Edinburgh Research Explorer Conducting research with vulnerable populations

It can be difficult for researchers to access research participants from vulnerable populations. Focusing on the single victim interviewee recruited for my human trafficking-related research, this article will examine the method employed to conduct research with her, which I term ‘case study by proxy’: a new hybrid qualitative methodological approach combining elements of the case study and interview by proxy methods. This may prove to be a valuable methodological tool for researchers studying vulnerable populations

In my doctoral research undertaken at the University of Edinburgh, I critically assessed how human trafficking is regulated across Britain, focusing on the UK's National Referral Mechanism (NRM): the support framework for trafficking victims.I conducted discourse analysis of UK, Scottish and Northern Irish government policies, carried out semi-structured interviews with stakeholders in England and Scotland, and studied the case of a domestic sex-trafficking survivor, for whom I used the alias 'Eve'.This article will discuss the complexities in accessing trafficking survivors for interviews and the hybrid methodological approach employed to conduct research with Eve that I term 'case study by proxy'.This approach, grounded in the empirical part of my study, has the potential to develop into a qualitative method that can mitigate some of the issues surrounding research with trafficking survivors and other vulnerable populations.This study was undertaken as part of my PhD in law at the University of Edinburgh and draws on my thesis.
Conducting research with human participants is a challenging task.This challenge multiplies when doing research with vulnerable populations (Langford and Bateman-House, 2020).First, there are obstacles related to maintaining robust research ethics throughout the study.Distinguishing between procedural ethics and ethics in practice, the former refers to the approval process by the ethics committee of the researcher's institution and the latter to the ethics surrounding the empirical research (Guillemin and Gillam, 2004).A key concern is the 'power imbalance between researchers and vulnerable groups as a researched population' and the fact that qualitative research with vulnerable populations usually covers sensitive topics that may cause harm, discomfort or distress to participants (Shaw et al, 2020: 279).Correspondingly, Antle (2017) emphasises that researchers should be aware that some questions may be sensitive for participants.
In my case, I sought and obtained full approval by the School of Law Research Ethics and Integrity Committee at the University of Edinburgh before the interviews.The ethics process helped me reflect early on the sensitive topics this research would explore, with the principle of minimising harm for participants being paramount for my research design (Mallia, 2018;Shaw et al, 2020;Drozdzewski and Dominey-Howes, 2021).Key measures to minimise harm included: the use of participant information sheets, informed consent forms and translators, if needed; full briefing with interviewees before the interview; giving participants the right to withdraw and retract their statements, even after the interview; holding interviews with victims in the supportive environment of a non-governmental organisation (NGO) (van Wijk and Harrison, 2013) or, if interviewing outside, having an NGO caseworker together to safeguard the process; the implementation of specific protocols of research if a victim experienced distress during the interview, as agreed after consultation with relevant NGOs; the opportunity for debriefing processes for myself and the interviewee by the NGO; and the possibility of conducting an interview by proxy, using a caseworker as interviewer.
Second, there are obstacles around accessing participants and/or earning their trust.The existence of several layers of gatekeepers should be considered carefully by prospective researchers (McCauley-Elsom et al, 2009;Quinn, 2015).Van Wijk and Harrison (2013) illustrate the importance of maintaining good relationships with gatekeepers, as this may mitigate sampling and access issues (Flanagan and Hancock, 2010;Lyberg et al, 2014), while gatekeeping organisations constitute a 'safe and controlled environment' for recruitment and research (van Wijk and Harrison, 2013: 578).Additional factors that may increase the unattainability of vulnerable populations may be the illegal character of some of the activities they might be engaged with, being 'on the run' from police, marginalisation and stigma (Goode, 2000).
The challenge of accessing trafficking victims for research was made more difficult for an outsider researcher, such as myself.Insider research is conducted with populations of which the researcher is a member (Kanuha, 2000).Contrastingly, outsider researchers are unrelated to the population under study (Corbin Dwyer and Buckle, 2009).Whether a researcher/interviewer possesses the insider or outsider status usually depends on their gender, age, culture/ethnicity, class and shared experiences with interviewees (Manohar et al, 2017).Both statuses present benefits and drawbacks.Insider status may increase the researcher's 'legitimacy', helping secure access and achieve trust with participants (Corbin Dwyer and Buckle, 2009: 58-9).This is due to the 'trust surplus' that exists between the researcher(s) and researched, based on the common factors and experiences they share, which enhance their communication and mutual understanding (Chhabra, 2020: 308).Still, insider status may lead to solipsism, self-centredness, subjectivity and partiality (Fay, 1996;Cammett, 2013;Chhabra, 2020).Contrastingly, outsider status may secure objectivity and critical-minded independence; still, it may hinder access, trust and communication (Cammett, 2013;Manohar et al, 2017;Chhabra, 2020).Bishop (2008: 148) adds that outsiders may lack the 'sensitive and … responsible manner' in conducting research compared to insiders.
In my case, I was an outsider not only towards the population I wanted to study but also towards the gatekeeping practitioners/organisations.This transformed me to an outsider on two levels, which significantly hindered my access to trafficking survivors.To mitigate this, I spent time liaising with gatekeepers, interviewing several of them not only to elicit crucial findings for my study but also to establish a strong bond with them in order to gain access to victims.Through this, I managed to recruit a domestic sex-trafficking survivor, for whom I use the alias 'Eve'.Eve was trafficked in England for ten years since early teenagerhood, until referred to the NRM, eventually receiving a positive conclusive decision on her status as a victim approximately three years before her interview.
From the start of our communication, Eve was keen to participate and have a platform to share her experiences.This aligns with Gray's (2015) observations when interviewing young males in custody with a background of trauma on how his participants started elaborating more freely on their lives only after he had informed them that their discussions could revolve around whatever they wanted.This gave them more control over the interview, making them more comfortable to contribute, something that would not have been the case had a standardised and more restrictive interview guide been used (Gray, 2015).Correspondingly, and in contrast with the semi-structured interviews with stakeholders, my plan regarding interviewing survivors was to employ more unstructured interviews in the form of narratives.These are more suitable for interviewees who reflect on their life, helping researchers elicit in-depth perspectives of how they experienced an event or time period, such as the trafficking or support-receiving experience (Miller, 2000;Riessman, 2008).
This also corresponds with Peterson's (2019) overarching notion: when someone is willing to listen and give a platform to another person to share their experiences, that person is -more often than not -going to honestly open up.This works for the benefit not only of the listener -in this case, the researcher -but also of the sharing individual's 'psyche', as 'people organise their brains with conversation' (Peterson, 2019: 250).This may also prove to be a key research benefit for the studied population: giving individuals the opportunity to open up about their experiences in a platform/setting that is not strictly counselling or support-based, and therefore professionally focusing on them as patients, but more research oriented, focusing on them as bearers of an experience that is unknown and thus invaluable to the researcher (Gale, 1992).This observation becomes important especially for vulnerable populations, who usually already have several experiences of professional support settings.The change of the talking platform to a more research-oriented, free-flowing one may be particularly welcome by them and beneficial or even cathartic in a therapeutic sense (Gale, 1992).
However, the main challenge with Eve was planning the interview.Coordinating three individuals' schedules (Eve, myself and the identified former caseworker of Eve who would safeguard the interview), Eve's understandable unwillingness to travel outside her remote area of residence and the fear of losing her if further delaying the interview were my key concerns.Since I had the ethical clearance to hold an interview by proxy, having predicted such obstacles with my supervisors in line with Quinn's (2015) advice to be flexible and anticipate the unexpected when researching vulnerable populations, we proceeded with an interview between Eve and her former caseworker, who resided much closer to her and was free to hold the interview on Eve's proposed dates.The interviewer's background also helped me decide more comfortably in favour of an interview by proxy.As Gomes et al (2016) explain, a key selection criterion for proxy interviewers should be their knowledge of the researched topic to mitigate the risk of negative impact on the study's validity.Cammett (2013) adds the need for interviewers to possess respect and openness, as well as an in-depth understanding of the study.Therefore, the choice of Eve's former caseworker as a proxy interviewer, made primarily on the basis of practical impediments around planning the interview, also fulfilled the aforementioned needs given the caseworker's background in supporting trafficking survivors and their personal, close and long-standing relationship of trust with Eve, which ensured Eve's safeguarding throughout the interview, as no issues that could cause discomfort or distress to her would be discussed.The result was a one-hour in-depth interview on Eve's trafficking and support experience.I chose to treat Eve's interview as a case study, as it gave me the opportunity to conduct an in-depth analysis of her single case, paying respect to her individual experience as a domestic sex-trafficking victim in a historic trafficking case starting from childhood/ teenagerhood to adulthood and an NRM service user.This agrees with the nature of the case study, which allows the researcher to dive into the unique, complex case of the single unit and conduct an intensive examination to elicit crucial details (Stake, 1995).
Still, the use of a proxy interviewer did not come without limitations.Ideally, I would have liked to personally conduct more interviews with Eve to establish a stronger bond and further dissect topics, such as Eve's traffickers' profile or family background.Additionally, achieving a fully research-minded/oriented setting over a counselling one was not possible due to the use of Eve's former caseworker as the proxy interviewer.Having a social work background may benefit the researcher/interviewer in some respects (for example, professional legitimacy and trauma understanding) but may also hinder their relationship with the interviewee/social work client (Roesch-Marsh et al, 2011).Still, the use of an interview guide containing questions prepared by me and my supervisors that addressed certain points of interest, the interviewer's longstanding experience and knowledge of Eve's case (Gomes et al, 2016), the established relationship of trust they had with Eve, and the fact that they were no longer Eve's caseworker were factors that helped mitigate the limitations just outlined, striking a crucial balance between maintaining a research-setting interview and safeguarding for Eve's needs as well as avoiding secondary traumatisation by asking questions that could have brought discomfort to Eve (Antle, 2017).Additionally, the proxy interviewer, based on their good understanding of my study (Cammett, 2013), asked their own well-targeted, probing questions, which I personally would not have considered.A good example was the interviewer's focus on the word 'rescue' and the question, "What do you think about the notion of rescue?You know, you've been rescued.That's it, isn't it?You've been rescued?", which led Eve to critically reflect on the benefits received by the NRM.
Thus, Eve's interview findings were powerful, her descriptions were detailed and sometimes shocking when elaborating on her treatment by her traffickers or the NRM, and her account reads so honest and natural, covering in good depth a wide array of life periods.Despite not achieving an even more biographical level of detail, Eve's account provided insights that either confirmed or contradicted stakeholders' interview findings, shed light on domestic survivors' experience of the NRM, and showcased Eve's unique relationship with her traffickers, together with the impact the NRM had on her psychological state and social reintegration. in a case study is not whether findings can be generalised to a wider group (for example, all victims) but whether they substantially help the research goals (Bryman, 2012).In other words, the key for a successful case study is not the generalisability but the depth of its findings.
I conclude that when conducting research with vulnerable and thus hard-to-reach populations, the researcher must anticipate a series of challenges in securing even a single interview.Provided that it fits with the research goals of one's study and particularly when the researcher is aiming to gain deeper insight into the details of a single case, the case study is therefore a good methodological option, as it relies on a single unit.My experience with Eve showed that the case study can be employed using a proxy interviewer, which then transforms this into what I term a 'case study by proxy'.This methodological approach, as it emerged grounded in my empirical work, evidences promising potential as an innovation in the field of qualitative research methodologies with vulnerable populations.Bearing in mind the aforementioned limitations, this approach can work particularly well when recruiting an external interviewer who is experienced in the research topic and trusted by the participant.To assess the benefits and limitations of this approach in greater depth, and therefore to evaluate its potential as a new methodology in the field, it would be well worth it to employ it on a wider scale in interviews with more vulnerable participants coming from various backgrounds.In conclusion, this approach may well satisfy the need to produce helpful findings for the researcher while maintaining robust research ethics by safeguarding the welfare of participants with respect to their rights and individual needs.