Lost or Gained in Translation? The Use of Interpreters during Qualitative, Semi-Structured Interviews

Post by Rachel J. Wechsler, DPhil student in Criminology, University of Oxford

Communication: one of the key aspects of qualitative research, but potentially one of its greatest challenges. This was certainly the case when I was carrying out primary research with female victims of sex trafficking in the Netherlands in 2014 and 2015. I did not want to limit the representativeness of my sample by only including English speakers, but I knew that using interpreters would pose its own set of challenges, not least of which was my limited funds as a doctoral student.   

I decided whether or not to use interpreters on a case-by-case basis, in view of potential participants’ English language abilities and interest in taking part in my research. I chose to use interpreters with some participants who did speak a moderate amount of English and may have been able to participate in an English interview because I wanted to ensure that they could express themselves in a nuanced manner. Furthermore, I wanted these participants to be able to focus on the substance of the interview rather than on finding the right words to articulate their thoughts in a language in which they were not entirely comfortable. 

When selecting interpreters, it is important to consider the characteristics of the research sample, particularly when dealing with vulnerable populations. As most of the participants in my study had experienced trauma at the hands of men, I made every effort to find female interpreters and was successful in doing so in all but one case. I also sought interpreters who had had previous experience working with human trafficking victims and were familiar with the level of patience and sensitivity that is often required when communicating with them.    

I observed that having the interpreters physically present during interviews appeared to facilitate their rapport with the participants, as compared with involving the interpreters via telephone or Skype. This is likely attributable to factors such as increased eye contact and opportunity for chit-chat prior to the interview when interpreters are physically present. Moreover, poor telephone or Internet connections, which can negatively impact the clarity and quality of communication, are not an issue during in-person interpretation.      

Recruiting and arranging interviews for non-English speakers was particularly challenging, as I did not want to risk overusing my goodwill with interpreters by calling them into the research site outside of the actual interviews. While gestures, body language, and drawings (of clocks, calendars, etc.) are quite useful for building initial rapport and scheduling interview appointments, such techniques are not sufficient to explain the substance of the research study.  For purposes of explaining my research to potential participants, I found it helpful to have interpreters translate a short blurb about the study into the relevant languages. Doing so allowed for greater accuracy and precision than using Internet translation mechanisms, such as Google Translate. However, using written materials only works with potential interviewees who are literate. In the few cases in which they were not, I arranged for interpreters to read the blurb to them over the telephone (to ensure correct pronunciation), which required a bit more coordination.     

Whilst offering clear benefits, using interpreters in qualitative research does raise certain concerns. One of the most significant is the risk that meaning will be lost during the interpretation process. I took precautions against this by asking the interpreters to translate word-for-word as closely as possible (rather than summarizing), request that participants pause during longer answers to allow for interpretation, and ask participants to repeat themselves or clarify if there was any uncertainty about what was said. At times, I needed to remind interpreters or participants of these procedures during the interviews, which occasionally disrupted the flow, but was important for data accuracy.    

Another issue is confidentiality. Participants may fear that the confidentiality of their responses could be compromised if a third party is listening to the interview and may be less forthcoming as a result. If an interpreter is from the same village or local area as the participant, the latter may be concerned that his or her personal information could be communicated to family, friends, or neighbours, leading to gossip and potential stigma. Yet, this concern must be counterbalanced with the need to use an interpreter who speaks the same dialect as the participant, in the interest of accurate and precise interpretation. This issue can often be addressed by using an interpreter from the same general region as the participant but from a different village, town, or city. Furthermore, it is vital to discuss the importance of maintaining confidentiality with interpreters prior to beginning any interview. 

A third concern is cost. Hiring interpreters can be prohibitively expensive, particularly if they must travel a long way to the research site. Luckily, my fieldwork was mainly conducted in Amsterdam, which is a diverse city that is home to many interpreters covering a wide array of languages. I was also fortunate enough to find interpreters who were supportive of my research and were willing to assist with interviews at no cost (or in one case, in exchange for editing her English writing). Nearly all were professional interpreters and the two who were not still had had considerable interpretation experience. 

During the course of my research, I conducted thirty-nine interviews in twelve different languages. Not a single participant was a native English speaker, but more than half were fluent or possessed a sufficient command of English for purposes of the interview. While the rhythm of these interviews differed from the ones involving interpreters, the data that emerged from both types tended to be similar in quantity and depth. The primary difference was procedural in nature: interviews involving interpreters tended to be much longer due to the repetition of all communications in another language. This required stamina, patience, and focus from everyone involved.    

Overall, I am satisfied with the quality of my data and believe that the thought and effort I put towards communication and interpretation issues were well worth it. The invaluable assistance I received from interpreters has enabled my research to paint a far richer and more representative picture of sex trafficking victims’ experiences with the criminal justice process in the Netherlands than would otherwise have been possible.  

Any comments about this post? Get in touch with us! Send us an email, or post a comment here or on Facebook. You can also tweet us.

__________

How to cite this blog post (Harvard style):

Wechsler, R. (2016) Lost or Gained in Translation? The Use of Interpreters during Qualitative, Semi-Structured Interviews. Available at: https://www.law.ox.ac.uk/research-subject-groups/centre-criminology/centreborder-criminologies/blog/2016/07/lost-or-gained (Accessed [date]).