Translators, a key communication tool?

Communication between an interviewer and interviewee is complex. Academic literature emphasises the challenge of language barriers, and this indeed was something which impacted each and every one of us, in more ways than we could imagine.
During our second interview a translator was used, a fellow student of our participant. To begin, we strongly felt that the tool of a translator would benefit the research process, helping the development of conversation and potentially enable us to build a better rapport. Nevertheless, it was only once the interview was completed and on reviewing the results that we became aware of the impact of a third-party member in the room. The rewording of reworded questions was an issue. Original questions were rephrased twice, once by the interviewer and then by the translator. This inevitably influenced the answers we received, unable to understand the Malayalam language it is unsure of the severity of this impact. Not only did the translator reword questions, but small conversations were had in-between questions, prompting answers to be given in a certain way or potentially completely unrelated and therefore not keeping on track of the task.
The use of a translator impacted the level of quality of our qualitative research, yet the interview would not have gone ahead without this third-party member being present. For future research, a genuine translator ought to be used, or if this is not possible a better selection of candidate from the group of people being studied, not merely a friend of the participant.

by Lauren Adams

Advertisements

Phrasing and negotiating questions when interviewing

When conducting interviews in Kerala, one particular struggle we had was asking appropriate follow up questions and knowing when to use prompts. Roulston et al (2003) note that students sometimes found it difficult to stay focused through the questions they asked when conducting an interview for the first time. In my experience, I found it difficult to ask appropriate follow up questions due to difficulty phrasing questions in a way that meant the participant understood what I was asking them. In addition, keeping focused on the research topic is important for obtaining useable data for analysis, this was difficult as participants brought up interesting facts about Kerala which took us off topic and made it difficult to get back on track.
Another difficulty of interviewing is recognising responses which suggest that a question should not be probed further. Brinkmann and Kvale (2015:161) discuss the suitability of using prompts in different situations and how to handle moments where the participant may feel uncomfortable answering a question. An example from our interview is when we asked our interviewee about why he chose to study at the university of Kerala, they responded that it had not been their choice. When we probed as to why, it became clear that this was an uncomfortable subject for them which was emphasised by their body language. Therefore, if we had been more experienced at interviewing we may have recognised that the participant was uncomfortable before prompting for a reason why.
Finally, another way in which Roulston (2003) suggest first time interviewers may struggle with taking advantage of silences to prompt further responses to a question. Brinkman and Kvale (2015) emphasise the importance of allowing pauses which gives the participant more time to think about their answer and add details. However, being comfortable with silences and knowing how to utilise them is a difficult skill that will come with practice.
Overall, the experience of conducting an interview was useful to be able to put learnt skills into practice. Being given the opportunity to conduct the interviews in Kerala was also rewarding as even though we were apprehensive about the prospect of conducting interviews in a foreign environment, we were able to learn a lot by being given the chance to conduct interviews that were more challenging than they would have been if the interviews had been conducted in Southampton.

Alice Meen

References
Brinkman, S. and Kvale, S. (2015) Interviews: Learning the Craft of Qualitative Research Interviewing, 3rd ed. London: Sage Publications.
Roulston, K., de Marris, K. and Leis, J, B. (2003) Learning to Interview in the Social Sciences, Qualitative Inquiry, 9(4), 643-668.

To be random or non-random: The sampling question

Sampling is used is research to reduce the 3 main factors that tend to constrain research. They are: Cost, Time and Effort (Etikan et al. 2016).

It can make research and the results more achievable, if we use a smaller proportion of the total population. From this, if we use a probability sampling strategy, we can use these results to create generalised picture of the wider population.

If we use a non-probability sampling strategy, it is the opposite. We, as the researchers, do not have any idea of the probability that a person in the population will be selected for the sample. Our findings, therefore, are not generalizable to the larger population.

There are criticisms to a non-probability sampling strategy, as being lazy and “theoretically weak” for collecting data (Moser, 1952, p.411). To an extent this is true, as participants sampled by the research team may not be suitable for the research question (Eitkan et al, 2016), or even create bias in the data (Moser, 1952).

On the other hand, at times, non-probability sampling may be the only reasonable course of action when a research topic is undertaken. For example, quota sampling is a great cheap and quick way to build up a participant base (Moser, 1952). Similarly, convenience sampling can help build up a sample size with participants who are relevant to the research question, if the sampling location is appropriate.

There are also drawbacks with probability sampling. It requires a complete sampling frame to create a representative sample. Those selected in the sample may be in vastly different geographic locations and would require more time and money to be questioned.

Overall, there is no clear answer. The consensus on the application of a sampling strategy is that it depends on how appropriate they are to the given research scenario.

“There are some types of survey on which nobody would suggest using a quota sample, while there are others on which random sampling may be impracticable” (Moser 1952 p411)

Micheal Hellinikakis

References
Etikan, I., Abubakar, S. M. & Rukayya, S. A., 2016. Comparison of Convenience Sampling and Purposive Sampling.. American Journal of Theoretical and Applied Statistics, 5(1), pp. 1-4.
Moser, C. A., 1952. Quota Sampling. Journal of the Royal Statistical Society, 115(3), pp. 411-423.

Confidentiality in Research

Research ethics is a crucial part of research in general as it’s the ethics of planning, conducting and reporting research. It is massively important in research that high standards of ethics are maintained. Firstly, as research is deemed a higher quality if high standards of research ethics are maintained. Furthermore, it is important to consider that when partaking in research the research team are representing themselves, their institution and the whole research community and if ethical standards are broken this can lead to research being invalid or to unintended consequences on the research team or even the participant.

When conducting our interviews in Kerala, I felt one of the biggest research ethics we potentially could break was confidentiality. Confidentiality can be defined as the concept of privacy regarding the information provided during research and the private details of the respondent (Oliver, 2010). Tolich 2004 describes two types of confidentiality, external confidentiality which refers to traditional confidentiality where the researcher acknowledges what the participant has said but promises not to identify them in the final report (Tolich, 2004). Whilst internal confidentiality is an issue which has not been widely explored as it refers to the ability for research subjects involved in the study to identify each other in the final publication of the research (Tolich, 2004). Moreover, internal confidentiality is not often considered by research ethics committees, unlike external confidentiality, which can make the participant more vulnerable as it can be questioned if their informed consent was truly informed (Tolich, 2004). This also was an issue when conducting research in Kerala as in both interviews I was present for there was always one other person present who was not the participant. We avoided issues of internal confidentiality because we are not publishing our findings and if we were to do interviews with others present in the future we would make them sign a form saying they would not breach confidentiality.

Callum Slade

References

Oliver, P. (2010). The Students Guide to Research Ethics (2nd Edition) . Maidenhead: Open University Press.

Tolich, M. (2004). Internal Confidentiality: When confidentiality assurances fail relational informants. Qualitative Sociology, 101-106.

 

Cultural differences we experienced whilst collecting data in India

When we were in India we conducted a study that explored the academic aspirations of the students at the University of Kerala.  This involved completing qualitative studies at the University after lectures had finished. When we were recruiting participants, we were asked to take part in another study. This study was investigating whether you could identify emotions when you could not understand the language spoken. This involved us listening to recordings and having to identify the intensity and type of emotion.

We identified some key culture differences between our study and the student from the University of Kerala’s study. Before starting the emotion study, we were given a participant observation sheet which was very brief – around a quarter to a third of a page long. This may explain why when we gave our participant information sheet to our interviewees they were shocked at the length of the document. Due to the length of our document we felt that some participants read it very quickly and therefore may have only skim read it. Israel (2015) supported this by saying that cultural differences mean that they may not have ever seen a western model of research before. This means that they may have found the document overwhelming.

We also did not have to give written consent to take part in the language study as he said that verbal consent was sufficient. This process seems unfamiliar to us and surprised us due to it feeling less formal than our normal research practises. At the University of Southampton, any research must gain ethical approval first and one requirement of research is that the consent form must be signed. If the consent form is not signed the interview cannot begin and consequently no data can be collected. When Catherine Reissman (2005) conducted some research in Kerala in collaboration with colleges in the 1990s on infertility and it seems she had a similar experience. She found that they did not use consent forms and that there were suspicions about having to sign a form (Reissman, 2005). This is because the action of signing a consent form can have implications in many non-western cultures (Reissman 2005). However, she observed that they did not have problem with the actual interview or being recorded (Reissman 2005).

The differences in research practices was not something we had considered or really discussed before our trip to India. When reflecting on our experience we had in India we now understand that there are differences in research practices across the world. This is important to consider for any future research and we feel is a vital skill we have learnt.

Sophie Ross

References:

Israel, M. (2015) Research Ethics and integrity for Social Scientists. 2nd Edition. London: Sage

Riessman, C.K., (2005). Exporting ethics: A narrative about narrative research in South India. Health:, 9(4), pp.473-490.

 

Cross Language Interviews and Interpreters

In qualitative research, it can be necessary to conduct cross language interviews. In this blog post I am going to discuss the importance of conducting cross language interviews and how interpreters could play a role in this research, following on from the issues of language barriers which have been discussed earlier on in this blog.

Cross language qualitative research is a key part of academic research in order for us to understand different cultures and contexts. Depending on the research question being undertaken it may be key to recruit participants who have English as a Second Language as well as those who speak no English. This is necessary in order for researchers to gather the unique perspectives of those people who are experiencing or have experienced the topic in question. In order to eliminate geographic barriers as well as technical barriers related to language differences it is beneficial to have members of the research team located where you want to conduct your research and with the necessary language abilities. However, this is not always a possibility, meaning language barriers can become a problem (Marshall, 1994; Squires, 2009).

Qualitative interviewing has its core in meanings. Therefore, it is essential to understand the meaning behind what your interviewee is saying and interpret them correctly. Patton (1990) argues that you cannot understand another culture without understanding the language of the people in that culture, due to the cultural assumptions which are embedded into languages. Because of this it is often essential to have an interpreter present for your cross-language interview even when the participant does speak English as a second language. Interpreters can act in numerous ways in an interview – either providing verbatim translations and acting solely as a conduit or acting more independently and taking on a more dominant role in the interview when appropriate (Kapborg and Bertero, 2002). Whilst translating, interpreters can also be used to overcome any cultural differences which may exist between the interviewer and interviewee.

Despite their advantages, interpreters may threaten the validity of qualitative interviewing. As mentioned, the core of qualitative research is in meanings. By introducing an interpreter to translate during the interview, you are introducing another interpretive act. This can damage the quality and validity of academic cross-language research as meaning can often get lost in translation.

Francesca Steeples

References:

Kapborg and Bertero (2002) Using an interpreter in qualitative interviews: does it threaten validity? Nursing Inquiry, 9(1): 52-56

Marshall, S. (1994). Interviewing respondents who have English as a second language: challenges encountered and suggestions for other researchers. Journal of Advanced Nursing, 19(3), pp.566-571.

Patton MQ. 1990. Qualitative evaluation methods. London: Sage

Squires, A. (2009). Methodological challenges in cross-language qualitative research: A research review. International Journal of Nursing Studies, 46(2), pp.277-287.

Van Nes et al (2010) Language differences in qualitative research: is meaning lost in translation? European Journal of Ageing. 7:313-316

 

 

Personal reflections on language issues in research

During our qualitative research on campus as the University of Kerala, we encountered several issues with language barriers. Communication with someone who thinks and speaks in a different language can cause interpretation to be inaccurate as well as cause a lack of understanding for the researcher and the respondent.

One of the first issues prevalent with language barriers was the trouble with gaining participant consent through the participant information sheet. The form was very long and wordy which was necessary, but this confused respondents. We didn’t know if the participant had fully understood what they were agreeing to creating confidentiality and ethical issues.

Participants often didn’t understand the question, so we had to probe but this is difficult to do without being leading. We had experience of trying to help a respondent understand us better but ended up pushing them into an answer. When asked about their experience at the University of Kerala we tried to further the respondents understanding of the question by asking if she felt proud, to which she answered yes. It could be said she was led into this response. Examples to further understanding should be avoided as they can lead responses. Probing if necessary, should be subtle and the basic sentence structure of the question should stay the same.

Issues with accent also caused lack of detail in the responses. Although, it is hard to assess whether a participant was non-responsive due to the language barrier or shyness. Lack of detail is particularly an issue with qualitative studies – many answers were very brief and the participant had nothing further to add even when probed. It is very easy for the respondent to interpret the question other than how it was intended and therefore give an unrelated answer. You can’t fully understand another culture without understanding the language of the people in that culture and this caused a severe lack of detail in our research.

Georgina Carter