Thematic Outline

Conducting Research with Survivors of Sex Trafficking: Lessons from a Financial Diaries Study in the Philippines

Laura Cordisco Tsai*

Columbia University School of Social Work, New York, NY 10027, USA

*Correspondence to Laura Cordisco Tsai, Columbia University School of Social Work,

1255 Amsterdam Avenue, New York, NY 10027, USA. E-mail: [email protected]

Abstract

Human trafficking is a significant human rights issue confronting social workers, and

yet rigorous research on human trafficking remains limited, particularly in the social

work field. Numerous ethical and methodological challenges arise when researching

human trafficking. Few studies have, however, examined the process of conducting

research with trafficked persons. This study explored the process of conducting re-

search with survivors of sex trafficking and their family members, focusing on rela-

tional dynamics that emerged in interviewers’ relationships with survivors. Findings

were taken from a six-month financial diaries study that was conducted with survivors

of sex trafficking and their family members in the Philippines, involving 352 inter-

views with survivors and family members. Field observations detailing the inter-

viewers’ interactions with research participants during the financial diaries were

analysed using thematic analysis. Four themes were identified in the analysis: the

trust-building process, multi-layered relationships and managing expectations, situ-

ational responsiveness and the emotional impact of the research process.

Recommendations for conducting research with trafficked persons are provided.

Keywords: Human trafficking, research ethics, qualitative methods, trafficking

research

Accepted: December 2016

# The Author 2017. Published by Oxford University Press on behalf of

The British Association of Social Workers. All rights reserved.

British Journal of Social Work (2018) 48, 158–175 doi: 10.1093/bjsw/bcx017 Advance Access Publication April 6, 2017

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

Introduction

Although over fifteen years have elapsed since the United Nations passed its Convention to define human trafficking (United Nations, 2000), rigorous empirical research on human trafficking remains limited. Global efforts to combat human trafficking have been largely driven by activist groups and government agencies (Zhang, 2009; Tyldum, 2010). Some scholars have critiqued the current evidence base around human trafficking as having emerged from dogmatic, politically driven usage of anecdotal information and weak research methods (Di Nicola, 2007; Zhang, 2009). While research pertaining to human trafficking has been growing, much of the research does not reflect academic standards in other fields (Salt, 2000). Trafficking research remains limited in part due to methodological challenges including: lack of clarity regarding the def- inition of trafficking, the hidden nature of the trafficking industry, prac- tical challenges associated with accessing trafficked persons and difficulties collecting accurate information (Cwikel and Hoban, 2005; Di Nicola, 2007; Zhang, 2009; Choo et al., 2010; Tyldum, 2010; Duong, 2015).

Scholars and practitioners have identified a myriad of ethical and safety issues associated with conducting research with trafficked persons. Many people whose experiences meet the UN definition of human traf- ficking do not self-identify as ‘trafficked’, raising the question of who owns the definition of human trafficking and who gets to decide whether a person is ‘trafficked’.

Additionally, gaining access to currently trafficked persons is challeng- ing and potentially dangerous for all involved, including trafficked per- sons themselves and researchers (Zimmerman and Watts, 2003; Tyldum, 2010). Ethically, it would be exceedingly difficult to justify conducting research with currently trafficked people without helping to prevent their ongoing exploitation and abuse, if they wanted such assistance; however, service providers would be better equipped to intervene than researchers in instances of current victimisation (Tyldum, 2010). Further, well-intentioned interventions on the part of researchers who do not have proper training or experience working with trafficked persons may cause more harm for research participants (Kelly and Coy, 2016).

In 2003, Zimmerman and Watts released World Health Organization (WHO) guidelines for interviewing trafficked women, outlining ethical and safety recommendations for research with this population. Recommendations included: careful selection and training of research team members; preparation of accurate and thoroughly screened referral information; demonstrating respect for women’s perspectives; not re- traumatising participants; not assuming trafficked women are traumat- ised; being prepared to respond to imminent danger; and using findings

Conducting Research with Survivors of Sex Trafficking 159

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

to further the development of interventions and policies to benefit traf- ficked persons (Zimmerman and Watts, 2003). Building upon this foun- dation, the United Nations Inter-Agency Project on Human Trafficking (UNIAP) published a guide to ethics and human rights in counter- trafficking work, including research with both male and female survivors (UNIAP, 2008). While these guidelines serve as a valuable foundation for scholars and practitioners who are embarking on research pertaining to human trafficking, further research is needed exploring ethical and methodological challenges faced in trafficking research. What should re- searchers be prepared to encounter when they conduct research on human trafficking? How can research pertaining to trafficking be con- ducted in a physically and emotionally safe manner for all parties involved? How can researchers build rapport and trust with trafficked persons? How can researchers ensure that they are collecting accurate data?

For researchers embarking on studying human trafficking, it is evident that clear ethical guidelines must be followed in the field. Ethnographers have noted that one essential aspect of preparing for ethical decision making in the field is the review of case studies that describe dilemmas encountered by previous researchers (VanderStaay, 2005). Few studies have been undertaken about the process of conducting research with people who have been trafficked. This study explores the process of con- ducting research with trafficked persons and relational dynamics that emerged in interviewers’ relationships with survivors during a financial diaries study with sex-trafficking survivors in the Philippines.

Method

This study addresses the following research question: What relational dy- namics emerged in interviewers’ relationships with trafficking survivors and their family members during a financial diaries study in the Philippines? As will be described below, descriptive field notes written by interviewers in a financial diaries study were analysed to answer this research question. As such, the current study presents the interviewers’ perspectives on their relationships with trafficking survivors and their family members during the study.

Parent study

Findings came from a parent study: a financial diaries study that was conducted with sex-trafficking survivors and their family members in Cebu City, Philippines (Tsai, 2017a). The financial diaries is a method- ology originating from behavioural economics that aims to generate

160 Laura Cordisco Tsai

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

a comprehensive understanding of the way people living in poverty man- age their finances (Collins et al., 2009). In the parent study, the financial diaries methodology was utilised to explore financial vulnerability among trafficking survivors upon exiting human trafficking and returning to live in the community. Many trafficking survivors face substantial fi- nancial difficulties when re-entering the community after being traf- ficked, including a lack of access to safe employment, lack of savings, responsibility to repay familial debts, and financial anxiety (Tsai, 2017b; Smith-Brake et al., 2015; Le, 2017). Financial stability is often a key priority for survivors upon community re-entry (Smith-Brake et al., 2015). Financial diaries study findings can be used to inform the design of reintegration services that are responsive to the financial needs of sur- vivors and their families (Tsai, 2017a).

Eligibility criteria for the study included the following: women aged eighteen and above who were formerly trafficked into sex work, had exited human trafficking and were residing in Cebu City, Philippines. To recruit participants, I partnered with two non-governmental organisa- tions (NGOs) providing services to trafficking survivors, as this is often the safest method of recruiting research participants with a trafficking history (Zimmerman and Watts, 2003; Tyldum, 2010). The referring agencies assessed through their own internal procedures whether pro- spective participants could be classified as ‘victims of trafficking’ accord- ing to the Filipino Expanded Anti-Trafficking in Persons Act of 2012 (Republic Act Number 10364, 2012). According to the Republic Act 10364 (2012), human trafficking is defined as:

The recruitment, transportation, transfer or harboring, or receipt of

persons with or without the victim’s consent or knowledge, within or

across national borders by means of threat or use of force, or other

forms of coercion, abduction, fraud, deception, abuse of power or of

position, taking advantage of the vulnerability of the person, or, the

giving or receiving of payments or benefits to achieve the consent of a

person having control over another person for the purpose of

exploitation which includes at minimum, the exploitation or the

prostitution of others or other forms of sexual exploitation, forced labor

or services, slavery, servitude or the removal or sale of organs (p. 1).

Although people can be trafficked into any industry, this study included only women who were previously trafficked into sex work, as this was the population primarily served by partner agencies. As per the above definition, adults who voluntarily engage in sex work are not con- sidered sex-trafficking victims. However, any person who is forced to en- gage in sex work through any of the means included in this definition (i.e. force, fraud or coercion) would be classified as a trafficked person (Republic Act Number 10364, 2012; United Nations, 2000). According to this legislation, any minor (below the age of eighteen) engaged in sex work is considered to be a trafficking victim regardless of consent.

Conducting Research with Survivors of Sex Trafficking 161

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

Definitions of human trafficking are highly contested. However, a com- plete discussion of the controversy surrounding these definitions is be- yond the scope of this paper. While all participants were adult women at the time of this study, some were trafficked into sex work when they were minors.

Among the thirty sex-trafficking survivors whose households partici- pated in the study, the average age of survivors was 21.5 (SD¼2.96). Half of survivors had at least one child and half had not completed high school (n ¼ 15). Most survivors were unemployed upon community re- entry (60 per cent); 26.7 per cent had self/causal employment outside sex work; 10 per cent had salaried employment; and one survivor re- ported working in sex work full-time during the financial diaries study.

The financial diaries study lasted six months, involving bi-monthly interviews with the households of thirty sex-trafficking survivors (total- ling twelve interviews per household). Interviews were held in a location of preference for participants (Liamputtong, 2000; Urada and Simmons, 2014). Structured questionnaires were used in each interview. Interviews lasted on average one hour. The first four study questionnaires ad- dressed household and family composition, household income sources, common household expenditures, use of financial instruments (such as savings, lending, borrowing and money guarding), attitudes about man- aging money, critical financial events and family norms around financial decision making. Interviews 5 through 12 involved completing a work- sheet recording all money coming into and going out of the household over the past two weeks, resulting in a continuous record of all financial transactions in the household for the final four months of the study. All interviews were conducted with the person who reported being the pri- mary household financial manager, as recommended by prior research (Cull and Scott, 2009). Trafficking survivors served as the study partici- pant in nineteen out of thirty households (63 per cent); in eleven house- holds (37 per cent), a parent of the survivor was the financial manager and therefore the study participant. Twenty-eight out of thirty house- holds (93 per cent) completed all twelve interviews. The study team lost contact with one participant at Interview 7 and one participant at Interview 9. Participants received 200 Philippine pesos per interview for the first six interviews, 225 pesos per interview for Interviews 7–9 and 250 pesos per interview for Interviews 10–12. Further details regarding the design of the financial diaries study can be found elsewhere (Tsai, 2017a).

Ethics approval was obtained from the Columbia Morningside Institutional Review Board and the Philippines Department of Social Welfare and Development. All participants gave written informed con- sent. All names used in the paper are pseudonyms. Two Filipina re- search assistants were hired to conduct all interviews. Interviews were conducted in Cebuano, the primary language spoken in Cebu Province

162 Laura Cordisco Tsai

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

in the Philippines. Both interviewers were bilingual in Cebuano and English, and had previous experience working with trafficking survivors. To ensure the safety and protection of participants, the interviewers par- ticipated in extensive training prior to study implementation. The inter- viewers were trained in establishing and maintaining rapport with participants; ethical issues in research; obtaining informed consent; confi- dentiality, data protection and management; conducting structured inter- views; dynamics of human trafficking and intimate partner violence; managing distress and conducting crisis referrals; detecting, handling and reporting adverse events; boundaries of collaboration with referring part- ners; and secondary trauma and self-care.

Data analysis

Immediately following interviews, interviewers wrote descriptive field observations for each interview. Field observations addressed, among other topics, nuances in communication with the participants (such as body language, affect, tone), the interviewer’s reflections on the inter- view process, observations of interactions between members of the household, disclosures made by the research participant outside the questionnaire and any challenges the participants or the interviewer may have faced during the interview process (Emerson et al., 1995). A total of 352 interviews were conducted during the financial diaries study; as such, 352 field observations were also recorded. For the current paper, all field observations were analysed to answer the abovementioned re- search question. In this study, analysis was not conducted of the individ- ual interviews with participants in the financial diaries study; rather, themes for this study were identified exclusively through analysis of the interviewers’ field notes.

The data were analysed using thematic analysis, which has been defined as a ‘method for identifying, analysing and reporting patterns (themes) within data’ (Braun and Clarke, 2006, p. 79). Although the rig- our of thematic analysis has been called into question, some scholars have argued that thematic analysis is a suitable method for studying understudied phenomena, especially when a distinct sequence of analytic stages is followed (Vaismoradi et al., 2013). An inductive approach to thematic analysis was utilised, meaning that the themes were developed from the data themselves rather than trying to link the data with an ex- istent coding frame (Braun and Clarke, 2006). Thematic analysis pro- ceeded in five stages, as articulated by Braun and Clarke (2006): familiarising oneself with data, generating initial codes, searching for themes, reviewing themes, and defining and naming themes. Three members of the study team participated in the data analysis process, including the Principal Investigator and two interviewers for the

Conducting Research with Survivors of Sex Trafficking 163

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

financial diaries study. First, the team familiarised itself with the data through repeated, active reading of the field observations. The three coders independently reviewed the field notes and conducted line-by- line coding all of field observations. When at least two of the three coders recorded a code representing the same concept in the same line, the coders came to an agreement on the exact wording of the code and the code was selected for the final list of codes. If only one coder noted a code, the three coders discussed the reasons for the discrepancy, agreed on a plan and added the code to the list, if the group collectively decided that the code should be added (Guest et al., 2012). Once the group finalised the complete list of codes, the group collectively re- viewed the codes and sorted the codes into conceptually similar catego- ries. Categories were compared and grouped together into potential themes. The team then reviewed all possible themes, comparing the themes to the coded data extracts relevant to that theme and reviewing the themes in comparison to the entire data-set. In the final stage, the group compared themes, defined and refined themes, and discussed the stories associated with each theme (Braun and Clarke, 2006).

Results

Four themes were identified during the analysis: the process of building trust with survivors, multi-layered relationships and managing participant expectations, importance of situational responsiveness among re- searchers and the emotional impact upon researchers. Each theme is dis- cussed separately below.

Process of building trust with survivors

At the beginning of the study, many trafficking survivors exhibited ap- prehension about disclosing information to the interviewers, limiting their answers and withholding sensitive information. Questions pertain- ing to household income proved to be particularly sensitive at first. For instance, one of the survivors whose household participated in the study, Geynalyn, was working in sex work full-time during the study. Her mother, Charito, was the primary financial manager and therefore the study subject. When asked about household income sources, Charito acknowledged that Geynalyn regularly contributed financially to the family, but declined to share the kind of work Geynalyn engaged in, simply stating that Geynalyn worked in a ‘service industry’. Charito dis- closed later in the study that Geynalyn was working in sex work and that the family was financially dependent upon the money Geynalyn contributed to the household. As the study progressed, Geynalyn

164 Laura Cordisco Tsai

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

revealed that she had returned to sex work because her stepfather was not contributing financially to the household and she felt responsible for her mother and younger siblings.

As the study progressed, trust grew between the interviewers and re- search participants. During the course of the six-month study, interviewers had the opportunity to gradually build relationships with participants, as interviewers were assigned to the same households for the duration of the study. Participants expressed directly to the research team that the ex- tended nature of the study helped them feel more comfortable with the interviewers. As Loreen, one of the survivors participating in the study, expressed: ‘At the beginning, it’s hard for us to trust anyone. But now since it [the study] took us a while already, now we can trust.’

As trust grew between interviewers and study participants, the ‘flood- gates started to open’. Survivors would often begin their meetings with interviewers updating them on recent events in their lives—primarily their relationships with their partners, family dynamics they were strug- gling with and topics that they felt they could not discuss elsewhere. Since the interviewers were trained to be empathetic listeners, they would listen to the survivors’ stories and challenges, and proceed to the interview questions once the survivors were ready to do so. The sur- vivors would often spend more time updating the interviewers on their lives than discussing the topics planned for the interviews. Many sur- vivors disclosed sensitive information to the interviewers without any prompts or questions from the interviewers. For instance, the study questionnaires pertained exclusively to the financial situation of the household. However, 40 per cent of participants voluntarily disclosed experiencing or witnessing family violence during the study without being asked any questions about violence.

Multi-layered relationships and managing participant expectations

Although the trust that grew between research participants and inter- viewers helped participants feel more comfortable with the research pro- cess, relationships between interviewers and participants became more complex as the study progressed. While interviewers, survivors and their family members related to one another as researcher to participant, their relationships extended beyond these roles. Some participants began to treat the interviewers as members of their family and/or friends. For in- stance, one survivor participating in the study, Noreen, asked an inter- viewer to become the godmother to her baby who was born during the study. Participants invited interviewers to attend family gatherings, such as birthday parties. Survivors and their family members also expressed care and concern for interviewers. The husband of one participant

Conducting Research with Survivors of Sex Trafficking 165

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

attempted to teach an interviewer how to use a gun and how to defend herself through martial arts, as the family was concerned for the inter- viewer’s safety when she travelled to their neighbourhood to conduct interviews.

Additionally, it became evident that some research participants began to see the interviewers as social workers rather than social work re- searchers. For instance, one survivor brought a friend to an interview and asked the interviewer to help her friend find a job; the interviewer then travelled with the survivor and her friend to an internet café and assisted the survivor’s friend in drafting a résumé prior to starting the interview. In some situations, participants explicitly shared their need for money with interviewers. One survivor frequently requested that interviews be held in advance of the allocated schedule in order to re- ceive her stipend earlier. Another survivor, Jasmine, shared her lack of money in most interviews—primarily for herself, but also for her father in prison. Although she never directly requested that the interviewer give or loan her money, the interviewer felt that Jasmine hoped to re- ceive financial assistance apart from the compensation that was provided for participation in the interviews. Given concerns about overstepping boundaries with research participants, the research team carefully dis- cussed all of these situations as a team as they arose and came to a group consensus (Eide and Kahn, 2008). In each case, the interviewer did not provide additional finances to the participants apart from the agreed-upon stipends. Instead, interviews were sometimes conducted over meals, with food being provided for participants (UNIAP, 2008). Interviewers brainstormed with the participants about other avenues for obtaining the funds they needed and referred them to services in the community when appropriate. However, these decisions proved difficult for the research team at times, as it can be difficult for researchers to spend time with someone who is distressed without offering direct assist- ance (Dickson-Swift et al., 2008).

Consistently with WHO guidelines for interviewing trafficked women (Zimmerman and Watts, 2003), the research team had prepared a pre- screened referral list prior to commencing the study. Given the diversity of needs expressed by survivors during the study, it was important to ar- range appropriate referrals for a wide array of services. Interviewers provided referrals to participants for numerous services, including emer- gency shelter, long-term housing, employment, medical, reproductive health and legal services, among others. While the research team saw these referrals as consistent with ethical guidelines on conducting re- search with trafficked persons, there was concern that such assistance could lead to role confusion. As noted earlier, two NGO partners referred survivors for participation in the study. During the study screen- ing and informed consent process, interviewers explained that they did not work for these NGOs, that the study was conducted independently

166 Laura Cordisco Tsai

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

from these organisations, that participants may receive no direct benefit from their participation and that the prospective participants’ decisions about participation in the study would in no way impact the receipt of services from these NGOs. However, once survivors and family mem- bers came to trust the interviewers, this was often accompanied by heightened expectations about the ways in which interviewers may be able to help participants and their families. In particular, some of the mothers of trafficking survivors who participated in the study had a diffi- cult time understanding that the interviewers did not work for the refer- ring NGOs or for the Philippines Department of Social Welfare and Development. As a result, it was important for the interviewers to revisit the informed consent procedure throughout the study and explain on nu- merous occasions the nature of their role and their independence from these organisations.

Importance of situational responsiveness among researchers

One of the most common challenges encountered during the research process was the instability of the participants’ housing situations. Survivors moved frequently during the study, making it difficult to main- tain contact with some participants, as others conducting research with trafficked persons have experienced (Choo et al., 2010). Interviewers learned to ask survivors about updates to their housing situation in every interview due to frequent moves. Survivors’ personal circumstances were often unstable, leading survivors to change their interview schedules fre- quently. Many participants were living at a subsistence level and had ir- regular employment/access to income. Therefore, when opportunities to earn an income presented themselves, these would be prioritised and interviews would need to be rescheduled accordingly. Additionally, among survivors who were mothers, lack of access to childcare was a constant challenge. Most interviews with mothers were conducted with babies/young children present, requiring flexibility on the part of the interviewers and impacting what topics could be openly discussed.

While the above situations required adaptability on the part of the re- search team members, situational responsiveness also was required in emergencies. Previous guidelines have stressed the importance of staff being prepared and trained to respond to emergency situations with traf- ficking survivors (Zimmerman and Watts, 2003; UNIAP, 2008). Prior to starting the financial diaries study, interviewers were trained in imple- mentation of a violence safety-planning protocol and responding to emergency situations. As a part of this training, interviewers conducted role-plays with the Principal Investigator exploring how to respond to case study scenarios, such as witnessing bruises on the participants, dis- closures of being afraid of one’s partner and disclosures of severe

Conducting Research with Survivors of Sex Trafficking 167

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

physical violence, among others. Even with careful training and prepar- ation prior to the start of the study, interviewers still encountered unex- pected situations that required careful and timely responses. For instance, one survivor, Jasmine, disclosed to an interviewer that she had experienced physical violence from her boyfriend in the past. Late one evening, the interviewer received a call from Jasmine, who said that she was arguing with her boyfriend. Jasmine asked the interviewer to stay on the phone, quickly telling the interviewer she wanted her boyfriend to know that someone was listening to the conversation (presumably to discourage him from becoming violent). Jasmine put down the phone and the interviewer quietly stayed on the line, hearing both the boy- friend and Jasmine yelling in the background. The interviewer heard the boyfriend yell ‘Are you trying to get help from Ms.?,’ after which point the call dropped. The interviewer did not know where Jasmine was and did not want to provoke the boyfriend’s anger by calling back immedi- ately. The interviewer followed the study emergency procedure, calling the Principal Investigator on call. The interviewer met with Jasmine to implement the safety-planning procedure as soon as it was safe to do so, provided referrals to domestic violence agencies and reached out to Jasmine’s social worker, with Jasmine’s permission. While the inter- viewer followed emergency procedures, she felt helpless in not being able to do more to assist the participant in the moment.

Another survivor, Jelena, was raped by her father during the study. Following her sexual assault, she fled her household (where she had lived alone with her father), becoming homeless and moving between different households for the remainder of the study. Jelena shared about her sexual assault during an interview, at which point the safety-planning protocol was administered, referrals to community services were provided and connections were made to her social worker, with Jelena’s permission. A community partner offered emergency shelter to Jelena, which she declined. Despite becoming homeless, Jelena expressed that she wanted to remain in the financial diaries study until the study com- pleted. The interviewer worked with Jelena to help her complete the study while staying in a different person’s home almost weekly towards the end of the study.

Emotional impact upon researchers

As other scholars conducting research on human trafficking and violence have found, the research process had an adverse emotional impact on interviewers (Coles et al., 2014; Easton and Matthews, 2016). Disclosures of sexual, physical and emotional violence caused the deepest emotional distress for the interviewers. Along with training interviewers about sec- ondary trauma prior to commencing the study, the Principal Investigator

168 Laura Cordisco Tsai

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

monitored interviewers for symptoms of vicarious trauma throughout the study and encouraged team members to monitor one another. Both interviewers experienced symptoms of vicarious trauma during the study, including anger, irritability, helplessness, insomnia and heightened pro- tectiveness. Funding was allocated in the budget to cover self-care activ- ities for both interviewers when needed.

In addition to being deeply impacted by participants’ disclosures of violence, the financial situation of participants caused emotional distress for interviewers. As interviewers built trust with participants, survivors and their family members disclosed a wide range of financial challenges, including several instances of husbands/boyfriends spending their income on alcohol, gambling and drugs while the women struggled to find enough money to feed children in the household. As articulated in Tsai (2017a), survivors adopted a range of strategies in order to survive with- out employment, such as taking donations from friends, staying in emo- tionally and/or physically abusive relationships and developing relationships with men online who would send money to them when needed. Even though the study team prepared a range of referral re- sources for participants, some participants chose not to avail of these services, raising questions about why survivors sometimes declined ser- vices. While the interviewers strove to maintain proper boundaries with participants, they cared deeply for and were concerned about the well- being of participants (Dickson-Swift et al., 2008). Interviewers expressed that they often wished they were in the role of service provider instead of researcher.

While the research process caused emotional distress for interviewers, the process also had a positive impact on the research team. Interviewers were impressed with the resilience they observed in partici- pants and encouraged when they saw survivors and their family mem- bers taking positive steps to achieve their goals. As the financial diaries study focused on the financial situation of survivors and their family members, interviewers reported learning from the financial management strategies of the participants and applying practices in their own lives that they learned from survivors and their families. Although both inter- viewers had prior experience working with trafficking survivors, the re- search process gave them greater insight into the day-to-day challenges faced by survivors upon community re-entry and fostered deeper em- pathy for survivors and their families.

Discussion

This study contributes to the nascent literature exploring the process of conducting research with survivors of human trafficking. As Eide and Kahn (2008) have articulated, ‘qualitative research requires a mutual

Conducting Research with Survivors of Sex Trafficking 169

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

standpoint, researcher to participant, human being to human being’ (p. 199). The primary aim of the interviews was not to provide a thera- peutic encounter for participants (Padgett, 1998). However, the skills involved in research interviewing overlap with skills required in thera- peutic work—including active listening, an ability to develop rapport and form an accepting relationship, and an attention to understanding the ex- periences of the other (Polkinghorne, 2005). Interviews may be viewed as ‘therapeutic’ in nature to research participants if they are listened to and responded to in an empathetic manner when they choose to disclose very sensitive or personal experiences (Birch and Miller, 2000; Dickson-Swift et al., 2008). Having the opportunity to speak with a warm, respectful, non-judgemental and compassionate person may be particularly important for human-trafficking survivors, who may feel they have been commodi- fied by their prior experiences (Cwikel and Hoban, 2005).

Critiques have, however, been raised about overemphasising emo- tional connections between researchers and participants, which can give rise to unclear boundaries (Dickson-Swift et al., 2009; Campbell et al., 2010). Findings from this study reflect the complexities of this issue. As trust is commonly broken during the trafficking process, rebuilding trust is central not only to survivors’ recovery, but also to conducting research with trafficked persons (Brennan, 2005; Easton and Matthews, 2016). Ethnographic methods and research designs that involve multiple inter- views are often necessary in order for trafficked persons to trust re- searchers (Tyldum, 2010; Yea, 2016). Survivors should be given the opportunity to set the pace of interviews (Zimmerman and Watts, 2003; Urada and Simmons, 2014), which is easier to accomplish over multiple interviews. However, as researchers and participants spend more time together, emotional connections grow. In the financial diaries study, the trust and connection that was fostered between interviewers and partici- pants at times contributed to role confusion, with survivors treating interviewers more and more like family members and social workers as the study progressed. As trust grew, survivors were more likely to ask the interviewers for help in other areas, as Yea (2016) also found.

These dynamics raise questions about how researchers can practically live out the principles of beneficence, non-maleficence and autonomy in research with trafficked persons. It is self-evident that the informed con- sent process must be handled in a conscientious manner at the study outset, as the concept of informed consent may be unfamiliar to traf- ficked persons (Cwikel and Hoban, 2005; Duong, 2015). Trafficking sur- vivors may be accustomed to sharing about their experiences with professionals, such as social workers, who offer tangible benefits (Brennan, 2005). Even if researchers explain verbally and in written con- sent forms that the researcher’s role is not to provide services, survivors may still expect some form of assistance from researchers (Brennan, 2005; Surtees and Brunovskis, 2016). Researchers should be aware of

170 Laura Cordisco Tsai

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

how expectations and understandings of the role of the researcher may change over the course of the relationship between the trafficked person and researcher (Yea, 2016; Goldenberg et al., 2015). In addition to ask- ing participants about their expectations for study participation at the study outset, it is recommended that researchers revisit these expect- ations and the informed consent process throughout the study. It is sug- gested that researchers develop a list of possible scenarios at the outset of the study with general guidance for field staff on how to handle such events (i.e. how to respond if a participant asks to borrow money, etc.). Roles and boundaries will likely need to be continually revisited throughout the study, involving a fine balance between maintaining an ethic of caring while managing the expectations of research participants (Dickson-Swift et al., 2009).

When research participants experience therapeutic benefit from par- ticipating in the study, researchers should provide linkages to alternative forms of support for participants upon completion of the study (Eide and Kahn, 2008). It is, however, prudent to be realistic about the level of support that can be provided by partners to survivors during and upon completion of the study so that no promises are made to traffick- ing survivors that cannot be fulfilled. Findings from this study stress the importance of a strong referral list for study participants at the outset, which must involve careful screening of all referrals to ensure the quality of services and adherence to human rights standards (Zimmerman and Watts, 2003; UNIAP, 2008). Although a range of services (both trafficking-specific and not trafficking-specific) were available for sur- vivors in the financial diaries study, in some contexts, researchers may not have sufficient referral networks due to unavailability of services for trafficked persons and/or poor quality of services (Surtees and Brunovskis, 2016; Goldenberg et al., 2015). In all interactions with refer- ring partners, it is essential that researchers engage survivors in discus- sions about what they want and need, respect survivors’ wishes and refrain from assuming that they know what is best for the trafficked per- son him/herself (Kelly and Coy, 2016; Zimmerman and Watts, 2003).

Researchers should expect the unexpected when conducting research on human trafficking. Findings reinforce the importance of careful pre- study planning and capacity building pertaining to emergencies and unex- pected scenarios. All research team members should be equipped to im- plement safety-planning procedures (Choo et al., 2010; Duong, 2015). Nonetheless, situations will likely arise that were not covered in pre-study preparations and procedural documents. Emergency on-call procedures, regular team debriefings and careful, consistent supervision of field re- searchers are essential in addressing issues as they arise in the research process, including responding to emergency situations, the appropriate use of referrals and boundaries of the researcher–participant relationship (UNIAP, 2008; Dickson-Swift et al., 2009; Choo et al., 2010). It is

Conducting Research with Survivors of Sex Trafficking 171

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

recommended that researchers prepare a contingency budget for emer- gency situations that may arise.

Additionally, the emotional hardship associated with conducting re- search on human trafficking should not be underestimated. In addition to vulnerability experienced due to a person’s trafficking history, survivors commonly experience social and economic vulnerabilities that precede their trafficking (Simkhada, 2008; Easton and Matthews, 2016) and con- tinue after they exit human trafficking (Smith-Brake et al., 2015; Tsai, 2017b; Le, 2016). Even if survivors’ trafficking histories are not discussed in interviews, the multiple vulnerabilities experienced by survivors means that researchers are likely to encounter traumatic material in interviews with survivors. Scholars in the field of sexual violence have argued for heightened understanding of the level of vicarious traumatisation experi- enced by sexual violence researchers (Dickson-Swift et al., 2009; Coles et al., 2014). Like clinicians, researchers focusing on violence encounter traumatic material; however, unlike service providers, the researcher’s role is not to directly assist the victimised person, which may exacerbate experiences of vicarious trauma and helplessness among researchers (Coles et al., 2014). It is important to be pro-active in preparing for and responding to vicarious traumatisation of research team members (Easton and Matthews, 2016). Research supervisors should be upfront in explain- ing difficulties associated with research on human trafficking, properly train team members regarding vicarious trauma and self-care strategies, and allocate funding and time for self-care strategies, as needed. As dem- onstrated in the field of sexual violence research, research teambuilding and a supportive work environment are critical aspects of helping to ad- dress vicarious traumatisation in the research process (Dickson-Swift et al., 2008; Coles et al., 2014). Research team members should be pro- vided with the opportunity to reflect on their experiences in a safe, non- judgemental environment, as has been recommended for domestic vio- lence service providers (Slattery and Goldman, 2009).

Limitations

This study presents relational dynamics that emerged between trafficking survivors and interviewers from the perspective of the interviewers, as recorded in the study field observations. It is possible that additional dy- namics may have surfaced during interviews that were not captured in field notes. Additionally, survivors’ perspectives on the research question are not reflected in this paper. Future research should elicit feedback from trafficked persons regarding their perspectives on the research pro- cess and their relationships with research team members. Further, par- ticipation in the financial diaries study was limited to female, adult survivors of sex trafficking and their family members. While the current

172 Laura Cordisco Tsai

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

study does not aim for representativeness, additional dynamics beyond those discussed in this study may emerge with other populations of traf- ficked persons.

Conclusion

Human trafficking has been given insufficient attention in the social work literature (Hodge and Lietz, 2007). Discussions regarding human- trafficking research often concentrate on concerns about the rigour of data produced. Expanding this conversation to include greater attention to process-oriented perspectives on trafficking research can lead to more ethically responsible and instructive outcomes (Yea, 2016). This study describes the complexities associated with field research with trafficking survivors, including the need for trust building over time, managing and revisiting roles and expectations throughout the research process, imple- menting proper pre-study planning and ongoing supervision pertaining to emergencies and other unexpected situations, and addressing vicarious traumatisation among research team members. Further research is rec- ommended to explore experiences of vicarious traumatisation among re- searchers studying human trafficking, the process of conducting research with other populations of trafficked persons (including men and people trafficked into forced labour) and trafficked persons’ perspectives on the research process.

Acknowledgements

Many thanks to Ivy Flor Seballos-Llena and Fe Tudtud for your assist- ance with data collection and analysis. Your dedication to the financial diaries study is deeply appreciated.

References

Birch, M. and Miller, T. (2000) ‘Inviting intimacy: The interview as therapeutic op-

portunity’, Journal of Social Research Methodology, 3(3), pp. 189–202.

Braun, V. and Clarke, V. (2006) ‘Using thematic analysis in psychology’, Qualitative

Research in Psychology, 3(1), pp. 77–101.

Brennan, D. (2005) ‘Methodological challenges in research with trafficked persons:

Tales from the field’, International Migration, 43(1-2), pp. 35–54.

Campbell, R., Adams, A. E., Wasco, S. M., Ahrens, C. E. and Sefl, T. (2010) ‘“What

has it been like for you to talk with me today?”: The impact of participating in

interview research on rape survivors’, Violence Against Women, 16(1), pp. 60–83.

Choo, K., Jang, J. O. and Choi, K. (2010) ‘Methodological and ethical challenges to

conducting human trafficking studies: A case study of Korean trafficking and

smuggling for sexual exploitation to the United States’, Women & Criminal

Justice, 20(1–2), pp. 167–85.

Conducting Research with Survivors of Sex Trafficking 173

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

Coles, J., Astbury, J., Dartnall, E. and Limjerwala, S. (2014) ‘A qualitative explor-

ation of researcher trauma and researchers’ responses to investigating sexual vio-

lence’, Violence Against Women, 20(1), pp. 95–117.

Collins, D., Morduch, J., Rutherford, S. and Ruthven, O. (2009) Portfolios of the Poor:

How the World’s Poor Live on $2 a Day, Princeton, Princeton University Press.

Cull, R. and Scott, K. (2009) Measuring Household Usage of Financial Services: Does

It Matter How or Whom You Ask?, Washington, DC, The World Bank

Development Research Group.

Cwikel, J. and Hoban, E. (2005) ‘Contentious issues in research on trafficked women

working in the sex industry: Study design, ethics, and methodology’, Journal of

Sex Research, 42(4), pp. 306–16.

Di Nicola, A. (2007) ‘Researching into human trafficking: Issues and problems’, in

Lee M. (ed.), Human Trafficking, Cullompton, Willan.

Dickson-Swift, V., James, E. L. and Liamputtong, P. (2008) Undertaking Sensitive

Research in the Health and Social Sciences: Managing Boundaries, Emotions and

Risks, Cambridge, Cambridge University Press.

Dickson-Swift, V., James, E. L., Kippen, S. and Liamputtong, P. (2009) ‘Researching

sensitive topics: Qualitative research as emotion work’, Qualitative Research, 9(1),

pp. 61–79.

Duong, K. A. (2015) ‘Doing human trafficking research: Reflections on ethical chal-

lenges’, Journal of Research in Gender Studies, 5(2), pp. 171–90.

Easton, H. and Matthews, R. (2016). ‘Getting the balance right: The ethics of research-

ing women trafficked for commercial sexual exploitation’, in Siegel, D. and de Wildt,

R. (eds), Ethical Concerns in Research on Human Trafficking, Heidelberg, Springer.

Eide, P. and Kahn, D. (2008) ‘Ethical issues in the qualitative researcher–participant

relationship’, Nursing Ethics, 15(2), pp. 199–207.

Emerson, R. M., Fretz, R. I. and Shaw, L. L. (1995) Writing ethnographic field notes,

Chicago, The University of Chicago Press.

Goldenberg, S. M., Mindt, M. R., Jimenez, T. R., Brouwer, K., Miranda, S. M. and

Fisher, S. C. B. (2015) ‘Structural and interpersonal benefits and risks of participa-

tion in HIV research: Perspectives of female sex workers in Guatemala’, Ethics &

Behavior, 25(2), pp. 97–114.

Guest, G., MacQueen, K. M. and Namey, E. E. (2012) Applied Thematic Analysis,

Thousand Oaks, Sage.

Hodge, D. R. and Lietz, C. A. (2007) ‘The international sexual trafficking of women

and children: A review of the literature’, Affilia, 22(2), pp. 163–74.

Kelly, L. and Coy, M. (2016) ‘Ethics as process, ethics in practice: Researching the

sex industry and trafficking’, in Siegel, D. and de Wildt, R. (eds), Ethical Concerns

in Research on Human Trafficking, Heidelberg, Springer.

Le, P. D. (2017) ‘“Reconstructing a sense of self”: Trauma and coping among re-

turned women survivors of human trafficking in Vietnam’, Qualitative Health

Research, 27(4), 509–19.

Liamputtong, P. (2000) Performing Qualitative Cross-Cultural Research, Cambridge,

Cambridge University Press.

Padgett, D. K. (1998) Qualitative Methods in Social Work Research: Challenges and

Rewards, Thousand Oaks, SAGE Publications.

Polkinghorne, D. E. (2005) ‘Language and meaning: Data collection in qualitative re-

search’, Journal of Counseling Psychology, 52(2), pp. 137–45.

174 Laura Cordisco Tsai

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

Salt, J. (2000) ‘Trafficking and human smuggling: A European perspective’,

International Migration, 38(3), pp. 31–56.

Simkhada, P. (2008) ‘Life histories and survival strategies amongst sexually trafficked

girls in Nepal’, Children andSociety, 22(3), pp. 235–48.

Slattery, S. and Goodman, L. (2009) ‘Secondary traumatic stress among domestic vio-

lence advocates: Workplace risk and protective factors’, Violence Against Women,

15(11), pp. 1358–79.

Smith-Brake, J., Lim, V. and Nhanh, C. (2015) Economic Reintegration of Survivors

of Sex Trafficking: Experiences of Filial Piety and Financial Anxiety, Phnom Penh,

Chab Dai.

Surtees, R. and Brunovskis, A. (2016) ‘Doing no harm: Ethical challenges in research

with trafficked persons’, in Siegel, D. and de Wildt, R. (eds), Ethical Concerns in

Research on Human Trafficking, Heidelberg, Springer.

Tsai, L. C. (2017a) ‘Family financial roles assumed by sex trafficking survivors upon

community re-entry: Findings from a financial diaries study in the Philippines’,

Journal of Human Behavior in the Social Environment Advance Access published

1 March 2017, 10.1080/10911359.2017.1288193.

Tsai, L. C. (2017b) ‘The process of managing family financial pressures upon commu-

nity re-entry among survivors of sex trafficking in the Philippines: A grounded

theory study’, Journal of Human Trafficking Advance Access published 23

January 2017, 10.1080/23322705.2016.1199181.

Tyldum, G. (2010) ‘Limitations in research on human trafficking’, International

Migration, 48(5), pp. 1–13.

United Nations (2000) United Nations Protocol to Prevent, Suppress, and Punish

Trafficking in Persons, Especially Women and Children, supplementing the United

Nations Convention against Transnational Organized Crime: Article 3 (a–d), G.A. res.

55/25, annex II, 55 U.S. (GAOR Supp. (No. 49) at 60, U.N. Doc.A/45/49 (Vol. I)).

United Nations Inter-Agency Project on Human Trafficking (UNIAP) (2008) Guide

to Ethics and Human Rights in Counter-Trafficking: Ethical Standards for Counter-

Trafficking, Bangkok, UNIAP.

Urada, L. A. and Simmons, J. (2014) ‘Social and structural constraints on disclosure

and informed consent for HIV survey research involving female sex workers and

their bar managers in the Philippines’, Journal of Empirical Research on Human

Research Ethics, 9(1), pp. 29–40.

Vaismoradi, M., Turunen, H. and Bondas, T. (2013) ‘Content analysis and thematic

analysis: Implications for conducting a qualitative descriptive study’, Nursing &

Health Sciences, 15(3), pp. 398–405.

VanderStaay, S. L. (2005) ‘One hundred dollars and a dead man: Ethical decision

making in ethnographic fieldwork’, Journal of Contemporary Ethnography, 34(4),

pp. 371–409.

Yea, S. (2016) ‘Trust, rapport, and ethics in human trafficking research: Reflections on re-

search with male labourers from South Asia in Singapore’, in Siegel, D. and de Wildt,

R. (eds), Ethical Concerns in Research on Human Trafficking, Heidelberg, Springer.

Zhang, S. X. (2009) ‘Beyond the “Natasha” story: A review and critique of current

research on sex trafficking’, Global Crime, 10(3), pp. 178–95.

Zimmerman, C. and Watts, C. (2003) World Health Organization Ethical and Safety

Recommendations for Interviewing Trafficked Women, Geneva, World Health

Organization.

Conducting Research with Survivors of Sex Trafficking 175

Downloaded from https://academic.oup.com/bjsw/article-abstract/48/1/158/3108817 by Adam Ellsworth, Adam Ellsworth on 16 February 2018

Copyright of British Journal of Social Work is the property of Oxford University Press / USA and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.