trying to move into nature conservation (her ideal career) at some point ("I'm not saying no, maybe one day"), she saw this as out of her reach at that time. Therefore, she presented it as a distant (and not necessarily plausible) possibility for the future, whereas the present was defined by focusing on hobbies and being content with having no responsibility at work. [32]

Such a strategy of disengaging oneself from one's work role is a very clear example of what GOFFMAN (1972) refers to as "role distance" and was a popular strategy among the group I was researching (i.e., highly educated migrants working in low skilled jobs) (TREVENA, 2010a, 2010b). [33]

I had not been in touch with Łucja since the first interview (but we had mutual friends who facilitated later contact). Therefore, when I went back to interview her in 2008, I did not know about the profound changes that had taken place in her life—both professionally and personally. By that time, Łucja had embarked on her dream career and had got engaged. I was completely taken aback and pleasantly surprised by the extent of these changes. I observed how these changes impacted on how Łucja recalled past events. As opposed to Louise RYAN and Magdalena LOPEZ RODRIGUEZ, I tried to avoid referring to what had been said in the previous interview in the initial stages of my follow-up interview; I did not want to immediately frame the follow-up interviews in relation to their key focus (i.e., work-related issues. Hence, my opening question was: "Last time we met was over three years ago. What has happened in this time that you think was important to you?" My interviewees were thus free to speak about any aspect of their lives they chose (not necessarily work). Therefore, it surprised me that Łucja immediately embarked on the story of changes in her professional life. She was now working as a project manager in a large London park. The job carried a lot of responsibility but she was happy and feeling confident about her professional future. Therefore, while my impression from our 2004 interview was that Łucja had chosen to build her world around "post-materialist values" (INGLEHART, 1997) and was generally content with this decision, her perception of her life at that time was now quite different:

"In 2005 I was on the verge of a nervous breakdown. At the beginning [of working in the clothes shop] it was okay. I was still learning the language a little and something else on the way. But later it stopped being okay. And this stage, when things were no longer okay lasted for quite long ... I suddenly reached enlightenment. I came to the conclusion that I simply must find some volunteering job in nature conservation". [34]

Łucja went on to tell me how she had gone to great lengths (including getting into debt) in order to get a volunteering post at a nature park in another part of England, and how she eventually managed to move into her ideal career by taking such a step. Her discourse around issues and values related to work had changed strikingly:

"I am very happy. Above all I am working in a sector I have always wanted to work in, which is nature conservation. And I am working with people, for me this is inseparable. (...) This work provides me with an excellent combination of everything I would expect from a job. And apart from this I feel I am developing all the time, that I'm learning, and this is probably most important to me". [35]

These stark differences between how Łucja spoke about work and its role in her life in the two interviews made me realise just how important spatial and temporal contextual issues and changing life circumstances are for how stories are framed and how one's life is understood. As BRANNEN (2013) notes, the past is retold and re-evaluated in relation to the present context and Łucja's case illustrates this point very well. The way she presented herself at the time of the first interview, her thoughts and her life was not a matter of a "role" and the story being a "creation"; rather, it was a matter of how she was preserving her identity or perhaps even sanity. Three years later, she was in her ideal job and work had become a pivotal part of her life and of her identity; therefore, her story was a reflection of this fundamental change in circumstances. [36]

Having presented three short case studies from our separate studies, in the next sections we now consider the implications of our experiences of re-interviewing in terms of risks, recall and ethics. [37]

A risk in carrying out repeat interviews is the challenge to earlier interpretations of the data. As noted, there is on-going discussion about how data generated from repeat interviews should be analysed and presented (THOMSON, 2007). Should each research encounter be analysed separately or should the transcripts be combined into one enlarged mass of data and analysed together? Indeed, this is a problem all three authors encountered when carrying out repeat interviews—especially as second interviews were not originally part of the research design. Because none of us had planned longitudinal research, we had analysed our original data separately, and presented and/or published papers on it (LOPEZ RODRIGUEZ, 2010; RYAN, 2011; TREVENA, 2010a; 2010b). Thus, re-interviewing participants after we had already written about them risked undermining our original findings. Indeed, as observed above, the changes between the two interviews were substantial. Nonetheless, each of us decided to treat the second interview as a related but separate research encounter, analysed in a separate stage, allowing for changes that may have occurred over time. [38]

For instance, in the case of Louise's interviews, while Ewa had initially presented herself as a very independent, self-reliant person who did not need other people—an "ivy person"—in the second interview she strongly highlighted the key role of family life and parenthood in facilitating settlement. While this may appear to challenge or even undermine the "ivy person" that Louise had already published about (RYAN, 2011), instead, we see this as an illustration of dynamism and fluidity over time. Similarly, in the case of Paulina's interview with Łucja, the first interview had given Paulina a completely different impression of Łucja's outlook on work and values as well as her career prospects than apparent in the second interview. Magdalena's experiences of interviewing Nina also draw attention to life's contingency. After their first interview, Magdalena was under the impression that Nina had a very good opinion of the English education system and had everything "worked out", only to discover a few years later that Nina felt badly let down by that system and her high hopes had been dashed. [39]

Does the new data from the second interview undermine our earlier findings? It should be noted that we are not undertaking repeat interviews as a way of "validating" our initial findings, instead we are interested in how repeat interviews highlight the contingency of individual accounts and interpretations. We are particularly mindful of temporal dynamics in the research process (NEALE et al., 2012), especially when collecting data over an extended period of time. Although our research was not designed as longitudinal, we argue that adopting a "timescapes" lens is useful in helping us to understand dynamic social processes and experiences (ADAM, 2000). Thus, we are not approaching time as some objective reality in which to assess change between two fixed points on a calendar. Rather, we are concerned with the subjective (micro) experiences of change in relation to inter-personal (meso) and wider structural (macro) contexts. [40]

In addition, as we discuss in the next section, carrying out repeat interviews raises specific ethical challenges. [41]

Repeat interviews raise a number of ethical considerations (NEALE & HANNA, 2012). A key consideration is anonymity. Because repeat interviews generate a considerable amount of data on one individual, it is easy to expose so much detail about them that their anonymity is jeopardised (NEALE & BISHOP, 2012). Therefore, in writing up our research results we took extra care to edit or remove particular details that might have made our interviewees recognisable. [42]

Another ethical issue arises from comparing contexts, situations and views over time. Repeat interviews encourage participants to reflect on how their lives have changed over time yet as PINI and WALKERDINE (2011) observed, replaying a certain version of someone's life back to them may be unsettling. This might be especially true where there may be an expectation (on the part of the interviewee, interviewer or both) of "progress", some improvement since the previous interview. In cases where interviewees feel such progress has not been achieved, feelings of discomfort or even failure may arise. Considering this, it is possible that those with a "good story" to tell may be more inclined to agree to be re-interviewed. Nevertheless, Paulina's experience of re-interviewing highly educated migrants working in low-skilled jobs demonstrated that this need not necessarily be the case. Still, she observed that study participants who had not made much "progress" in the years between the two interviews indeed often displayed discomfort and were not willing to speak about certain issues. On the other hand, some participants thanked her for "forcing" them to re-consider their situation: "You made me think about things I was usually avoiding to think about, you made me reflect on my position, which I had been avoiding for some time". This, in turn, raises issues around how the interview encounter may impact on the participants' perception of their own life and, ultimately, certain steps they take. [43]

With regards to relations with study participants, repeat interviews also raise specific challenges around the ethical representation and use of data. After the first round of research, Paulina had not envisaged follow-up interviews and hence was not thinking about keeping professional boundaries over an extended period of time. As Paulina had much in common with many of her interviewees, she became friends with several of them and maintained contact, including visits to one another's houses. Several participants were very interested in the research results and repeatedly told her they would like to see these. Therefore, a few months after the first round of interviews, Paulina sent them a conference paper, written on the basis of her analysis. One of the interviewees became very offended by what was written and stated that they had "nothing to do with these people" and that Paulina was wrong in her thinking. This created at least two ethical issues which Paulina is still working through. First, when dealing with potentially sensitive issues, should a researcher share the results of analysis with the study participants (especially if there is a direct request from participants)? Second, should a researcher (and if so to what extent) consider the study participants' feedback on interpretations of the data (see MORSE, HUTCHINSON & PENROD, 1998)? These challenges are particularly pertinent in repeat interviews where researchers and participants meet on more than one occasion and maintaining a positive and friendly relationship is key to the success of a study. [44]

While all research encounters raise dilemmas about how much of themselves researchers should share and reveal to participants (NOWICKA & RYAN, 2015), this is particularly complicated when it comes to repeat interviews. Meeting participants over a period of time may lead to growing trust and familiarity between researchers and interviewees. Hence, the issue of how to maintain professional boundaries while managing such relationships over time becomes especially salient (NEALE & HANNA, 2012). This may be an even greater challenge when carrying out ad hoc longitudinal research. For instance, following their first waves of interviews, Magdalena LOPEZ RODRIGUEZ and Paulina TREVENA had stayed in touch with some of the original study participants; over time these relationships had in fact evolved into friendships. While nurturing long-term relationships is generally seen as a key strength in qualitative longitudinal research, in our case it brought about additional challenges, such as the impact of revealing one's own experiences and beliefs on subsequent research encounters. [45]

In these terms the experiences of Magdalena are particularly interesting. As noted, she did not originally intend to do repeat interviews and due to the multiple connections established following the first interviews, several of her interviewees had become her on-line "friends". At the time, Magdalena had not considered the possible impact of this on her interviewees' attitudes and perceptions. Magdalena had been posting about her own children's educational experiences of schooling across different cultures, and frequently shared media articles about social and educational issues. In effect, those respondents who had become her social media "friends" were generally aware of her stance on education, and to some extent familiar with how she personally coped with problems around her children's schooling. The obvious question that follows is if and how this influenced the responses and discourse constructed during the second interview. [46]

Revealing her identity as a mother, facing similar issues as the participants facilitated feelings of trust and having a common ground. Magdalena had occasionally interacted with Nina between the two interviews when they dropped off their children at the Polish Saturday school. It was evident during their second interview that Nina had formed assumptions about how much Magdalena knew and what she thought about various topics tackled in her own study. Like Nina, other mothers who became Magdalena's friends between the first and second interview frequently expressed assumptions of shared background or views by saying "it was like that with us ... you remember" or "well, you know what I am talking about". However, in the initial interviews, before they learned about Magdalena's personal situation, the interviewees would speak about their experiences as if portraying a whole new world to her. Therefore, revealing her full identity influenced the quality of the data as interviewees assumed that they had common experiences and expectations, and that rhetoric around schooling did not need to be spelled out. [47]

Magdalena's is a clear example of the complex and dynamic positioning that goes on in interview encounters (LIN, 2014; RAZON & ROSS, 2012), especially where the implications on subsequent interactions of revealing the researcher's personal circumstances and views were not initially considered. The participants are performing a particular role in the interview encounter but they are also placing and positioning us in particular ways based on their growing knowledge about us as individuals—both professionally and personally. As exemplified above, this may influence their discourse to some degree. In the case of ad hoc repeat interviews, our respondents might have learnt more about us than we would have wished to reveal had we envisaged further research interactions. Therefore, unplanned qualitative longitudinal research raises particular issues around revelation and requires a highly reflexive approach towards the interview encounter as well as analysing data. [48]

As argued elsewhere in the literature (HENDERSON et al., 2012) longitudinal research, especially repeat in-depth interviews, offers many advantages to researchers. Rather than a single moment in time, it attends to temporality (THOMSON & McLEOD, 2015, p.247) and so captures the dynamics of participants' (and researchers') changing life experiences. Coming together to do this article arose from a conference on migration⁵⁾. Writing together and sharing our experiences has pushed us to be more reflexive and to think more critically about the re-interviewing process. Comparing and contrasting our experiences has been very fruitful for us and led to new insights. [49]

At the start of this article we asked some key questions about how to analyse data from repeat interviews and how second interviews may be framed in relation to first interviews. We sought to explore those questions by bringing together our different experiences of interviewing and re-interviewing Polish migrants in London. What does our combined experience add to existing knowledge on repeat interviewing particularly when those are unplanned, ad hoc and somewhat opportunistic? Hence, the key contributions of this article relate firstly, to reflections on doing unplanned, follow up interviews and secondly, to research in migration studies. [50]

Firstly, while much has been written about the advantages of qualitative longitudinal research, our article seeks to contribute to knowledge by presenting our critical reflections on undertaking unplanned, ad hoc and fairly opportunistic repeat interviews after an interval of several years. As we discussed here, there is a particular challenge of collecting new data after the original corpus of data was analysed and published. While this risks undermining original findings, we argue that repeat interviewing, nonetheless, is a worthwhile process. Although these are separate research encounters, second interviews are always to some extent framed in relation to the first interview both by us and by the interviewee—though perhaps in different ways. Using a Goffmanian approach, we are reflective about how self-presentation may change between interviews. In terms of analysis, we suggest a phased approach whereby each stage of data are analysed separately and later synthesised to allow new insights to emerge by paying attention to changes and shifts over time. We are interested in consistencies but also inconsistencies and even contradictions in the data. In this way we do not seek to "validate" our earlier findings but rather to appreciate the complexity and multi-dimensionality of personal narratives within changing temporal and spatial contexts. This attests to the contingency of data and the non-linearity of life stories told over different interview encounters. [51]

Secondly, we suggest repeat interviews have a particular contribution to make to migration research. Migration involves a series of personal transitions, often in contexts of changing structural circumstances (e.g., political or economic situations in both countries of origin and destination). However, in most research projects, migrants are only interviewed once, often at the early stages of their migration trajectory. Attending to the complexity of temporality, we have illustrated how participants' stories may unfold in unanticipated ways. Time is not just a backdrop on which things happen (ADAM, 2000). Going beyond a simple, quantifiable clock time, we have used a longitudinal approach to take the subjective, contextual dimensions of temporality seriously. Thus we are attentive to how participants present their personal (micro) narratives, through changing interpersonal (meso) life events, while being mindful of how these are located in the wider structural (macro) contexts. Hence, migrants are active agents engaged in dynamic social processes within broader institutional structures. As we show, predictions and plans made during early interviews are not necessarily a good indication of how migrants' lives will evolve over time. Therefore, we urge circumspection in drawing firm conclusions about migratory plans and experiences based on one interview. That is not to suggest that two interviews spaced over three or even eight years is sufficient to capture a life time, but as we show, even two interviews can add a sense of dynamism. In her work Maggi LEUNG (2015) recounts repeatedly interviewing a Chinese migrant in Germany over many years and how each interview added new information and layers to his story. [52]

In conclusion, we strongly urge migration researchers to consider adding a longitudinal dimension to their projects. While we recognise that funding constraints place limits on what is possible, every effort should be made to follow up a sample of participants. Clearly, our approach is not without its challenges, many of which stem precisely from its unplanned nature. Any researcher embarking on unplanned, ad hoc, repeat interviews, needs to be mindful of practical, ethical and methodological challenges. Nonetheless, while repeat interviews carry some possible risks, as long as we accept the contingency of individual accounts and interpretations, the benefits of repeat interviews outweigh the risks. [53]

1) The names of all participants have been changed to protect their identity. <back>

2) In the UK, a Statement is a document which sets out a child's special educational needs and any additional help that the child should receive. The aim of the Statement is to make sure that the child gets the adequate support to enable them to make progress in school. <back>

3) Ofsted (Office for Standards in Education, Children's Services and Skills) inspects and regulates the services of educational providers in England. <back>

4) RICOEUR's (1984) concept of "human time" consists of a complex experience in which phenomenological (time experienced in terms of the past, present and future) and cosmological (linear) time are integrated. <back>

5) This conference was organised by Professor Anne WHITE at University College London in 2015. <back>

Adam, Barbara (2000). The temporal gaze: The challenge for social theory in the context of GM food. The British Journal of Sociology, 51(1), 125-142.

Bradby, Hannah (2002). Translating culture and language: A research note on multilingual settings. Sociology of Health & Illness, 24, 842-855.

Brannen, Julia (2013). Life story talk: Some reflections on narrative in qualitative interviews. Sociological Research Online, 18(2), 15. http://www.socresonline.org.uk/18/2/15.html [Accessed: April 1, 2016].

Corden, Anne & Millar, Jane (2007). Time and change: A review of the qualitative longitudinal research literature for social policy. Social Policy and Society, 6(4), 583-592.

Fathi, Mastoureh (2014). Reflections on qualitative longitudinal research and two studies with Iranian migrants in the UK. In Lucy Hadfield, Ester McGeeney & Fiona Shirani (Eds.), New frontiers in qualitative longitudinal research: Perspectives of doctoral and early career researchers (pp.92-103). Southampton National Centre for Research Methods Working Paper 9/14, http://eprints.ncrm.ac.uk/3353/2/qlr.pdf [Accessed: April 24, 2016].

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: December 10, 2015].

Gaspani, Fabio (2014). (Spatio-) temporal dimensions of social research: Researchers' qualitative longitudinal experience. In Lucy Hadfield, Ester McGeeney & Fiona Shirani (Eds.), New frontiers in qualitative longitudinal research: Perspectives of doctoral and early career researchers (pp.64-79). National Centre for Research Methods Working Paper 9/14, http://eprints.ncrm.ac.uk/3353/2/qlr.pdf [Accessed: April 24, 2016].

Goffman, Erving (1961). Encounters: Two studies in the sociology of interaction. London: Penguin books.

Goffman, Erving (1963). Stigma: Management of a spoiled identity. Englewood Cliffs, NJ: Prentice-Hall.

Goffman, Erving (1971). Presentation of the self in everyday life. London: Penguin books.

Henderson, Sheila; Holland, Janet; McGrellis, Sheena; Sharpe, Sue & Thomson, Rachel (2012). Storying qualitative longitudinal research: Sequence, voice and motif. Qualitative Research, 12(1), 16-34.

Hoang, Lan Anh & Yeoh, Brenda S.A. (2015). Transnational labour. Migration, remittances and the changing family in Asia. Basingstoke: Palgrave Macmillan.

Inglehart, Ronald (1997). Modernization and postmodernization: Cultural, economic, and political change in 43 societies. Princeton, NJ: Princeton University Press.

Krings, Torben; Moriarty, Elaine; Wickham, James; Bobek, Alicja & Salamo, Justyna (2013). New mobilities in Europe: Polish migration to Ireland post-2004. Manchester: Manchester University Press.

Leung, Maggi W.H. (2015). "Talk to her, she is also Chinese": A reflection on the spatial-temporal reach of co-ethnicity in migration research. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 16(2), Art. 13, http://nbn-resolving.de/urn:nbn:de:0114-fqs1502130 [Accessed: December 10, 2015].

Lopez Rodriguez, Magdalena (2010). Migration and a quest for "normalcy". Polish migrant mothers and the capitalization of meritocratic opportunities in the UK. Social Identities, 16(3), 339-358.

Mauthner, Natasha S. & Doucet, Andrea (2003). Reflexive accounts and accounts of reflexivity in qualitative data analysis. Sociology, 37(3), 413-431.

McKie, Linda; Gregory, Susan & Bowlby, Sophia (2002). Shadow times the temporal and spatial frameworks and experiences of caring and working. Sociology, 36(4), 897-924.

Morse, Janice M.; Hutchinson, Sally & Penrod, Janice (1998). From theory to practice: The development of assessment guides from qualitatively derived theory. Qualitative Health Research, 8(3), 329-340.

Neale, Bren; Henwood, Karen & Holland, Janet (2012). Researching lives through time: An introduction to the Timescapes approach. Qualitative Research, 12(1), 4-15.

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 [Accessed: April 27, 2016].

Pilkington, Hilary (1998). Migration, displacement, and identity in post-Soviet Russia. London: Psychology Press.

Pini, Maria & Walkerdine, Valerie (2011). Girls on film: Video diaries as auto-ethnographies. In Paula Reavey (Ed.), Visual methods in psychology: Using and interpreting images on qualitative research (pp.139-152). London: Routledge.

Razon, Na'amah & Ross, Karen (2012). Negotiating fluid identities: Alliance building in qualitative interviews. Qualitative Inquiry, 18(6), 494-503.

Ricoeur, Paul (1984). Time and narrative, Vol. 1 (trans. by K. Blamey and D. Pellauer). London: University of Chicago Press.

Ryan, Louise (2011). Migrants' social networks and weak ties: accessing resources and constructing relationships post-migration. The Sociological Review, 59, 707-724. doi:10.1111/j.1467-954X.2011.02030.x

Ryan, Louise (2015b). "Inside" and "outside" of what or where? Researching migration through multi-positionalities. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 16(2), Art. 17, http://nbn-resolving.de/urn:nbn:de:0114-fqs1502175 [Accessed: April 20, 2016].

Ryan, Louise & Golden, Anne (2006). "Tick the box please": A reflexive approach to doing quantitative social research. Sociology, 40(6), 1191-1200.

Ryan, Louise; Sales, Rosemary; Tilki, Mary & Siara, Bernadetta (2008). Social networks, social support and social capital: The experiences of recent Polish migrants in London. Sociology, 42(4), 672-690.

Thomas, Felicity; Aggleton, Peter & Anderson, Jane (2010). "Experts", "partners" and "fools": Exploring agency in HIV treatment seeking among African migrants in London. Social Science and Medicine, 70(5), 736-743.

Thomson, Rachel (2007). The qualitative longitudinal case history: Practical, methodological and ethical reflections. Social Policy and Society, 6(4), 571-582.

Thomson, Rachel & Holland, Janet (2003). Hindsight, foresight and insight: the challenges of longitudinal qualitative research. International Journal of Social Research Methodology, 6(3), 233-244.

Thomson, Rachel & McLeod, Julie (2015). New frontiers in qualitative longitudinal research: An agenda for research. International Journal of Social Research Methodology,18(3), 243-250.

Trevena, Paulina (2010a). Konstruowanie tożsamości jednostkowej przez wykształconych migrantów w sytuacji sprzeczności pomiędzy jaźnią subiektywną a odzwierciedloną [Identity construction of highly educated migrants in the situation of conflict between subjective and looking-glass self]. In Krzystof Konecki & Anna Kacperczyk (Eds.), Procesy tożsamościowe. Symboliczno-interakcyjny wymiar konstruowania ładu i nieładu społecznego [Identity processes. The symbolic-interactive aspect of constructing social order and disorder] (pp.243-265). Łódź: Wydawnictwo Uniwersytetu Łódzkiego.

Trevena, Paulina (2010b). Degradacja? Koncepcje socjologiczne, percepcja społeczna, a postrzeganie własnego położenia przez wykształconych migrantów pracujących za granicą poniżej kwalifikacji [Declassing? Sociological concepts, social perception and self-perception of own position by educated migrants working below their level of qualifications]. In Halina Grzymała-Moszczyńska, Anna Kwiatkowska & Joanna Roszak (Eds.), Drogi i rozdroża. Migracje Polaków w Unii Europejskiej po 1 maja 2004 roku. Analiza psychologiczno-socjologiczna [Roads and crossroads. Migrations of Poles in the European Union after 1st May 2004] (pp.146-160). Kraków: Nomos.

Trevena, Paulina (2013). Why do highly educated migrants go for low-skilled jobs? A case study of Polish graduates working in London. In Birgit Glorius, Izabela Grabowska-Lusinska & Aimee Kuvik (Eds.), Mobility in transition. Migration patterns after EU enlargement (pp.169-190). Amsterdam: Amsterdam University Press.

White, Anne (2011). Polish families and migration since EU accession. Bristol: Policy Press.

Louise RYAN, Dr., is professor of sociology and co-director of the Social Policy Research Centre, Middlesex University, London. Her research interests encompass gender, migration, ethnicity, religion and identity, social networks, social movements and transnational families. She teaches research methods and migration. Her most recent book is "Migrant Capital: Networks, Identities and Strategies (2015) co-edited with Umut EREL and Alessio D'ANGELO (Palgrave publishers).

School of Law
Middlesex University Hendon Campus
The Burroughs
London NW4 4BT, United Kingdom

Magdalena LOPEZ RODRIGUEZ (BA Cultural Anthropology, MA Anthropology of Development, MRes Social and Educational Research, MPhil) is a research fellow currently researching perspectives of Polish migrant mothers on their children's education in the UK as well as Polish immigrant organisations in Ireland. Her research interests focus on analyses of cultural and social capital in migratory situations. She has extensive experience and expertise in qualitative research methods.

Paulina TREVENA, Dr., is research associate at the department of Central and East European Studies, University of Glasgow, currently working on a project on Social Security and Migration in Scotland (http://www.gla.ac.uk/research/az/gramnet/research/ssamis/). Her research interests centre on the social world and social position of migrants, and issues relating to labour market positioning, social and occupational mobility, education, and well-being in particular. She has published a number of journal articles and book chapters on these issues.

Central and East European Studies
University of Glasgow
8-9 Lilybank Gardens
Glasgow, G12 8RZ, United Kingdom

Ryan, Louise; Lopez Rodriguez, Magdalena & Trevena, Paulina (2016). Opportunities and Challenges of Unplanned Follow-up Interviews: Experiences with Polish Migrants in London [53 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 17(2), Art. 26,
http://nbn-resolving.de/urn:nbn:de:0114-fqs1602266.

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

ISSN 1438-5627

Creative Commons Attribution 4.0 International License