1The literature review above allowed me to develop the following research question:

2How do refugee women in Geneva use negotiations of identity and processes of boundary work to affect their sense of belonging and unbelonging given their translocational positionality?

3Consequently, the concepts and theoretical tools I focus on in this research are firstly the notion of experiences of settlement and integration as conceptualized through the lens of boundaries (Wimmer, 2004; 2008), especially the various strategies of boundary making and identity negotiation which are part of processes of belonging (Anthias, 2016: 177).

4Second, my primary concern was to understand how the various identities and social locations of the women affect these experiences of settlement and integration. The concept of translocational positionality (Anthias, 2002; 2008; 2012; 2016) permits this understanding of various positionalities among different contexts and spaces. Thus, the interplay of gender, class, ethnicity and other factors needs to be considered when focusing on the refugee women’s identities and should be seen as embedded in a transnational space made of hierarchical and social boundaries.

5Defining a research design allowed me to undertake the empirical study of my research question. First, the key concepts of the research question must be defined. Second, to conduct this research the most appropriate methods were qualitative, mostly content analysis and in-depth interviews.

6The first key concept to be defined is the notion of identity which must be understood with a transnational frame, given the transnational space and practices this entails (Vertovec, 2001). Anthias (2008: 7) gives some elements of the notion of identity: notions of the self, the core self and the aspirational one, but also notions of categorizations. Identity also entails forms of practice and performance. Shared spheres of similarity with others are important parts of individuals’ identities. Moreover, and importantly, there is the notion of identity as “a site of struggle, relating to strategies of power, recognition, representation and redistribution” (Ibid.). Finally, identity needs to be understood as a process (Anthias, 2012: 107). This leads to the idea of negotiation of identities understood as a “social practice that can be understood as the combination of discourses, performances and (dis-)identification strategies constructed on a multiplicity of cultural frameworks and reference systems” (La Barbera, 2015: 7). Social interactions are moreover fundamental in the construction of identities because “identity as a process, story and discourse, is always spoken from the position of the Other” (own translation in Hall, 2013: 62 in Zodogome, 2017: 85).

7Negotiation of identities is closely related to the concept of boundary work and the processes it involves. Thus, some strategies used are individual boundary crossing (Dahinden et al., 2012), social identities reconstruction (Holtz et al., 2012), normative inversion (Wimmer, 2008; Dahinden et al., 2012) and boundary shifting (Wimmer, 2008).

8The last key concept of the research question is the notion of a sense of belonging and unbelonging. According to Yuval-Davis (2006b: 199), belonging is about “feeling at home” and “feeling safe”, and is a dynamic process involving the interaction between organizational and subjective factors. In fact, belonging entails processes of inclusion, exclusion, access and participation which give sense to how people feel about their locations in the social world (Anthias, 2008: 8). Finally, belonging must equally be understood within a transnational frame, as individuals can feel various senses of belonging in multiple social spaces (Anthias, 2016: 183).

3.1 Methodology

9As my field research mostly focused on the perceptions of refugee women, I undertook in-depths interviews – using a biographical narrative focus – with my research partners during the period of February-March 2019. In fact, to examine how individuals understand and interpret their place in the world and sense of belonging, narrative accounts by the actors themselves are seen as the most accessible for social researchers (Anthias, 2002). Furthermore, biographical narratives allow for an understanding of identity processes as linked to concrete social and cultural situations and contexts (Al-Rebholz, 2015: 67).

10To constitute my field and choose my research partners, I used the principle of theoretical sampling (Glaser, 1992). The aim was consequently not to gain an accurate statistical representation but, by a detailed analysis of some pertinent case studies, to understand deeply the phenomenon I am interested in. Focusing on individual women who represented a range of relevant situations among refugee women allowed me to choose relevant case studies. For instance, I tried to interview women who have different residence status (N, F, B, C or Swiss citizenship), age (23-60), time of residence (3-20 years) and country of origin in the Middle East (Syria, various part of Kurdistan, Iraq and Iran). I was able to obtain an interesting range of situations and experiences using this principle. In order to find my research partners, I used my personal contacts, collaborated with two refugee organizations (Camarada and Agora1) and used the “snowball” principle (Biernacki and Waldorf, 1981).

11I interviewed 14 women over the period of 2 months. The interviews usually lasted from 1 to 2 hours, with some exceptions: three were about 13 minutes and one was extended in 3 periods of 2 hours. We would meet either in a café, in the woman’s apartment or in the local of one of the partner’s organizations depending the woman’s preference. I always started our encounter by explaining my subject and interest, confirmed the confidentiality of the interviews (every name was changed in the analytical section) and asked if they agreed for me to record the interview (I usually transcribed and translated the interviews right after they were completed). I used the biographical narrative interview method, which entails three steps (Rosenthal, 1993: 60). First, I asked a broad question regarding the person’s life history and let the woman talk. Afterwards, I asked narrative questions on topics and biographical themes that she mentioned in her story. Finally (possibly during a second interview) I asked some specific questions about issues that had not been addressed in the first and second parts. I had an interview guideline with me (annex 1) which allowed me to guide the women towards subjects and topics I was interested in. However, I mainly tried to let them talk and tell me about their journeys and experiences when arriving in Geneva.

12I was also interested in gaining some insights and points of view from professionals working in the field. Indeed, the importance of the context and of the institutional framework cannot be denied, as negotiation of immigrants’ identities and strategies should be assessed in reference to the subjective meanings, social locations of the subjects as well as “products of the institutional framework” (Al-Rebholz, 2015: 67). Consequently, I conducted three interviews with practitioners. First, I was able to interview a social worker from the Hospice Général.2 The discussion we had was very broad, but I gained some interesting insights from her professional experience. Second, I had an informal discussion with two women working and volunteering with Agora. One of them is a chaplain and the other one has been volunteering for more than 20 years, spending one afternoon a week talking with refugee women. Finally, I conducted an interview with one professional from Camarada who explained to me the work of the association and the challenges many women face. To complete my understanding of the institutional context, I undertook secondary literature research on governmental policies and practices.

13To analyze my materials, I used the open coding method for grounded theory (Charmaz, 2006). I started with an initial open coding phase, where I “read all my notes, transcripts line-by-line to identify and formulate patterns and themes” (Ibid.: 47-48). After that, I tried “synthetizing and explaining larger segments of data” (Ibid.: 57-58) in the focus coding part. Finally, the theoretical coding allowed me to “conceptualize how substantive codes may relate to each other” (Ibid.: 63-64).

3.2 Limitations and ethics considerations

14Ethics considerations are essential, as the population I am interested in can be considered as “vulnerable”. The importance of traumas certain individuals might have been through cannot be forgotten, for instance. These considerations led me to interview women who had been settled in Switzerland for some time and had not just arrived. Furthermore, I considered that interviewing women who had arrived some time ago was beneficial for this study. In fact, the social processes I am interested in take time and cannot be apprehended within a limited time period.

15Limitations of the study entail considerations of representativeness. As the study uses qualitative methods, especially in-depth interviews, I was not able to interview enough women to satisfy what statistical representativeness would ask for. However, as stated before, the aim was not to gain an accurate statistical representation but rather to understand deeply the phenomenon I am interested in through a detailed analysis of some pertinent case studies.

16Finally, it is important to provide some reflections on the interviews I undertook. Interviews cannot be neutral and, as social interactions, their development largely depends on both (or all) actors’ strategies, cognitive resources and the context in which the interview takes place. The interview can be understood as an intercultural meeting (Briggs, 1986 in Zodogome, 2017: 26) where different norms that are sometimes incompatible are encountered. Indeed, my research partners had very different behaviors towards me depending on their age, the location of our meeting and so on. For instance, I interviewed three young women who quickly considered me almost as their friend and easily shared their stories. On the other hand, when the woman was older, she would consider me as young and as if I could have been her daughter. Consequently, our interactions were very different depending our various social positioning. Social desirability also needs to be considered when looking at my interviews. As I asked questions about exclusion, inclusion, and the Swiss society and system, my position as a Swiss citizen most likely affected some of the responses.

17Moreover, my analysis focuses on the discourses of my research partners. Discourses are gateways into reality; however, they are distorted by the point of view of the person being interviewed. There can be differences between reality and what the person is saying. Consequently, I did not have access to my research partners’ actual practices and experiences but to their opinions on these. However, this is also a strength of this study. Indeed, my main interest was in the perception of well-being, of belonging and unbelonging of the women as these were created by their social interactions and experiences of inclusion/exclusion. Consequently, not only the actual social interactions or experiences of inclusion/exclusion interested me but also the women’s perceptions of these and how they were affected by them.