1 Introduction

Social gerontology has often been criticised for being “rich in data, but poor in explanations” (Daatland 2002: 1), particularly by the tendency to under theorise and instead favour a problem-solving focus, guided by pragmatic dimensions of policy and practice (Hendricks et al. 2010). Critics also contend against the discipline’s tendency to overlook the heterogeneity of subjective lived-experience and marginalise ageing differences by prescriptive (Western) conceptualisations of ‘successful’ or ‘healthy’ ageing (Lamb 2014; Martinson and Berridge 2015; Stowe and Cooney 2015). This criticism also extends to the often lack of sensitivity to cultural diversity and ethnicity. As Phillipson (2015), Torres (2015, 1999) and Wray (2007, 2003) have argued, ethnic, historical, and cultural diversity remains under-researched although mainstream social gerontology has started to uncover (Tannistha 2017; Zubair and Norris 2015) the role culture plays in shaping people’s understandings of quality of life in later life, family support, and agency, for instance. These conceptual indicators usually depart from dominant British/North American value orientations and theoretical frameworks that are often applied, uncritically, across ethnic and culturally diverse groups. This means that, although social gerontologists sometimes acknowledge the importance of culture to those understandings, the incorporation of such acknowledgement into the arguments they propose is still lacking. I advocate the relevance of a culturally sensitive and reflexive approach to social gerontology research on non-Western societies (e.g., of Africa, Asia, Latin America, the Middle East,). Such an approach is conducive to empirically and theoretically rich social gerontology, which is needed to grasp cultural differences and more specifically how cultural value orientations—embedded in socio-economic, political, religious, and gendered foundations—shape our understandings of old age and ageing.

On this basis my research question became one of methodological concern: selecting the method for data generation and harmonising theoretical ideas and procedures for doing my research on ageing in Mexico (Stanley and Wise 2008). Specifically, how can I go about collecting, analysing, and interpreting the experiences of Mexican women by recourse to theories and concepts grounded in a WesternFootnote 1 context, without falling into a normalisation of Western values? How can I address the specific cultural and social contexts in which the women in my study make sense of ageing and old age? I needed an approach by which to emphasise women’s biographies and locate their personal ageing experience within wider socio-cultural and structural settings to identify both individual agency and structural aspects of ageing and how these influence each other. My literature review suggested that these requirements would be met by establishing a methodological framework informed by theories of feminist social gerontology, cultural variation, and sociology of religion along with a biographical approach to data enquiry and a broadly thematic narrative analysis.

Feminist research has been instrumental in the development of alternative methodologies, especially biographical and life course perspectives across the social sciences; contributing to giving a voice, raising awareness or empowerment amongst vulnerable or marginalised communities (Bernard et al. 2000; Roberts 2002). Biographical research departs from an analytical approach focused on “people’s search for meaning and their attempt to make sense of their lives and identities” (Wilkinson 2000: 438 cited in Reeve et al. 2010, emphasis added). Given that little is known about the actual experience of being an old woman in early twenty-first century Mexico, my study aimed to fill the knowledge gap concerning the lived-experience of ageing Mexican women. Given my endeavour, a biographical/life story approachFootnote 2 appeared the most viable to the nature of this research enquiry. “Life stories are central to human development, the interaction between generations, and integrity in late life.” (Atkinson 1998: 18); they also allow researchers to understand the meanings and interpretations individuals ascribe to their lived-experience, and work more sensitively with the research participants (Liamputtong 2010). Therefore, a biographical/life story approach can serve as the data generation method within a cross-cultural, cross-language gerontological research such as mine. Its emphasis on the intersection of individual lives and wider social and structural ‘contexts’ made it a suitable approach to explore the meanings of old age and the experience of growing old. It is precisely such intersection that uncovers both the patterns of women’s lifetime gendered and social inequalities and their sites of agency and empowerment. With ‘contexts’ I am referring to the physical, historical, cultural, and ideological milieu in which the participants’ lives are lived and to the ways in which they have shaped their context.

For the data analysis and interpretation, I needed an approach to help me looking at ‘biographical ageing’ (Randall and Kenyon 2001)—how meanings of ageing are constructed and re-constructed. A narrativeFootnote 3 approach was particularly suitable to these endeavours. Ray (2007) notes that most narrative gerontology scholars focus on narrative as texts (derived from in-depth interviews) with which they try to elicit retrospective accounts of participants’ lives (personal narratives) to then analyse the individual narratives for themes and patterns (cultural narratives) in relation to constructions of the ageing self (Kaufman 1994; Koch 2000; de Medeiros 2005).

My analytical approach is not that of ‘narrative analysis’, in which the researcher focuses on the storytelling process, the act of ‘telling’ a narrative—the how (Creswell 2012; Riessman 2008). Rather, my approach is much similar to what Riessman (2008) defines as ‘thematic-narrative analysis’, in which attention is given to what is said. Therefore, my focus is on the ‘told’ elements to identify themes and patterns in a narrative, rather than on the act of ‘telling’ a narrative. However, given the cross-language nature of my study, specifically the complexities of Spanish–English translation, in my approach to the participants’ narratives I did interrogate both intention and language. As van Ness and colleagues argue, “giving words to experiences is a complicated process as the meaning of experiences is often not completely accessible for subjects and difficult to express in language” (van Nes et al. 2010: 314). Translation issues originate further challenges to the validity of the study, as both interpretation and representation of meanings are central to narrative research. Thus, my aim is to provide a methodological account of how I conducted biographical research on ageing Mexican women, a reflection on the insider/outsider dynamics, and the decisions I made in my dual role as researcher and Spanish–English translator. I hope this exercise could be useful for others undertaking auto/biographical research, particularly earlier career researchers and postgraduate students dealing with the complexities of conducting cross-language research.

Below I describe the study aims and how I entered the field, gained access to my participants, and dealt with ethical issues. I then further elaborate on my positionality as insider/outsider and on the details of how I approached the data for analysis and interpretation, highlighting the dilemmas and impact of moving across Spanish and English. Finally, I give some consideration as to why is important to explicitly report how our study participants’ narratives are analysed, interpreted, written-up and reported, especially when this is done in a different language than the one used during data generation.

2 A study of old(er) Mexican women

The focus of the study on which this article draws from was to explore how Mexican women construct their personal narratives of ageing and later life. The research methodology involved conducting multiple semi-structured interviews with 32 women (of which 19 were interviewed twice over a period of 5–7 months) aged 60 and older from a wide range of social, economic, and cultural backgrounds living in the metropolitan area of Monterrey, in northeast Mexico. By examining these personal narratives, the study sought to provide valuable insights into the experiences of ageing and later life for Mexican women and contribute to a more comprehensive understanding of the complexities of ageing across diverse contexts.

When I set out my research project the term I used to refer to my study subject was ‘older women’. However, as I became more engaged with the literature of social gerontology and the way some feminist scholars (Calasanti and Slevin 2007, 2001; Calasanti 2004), approach the study of old age and age relations, I started to question my own choice of language. I found myself indistinctively writing ‘old’ and ‘older’, and constantly reflecting on the political implications of such words. Andrews (1999) has advocated for calling old people old, as this acknowledges the challenges and possibilities their years embody. Calasanti (2008, 2004) has also expressed her concern in reclaiming the positive valuation of the term, while Gullette (2017, 2018) has consciously opted for using ‘old’ people in her work on age.

Most of the women I interviewed preferred the term adultos mayores (older adults) or personas mayores (‘older people’ but also translated as ‘greater people’), very few openly embraced the terms vieja or anciana (old woman) and used them in a positive way. However, the cross-language nature of my study made choosing a term more difficult. I was caught in a dilemma. I wanted to subscribe to Andrew’s and Calasanti’s reclamation of the term ‘old’, but at the same time I needed to use an English word that was closer to my participants’ own chosen category: ‘older’. I opted for using the latter but felt that by doing so I was reinforcing society’s ageist attitudes.

While ‘old’ is socially constructed, reified and stigmatised, as are many other terms for oppressed groups, using the term ‘older’ conveys that the old are more acceptable if we think of them as more like the middle-aged (Calasanti 2003: 16).

Given that my intention is to subscribe to a feminist critique of ageism and age relations (Calasanti and Slevin 2007; Cruikshank 2013; Gullette 2004), I opted for using the category ‘old’ rather than ‘older’. As Caissie (2011: 131) notes: “[b]y using [the term old] in positive ways we are saying that there is nothing wrong with being old.”. ‘Older’ then will only appear whenever is part of a direct quote or when using the term ‘older adults’.

In the beginning, I was too focused on eliciting pre-coded definitions that I was missing the opportunity to explore the meanings and understandings participants themselves made of ageing and old age. Therefore, with an exploratory approach, my empirical research focused on the women’s own constructs and moved away from using pre-determined conceptual frameworks (e.g., ‘successful ageing’) and comparing them with the women’s lay definitions. By doing so, I was in fact following a feminist methodology,Footnote 4 the one I initially intended to do, which allowed participants to tell me their auto/biographies in their own terms, whilst also allowing new understandings of ageing and old age to emerge. The present article draws on research informed by feminist methodology; one concerned with the production of ‘unalienated knowledge’ (Stanley 1990) about ageing. Drawing on Ray (1999: 173), this should be characterised by: (1) challenging the scientific paradigm by being personally involved and critical, and also overtly political (i.e. advancing an agenda towards the empowerment of both researchers and researched), and (2) pursuing alternative ways of form and style to report scholarly findings; equally involved and critical (e.g., life-history; narrative analysis, auto/biography).

By undertaking a feminist methodology, the Mexican women in my study had the opportunity to describe and examine their own ageing process and the issues they deem important to it. Certainly, we need an approach that considers cultural and personal lived-experience, as well as socially-constructed identities, gender differences and inequalities, with a view of uncovering alternative—non-Western forms and images of ageing and old age.

3 Exercising reflexivity in the field

Following Frankland and Bloor’s (1999) argument that piloting provides the researcher with ‘a clear definition of the focus of the study’ and given the potential translation-related issues, I decided to undertake a pilot study prior to fully embarking on fieldwork. Whilst data from the pilot interviews are not included in the main study, their completion helped me to identify whether the interview questions were ambiguous or difficult to understand. It also uncovered important cross-cultural and formal consent issues. For instance, the pilot participants found the consent form—which I initially developed in English and then translated into Spanish—‘hard to understand’ and preferred a verbal explanation. The pilot interview participants’ comments helped inform the final versions of the research instruments and consent form. Most of the study participants preferred a detailed verbal explanation instead of reading the study information and consent forms on their own; they all gave verbal consent once they understood both the purpose of the study and what their participation entailed. Except for one participant who was not comfortable signing the form, the rest of the participants all agreed to sign the consent form.

The participant’s reluctance to sign an informed consent helps to illustrate some of the challenges we may encounter while conducting cross-cultural research, particularly with issues of trust and mistrust. According to Liamputtong (2008), cross-cultural researchers have often argued that standard written consent forms can be intimidating, awkward and disconcerting to many cultural and ethnic minority groups, particularly those at the intersection of various vulnerabilities. For some of these groups the request to sign a consent form may in fact—counterintuitively—generate fears about the loss of confidentiality and anonymity that researchers try to give reassurance about with the very consent form (Birman 2006). When conducting research in a cross-cultural context, researchers must take a flexible and sensitive stance to accommodate the local cultural practices of the individuals or communities they are working with; the emphasis then should be on building trust, reciprocity, and rapport than on securing written informed consent (Liamputtong 2008). In the practice of qualitative research consent is a process constantly exercised by the participants and facilitated by a reflexive researcher throughout the project and not merely taking place at the outset with a signed form (Rooney 2013).

In accessing potential participants, I opted for employing opportunity sampling. This strategy involves on-the-spot decisions, taking advantage of events as they unfold during fieldwork. This is a flexible design that fosters a continuous reflection and preliminary analysis and allows developing the sample during the fieldwork (Patton 2014). Thus, my participants were ‘hand-picked’ with specific criteria in mind, which allowed me to achieve a wide variety in the sample (Denscombe 2007). Since I set out to study the ageing experience of broadly (lower and upper) middle-classFootnote 5 Mexican women, and so, being part of the Mexican middle-class myself, everywhere I went became a potential recruitment site. I found participants through acquaintances, taxi drivers, customers and attendants in various shops, gyms, in medical doctors’ waiting rooms, physiotherapy centres, local churches, and public parks. To a lesser extent, I also found participants through snowball sampling.

I contacted and visited several public and private care homes. The main selection criterion was the accommodation price. I excluded the ones I deemed ‘expensive’ or were not a reflection of the people I wanted to include in the study. Some of the places I contacted showed no interest whilst others were ‘suspicious’ of the nature and impact of my project, it seemed to me they were concerned about whether I was going to report administrative or quality of service issues. Eventually, I obtained a positive response from a private care home and interviewed 12 of its residents.

The care home’s agreement to participate was not without certain restrictions. The staff functioned as information gatekeepers: they selected the residents who would take part using their own criteria. This is important because I was not able to recruit participants amongst the very old residents who had less mobility or las encamadas [bedridden] as the staff called them. I was not given any reason for such restriction; it seemed that in the eyes of the staff, the stories of those women were not of the same value as those of the non-dependent, non-disabled, or younger residents. This sampling restriction is already a finding as it uncovers an image of the kind of marginalisation that older people could suffer within an institutional setting, and how easily their voices remain unheard, likely because of enfeebling and ageist attitudes on the part of administrative and care staff.

I obtained a sample of 32 women: 20 were living in the community (either alone or with relatives) and 12 in a care home in the metropolitan area of Monterrey. Certainly, the final sample included differences of age (60–89), education level, residential context (community, care home), marital status (including differing lengths of widowhood in the case of the widowed participants), socio-economic resources and health status, to name a few. My sample, although not representative, offered a heterogeneous image of women ageing in Mexico. I was not interested in generalising from the experience of the participants but in learning from their experience (Chambers 2005). In my study I aimed for diversity rather than homogeneity, which I argue enables the complexities of women’s ageing and old age to be further explored.

3.1 Protecting women’s stories: trust building and ethical issues

As the researcher it was my duty to ensure the rights, privacy, and well-being of the participants (Berg 2007) whilst they took part in my study. It was also my duty to consider ethical issues such as informed consent and data confidentiality (Banister et al. 2011; Berg 2007). Each participant, the administrative staff of the care home and at times relatives of the participants were informed about the purpose of the study, the content of the interview schedule, what exactly their participation would involve, potential benefits and risks, and that the work was intended for educational purposes.

Each participant was informed that their participation was voluntary: they could decide which questions they would answer, pause, postpone, or entirely withdraw of the study at any time and for any reason or no reason at all. While the staff members of the care home were very friendly and helpful, at times they would tell the potential participants that they ‘must answer my questions’. On such occasions, I made a much stronger emphasis on the voluntary nature of their participation.

Indeed, conducting culturally sensitive and ethically responsible research practices has more to do with ‘empowering research participants’ (Ramazanoglu and Holland 2002) than with simply obtaining a signed informed consent. The research experience resulting from building trust between researcher and participants allows a more balanced power relation, which is paramount while committed to a feminist methodology.

The use of a semi-structured interview format and participant observation generated both the development of rich data and personal relationships. Particularly, the use of a follow-up interview gave both participants and me an opportunity to reflect on the issues discussed in the first interview. This enabled clarification and validation of earlier findings, and collaboration on the initial themes and constructs with the participants.Footnote 6 During the course of the fieldwork, I made regular visits to the care home, at least twice a week, which allowed me to immerse myself in the ‘community’ setting, observing and taking part in group activities and social dynamics of the participants (e.g., games, arts and crafts, meals, cinema). This helped me gain a nuanced understanding of the participants' experiences of ageing in a care home—an insider’s perspective of their ‘social world’. This also provided me with opportunities to become more acquainted with the residents; the women who had agreed to be interviewed or those who were at least considering taking part in the study. Through these interactions, I was able to establish a level of trust and rapport, and while some of these interactions did not contribute to the study data, they provided valuable insights into the lived-experience of the residents. These interactions also contributed to developing friendly relationships, which continued for years after completing my study; unfortunately, most of these women have now passed away.

Becoming so close with some of the participants living in the care home affected the way they shared information with me. They would confide in me very personal and emotionally charged accounts, but they would explicitly ask me not to include specific feelings, actions, or events in my study report. Ethically, this represented a dilemma. At the time, I felt the restriction was unfair as I deemed sharing those stories with the final readers of this study necessary to have a complete picture of the participants’ ageing reality, but eventually I understood that such omissions did not have such an impact on the overall findings and the narratives of the participants.

[my] primary obligation is always to the people [I] study, not to the project or a larger discipline. The lives and stories [I] hear and study are given to [me] under a promise, that promise being that [I] protect those who have shared them with [me] (Denzin 1989: 83)

With one exception, all the women I interviewed were very enthusiastic and I felt very welcomed into their lives, or more exactly, into a fraction of their lives (see Online Appendix A—Thick description—notes of first care home interview). Whilst many of them seemed to forget it was a research ‘interview’ and were relaxed and eager to talk, few participants kept making comments regarding the utility of their accounts: ‘I don’t think this is useful for your study, is it making sense to you?’ (emphasis added). Feminist researchers, amongst others, have commented on methodological issues regarding the necessity of developing an intimate and non-hierarchical relationship between researchers and researched and the positive impact of sharing gender identity and experiences (Finch 1984; Oakley 1990). I genuinely tried to get to know the women in my study; I enjoyed listening to their stories and many of them told me how much they appreciated having someone to talk to about their ideas and feelings. One woman said: “I loved this; can I tell my friend? Can she talk to you as well? This would do her so much good”. This woman’s friend is one of the participants.

Certainly, I could claim to have developed a positive rapport and trust with my participants; I even become good friends with a few of them, often joining in activities of mutual interest or simply social encounters. I did not, however, actively set out or employed a methodological ‘friendship approach’ from the start of my fieldwork but I purposely tried to cultivate a ‘dialogical relation’ underpinned by emotional reflexivity and empathy (Smith et al. 2009). This relationship requires genuine reciprocity, continually moving from “studying ‘them’ to studying us” (Tillmann-Healy 2003: 735), and ethics of emotional care not only in relation to our participants but also to ourselves (Owton and Allen-Collinson 2013).

Overall, sharing the same gender, language and cultural background with the participants helped generating trust and a positive relationship. However, valuing a reflexive stance I will further elaborate on my positionality within the study.

4 Me and Them/ I and We: “insider/outsider” dynamics

Some feminist researchers have argued for a reflexive sociology in which sociologists locate their own social identity and values within their work (Reay 1996; Roberts 1990; Stanley and Wise 2008). Certainly, the appeal of qualitative enquiry “lies in its intimacy with social facts and its reflexivity between researchers and [researched participants]” (Wang and Liu 2021: 985). Because I was doing a qualitative study that involved issues of language translation and the use of women’s life stories, I felt committed to locating myself within my research. Drawing on Roberts (2002) ‘telling my own story’, locating my role as a young Mexican woman studying and reconstructing the lives of older Mexican women is fundamental as it provides the reader with an insight into the subjective interpretations—including judgments—across the research process from design to the writing of findings. Although how much my personal story in terms of my age, gender, social class, ethnicity, or religion influences what the participants shared with me and the whole study cannot be assessed, it is important to reflect on such an influence and make visible my identity within the research. I admit that prior to entering the fieldwork, my standpoint as an ‘insider’ was naïve—I envisioned only the advantages of doing research in and about my own culture and hardly reflected on the effects or challenges of my ‘insiderness’. In what follows, my intention is to provide a reflexive account of my insider/outsider positionality.

As a Mexican national I share the same ‘culture’, language, religious heritage, and gender socialisation with the participants. Yet, being a Mexican-Japanese of second-generation I am aware of a degree of cultural variation. For instance, I have been brought up with a diet consisting of a mixture of Japanese and Mexican food, and part of my personal identity is being knowledgeable and fond of both cultures. Furthermore, whenever I meet a new person or simply when I am asked my name, I am usually asked to explain why I have a full Japanese name whilst my physical features are not quite Japanese. Most people would usually follow by saying something like ‘Oh, yes I can see it now, your eyes have that almond shape, but you look Mexican’. Most of the participants did the same. Some women even showed great curiosity and asked me more details about my Japanese heritage and very few did not make any comments about it. However, I believe most participants saw me as a young Mexican woman. I was born and raised in the Northwest of Mexico. I am knowledgeable of the region, the accent and nuances of the language, and the common customs of the people. In short, I am a mujer norteña [Northerner woman] and so are the women I interviewed, either by origin or adoption. Nonetheless one cannot generalise about northerners, especially because there is a cultural diversity within the region, the sense of identity of a Northerner differs from that of those from another part of the country. I would not have felt like an “insider” had I undertaken my fieldwork with women living in a southern community in Mexico.

The city of Monterrey is well-known for being the industrial heart of the country and is considered a more ‘westernised’ Mexico. In contrast, in social and economic terms the south is significantly underdeveloped; when compared to the north, it shows higher rates of poverty, mortality, and illiteracy (INEGI 2021). Undeniably, this divide underpins people’s ways of life and their understanding of being in the world, in which a basic common culture is embedded.

The women in my study and I share a Mexican culture based on a non-Indigenous ethnicity. According to Bonfil (2007), in Mexico most of the popular classes and sectors are Indigenous people. In contrast, some upper-class sectors derived from the Spanish colonisers and tend to conserve non-indigenous cultural forms. Although my study sample is not representative of marginalised cultural minorities or deprived sectors of Mexican society, I am unable to determine with certainty the participants’ or my own precise ethnicity. Thus, the participants and I share being the product of mestizaje (ethnic mixture). We share being immersed in an imaginary Mexico that imitates the West and denies the reality of the remaining Indigenous communities, where industrialisation, urbanisation and ‘middle class-isation’ conditions cultural dynamics that prevail over what Bonfil (2007) calls the ‘Mexico profundo’, in touch with its original Indigenous culture.

I have a similar socio-economic background to that of the participants: like most of the women in my study, my parents did not have higher education and instead they both undertook vocational training. My mother’s experience mirrors most of the participants’ own experiences regarding limited access to formal education and thus I found myself easily relating to their stories and lived-experience.

The participants and I share the same religious tradition—another ideology imposed by the West. Like most of the women interviewed I too had a Catholic upbringing, though my belief and belonging is no longer rooted in church membership. I am aware that the way I practice my faith might seem a little lax or unorthodox to both devout practitioners and atheists. Indeed, I identified with many of the participants who talked about expressing their faith by means of prayer and not necessarily by mass attendance, with the difference that most of them considered themselves to be highly spiritual. Yet, for many of these women not going regularly to church services derives from restricted health and mobility and/or lack of transport or social support, whereas in my case is more of a personal and active choice.

I should mention that I did not openly disclose my religious belief to the participants and only commented about it when I was asked directly. This is not something I planned; still, in hindsight, I wonder whether I should have been more explicit about this aspect and if by doing so I would have recognised sooner the women’s accounts regarding their faith and religious beliefs as their resource for making sense of their world and not merely as linguistic features of Mexican culture. This aspect is particularly important, as it might have affected my analysis and interpretation of the findings.

I recognise that belonging to the same culture, speaking the same language, sharing beliefs, and having a similar social status have both positive and negative effects on the research process and outcomes. Having taken for granted the participants’ religious belief is the perfect example of this. Now, I believe what helped the participants and me to feel comfortable, establish trust and gain more intimate insights during the interviews was our shared gender. As Hayfield and Huxley (2015) note, a researcher can simultaneously hold a position as insider and outsider with both positions being advantageous as degrees of both distancing and empathy can aid the research process.

Considering all the aspects discussed above my researcher’s identity and positionality could be deemed as ‘insider’, and thus I could easily construct the participants’ stories in terms of ‘us’. Paradoxically, being an ‘insider’ might also increase awareness of both researcher and researched regarding the social divisions between them and make evident the fluidity and complex shifting nature of insider and outsider positionalities (Adu-Ampong and Adams 2020; Ganga and Scott 2006; Savvides et al. 2014), which in turn structure the interaction dynamics and impact on the generated knowledge. By being an insider, I realised that there was a considerable social distance between ‘me and them’. This distance stems from a combination of several factors. For instance, there is an age gap. I turned 30 two months into the fieldwork. At some point during the interview, the participants usually made a comment about my age. Most of them were under the impression that I was in my mid-twenties. My socio-cultural position shifted when the women learned of my ‘childbearing’ age and could be probed about this (Caretta and Jokinen 2017)—the first questions they would ask were: “Are you married?”, “do you have any children?” Before my negative response, they would usually make comments about the age they entered their own marriages and/or had their children, which oftentimes were followed by statements about changing social practices. Although not all the participants had experienced marriage and/or motherhood, they all certainly had ‘caring’ responsibilities at some point in their lives. Indeed, not sharing such life course events and knowledge was another distancing factor between them and me.

Following the questioning about my age and ‘gendered’ life experience there was a certain amount of interest in knowing my reasons for being single. Typically, the participants thought it was good I was pursuing higher education and commented on how fortunate I was for having such an opportunity. Some also reflected on how by having a higher education I was increasing my chances to land a good job and a good husband. A few others told me it was good I was not yet responsible for a household and husband/family. Several of them gave me [unsolicited] advice on love-relationship matters and the importance of saving my own money for ‘a good old age’. One woman said to me: “Now, it’s OK ponerte tus moños [Mexican idiom meaning to be picky; equivalent to British idiom ‘putting on airs and graces’] after all, you are only 30 […] however at 40 you should just settle for whatever is available! [Laughs]”. Certainly, through these initial comments, their personal advice, and questions I was not only gaining insight into the participants’ values and how they had adapted their own worldview, but I was also reflecting on my own life and generational context.

In my sense of identity as a middle-class, Catholic, university-educated, young Mexican woman living abroad, I was in a privileged social location in comparison with most of the participants. Such location derives from the interplay of structural factors, family resources, dispositions, and individual choices. The fact that I had shared aspects of my life with the participants made both the interview process and writing about their lives more ‘balanced’ as well as charged with personal meaning. Contrary to a positivist perspective, where the researcher holds a position of power, my approach was to see my interviewees as the experts on the topic and not merely as objects of study, under surveillance (Oakley 1990) and to follow the principles of reflexivity, relationality, mutuality, care, sensitivity, and respect (Cole and Knowles 2001).

I could claim that during this process, particularly during the interviews, I achieved a relatively intimate and non-hierarchical relationship (Oakley 1990) with the participants. As Stanley (1990) would argue, the research accounts that form this study are products that merge my autobiography and the biographies of the participants. Citing Reay (1996), from where I am socially and culturally located, and how I come to understand what I do, personally and professionally, certain aspects of the women’s life stories are much more prominent than others and thus my interpretation is biased and remains an imperfect and incomplete process. Paradoxically, as Bornat (2002) would argue, this emphasis on reflexivity—on subjectivity—may have shifted the ‘balanced’ relationship I claim to have achieved with the participants. Ultimately, my aim was not to achieve ‘objectivity’ but to try and make sociological sense of my data. In the following section, I will explain this ‘making-sense’ process whilst uncovering my concerns of representation and unequal power relations in the context of cross-language research.

5 Interpreting Mexican women’s stories

My analysis included several layers. Preferably, the analytic process starts during the data generation and involves an inductive and comparative approach (Lofland and Lofland 1995). Immediately after each interview, I audio-recorded myself in Spanish reflecting on the interaction with the participant and the contents of the interview in general. Afterwards, I would listen to both recordings and make notes on emergent ideas for subsequent analysis.

5.1 Constructing stories

According to Riessman (2008: 29), the process of transcribing “[…] is deeply interpretive as […] is inseparable from language theory”. Considering the cross-language issues ahead, I opted for constructing a ‘detailed’ transcript of each interview. Informed by the work of Riessman (1987) and the transcription convention of Silverman (2000), my detailed transcription consisted of displaying speech and conversation features (e.g., emotion, pauses, hesitations, interruptions, shifts in pitch, and overlapping talk). The notes taken during the interview complemented this representation of the verbal exchange with the participants. At this stage, I had to make an important methodological decision: when and how translation should take place. Hiring a professional Spanish–English translator was out of the question due to financial cost, whereas translating all the material by myself might have been a time-consuming process; I did not feel qualified enough to do such work. Since my research was concerned with eliciting old Mexican women’s voices, I was afraid that making and working with translated interviews would erase the very Mexican context I was interested in preserving in the first place.

As Riessman (2008) would argue, my interview transcripts were merely representations or imitations that cannot quite capture the dynamic movement of words and gestures, let alone Mexican Spanish dynamic talk: I decided not to translate them into English. I believe in this way I remained closer to the data, paying attention to the women’s particular words and lexical choices. Nonetheless ‘incomplete, partial and selective’ (Riessman 2008), my Spanish transcripts were ready for further analysis/interpretation. For my own notes and reporting findings, excerpts in Spanish were selected and translated into English.Footnote 7

5.2 Working from Spanish into English

Meanings are constructed in and not just expressed by language (Barret 1992; Derrida 1991) and this in turn generates a particular social reality. Choosing which word ‘represents’ the view of the participants and ‘communicates’ best in a sentence in another language becomes a crucial issue. Therefore, language itself.

carries accumulated and particular cultural, social, and political meanings that cannot simply be read off through the process of translation and organises and prepares the experience of its speakers (Temple and Edwards 2002: 5).

As a multilingual person I am aware of the various possibilities and challenges that might derive from undertaking a cross-language research, and the accompanying translation and interpretation processes. As a native Spanish speaker, I share with the participants a similar way to make sense of the world and attain and process knowledge. In our conversations Spanish language is not just a tool or technical label both the participants and I use for conveying abstract concepts, but it also incorporates values and beliefs with which we co-construct meanings of social interaction.

Working across languages in qualitative research intensifies already existing issues of representation, voice, and authority (Hole 2007) between the researcher and researched. Since the participants organise their lived-experience and construct meaning through their own language an English translation can potentially distort or misrepresent their ideas and construct new meanings (Derrida 1991). Therefore, the positionality and authority of my dual role as researcher-translator, constantly making the decisions regarding issues of selection and interpretation between Spanish and English, requires special consideration.

5.3 Positioning myself as researcher-translator

Within cross-cultural and cross-language qualitative studies it is common to see both translation issues and the role of interpreters/translators not being fully addressed as part of the research process. When I was conducting my literature review for my study on ageing women, I found several instances (e.g., Lock 1993) of seminal studies with non-Western communities where the theoretical work is at the front whereas the translation dilemmas remain in the shadows “as if interviewees were fluent English speakers or as if the language they used is irrelevant.” (Temple and Young 2004: 163). As a reader I wondered about when and how the collected data was transcribed and translated into English and whether there was an interpreter/translator, and if so, how deeply involved they were in the interpretation process.

Besides the epistemological and ontological issues related to translation across languages, the choice of when and how to translate will be always determined by the resources available to the researcher, in terms of language fluency, time availability and the funding designated to the research project. Translation issues had an impact on the overall methodology of my study; thus, they deserve to be addressed openly and in detail. In doing so, my decisions and the whole translation process should become transparent to the reader. Indeed, discussing the translation process could help promote reflexivity and accountability in cross-cultural research (Kim 2012; Turhan and Bernard 2021); below I focus on how I tackled such issues. This was not a linear process and I dealt with issues of translation and interpretation not only during the design of research instruments and throughout the data analysis process but also during the writing-up of the findings.

Being both the researcher and translator was an advantage as I did not rely on a third person’s translation and interpretation of what the participants said. Nevertheless, during this process, new meanings, or ‘in-between’ forms of understanding (Smith 1995 cited in Twyman et al. 1999) emerged from a variety of texts, such as the audio-recorded conversations, the transcripts, my memos, and margin notes, translated excerpts, descriptions and interpretations, and thus new insightful material could be produced.

When the researcher is also the translator the quality of translation is influenced by three factors: the researcher’s autobiography; his/her knowledge of the language and culture of the study subjects; and the researcher’s fluency in the language of the write-up (Birbili 2000). Initially, I assumed that speaking the same language as the participants meant that I would easily locate and grasp the cultural context in which they constructed concepts and meanings. Interestingly, it was in fact the translation process, my dual role, which gave me the opportunity to uncover ambiguities and interrogate specific word choices in Spanish and to rethink about cross-cultural meanings and verify the validity of my own interpretations. Some frustration came at times when I found myself unable to find equivalent words in English that would convey a similar meaning or intention to the participants’ words. I dealt with this issue by engaging in discussions with colleagues who are native English speakers and Mexican colleagues who speak fluent English. The following excerpts illustrate the intricacies I faced while working on the Spanish–English translation of selected accounts:

[Original excerpt (Cecilia, age 80)] Pues yo todavía me siento fuerte y me siento bien ¡No tengo nada! Bueno, me dan reumas a veces y me duele la rodilla, ¡pero es cuestión del clima y los huesos viejos y pues [pausa] ya no hay remedio, más que batallar!

[Translation] Well, I still feel strong, I feel good! There is nothing wrong with me! Well, sometimes I get rheumatism and my knee hurts, but it’s because of the weather and my old bones and so [pause] there is no other choice but to soldier on!

[Original excerpt (Jacinta, age 75)] No me gusta mucho batallar niños, ¡no! no soy muy niñera, descartado, mas bien. No me gusta mucho batallar a mí. Batallo con mi trabajo, que es mi profesión, pero con niños no, que me encarguen un niño me pone de mal humor. No, no tengo mucha paciencia, ¡para eso no!

[Translation] I don’t like to trouble myself with looking after kids, no! I don’t like kids! Absolutely no! I can’t be bothered! I work hard at my job, my profession, but no, if I ever have to look after a child that puts me in a bad mood. No, I don’t have patience for that, no!

Both women use the word ‘batallar’ (literal trans. ‘to battle’) in their accounts, however the meaning and intention they ascribe to the word is subtly different. The first one is describing herself as healthy despite feeling pain while also recognising that she needs to face or deal with her situation. I was struggling to find an expression in English that could capture the original intention. Whilst “there is no other choice but to deal with it” could have worked fine, it does not convey the bellicose attitude expressed by Cecilia’s ‘batallar’. A native English speaker I often discussed my findings and translation issues with, solved my dilemma with the expression ‘soldier on’.

In the second excerpt, Jacinta is talking about her resistance to being a caregiver, and her preference for her job. Her use of ‘batallar’ has less to do with a ‘battle’ or a ‘fight’ and more to do with a decision to remain trouble-free. She also uses ‘batallar’ to convey how dedicated she is to her job, which I translated to ‘I work hard’. My translation process was directed “towards obtaining conceptual equivalence without concern for lexical comparability” (Birbili 2000: para. 6). I was focused on capturing the intended meaning or concept behind the words/expressions used in Spanish and finding equivalent ways to convey them in English, as opposed to preserving word-to-word correspondence. This strategy proved to be useful. By considering the cultural, contextual, and semantic aspects of the original content, I have achieved plausibility in the English representations of the participants’ accounts (Online Appendix B shows a comparison of translated excerpts with my comments on translators’ specific choices and the impact of cultural and contextual knowledge on the translation).

Another important decision regarding the translated quotes is that I opted for preserving particular words in Spanish whenever I considered there were no adequate equivalents in English (e.g., viejitos [old people], Diosito [a diminutiveFootnote 8 of Dios but an endearing word]); this was also an attempt to render the Mexican flavour of the women’s accounts. This strategy then is not about finding the exact meaning of words or literal translation but provides foundations on which differences and context may be discussed. For example, regarding the concept of health many of the participants talked about ‘tener salud’ [having health], as opposed to ‘estar sano’ [being healthy]; I opted for keeping the former phrase, since this gave me an opportunity to analyse the way these Mexican women construe such concept. Different was my approach in translating the way they expressed their age: ‘tengo N años de edad’ [I have n years of age]. I translated it into the English phrase ‘I’m N years old’ because in this instance the use of ‘tengo’ [have] is inherent to the linguistic features of Spanish language itself rather than to the women’s subjective constructs of meaning. Regarding issues of translation the emphasis should be on

understanding […] the way language is tied to local realities, to literary forms and to changing identities. […] the process of meaning transfer has less to do with finding the cultural inscription of a term than in reconstructing its value (Simon 1996: 137–138)

I have presented some of the dilemmas that arose while working with data collected in a language other than English, as well as the decisions I made to tackle representational concerns. With the translated excerpts from the interviews I tried to re-construct the Spanish speech of the participants into plausible and ‘vivid’ English accounts. I adopted a ‘subtle realist’ perspective (Hammersley 2002) in which the focus is centred on the “assumptions and processes underpinning [knowledge] production” (Doucet and Mauthner 2002 cited in Reeve et al. 2010: 181). This means that the women’s accounts are the product of my own view and interpretation of what happened in the interviews along with my English proficiency levels, and therefore they should not be considered as ‘true’ accounts but as selected and edited versions offering insight into personal lived processes (Online Appendix C summarises the process of knowledge production from Spanish into English).

6 Conclusion and future directions

In this article, I offered a reflexive account of the overall research process and decisions I made while doing and writing-up a qualitative exploration into the lived-experience of old Mexican women. My aim was to enable public examination of the impact of the selected methods of recruitment, data generation and analytical/interpretative approaches as well as my position as insider/outsider, how I interacted with the participants in the field, and the way I approached translation issues arising during the data interpretation and writing-up stages.

During my fieldwork I was keen to provide any personal information my interviewees wanted to know about me. I believe sharing aspects of one’s own biography can help building rapport and a more balanced relationship between the researcher and the research participants. Here, I have sought to make visible my identity as a middle-class, Catholic, university-educated, young Mexican woman living abroad, to give readers an insight into the subjective interpretations throughout the research exercise. I highlighted my dual role as researcher and Spanish–English translator and provided details about the methodological choices and decisions undertaken regarding issues of words selection and interpretation across languages. Within a cross-language research project, who, how and when the translation has been conducted matters! (Temple 2008). I contend that researchers need to systematically address translation issues and make explicit the dilemmas and consequences of moving between languages in the way in which participants’ narratives are analysed and interpreted, written-up and disseminated. By providing this reflection, readers might be able to fully understand the processes involved and the implications for knowledge production; the auto/biographies that were co-created between myself and the women in the study.

Certainly, accounts of reflexivity raise difficult questions concerning the time and space that can be allocated to meet journals requirements (Newton et al. 2012). Future research could explore the challenges faced by Ph.D. students when conducting cross-cultural/cross-language research, especially the difficulties in language interpretation, translation, or navigating cultural nuances. Investigating the extent to which students share their dilemmas with their supervisors and exploring strategies to effectively navigate such issues could provide valuable insights for improving support and mentorship in cross-cultural/cross-language research. Particularly, the literature discussing cross-language research strategies in detail is still scarce, and preponderantly qualitative. Researchers conducting cross-national mixed-methods or purely quantitative studies could also fully discuss cross-language issues (e.g., translation/adaptation of survey instruments) and the strategies employed to ensure that linguistic equivalences capture the cultural and contextual differences across the languages and cultures involved in the project.

I contend that if we want our cross-language research to be recognised as valid, ethically sound, and reliable we must prioritise engaging in a critical and reflexive account of the research process, methods, and outcomes and the potential advantages and challenges of the researcher’s position as both insider/outsider and translator or those faced when working with interpreters/translators. By making these methodological and reflexive accounts explicit more often, we could contribute to develop an expectation of critical reflexivity on both the process and the co-produced knowledge within reported cross-language qualitative research.