Conducting post-disaster research with refugee background peer researchers and their communities

Abstract

This paper incorporates peer researchers from refugee backgrounds to deconstruct their experiences of conducting interviews and focus groups with refugee communities in a post-disaster environment in the Canterbury region of New Zealand. The associated dynamics illustrate the contextual intricacies of recruitment, building relationships, the politics of interpreting and engaging with people’s lived experiences in respectful and safe ways. The peer researchers’ experiences highlight several methodological and ethical complexities to critically examine the role of ‘insiders’ and ‘outsiders’ as a continuum when working in post-disaster contexts with culturally and linguistically diverse populations.

Keywords

Insider–outsider refugee disaster culture diversity peer research

Introduction

There are numerous autobiographical, fictional, artistic and academic accounts that document the settlement experiences arising from forced migration. These insights offer understanding into the sheer diversity of what might be problematically termed the ‘refugee journey’. This journey is complex and often contested where questions of ‘truth’ (Eastmond, 2007), ‘authenticity’ (Marlowe, 2009) and ‘front-stage and back-stage’ narratives (Miller, 2004) about people’s experiences come into play. Researchers who are outsiders to communities that are culturally and linguistically diverse may not have access to specific groups and particular forms of inquiry due to relational and contextual dynamics. As questions of audience and narrator come into focus, the associated interactions between participants and researchers can have significant impacts on the quality of data collected, particularly in a post-disaster environment.

The Canterbury region of New Zealand has experienced four major earthquakes and more than 12,000 aftershocks since the first major 7.1 magnitude earthquake in September 2010. The most devastating earthquake occurred 22 February 2011 when the city of Christchurch experienced extensive damage to the central business district and surrounding areas and resulted in 185 fatalities. Christchurch represents one of New Zealand’s principal refugee resettlement sites and several reports have recently documented both the challenges experienced and the capacities that these communities demonstrated in the wake of these significant events (Christchurch Migrant Inter-Agency Group, 2011; Marlowe, 2013a; Marlowe and Lou, 2013; Osman et al., 2012; Wylie, 2012). As a research team working with refugees in Christchurch, we present this reflective paper on the methodological and ethical implications of conducting research with and as peer researchers on an insider–outsider continuum in a post-disaster context. We use the term ‘peer researcher’ in two primary ways: (1) to acknowledge that two of the authors are from refugee backgrounds with close links to refugee communities and; (2) that they were also peers in the research project in terms of recruitment, data collection, analysis and dissemination of the study. This paper thereby highlights a number of methodological, ethical and pragmatic considerations of working with peer researchers within the contexts of forced migration and disaster-based inquiry.

Refugees, resettlement and the role of ‘insiders’

The 1951 United Nations Convention Relating to the Status of Refugees formally defined a refugee as:

a person who is outside his or her country of nationality or habitual residence; has a well-founded fear of persecution because of his or her race, religion, nationality, membership of a particular social group or political opinion; and is unable to avail himself or herself of the protection of that country, or to return there, for fear of persecution. (United Nations High Commissioner for Refugees [UNHCR], 2013: 114)

The UNHCR (2013) currently estimates that there are more than 10.5 million refugees worldwide. Less than 1% of this number will be offered an opportunity for resettlement every year in the 26 countries that currently have formal resettlement programmes such as the United States, Australia, Canada, United Kingdom, Scandinavian countries and New Zealand (UNHCR, 2013). Whilst resettlement pathways support only a minority of refugees globally, these countries and the associated resettlement programs provide a durable solution towards a life of relative safety and security from previous experiences of persecution. Within New Zealand, more than 50,000 refugees have been resettled since World War II, with more than 7000 people from 55 countries settling in the last decade alone (Mortensen et al., 2012).

Conducting research with refugees can raise particular methodological and ethical considerations as numerous authors have noted the potential vulnerability of participants and the ease by which they may come to harm or have their experiences misrepresented (Gifford et al., 2007; Mackenzie et al., 2007; Marlowe, 2013b; Minkler, 2004; Pittaway et al., 2010; Pupavac, 2008; Schweitzer and Steel, 2008). Jacobsen and Landau (2003) discussed the ‘do no harm’ imperative of research with refugees and cautioned that refugee-related research has, at times, obfuscated details of the study design and may be guided by preconceived understandings at the expense of maintaining transparency and rigorous methodological and ethical standards. To avoid this, they suggest beginning with ‘greater conceptual clarity’ and highlight the need to remain critical of the processes and underlying implications behind refugee research (Jacobsen and Landau, 2003: 187).

Research with culturally and linguistically diverse populations often requires additional thought in the recruitment process, establishing informed consent and the use of interpreters to better ensure that there is no coercion and that subsequent interpretations of data accurately capture people’s lived experiences and meanings. Part of the associated complexity relates to the authenticity of knowledge obtained, particularly when working with diverse groups. Birman (2006) discussed the ethical and methodological importance of ‘cultural insiders’ to be involved in cross-cultural research. Such persons possess both the linguistic and cultural expertise to allow them to access the specific cultural communities, which would be difficult for even ‘sensitive and knowledgeable outsiders’ (Birman, 2006: 172; Dona, 2007; Ryan et al., 2010).

Court and Abbas’ (2013) case study of the collaborative process between two researchers from different ethnic backgrounds when interviewing a family from an ethnic and religious minority provides an example of how a cultural insider researcher may add layers of understanding. The case study described how Court, an English- and Hebrew-speaking Israeli-Canadian woman, worked together with Abbas, a Hebrew- and Arabic-speaking Israeli Druze woman, to interview two Arabic-speaking Druze women. Abbas provided not only a key linguistic ability that allowed the interview to take place, but her understanding and personal experiences within the culture and religion of the interviewees meant she was able to provide insights into what was said and not said during the interview and follow up on points a cultural outsider may have missed. At the same time, she straddled a delicate balance of wanting to protect and respect her own culture and religion while being an academic researcher seeking to further her own career. This, in turn, affected how she approached translating some of the traditional opinions of the interviewees that could have been construed as casting the culture in a negative light. Cross-cultural discussions between Abbas and Court became an important place to clarify these competing roles and obligations, as well as cultural differences in understandings of ethics and possibilities for coercion.

Whilst there are a plenitude of studies that examine the roles of ‘insiders’ and ‘outsiders’ in research contexts with diverse populations, there is increasingly a recognition that these concepts are located more along a continuum rather than as discrete binaries (Carling et al., 2013; Collet, 2008; Merriam et al., 2001). Kusow’s (2003) ethnography of conducting research as a ‘native ethnographer’ demonstrates that what makes a researcher an insider or outsider is contextual, relational and dynamic. At the same time, the authors above also note the methodological importance (even imperative) of ‘insiders’ in community-based research with migrant and ethnic minority groups (see also Thomas et al., 2000).

The peer researcher’s interpretation and views are therefore an important component of the knowledge constructed in the research process but are also subject to the bias Jacobsen and Landau (2003) described. Furthermore, the identification of any insider researcher must be prefaced with a critical discussion of what constitutes ‘insider’ status (Birman, 2006; Dona, 2007) as the insider–outsider binary is actually more of a continuum that shifts depending on time, relational and interpersonal dynamics. A person’s relationships, therefore, may be multiple within such small communities and means that an insider will likely move along the continuum depending on the content being covered and the relational contexts between the researcher and the participant(s) concerned. The fact that it is usually not possible for the ‘outside’ researcher to understand the multiple nuances and meanings that are communicated through cultural, ethnic and other identities also provides a further imperative to remain cognisant of this dynamic.

Ghorashi (2008) notes that there is a need to go beyond expressed words with refugee groups to understand the ‘untold’ which is often not accessible by outsiders. Though insiders may have greater access to the unspoken, it is also important to recognise that broad terms such as ‘Afghan’ may possibly suggest a common language (Dari, for instance) and country of origin, but it does not capture the ethnic diversity within such nationalistic designations where there may be histories of inter-ethnic conflict—for example, between the Hazara and Pashtun groups from Afghanistan. It is in this sense that having an Afghan peer researcher could be potentially problematic as his/her ethic group could be perceived by another Afghan as the group responsible in some way for their forced migration experiences. Such inter-ethnic dynamics led Jacobsen and Landau (2003) to note the value of having interviewers participating who spoke the same language but who were of a different nationality. Part of this conclusion is informed by the fact that most refugees and other populations of concern are created by intra-country conflicts rather than external ones that are often defined across ethnic, cultural and/or religious identities (see UNHCR, 2013).

In addition to the differences across and within ethnic groups, refugees in a resettlement context have additional layers of diversity in terms of gender, age, time resettled, acculturation experiences, size of community and other factors that could mean differing reactions to research projects and researchers (Birman, 2006; Edwards and Alexander, 2011; Sulaiman-Hill and Thompson, 2011). Sulaiman-Hill and Thompson (2011), for instance, found significant differences in how Afghan refugees responded to recruitment and interviewing methodologies, depending on factors like gender, length of resettlement and whether they were settled in Australia or New Zealand. Many Australian participants preferred minimal contact or the use of written questionnaires only, and researchers faced difficulties and feelings of suspicion in their attempts to locate enthusiastic persons to initiate snowball sampling. New Zealand participants, on the other hand, welcomed researchers with invitations to have meals and were eager to express themselves and help other refugees through their own participation, especially in the case of more recently arrived women. The authors speculated that this marked difference may have been due to the generally longer resettlement period of the Australian population in combination with the more contentious nature of refugee issues in Australian society, both of which lead to greater suspicion of research aims. Conducting research with refugee groups therefore means moving beyond, but not completely dismissing, archetypes of particular groups to recognising the dynamic and relational factors between researchers and participants that influence levels of access, insight and authenticity.

The disaster context

Aside from the relational and dynamic complexities when working with refugee populations, conducting research in a post-disaster site presents additional considerations. Zakour and Harrell (2003) acknowledged that disasters can exacerbate vulnerabilities for communities that already have markers of disadvantage such as poverty, higher rates of unemployment and poor health outcomes. When a disaster occurs, communities placed in marginalised positions may find themselves pushed even further out from particular centres of institutional, economic and social support. While we must remain mindful that refugees are not inherently vulnerable and that many have incredible sources of resilience that support them through times of adversity, we must also consider what research in disaster settings might mean for the ‘do no harm’ imperative.

In addition, the disaster context may not always be the central consideration in conducting post-disaster research as classical disaster research suggests. As Stallings (2003) asserted, there is little, methodologically, to differentiate disaster research from more conventional qualitative or quantitative research methods, though disaster research is affected by the context in which the research occurs – most specifically, before (if there is a warning), during and throughout the recovery processes of a disaster. In Tierney’s (2007) critique of past and current trends in sociological disaster research, she argued that the classical perspective on disasters as events with a beginning, middle and end artificially separates the disaster scenario from the larger societal timeline and fails to address the impact of wider social, cultural, political, economic and environmental factors. A disaster event can be alternatively thought of as another factor amidst existing inequities possibly demarcated by socioeconomic status, race and ethnicity, linguistic capacities and gender.

It is notable that the refugee and disaster research literature often converge on principles of local engagement, reciprocity and rapport building. Multiple papers discussing qualitative studies involving refugee and cross-cultural research acknowledge the importance of building positive rapport and relationships with participants (Birman, 2006; Collie et al., 2010; Fawcett and Hearn, 2004; Sulaiman-Hill and Thompson, 2011) as the barriers to accessing refugee groups has been well documented (Harrell-Bond and Voutira, 2007). In this sense, relational contexts and building rapport can be incredibly important parts of accessing particular communities and having the support of key leaders. However, these relationships are dynamic, multiple and, at times, contested.

As local, refugee background peer researchers (Authors MO and ZA) who had also experienced the Canterbury earthquakes, they were arguably in better positions to access participants and their associated experiences than the outside researchers (JM and LL). However, because they were local figures already known by the community in other ways, additional ethical parameters around negotiating multiple roles required consideration to better ensure the safety of participants and peer researchers and that clear process of informed consent without coercion were followed. As such, MO and ZA involvement as refugee background peer researchers in this study highlighted ethical and methodological concerns and opportunities in several areas – language and communication, recruitment and access, engaging with different demographics and the need to maintain and function in different roles within their respective communities.

Study design

Central to this study were two former refugee background peer researchers (MO and ZA) who were employed and trained to recruit participants and conduct semi-structured focus group and interview discussions. This project involved conducting interviews and focus group discussions with various refugee background community members to ascertain their perspectives and responses to the Canterbury earthquakes. MO identifies as a male Somali New Zealander and has lived in Christchurch for more than 10 years and ZA is a female from Afghanistan who has lived in Christchurch for 3 years. Both have university degrees and had previous experience with conducting research at the time they joined the research team. JM delivered additional research training in relation to the specific ethical, recruitment and data collection practices and methods to be used in this study, such as facilitating focus groups and probing for more detailed responses in the semi-structured interview format.

The recruitment process was done via a third-party approach, predominantly through community leaders from refugee backgrounds. Those interested in participating then made direct contact with the research team. Twenty-seven participants from refugee backgrounds took part in the semi-structured interviews and there were a total of 10 focus groups with 74 participants that occurred from November 2012 through March 2013. All of these discussions were audio recorded, transcribed and analysed through a process of initial and focused coding and writing memos to develop key themes as outlined by Saldaña (2009). Most participants came from four primary countries of origin – Afghanistan, Ethiopia, Somalia and Bhutan – and included representations of men and women ranging from 20 to over 50 years of age. A $30 grocery voucher was offered to all participants and the study received ethics approval from the associated tertiary ethics committee. Consent forms and participant information sheets were translated into three additional languages – Dari, Somali and Nepali – and interpreters were available as needed.

Peer researchers as insiders and outsiders

The sections that follow present the most salient ethical and methodological implications that MO and ZA encountered as inside researchers and at other times as outsiders. Their experiences as former refugees, cultural understandings and experience of the Canterbury earthquakes provided them with an insider status with a number of participants. However, Christchurch is home to four fairly large communities (Afghan, Bhutanese, Somali and Ethiopian) and several smaller groups which are all characterised by their own cultural and linguistic diversity. Thus, it was necessary to consider this insider status on a continuum along which identity and context shifted their status with respective communities and individuals.

Recruitment and access

As members of the local Christchurch community and specific ethnic groups, MO and ZA gained greater access and trust than would have been possible for JM and LL who do not live locally or come from refugee backgrounds. A mediated or direct relationship was critical for recruitment as the peer researchers primarily contacted community leaders to arrange the distribution of information about the study as a form of third-party recruitment. How smoothly the process went very much depended on the relationships they had within specific communities. They found that recruitment was most successful when community leaders supported the project and then disseminated information about the study to their associated networks. Numerous participants from different ethnic groups spoke of this endorsement of a respected leader or elder from their community as being a critical element that influenced their decision to participate. In this sense, being able to work across ethnic groups and sub-groups, and at times language barriers, to gain the trust needed for this level of support was a primary consideration. This process, however, was not without several complexities.

Obtaining access to the communities themselves was just the first step in the process and peer researchers encountered further difficulties engaging community members. As ethnic groups were not uniform, some sub-groups – such as Pashtun and Tajik Afghans – were more reluctant to participate, which raised the potential for participants to skew towards more enthusiastic sub-groups. It was also sometimes difficult to identify community leaders at meetings and forums, as views and loyalties within the communities varied. Since recruitment was primarily achieved through community leaders, this meant some groups within particular communities were more difficult to reach or were even invisible. For instance, one particular sub-group asked to organise their own focus group of people who would be less likely to be contacted by the ‘community leader’ due to internal tensions and political contestations. This dynamic meant that we needed to offer several focus groups and individual interview opportunities so that people who were interested in participating had a greater opportunity to engage. Both peer researchers worked with members of the Somali and Afghan communities which meant as peer researchers they were sometimes more outsiders than insiders and vice versa. This approach provided multiple pathways of access to particular communities and an awareness of the politics occurring within these groups loosely defined by country of origin.

Gender was also a significant consideration as interactions between men and women were highly prescribed within some of the cultures and ethnic backgrounds that participants identified. ZA noted that, as a woman, she needed to consult with Afghan men to arrange for another family member to be present during interviews as it was culturally inappropriate for them to speak alone. On the other hand, accessing Afghan women as participants was something that was not possible for MO as a male. Even within their own or similar respective cultures, the peer researchers understood customs around gender that influenced the way interview and focus group arrangements were made and how the interviews themselves were conducted so that participants would more likely find this process resonant and appropriate.

Maintaining multiple roles in the community: Language and communication

Linguistic competencies and shared languages were a central consideration on whether a peer researcher was viewed as an insider or outsider. As each peer researcher could only speak the language that was common to their community, they often had to use English or participant chosen interpreters. In terms of being able to communicate with participants and eliciting their responses, peer researchers noted the politics around interpreting, audio recording and facilitating the interviews and focus groups. Their language skills allowed them to provide explanations in another language when conducted within their own community groups and spoken languages to check for some participants’ understanding. However, as the peer researchers were not fluent in all languages spoken and JM and LL did not speak any of these other languages, interviews were generally conducted in English and professional interpreters were offered. Despite having this resource, participants refused these interpreters in every instance. Even when interpretation was necessary, participants requested to have family members as interpreters often because they did not have much trust in professional ones. Within this study, we decided that a participant’s choice to use their own interpreter was one that they were freely making and recognised their concerns about confidentiality as there was a reasonable likelihood that participants personally knew the locally based professional interpreters.

Another significant issue was the discrepancy between what participants would say on and off the record. When the audio recorder was on, participants spoke politely about social service organisations doing as much as they could for local communities in the wake of the earthquakes’ devastating impact. However, when the recorder was off, participants revealed that they in reality held higher expectations for several of these organisations (usually generalist services rather than refugee focussed) and were critical of what had happened in their situations and experiences. This happened despite peer researchers’ attempts to elicit more of these critical responses during the recorded interviews, leading them to believe that participants distrusted that information would be held in confidence. Several participants also articulated a fear that speaking ‘on the record’ might disadvantage their communities in the context where important resourcing decisions were still being made and they did not want to be seen as ungrateful or problematic to those who were coordinating earthquake response and recovery initiatives. This dynamic was not universal as the peer researchers found their own community groups seemed more comfortable to be critical of the earthquake response when compared to talking with participants from different backgrounds. These differing responses highlight this insider–outsider continuum where the political, cultural and social contexts play a key role in much of disaster and refugee focussed research.

Since peer researchers were also active members of their communities, it was quite likely or even inevitable that they would encounter individuals they knew in other settings through the course of conducting this research. On the one hand, their status as former refugees and their familiarity with organisations and communities facilitated the recruitment and interviewing processes as they already possessed a certain amount of trust that encouraged participation. On the other, they had to manage their roles as community members and as researchers in a way that did not affect the rigour of the process. In one case, a participant from the Somali community responded to one of MO’s questions by asserting ‘But you already know this!’ to which he replied, ‘Pretend I don’t know!’ Such comments demonstrate the importance of self-awareness and the need to ensure as much as possible that participants are speaking to their experience as opposed to the assumptions of the researcher, especially when the interaction involves previously established relationships.

These experiences highlight that the shared experience of MO and ZA coming from refugee backgrounds was a key resource in accessing data that would have otherwise been unattainable. However, participants had different lines of demarcation defined possibly by culture, gender, age, shared experience and linguistic competencies that influenced whether a peer researcher was an insider, outsider or something in between. This paper now considers the ramifications and opportunities of conducting qualitative research with culturally and linguistically diverse groups and peer researchers.

Working as and with peer researchers from refugee backgrounds

The methodological opportunities of working as and with peer researchers from refugee backgrounds is clear. This partnership brought together the team’s strengths where the methodological and analytical expertise of researchers who were not locally based or from refugee backgrounds were powerfully complemented by the peer researchers’ nuanced cultural understandings and established relationships. It also suggests that there is not just insider or outsider research, but rather a continuum along which both the researchers and participants carry different forms of identity with respect to culture, ethnicity, linguistic capabilities, gender, age, setting and many other factors. In this sense, there is not just one authentic account of a person’s perspectives and lived experience. As a team, we were able to discuss and debrief the tensions of recruitment, data collection and analysis through ongoing supervision (via face to face, telephone and Skype) between the peer and outside researchers. This was not a one-way interaction but a dialogue where multiple knowledge bases – cultural, methodological and ethical – were discussed to find ways to respond to particular dilemmas regarding linguistic diversity, community leader involvement and being aware of the possibilities for coercion and other power dynamics. In this sense, there were times that the peer researchers almost had to step back from their cultural understandings and experience to become a curious but knowledgeable inquirer to ensure that participant voices were heard. They had to get participants on board with making shared knowledge more explicit during the interviews and probe for more details even when they knew the answers, which required a critical awareness for what information needed to be collected. There were other times as outside researchers that we needed the input of the peer researchers to ensure that our analysis and ensuing interpretations were relatively accurate.

These experiences leave us to conclude that there is not just a dichotomy of insiders or outsiders, nor is there such thing as ‘authentic’ knowledge. However, there are certainly deeper levels of authenticity that one can access that are often contingent upon relational and contextual factors. Deeper layers of access also raise additional pragmatic, ethical and methodological questions. As Pittaway et al. (2010) note, groups like refugees are often highly cognisant of the power researchers have to effect change, as well as the personal risk they take to engage with such persons. They are also aware of the power discrepancy between themselves and the researcher and may try to provide the researcher with what researchers ‘want’ to hear or be suspicious of the reasons behind the research, even when all steps are taken to engage local community members in the research process (see Guerin and Guerin, 2007). The fact that peer researchers from refugee backgrounds are more likely to be seen as insiders (due to their forced migration experiences, linguistic capabilities, local connections and cultural background) offers important relational and methodological resources to a particular project. These resources can be complex and the additional research relationship that a person has with a potential participant needs careful consideration.

Our analysis has links with other qualitative and ethnographic literature that document the complexities of the insider–outsider continuum (Carling et al., 2013) and reinforces the fact that researchers are also key actors in the social and political settings that research occurs (Charmaz, 2006; Rock, 2001; Tuhiwai Smith, 1999). Drawing upon symbolic interactionism as a theoretical lens, Snow (2001) maintains that people’s meaning making processes and how researchers might capture this information highlight that the politics of representation occur in relationship between researchers and participants. The insider–outsider continuum illustrates that people’s stories can be shared, experienced and interpreted in multiple ways and requires awareness and explicit exploration of discursive notions of power and meaning in community-based study. The binary assumption that anyone on a research team, including peer researchers, is necessarily an insider or outsider is one that therefore needs cautious critique and examination.

It is useful to return to Court and Abbas’ (2013) earlier discussion around the multiple roles and obligations Abbas faced as a cultural insider. As with Court and Abbas, an ongoing cross-cultural dialogue in this study allowed for both refugee background and non-refugee background members of the research team to become conscious of multiple cultural lenses and raise points the other side may have overlooked. For example, JM was able to discuss with MO and ZA how to probe more into certain areas based on completed transcriptions and how to interpret findings that included the complexities of the informal, off-the-record conversations with participants to ensure that the ensuing analysis and interpretations were more likely to be accurate (and importantly, ethically informed) so that information shared in confidence was not unduly reported in the dissemination process. The extended time period of the research project and the ongoing transcription and review process of the interviews allowed for there to be ongoing discussions on data collection techniques as interviews proceeded, which peer researchers found useful in informing subsequent interviews. The interviewing process was thus refined over time as MO and ZA brought up more challenging situations and were asked to focus on particular areas along the way. As there were times that they were more insiders and at other times outsiders, the role of supervision shifted to respond to the associated relational dynamics between peer researchers and participants.

The associated issues of using interpreters where communities are relatively small and the persons available for ‘professional’ interpreting are known within the community are well documented (Hugman et al., 2011; Mackenzie et al., 2007; Temple, 2002). Though not a specific focus for this study, the peer researchers believed that a possible concern for some participants – though not directly articulated – was that a registered interpreter had to report any family violence concerns and cause the involvement of Child, Youth and Family which is New Zealand’s statutory child protection agency. These concerns about the roles of child protection services are common in other studies (Deng and Marlowe, 2013; Lewig et al., 2010). It is also important to distinguish between the use of interpreters in health and other professional settings as opposed to research contexts. Though professionally trained interpreters have an important and sometimes essential role in certain situations, there is also a growing recognition that people choosing their own interpreters (family or friends) is not necessarily problematic in a number of situations (Gray et al., 2011).

An awareness of power represents an important consideration when working with peer researchers. Whilst the lead author hired the peer researchers and thereby had a degree of power within the research team, the ensuing relationship was made as collaborative as possible. Supervision sessions (often through Skype) were conducted where we all sought advice to effectively respond to issues rather than being a forum for providing one way advice or instruction. For instance, when the issue of using interpreters arose, the decisions about how to proceed were negotiated collectively and through discussion. As it became clear that we needed to work with different leaders within the same ethnic community to get better representation, the process of managing community-based relationships was discussed openly to determine the best way forward. We were often in a better position to respond to and successfully navigate issues with recruitment as MO and ZA had knowledge about particular cultural nuances and political dynamics. In relation to the study’s analysis, we discussed if the abstraction of the data were appropriate through online discussions, email correspondences and meeting in person. We have delivered multiple presentations where the peer researchers were named contributors and provided their direct feedback in these forums. The same process has occurred with this paper as well. This collaborative and dialogic approach has provided valuable methodological, ethical and pragmatic insight, which has not only incorporated reciprocity and mutual learning but also rigour in the research process.

The need to consider the complexities of conducting research in post-disaster settings with resettled refugee communities is clear. It is worth noting that many of the interviews and focus groups occurred nearly 2 years after the devastating February 2011 earthquake. Chronology is important where the immediate-, short- and medium-term response and recovery periods may have people experiencing higher stress. However, though the research team was ready to respond to issues of safety and re-traumatisation, there were few indications that participants were additionally upset by speaking about the disaster circumstances. While participants acknowledged the stress, fear and difficulties that they experienced as a result of the earthquakes, they were far more interested in discussing their experiences and making suggestions for changes around longer term issues such as education and employment. These statements support Stallings (2003) and Tierney’s (2007) arguments about research not becoming dominated by nor dismissive of the immensity of a disaster context. The disaster literature demonstrates how inequality and oppression place some groups of people at higher risk of negative outcomes (Adrich, 2012; Klinenberg, 2002). In this sense, social work provides a useful knowledge base to critique wider structural forces and environmental contexts when examining the impacts of disaster (see Hölscher and Bozalek, 2012; Zakour and Harrell, 2003). Thus, considering the balance between people’s capacities and existing vulnerabilities are a key component of a disaster analysis. And so too are the multiple contexts of participants' experiences that relate to daily life (spiritual, cultural, social, economic, institutional) which may or may not have direct relevance to a disaster.

Finally, it is necessary to recognise that research into forced migration is often written by people who have not had these experiences. A similar comment can be levelled at much of disaster-related research which involves researchers coming from outside locally based and impacted communities. Recruiting and training people who may have ‘insider’ status on a number of levels to conduct research that is ethically informed and methodologically rigorous can not only help ensure that the ensuing analysis and interpretation is more likely to be accurate but also helps build local capacities. These considerations for capacity building and reciprocity, as Hugman et al. (2011) argued, should be central in research with refugees. This view is in line with the goals of social work research in disasters to benefit current and future preparation and recovery efforts (Zakour, 1997; Zakour and Harrell, 2003). Overall, both peer researchers felt being representatives of the research project raised their statuses within the communities and they were able to gain a deeper understanding of the associated communities, including their own. They were especially positive about the initial training and linked this and the subsequent interviewing experiences to a sense of empowerment around being able to conduct rigorous academic research. Both have subsequently pursued further academic studies in medical school and public health.

Conclusion

A person’s insider status occurs at different levels as people’s experiences, history and meaning-making endeavours interact with various actors that may include refugee background communities, refugee-based organisations, media, government agencies and the wider society. It is within these unique contexts that insider and outsider researchers can provide valuable lenses to ask a number of critical questions about who the players are in any particular disaster scenario and what specific meanings different groups might have about their associated experiences. The importance of having locally informed knowledge, relationships of trust and an awareness of context represent critical considerations for conducting research with refugee communities that offer valuable opportunities for insight, access and accountability. These relationships and sources of information, however, are neither static nor uniform. The model of using refugee background peer researchers is one that has great potential, but it also requires an analysis of the potential tensions inherent in negotiating multiple roles along the insider–outsider spectrum. This complexity is illustrated through Guerin and Guerin’s (2007) notion of research being a spiralling process of encountering layers where relational and contextual dynamics either constrain researchers or allow them to peel back particular layers of meaning and experience. Qualitative research with refugee groups and in disaster settings highlights the dynamic contexts in which research occurs and the interpersonal interactions between researchers and participants possibly defined by shared experience, ethnicity, culture, linguistic capabilities and gender.

Footnotes

Acknowledgements

We would also like to acknowledge Anne Marie Reynolds and Ahmed Tani for their support and guidance when conducting the study.

Funding

This research was funded by a University of Auckland Faculty Development Fund grant (3701920).

References

Aldrich

(2012) Building Resilience: Social Capital in Post-disaster Recovery, Chicago, IL: University of Chicago Press.

Birman

(2006) Ethical issues in research with immigrants and refugees. In: Trimble

Fisher

(eds) The Handbook of Ethical Research with Ethnocultural Populations and Communities, London, UK: SAGE, pp. 155–177.

Carling J, Erdal MB and Ezzati R (2013) Beyond the insider–outsider divide in migration research. Migration Studies. Epub ahead of print 15 October 2013. DOI: 10.1093/migration/mnt022.

Charmaz

(2006) Constructing Grounded Theory, London: Sage Publications.

Christchurch Migrant Inter-Agency Group (2011) Christchurch Migrant Inter-Agency Group: Lessons Learned Following the Earthquakes of 22 February 2011, Christchurch, NZ: Christchurch Migrant Inter-Agency Network.

Collet

(2008) Confronting the insider-outsider polemic in conducting research with diasporic communities: Towards a community-based approach. Refuge 25: 77–83.

Collie

Liu

Podsiadlowski

(2010) You can’t clap with one hand: Learnings to promote culturally grounded participatory action research with migrant and former refugee communities. International Journal of Intercultural Relations 34: 141–149.

Court

Abbas

(2013) Whose interview is it, anyway? Methodological and ethical challenges of insider–outsider research, multiple languages, and dual-researcher cooperation. Qualitative Inquiry 19: 480–488.

Deng

Marlowe

(2013) Refugee resettlement and parenting in a different context. Journal of Immigrant and Refugee Studies 11: 416–430.

10.

Dona

(2007) The microphysics of participation in refugee research. Journal of Refugee Studies 20: 210–229.

11.

Eastmond

(2007) Stories as lived experience: Narratives in forced migration research. Journal of Refugee Studies 20: 248–264.

12.

Edwards

Alexander

(2011) Researching with peer/community researchers: Ambivalences and tensions. In: Williams

(ed.) Sage Handbook of Qualitative Research, London: SAGE, pp. 269–292.

13.

Fawcett

Hearn

(2004) Researching others: Epistemology, experience, standpoints and participation. International Journal of Social Research Methodology 7: 201–218.

14.

Ghorashi

(2008) Giving silence a chance: The importance of life stories for research on refugees. Journal of Refugee Studies 21: 117–132.

15.

Gifford

Bakopanos

Kaplan

(2007) Meaning or measurement? Researching the social contexts of health and settlement among newly-arrived refugee youth in Melbourne, Australia. Journal of Refugee Studies 20: 414–440.

16.

Gray

Hilder

Donaldson

(2011) Why do we not use trained interpreters for all patients with limited English proficiency? Is there a place for using family members? Australian Journal of Primary Health 17: 240–249.

17.

Guerin

(2007) Research with refugee communities: Going around in circles with methodology. The Australian Community Psychologist 19: 150–162.

18.

Harrell-Bond

Voutira

(2007) In search of ‘invisible’ actors: Barriers to access in refugee research. Journal of Refugee Studies 20: 281–298.

19.

Hölscher

Bozalek

(2012) Encountering the other across the divides: Re-grounding social justice as a guiding principle for social work with refugees and other vulnerable groups. British Journal of Social Work 42: 1093–1112.

20.

Hugman

Pittaway

Bartolomei

(2011) When ‘do no harm’ is not enough: The ethics of research with refugees and other vulnerable groups. British Journal of Social Work 41: 1271–1287.

21.

Jacobsen

Landau

(2003) The dual imperative in refugee research: Some methodological and ethical considerations in social science research on forced migration. Disasters 27: 185–206.

22.

Klinenberg

(2002) Heat Wave: A Social Autopsy of Disaster in Chicago, Chicago: University of Chicago Press.

23.

Kusow

(2003) Beyond indigenous authenticity: Reflections on the insider/outsider debate in immigration research. Symbolic Interaction 26: 591–599.

24.

Lewig

Arney

Salveron

(2010) Challenges to parenting in a new culture: Implications for child and family welfare. Evaluation and Program Planning 33: 324–332.

25.

Mackenzie

McDowell

Pittaway

(2007) Beyond ‘do no harm’: The challenge of constructing ethical relationships in refugee research. Journal of Refugee Studies 20: 299–319.

26.

Marlowe

(2009) Accessing ‘authentic’ knowledge: Being and doing with the Sudanese community. The Australian Community Psychologist 21: 39–49.

27.

Marlowe

(2013a) Resettled refugee community perspectives to the Canterbury earthquakes: Implications for organizational response. Disaster Prevention and Management 22: 434–444.

28.

Marlowe

(2013b) Going slowly slowly: An ethnographic engagement with the south Sudanese. Studies in Symbolic Interaction 40: 153–173.

29.

Marlowe

Lou

(2013) The Canterbury earthquakes and refugee communities. Aotearoa New Zealand Social Work 25: 58–68.

30.

Merriam

Johnson-Bailey

Lee

M-Y

(2001) Power and positionality: Negotiating insider/outsider status within and across cultures. International Journal of Lifelong Education 20: 405–416.

31.

Miller

(2004) Beyond the frontstage: Trust, access, and the relational context in research with refugee communities. American Journal of Community Psychology 33: 217–227.

32.

Minkler

(2004) Ethical challenges for the “outside” researcher in community-based participatory research. Health Education and Behavior 31: 684–697.

33.

Mortensen

Rainger

Hughes

(2012) Refugee Health Care: A Handbook for Professionals, Wellington, NZ: Ministry of Health.

34.

Osman

Hornblow

Macleod

(2012) Christchurch earthquakes: How did former refugees cope? New Zealand Medical Journal 125: 113–121.

35.

Pittaway

Bartolomei

Hugman

(2010) ‘Stop stealing our stories’: The ethics of research with vulnerable groups. Journal of Human Rights Practice 2: 229–251.

36.

Pupavac

(2008) Refugee advocacy, traumatic representations and political disenchantment. Government and Opposition 43: 270–292.

37.

Rock

(2001) Symbolic interactionism and ethnography. In: Atkinson

Coffey

Delamont

(eds) Handbook of Ethnography, Thousand Oaks, CA: Sage Publications Inc.

38.

Ryan

Kofman

Aaron

(2010) Insiders and outsiders: Working with peer researchers in researching Muslim communities. International Journal of Social Research Methodology 14: 49–60.

39.

Saldaña

(2009) The Coding Manual for Qualitative Researchers, Thousand Oaks, CA: Sage.

40.

Schweitzer

Steel

(2008) Researching refugees: Methodological and ethical considerations. In: Liamputtong

(ed.) Doing Cross-Cultural Research, Netherlands: Springer, pp. 87–101.

41.

Snow

(2001) Extending and broadening Blumer’s conceptualization of symbolic interactionism. Symbolic Interaction 24: 367–377.

42.

Stallings

(2003) Methods of disaster research: Unique or not? In: Stallings

(ed.) Methods of Disaster Research, Philadelphia: Xlibris Corporation, pp. 21–44.

43.

Sulaiman-Hill CMR and Thompson SC (2011) Sampling challenges in a study examining refugee resettlement. BMC International Health and Human Rights 11(2): 1–10. Available at: http://www.biomedcentral.com/1472-1698X/1411/1472.

44.

Temple

(2002) Crossed wires: Interpreters, translators, and bilingual workers in cross-language research. Qualitative Health Research 12: 844–854.

45.

Thomas

Blacksmith

Reno

(2000) Utilizing insider-outsider research teams in qualitative research. Qualitative Health Research 10: 819–828.

46.

Tierney

(2007) From the margins to the mainstream? Disaster research at the crossroads. Annual Review of Sociology 33: 503–525.

47.

Tuhiwai Smith

(1999) Decolonizing Methodologies, London: Zed Books.

48.

United Nations High Commissioner for Refugees (2013) UNHCR Global Appeal 2012–2013, Geneva: United Nations High Commissioner for RefugeesAvailable at: http://www.unhcr.org.

49.

Wylie

(2012) Best Practice Guidelines: Engaging with Culturally and Linguistically Diverse (CALD) Communities in Times of Disaster, Christchurch, NZ: Community Language Information Network Group.

50.

Zakour

(1997) Disaster research in social work. Journal of Social Service Research 22: 7–25.

51.

Zakour

Harrell

(2003) Access to disaster services: Social work interventions for vulnerable populations. Journal of Social Service Research 30: 27–54.