Narrative Depictions of Working With Language Interpreters in Cross-Language Qualitative Research

What Is Already Known?

Decisions around how to work with language interpreters in cross-language qualitative research have implications on how the research is conducted and the research findings.

Increased involvement of interpreters in qualitative research requires researchers to negotiate their epistemological views with practical and contextual considerations.

What This Paper Adds?

The experience of working with interpreters is complex and nuanced, evolving over the course of researchers’ careers: working with interpreters in early career stages entails working amid uncertainty, being adaptable and reflexive contemplation; later career stages are characterized by prior experiences and greater seniority and resources.

Early career researchers require their own practical field experiences working with interpreters to complement their theoretical understandings and facilitate epistemological self-discovery.

Cross-language research strategies can be strengthened by accounting for financial, time and other resource constraints that are likely to arise and revisiting the strategy on an ongoing basis throughout the research.

Introduction

The process of cross-language interpretation requires the interpreter to navigate receiving raw data in the source language, conceptualizing and understanding the meaning of the data, and then reexpressing the meaning within the cultural context of the target language (Esposito, 2001). During qualitative data collection activities (such as open-ended, close-ended, or focus group interviews), interpreters work within dynamic environments to convey information and meaning between researchers and participants; their role may also extend to other stages of the research such as participant recruitment, transcription and translation activities, data analysis, and/or reporting (McLean & Campbell, 2003; Shimpuku & Norr, 2012).

Traditionally, language interpretation has often been viewed as a value-free process. When the interpreter role is framed as that of a passive instrument, researchers remain concerned with minimizing or eliminating bias (Hennink, 2008), aiming to get as close as possible to a correct version of the translation through standardized protocols such as back translation (Brislin, 1970; Chen & Boore, 2010).

A growing body of literature over the past decades, however, has sparked debate and discussion about the merits of more inclusive approaches to working with interpreters in research, and the potential implications of language interpretation for research findings (Chidlow, Plakoyiannaki, & Welch, 2014; Plumridge et al., 2012). Researchers oriented toward social constructionist views generally see interpreters as part of the knowledge production process, and question how far and in what capacity to involve interpreters in research (Temple & Young, 2004). By acting as translators, cultural brokers, mediators, and/or gatekeepers, language interpreters are acknowledged for inherently shaping qualitative research outcomes and processes (Caretta, 2015) and may be viewed as collaborators that enrich the research (Larkin, Dierckx de Casterle, & Schotsmans, 2007).

Further to researchers’ theoretical orientations, the increased involvement of interpreters in qualitative research is subject to practical and contextual considerations. Timing and the availability of finances across various stages of a research project are two factors that determine interpreter involvement (Turner, 2010). The availability of suitable interpreters, their fluency in the source and target languages, their experience and familiarity with qualitative methods, their prior knowledge of the subject area, their status and position in the community, and their relationships with other members of the research team also affect their contribution to the research. These types of methodological realities raise questions about how qualitative researchers reconcile epistemological convictions with possible resource and personnel constraints. While previous research has interrogated the merits of various approaches to cross-language interpretation, fewer studies have explored in detail the complex process of working with interpreters from the perspective of the researcher.

This article draws on narrative methods to portray how two researchers’ theoretical perceptions about the interpreter role became intertwined with the practical realities of conducting research. Building on a concise overview of literature about the conceptualization of the interpreter’s role in research, I present the experiences of a qualitative researcher who has grappled with the role of the interpreter in her research for more than 20 years, and that of myself, a qualitative researcher new to cross-language research, and currently engaged with interpreters in a research study in Jimma Zone, Ethiopia. The objective of this article is to extend methodological discussions about how researchers at different career stages navigate the realities of working with interpreters in cross-language qualitative research.

Background Literature

A review of background literature demonstrates how researchers and practitioners have characterized approaches to working with interpreters (i.e., according to the extent of autonomy in the interpreter role). It also presents considerations to enhance the quality of research that involves interpreters. This body of literature begins to elucidate the complexities of working with interpreters and provides a theoretical and empirical basis for the present study, which addresses the practicalities of implementing a desired approach for working with interpreters in research.

Approaches to Cross-Language Interpretation

Approaches to working with interpreters may be characterized by the level of autonomy granted to the interpreter (Figure 1). Baker (1981), among the first to discuss the role of interpreters, described a continuum ranging from “verbatim style” of interpreting (i.e., interpreter as a technical tool), to “independent intervention” where the interpreter dominates the interaction. Similarly, Edwards (1998) distinguished between working “through” or “using” interpreters (lower autonomy) versus working “with” them (higher autonomy). Shimpuku and Norr (2012) identified three major patterns of how researchers in Tanzania worked with interpreters, which can be ranked according to the level of autonomy granted to the interpreter: “Invisible assistance” was observed in cases where researchers provided scant information about the role of the interpreter (assumed low autonomy); “independent fieldwork” occurred when interpreters, under the directives of the researchers, collected data in the field without the presence of the researcher (increased autonomy) but were not involved in other stages of the research process; and “integrated collaboration” involved interpreters being involved in the research process beyond data collection (e.g., recruitment, transcription, translation, and analysis) and having ongoing interaction with the researcher throughout the study (high autonomy; Shimpuku & Norr, 2012).

OPEN IN VIEWER

Figure 1.

Conceptualizing interpreter roles in research by level of autonomy.

Using Interpreters in Verbatim Translation

Across the continuum of interpreter autonomy, various protocols have been applied in contemporary cross-language research. Where interpreters conduct verbatim translation, the role of the interpreter may be relatively passive (e.g., using on-site professional translators during focus group discussions to allow the researcher to listen on the discussion, Esposito, 2001) or relatively active (e.g., having interpreters provide summaries of participant responses to the researcher throughout an interview and allowing the interpreter to ask some follow-up questions or probes; Williamson et al., 2011). Invariably, even when assigned a passive role, interpreters have been found to influence data collection in ways such as editing information, initiating information seeking, exerting control of the interview, and taking over the interviewer and/or respondent roles (Murray & Wynne, 2001; Suurmond, Woudstra, & Essink-Bot, 2016).

Researchers have proposed strategies to enhance the quality of data collection in low autonomy contexts. These include providing careful instruction to interpreters prior to the interview (Freed, 1988; Murray & Wynne, 2001; Plumridge et al., 2012; Suurmond et al., 2016), debriefing with interpreters about the interview (Murray & Wynne, 2001; Suurmond et al., 2016), and using mixed methods research to validate study findings (Suurmond et al., 2016). A number of researchers recommend closely monitoring the accuracy of translations and/or comparing multiple versions of transcripts/recordings to check for consistency (Chen & Boore, 2010; Esposito, 2001; Murray & Wynne, 2001; Suurmond et al., 2016; Williamson et al., 2011), though the procedure to reconcile discrepancies between multiple translations is not always apparent and often relies on the availability of bilingual subject matter experts. For instance, Chen and Boore (2010) proposed consulting with an expert panel committee with language, cultural, subject, and methodological expertise to reach a final agreement on issues that arise during translation.

Working With Interpreters

Increasingly, qualitative research scholars have advocated that language interpreters be brought on as collaborators across multiple stages of research projects and that their contributions and influences be discussed openly in research reports (Ingvarsdotter, Johnsdotter, & Ostman, 2010; McLean & Campbell, 2003). A collaborative approach to working alongside interpreters requires that interpreters are given extensive training, especially at the onset of the research but also on an ongoing basis throughout the research (Adamson & Donovan, 2002; Edwards, 1998; Plumridge et al., 2012). Edwards (1998), favoring a highly autonomous interpreter role, argued that working closely with interpreters was preferable for qualitative research, as it makes visible the ways that disparities in power operate in the research context—a topic that is largely neglected in research reports. Edwards recruited interpreters from the communities where research was being conducted, involved them at multiple stages of the research, entrusted them with participant selection, and had them independently conduct interviews and transcribe interviews into English (Edwards, 2013; Edwards, Temple, & Alexander, 2005). In addition to their role in interpreting, these individuals took on tasks that may be typical of research assistants. Recognizing the large role that the interpreters had in the research process, she suggested that researchers do exit interviews with the interpreters to better understand how the positionality of the interpreter may be reflected in the research findings (Edwards, 1998). This recommendation has been echoed by others (Hennink, 2008; Murray & Wynne, 2001) but is not a widespread practice in cross-language research.

Reporting the Role of the Language Interpreter

Increasingly, qualitative researchers contend that the contributions of the interpreter should be visible in reports of research findings and openly discussed for the sake of contextualizing the findings. Accordingly, researchers are called upon to engage in reflexive contemplation of the impact of the interpreter at all stages of the research (Adamson & Donovan, 2002; Ingvarsdotter et al., 2010). Strategies to promote high visibility of the interpreter in research reports include asking interpreters to make translations in third person voice (to situate themselves within the data; Edwards, 1998; Ingvarsdotter et al., 2010), citing field notes, and integrating the opinions of the interpreters, when appropriate (Shimpuku & Norr, 2012). Criteria have been put forth for evaluating how interpreter contributions are reported in cross-language studies (e.g., Wallin & Ahloström, 2006; Squires, 2009), although some find that prescriptive approaches give a false sense of methodological rigor and argue instead for rich contextual descriptions (Ingvarsdotter et al., 2010).

Method

Study Design

This qualitative study was designed in two sequential stages. Stage 1 was an in-depth exploration of the experiences, advice, and epistemological positioning of a mid-career qualitative researcher, encompassing both her early and later career work with interpreters. Stage 2 of the research explored how I internalized and then applied these learnings and considerations in my subsequent fieldwork with a team of interpreters in Ethiopia. The findings at each stage are presented in the form of a three-part narrative. The research was approved by the research ethics board at the University of Ottawa and carried out over a 6-month period in 2016. The participant provided informed, written consent to participate in the study; she also consented to the use of her full name in publications of the research findings.

Stage 1

Participant recruitment and data collection

Participant selection criteria for Stage 1 required that the individual had extensive experience working with language interpreters in qualitative research at multiple points in their career and had authored or contributed to academic publications about the methodological implications of working with interpreters.

Dr. Rosalind Edwards, professor of sociology at the University of Southampton in the United Kingdom, met these criteria and agreed to participate. Dr. Edwards has previously worked with language interpreters in research with homeless mothers and fathers in the United Kingdom (Edwards, 1992) and research with minority ethnic groups living in Manchester and London (Edwards et al., 2005). She has published about the conceptualized role of the interpreter in research, with a focus on themes surrounding power and trust (Edwards, 1998, 2013; Edwards et al., 2005; Temple & Edwards, 2002).

The primary form of data collection in Stage 1 was a semistructured interview. I conducted one, 45-min interview through a video call. The interview guide covered five areas of questioning: selecting interpreters, working with interpreters during research design, working with interpreters during data collection, working with interpreters during analysis, and advice or strategies. To supplement the interview data, I read in detail all of Dr. Edwards’s academic publications that addressed working with interpreters, and we exchanged follow-up communications by e-mail.

Data analysis

The analysis of Stage 1 data drew from narrative approach (Elliot, 2005), which is appropriate to capture “detailed stories or life experiences of a single individual” (Creswell, 2013, p. 73). I used a re-storying process to construct Parts 1 and 2 of the narrative, portraying the themes and experiences of the participant at early and later points in her career. Through re-storying, I retained an active role in composing and crafting her story. The re-storying process has been applied across several disciplines and is considered a transparent, accountable, and replicable approach for the cocreation of narratives between the participant and researcher (Nardi, 2016).

I adapted steps for re-storying problem-solution-structured narratives (Ollerenshaw & Creswell, 2002). First, I transcribed the recorded interview and read the transcript several times, making preliminary margin notes to synthesize the content and themes. Next, I color-coded parts of the transcript that pertained to the participant’s early and later career experiences (Parts 1 and 2). After separating and reviewing the content within each of the two parts, I developed categories that corresponded with elements of the plot structure (Table 1). I organized the content by grouping data (i.e., quotes) together within like categories, and then arranged the categories, and the data within each category, in chronological order.

OPEN IN VIEWER

Table 1.

Categories for Data Analysis to Guide the Re-storying Process in Part 1 and Part 2 of the Narrative (Stage 1).

Early Career (Part 1)	Later Career (Part 2)
Preliminary steps	Initiation of idea
Description of project	Description of project
Conducting the interviews	Working with interpreters
Analysis/reporting	Conducting research
Sparking interest	Analysis/reporting
	Reflections on the process

Finally, following the structure that I had established in the re-storying process, I wrote the narrative. To retain the voice of the participant and draw upon her verbatim phrasing as much as possible, I presented the narration in first-person voice and adopted a limited omniscient perspective. My voice and presumptions are evident in the writing, as I used creative license to shape the pace, structure, and design of the narrative. Thus, I consider Parts 1 and 2 of the narrative to reflect the experiences of a so-called “Protagonist 1.”

To facilitate the participant’s additional input into the narrative—and to strengthen the trustworthiness of the research—I invited her to review a draft of the narrative as a form of member checking. The participant’s response was favorable and encouraging:

I am perfectly happy with you as the storyteller and the story you are telling about me, and with you as the mirror and the image that you are reflecting that isn’t quite accurately me but isn’t not me either (N.B. Edwards, March 19, 2016, personal communication).

Stage 2

Context and documentation practices

Stage 2 draws from my subsequent fieldwork that commenced a few months after completing the Stage 1 data analysis and write up. At the time of fieldwork, I was in my first year of a PhD program in population health at the University of Ottawa. This was my first time doing fieldwork that involved qualitative data collection and my first experience working directly with language interpreters in a research context. On a previous visit to Ethiopia (about 5 months prior to the fieldwork), as part of research team meetings, I had visited some of the fieldwork sites.

The fieldwork entailed working with a team of nine language interpreters to collect qualitative data for a rapid qualitative assessment as part of the ongoing research study titled the Safe Motherhood Project. The Safe Motherhood Project is a collaboration between researchers at the University of Ottawa and Jimma University, focusing on the implementation and scale-up of maternal and child health initiatives in Jimma Zone, Ethiopia. One of the key interventions of the Safe Motherhood Project involves holding information, education, and communication (IEC) workshops in rural communities. The primary purpose of the rapid qualitative assessment was to gather data to inform the design and delivery of the IEC workshops (Bergen et al., 2018).

The data collection for the rapid qualitative assessment included 12 focus groups with male and female community members and 24 interviews with key informants (religious leaders, community health workers, and members of the male and female development armies ¹ ). All data collection took place in Afan Oromo, the local language. Nine individuals (seven males and two females), fluent in both Afan Oromo and English, were employed to assist with this research. These individuals were graduate students, researchers, and faculty members from Jimma University. For several individuals, this was their first experience doing qualitative data collection. Their scope of work included participating in a 2-day induction workshop, conducting interviews and/or facilitating focus groups, preparing written transcripts in English, and participating in debriefings and exit interviews.

I contributed to the fieldwork of this rapid qualitative assessment by coleading the induction workshop for the interpreters, supervising data collection activities in the field, and conducting daily debriefing sessions and exit interviews with the interpreters. In total, I engaged with the interpreters regularly over a 6-week period.

Throughout fieldwork, I journaled and kept extensive field notes about the experience of working with interpreters, attending to how my perceptions and expectations evolved. In addition to my personal notes, I produced regular updates for the research team, detailing the challenges and successes of the fieldwork. Together with the two other data collection supervisors, I compiled an internal report at the end of the data collection to summarize the strengths and weaknesses of the experience, and make recommendations for upcoming, larger scale qualitative data collection activities. The internal report also reflected interpreters’ opinions and feedback, as expressed in the daily debriefing sessions and exit interviews.

Personal narrative composition

The documentation that I produced during and after the fieldwork informed Part 3 of the narrative. Given that this part of the narrative pertains to my own experiences, I adopted a more holistic and intuitive approach to analysis and composition than in Stage 1. To compose Part 3, I first reviewed the documentation from my fieldwork and noted the most salient aspects of this experience. I identified three categories that captured what I deemed to be most significant: (1) my initial expectations, (2) the evolving role of the interpreters, and (3) developing a sense of partnership. I then organized my notes according to their corresponding category. Finally, I drafted Part 3 of the narrative, following the structure provided by the three categories. As both the subject and writer of the story, I employed a first-person voice.

Part 3 is based on my experiences of doing fieldwork, which were rich and complex. I made decisions about how to condense and focus the content of these experiences to create a narrative that I perceived to be genuine at the time of writing; however, the stories that I have told and will tell about these experiences are dynamic over time. Thus, I considered Part 3 of the narrative (Stage 2) to be that of “Protagonist 2.”

Findings

Part 1: Thinking About Things

(An early career narrative of Protagonist 1)

That first project really got me thinking about things. It was early in my career and I was doing a piece of contracted research for a government department. It was very much a top down agenda: We were evaluating projects that focused on homeless people and homeless families. A number of those people were immigrant or refugee families who didn’t speak English. They used interpreters to access government-funded services.

And so I worked with the interpreters. The interpreters were those delivering the services or sometimes volunteers who worked with the services. I’d reviewed the literature and so on, but really, I was working with the interpreters in an ad hoc way. There was no ongoing relationship. I would make the arrangement and then meet up with them 10 minutes beforehand. You’d go, you’d explain what the project was about and you’d do the questioning. They didn’t need the interview schedule because they’re doing the interpreting. They were paid for the hour and then that was it. If I wanted to ask them further questions after the interview or go back and check anything with them, then that was their good will.

It was interpretation in the pure sense. That classic triangle: the person being interviewed, the researcher, and the interpreter. As soon as you are there, in that triangular relationship, you’re posed in a particular way. Everyone is posed in a particular way. You trust that the interpreter is asking the kinds of questions that you are wanting them to ask. You’re not always sure what’s going on at various points in time. You’re also marginalized in a way, you know. Who do you look at? All of those sorts of things. But you are there, so in that sense you’ve got that very real feeling. You’re actually in the situation.

I thought of it as a relational exchange, a three-way exchange. It wasn’t just like there’s data out there and I’m going to go collect it. I can’t collect it directly. Someone else is doing it for me and they are telling me the truth. Their truth. They were always implicated in everything that was said. There were times when I felt that there was a particular gloss being put on something, but I wasn’t there to cut the interpreter out of it. I was interested in making the interpreter visible. There isn’t just one direct interpretation that is always correct. So I didn’t do the back translation stuff to check the veracity of the interpretation. But I would go back and check the quotes with them.

And the people being interviewed, they aren’t passive in the situation either. They want to try to communicate directly with you, however basic their language skills. So, it was a really a sort of relational interchange.

And that was my first experience conducting research with interpreters. It was very interesting. It got me thinking about things.

Part 2: Usurping the Triangle

(A later career narrative of Protagonist 1)

The second project, it bubbled around in the back of my head for a while. At that time, the literature was mostly about the interpreters’ experiences or about service providers. There was very little about the experiences of the people who were relying on interpreters to access services. I had two colleagues who were interested in the project: One had also worked with interpreters previously—that was how we found each other—and the other was an expert in race, ethnicity, and ethnographic methods. This idea, it was based on our reviewing the literature, not being very happy with it, and then trying to think about, well, what’s the ideal? What would we like, as researchers? We had a lot of experience, decades of experience between the three of us, doing qualitative research.

At this point, I was in a different career phase—more established and so on. I wanted to work with interpreters in a more inclusive, more ethical way. I didn’t want to simply use interpreters, I wanted to include them. I wanted them to get something out of it as well. They would be part of the process as much as possible.

So, we had the idea, we designed the study and we got the funding.

As soon as we got the funding, that’s when we advertised for the interpreters. This project built in roles for them as full-time or part-time research assistants. We brought them in so that they were inducted into the project, and they were involved in developing the research instruments and thinking through how they might go about accessing people and so on. It was a lot of induction into the project, into the purpose of it, into how to do qualitative interviewing, into how we’re going to agree on working practices, and so on. These things take a lot of resources to do them as well as you’d like to. It’s a big investment of time on everybody’s part. And finances. We were treating these people with respect in terms of their salary. And they were employed through the university so they had access to all of the staff development and space.

We had already decided on particular language groups that we were interested in, so we went out to look for interpreters in those languages. We looked for people with a reasonable level of English and obviously fluency in their first language. We were looking for people who we felt would have an understanding of the research process, who had the time to do it. But it was also what they didn’t have as well. We wanted people who didn’t have a link to a professional service where they were gatekeepers in a way to the people that they were interviewing. Because we didn’t want that possibility that people felt they had to take part otherwise they might not get the service. One or two of them were professional interpreters but they weren’t linked to a service, so that was ok.

The interpreters, they’re not trained as researchers originally. I mean, they would’ve felt much more comfortable with a survey with tick boxes. But we wanted this more open approach. We were interested in people’s life stories. And so that’s quite a lot to ask of people who haven’t gone through research training. For whom qualitative research might be a new or different idea. It’s quite a lot to ask them. So we did keep it semistructured with defined questions rather than just topic areas that we might have…I mean, if we’d done it ourselves we might have just gone with topic areas. We wanted to leave space for other people’s agendas—the agendas of the people who were being interviewed—to come through. But, you know, you can’t work with these very open stories all the time. So it’s an issue that’s part of the solution.

I would meet up with the interpreters every 1 or 2 weeks and just talk through with them how things were going and how they might handle things. It was much more of an ongoing relationship that we had with the interpreters. We had brought them in to work as partners. I mean, in a way it was a hierarchical relationship in terms of the direction of the project and suggestions for how to handle things. I don’t want to sound too partnership-y and equal. They were working on our project, if you like. I’m not a completely inclusive researcher willing to hand over everything. Doing research is a big part of my job and I do feel like I should exercise some expertise as well. Part of that expertise is being able to listen and identify issues and bring them into the research.

But they knew what was going on, the interpreters. They had different areas of influence on the project. They knew what was going on in the field and they were the experts there. And they had their particular interests that they pursued as well.

The interpreters weren’t directly involved in full analysis but we (myself and my two colleagues) did speak to them about their understanding of what did come out across all of their interviews. We took their comments seriously. And then we did an exit interview with them that, in fact, turned them into research data. Because then we started using it and analyzing what they had told us as well. The interpreters, they were always present in terms of how the language was presented. We would be very clear that it was an interpretation. They were in the reports to the extent that they wanted to be. One interpreter did not want much about himself at all—nothing that could possibly identify him. So, he came up with a pseudonym, whereas the other researchers were very proud to have their names on the report to the funder. But the academic pieces weren’t of great interest to them. They didn’t feel that this was in their interest or something that they could relate to.

Part 3: Applying What I Know and Learning as I Go

(An early career narrative of Protagonist 2)

It was my second time in Ethiopia, so I knew a little of what to expect, but there was still a lot of uncertainty when we started the fieldwork. It was unknown territory for me—both doing data collection in rural Ethiopia and working with a team of interpreters—so I didn’t have rigid expectations for how things would go.

But I did have some ideas at the onset.

I wanted to establish positive working relationships, and maybe even friendships, with the nine interpreters that we employed for our study. They were graduate students, early career researchers, and faculty members; so on some level, we could relate to each other. We all had a vested interest in gaining skills and experience doing qualitative research. I saw them as allies. I understood that cultivating a sense of mutual understanding and trust would add strength and legitimacy to our research.

I wanted to determine how the nine interpreters would go about bridging the gap between researchers and participants and to take full advantage of wherever their strengths lay. Others have described interpreters as being both insiders and outsiders to the research process. I suspected that would be the case with our interpreters as well, though I didn’t know the specifics of how that would play out. I assumed they would be excited about our research topic. I thought they would have spent time in rural communities and know people in similar life situations as our participants through their families or social networks.

I planned to be engaged with the interpreters and wanted to help them develop confidence and feel appreciated in their role. I wanted to do daily debriefing sessions and exit interviews to actively seek their impressions about the research process and the data that we’d collected.

And then things officially began.

We started with a 2-day induction workshop, during which we spent a great deal of time reviewing the research tools. One reality that I had overlooked beforehand was that the interpreters all shared a commonality that I and other members of the research team did not: expertise in the local language. The ensuing discussions, clarifications, and debates surrounding how best to articulate the themes of the research gradually shifted into the local language, where they were most comfortable. I encouraged these exchanges, but at the same time I was sidelined, unable to weigh in on the advice and information that passed between them. I relinquished control over the research tools that we had spent months carefully and precisely crafting and that took some getting used to.

Then it was out to the field. For 6 days. Our daily schedule soon became routine: leaving the university campus in the early morning, driving up to 3 hr to a designated rural community health post, simultaneously conducting interviews and focus group discussions, and then leaving as quickly as we’d arrived, debriefing in back of the van on the drive back to the university campus.

The interpreters were quick to take ownership of the research process. They made arrangements for where to sit as they talked to the participants, determined how to navigate interruptions or distractions, and supported each other in overcoming difficulties. Fluent in the local language, they interacted with drivers, staff at the health posts, and passersby.

It took awhile for the interpreters to overcome their initial shyness toward me. No, I wasn’t there to hover over their shoulders or check up on how they were performing, though I suppose it might have seemed a little like this at first. I wanted the interpreters to ask me questions when they were unsure and to open up about their challenges. But I first needed to demonstrate that they could trust me.

Around the fourth day of data collection was when we hit our stride. The interpreters had bonded as a group, and we had all become less shy around each other. In the debriefing sessions, I became more comfortable asking them to switch to English so I could be part of the conversation. It also gave them a chance to practice talking about the research topic in English. They came to see that I would not judge or reprimand them if things didn’t go well. Our debriefing sessions became increasingly honest, productive, and insightful.

The downtimes during the data collection activities were important. In these moments, we could really talk and I learned about the interpreters as people. I was surprised that many of them had not spent much time in rural communities and that they, like me, were re-evaluating some of their own preconceptions. I learned which interpreters relied on humor to establish rapport with participants and which were masterful in generating small talk that put participants at ease. We discussed politics, family life, Ethiopian culture, and sports.

Finally, having finished the data collection, I held one-on-one exit interviews with each interpreter. The exit interview consisted of two questions, which the interpreters knew ahead of time: What were your impressions of the participants’ responses? How did you find the process of collecting data for this study? I didn’t audio record these sessions as I considered them a form of field notes. It was the final scheduled opportunity to exchange any lingering impressions and feedback about the data collection experience. And it brought me a lot of clarity as I moved forward into data analysis and reporting.

I can’t say I would do anything radically different, but there are, naturally, many areas for improvement. For one, I would initiate explicit conversations with the research team about the interpreter role early on, ensuring that we reach a common understanding of our desired approach early in the research process. I would invest time during the induction process to help the interpreters understand their role in the project and to become familiar with the broader context of the study. And I would remember, at all times, that relationships are key—both between interpreters and researchers, and also among the group of interpreters—but take time to mature and develop.

Discussion

Working With Interpreters in Early Career: Learning Through Experience

For both protagonists, early career experiences (captured in Parts 1 and 3 of the narrative) shared several commonalities. Early career was an exploratory period, marked with uncertainty, adaptation, and epistemological self-discovery. In Part 1, Protagonist 1 described a low-autonomy approach to working with interpreters: She interacted with interpreters in an ad hoc way, as per the arrangements of her research contract. When she sensed that the interpreters were placing their own gloss on the interpretation, she responded by making the interpreter visible, thereby increasing the interpreter autonomy (and aligning more closely with her epistemological positioning).

After having considered Protagonist 1’s experiences, and done a comprehensive review on literature, Protagonist 2 had a stronger epistemological preference when she began the research. Protagonist 2 approached her research intending to use an approach centered on relationship building and a higher autonomy role for the interpreters. For her, working amid uncertainty meant being patient and allowing for trust to develop over time and with increased researcher–interpreter engagement. Similar to the early career experience of Protagonist 1, Protagonist 2 had to adapt to the existing research design: her response to navigating budget and resource constraints was to conduct exit interviews with the interpreters (an activity with minimal resource costs), which increased their visibility in later stages of analysis and reporting.

The early career experiences of the two protagonists suggest that, while there is value in considering others’ experiences, researchers require their own practical experiences working with interpreters to inform and develop their epistemological positioning. As a complement to theoretical components of their education, new researchers stand to benefit from authentic field experiences early in their career that expose them to the realities of working with interpreters and encourage reflexive contemplation of the process. Heller et al. (2011) advocated the process of reflexive journaling for junior researchers and noted the merits of engaging in group reflection and debate. By writing down experiences and sharing them with colleagues at similar career stages, the participants gained insight into their own positionalities with regard to the research process.

Experience, Seniority, Resources

As researchers gain experience and seniority, and as they have greater input into the research design, they may be able to more readily realize their preferences for working with interpreters. However, although certain resource-related limitations may (or may not) be alleviated, the later career experience presented in this study suggests that even experienced researchers may still grapple with determining what constitutes their preferred approach for working with interpreters—a notion that is pluralistic, and subject to shift over time or depending on the research context.

In this study, at early career stages, both protagonists were constrained by predetermined research designs, low seniority on the research team, and, consequently, their lack of influence over the time and budget allocated for interpretation tasks. In Part 1 of the narrative, Protagonist 1’s ability to realize a higher autonomy approach to working with interpreters was hindered by a research design that involved the use of verbatim style interpreting, limited contact with the interpreters before or after the interview, and top-down agenda of the research project. Protagonist 2 worked with a team of interpreters who were already graduate students, researchers or faculty members and consequently the resources that had been allocated for prefieldwork training were minimal (assuming that the interpreters already had exposure to qualitative research procedures). Protagonist 2, however, reflected that a more comprehensive prefieldwork training workshop was warranted to provide background information about the study, anticipate issues in the field and strategize possible responses, build rapport between the interpreters and the research team, and discuss in detail the protocols and expectations for the interpreters.

In her later career, Protagonist 1, with more seniority, resources, and power, was better positioned to realize her preferred approach to working with interpreters (which was informed by her earlier experiences). Her prioritization of relationship building, interpreter inclusivity, and interpreter visibility suggest a reaction to certain aspects of her earlier career experiences, where she held less power. For example, in contrast to her early career experiences, she included the interpreters across several stages of the research process and invested resources in developing their capacity to work in a highly autonomous manner. Protagonist 1, however, encountered a limit to the extent to which it was feasible for her to include interpreters, particularly in analysis and writing-up stages. Although she had the resources (and initial intention) to facilitate greater interpreter involvement, as she realized her own desire to maintain control of the research being compromised, her epistemological positioning ultimately led her to limit the level of autonomy that she granted to the interpreters.

The research design described in Part 2 facilitated greater inclusion and visibility of the interpreter through a collaborative approach that trained interpreters in research methodology, encouraged the interpreters’ inputs and feedback throughout the research process, and conducted exit interviews with the interpreters. Substantial resources were required to train and employ the interpreters, factors that have been acknowledged as notable considerations for inclusive approaches (Baker, 1981; Temple & Young, 2004; Turner, 2010) that should be accounted for in the budget and design of the research project (Irvine, Roberts, & Bradbury-Jones, 2008). Protagonist 1’s later career experiences demonstrate the ongoing evolution of epistemological positioning over the course of her career.

Researcher–Interpreter Relations

Issues surrounding researcher–interpreter relations emerged throughout all parts of the narrative, but especially in the second and third parts where the protagonists prioritized a more inclusive approach. Although the protagonists intended these relations to be trusting, honest, and mutually beneficial, they encountered challenges, including the perceived loss of power and authority, difficulties bridging cultural gaps, and difficulties balancing priorities for researchers and interpreters. Power differentials between the researcher and interpreter were shaped by factors such as age, experience, and career stage. Strategies to account for these differences through selective matching of researcher and interpreter characteristics have been suggested (Chiumento, Rahman, Machin, & Frith, 2017) though the opportunity to do so may not be available for early career researchers. Although, in Part 2, Protagonist 1 had more career experience to draw from, she also held a more senior position than the interpreters in her study, which may have made it more difficult for them to relate to each other on a personal level. In Part 3, Protagonist 2 acknowledged that the interpreters had similar professional and academic levels of seniority as her and saw herself potentially developing friendships with them; however, she did not account for barriers such as different cultural backgrounds and the variable level of proficiency in the common language (English). A heightened awareness of potential issues and ongoing reflexivity are key strategies that can help researchers navigate challenges.

Acknowledging interpreters as crucial contributors to the research process, other scholars have advocated that significant time and effort be dedicated to familiarize the interpreters with the research aims and questions and develop a common understanding of how the research will be conducted (Adamson & Donovan, 2002; Liamputtong, 2008). Turner (2010) emphasized the importance of researchers extending friendliness to the interpreters with whom they work, which she noted, should be balanced with professionalism and avoiding appearing arrogant. The findings of this research suggest that interpreter induction activities could be broadened to encourage informal interaction and rapport building between researchers and interpreters, especially at the onset of the study.

Another key aspect of researcher–interpreter relations is trust. In general, given their control and power in delimiting the nature of the participant–interpreter interaction, the researcher holds a larger share of the onus for promoting the conditions for trusting relationships, while simultaneously achieving the objectives of the study. Protagonist 1’s transition toward greater emphasis on trust-building throughout interpretation experiences was likely linked to her desire to adopt a high autonomy approach to working with interpreters in Part 2; in turn, her trust-building efforts may have enabled her to relinquish a certain amount of control over the research process to the interpreters.

Strengths and Limitations

In this study, I contrast the early and later career experiences of a prominent qualitative researcher and experiences a more novice researcher (myself) and draw links between the stories. The study draws from a narrative methodological approach, where a smaller sample size is appropriate to capture the in-depth experiences of a smaller number of participants. Indeed, Creswell (2013), acknowledging the diverse forms and applications of narrative research, argues that “narrative research is best for capturing the detailed stories or life experiences of a single individual or the lives of a small number of individuals” (pp. 73, 74). While the use of a larger sample size may offer different and interesting insights, it would fundamentally change the nature of my research question and the aim of this study, which rely on the detailed analysis of two researchers’ experiences.

Conclusion

While early career researchers can benefit from the mentorship of experienced researchers and detailed theoretical study, practical field experiences are crucial in shaping their epistemological views. In my experience, the opportunity to work with interpreters for the first time in a small-scale rapid qualitative assessment was beneficial for exploring my own epistemological positioning and experiencing some of the practical realities; this helped me to prepare for larger scale, more resource-intensive research activities, as subsequent data collection activities could be undertaken with a better-informed view of how to effectively work with interpreters in that setting.

Establishing a cross-language strategy by research teams at the onset of designing a study promotes a unified approach to the level of inclusion and visibility granted to the interpreter and removes the uncertainty of relying on ad hoc decisions as the research progresses. It may also strengthen the way the interpreter is introduced to the research context and clarify expectations surrounding their role. When appropriate, interpreters may also contribute to the development of a cross-language strategy.

The findings of this study offer insight into useful components of a cross-language research strategy. A cross-language strategy should specify, as much as possible, the epistemological positioning favored by the researchers, the role of the interpreter throughout the research process, and the planned use of methods to gather interpreter feedback (e.g., debriefing sessions or exit interviews). Ongoing reflexive journaling and group reflection are strategies to help mature and articulate appropriate approaches. Practically, adequate resources should be allocated to support all aspects of the desired approach. As noted in the findings presented in this article, alterations to a cross-language strategy may sometimes be necessary to adapt to context, resource availability, and/or emerging challenges. Researchers should consider, from the onset, how the role of the interpreter will be reported alongside the findings of the study. Regardless of the interpreter role, however, comprehensive and transparent reporting about the interpreters and their contributions—as well as the rationale behind decisions—is paramount to establishing trustworthiness (Wallin & Ahlstrom, 2006).

Temple and Young (2004) argued that issues surrounding epistemological and methodological aspects of cross-language research have been largely neglected in academic literature. Few studies have explored the complex process of working with interpreters from the perspective of the researcher. This study adopted a novel approach, using narratives to depict three distinct sets of experiences, adding insight into how researchers approach diverse research scenarios that are more/less amenable to their epistemological positioning. The findings of this research highlight the evolving nature of how researchers approach working with interpreters over the course of a research project but also at different career stages. Consideration of the interpreter role is particularly important when establishing the research design and for the development of a cross-language research strategy but also on an ongoing basis throughout the research.