Dependency, partiality and the generation of research questions in refugee education

Marianne Turner and Farida Fozdar
Murdoch University

From the late 1990s to the mid 2000s refugees who gained entry to Australia came mainly from the African region, in particular Sudan. This contrasts with earlier intakes from Asia, the Middle East and the former Yugoslavia. The change in regional focus sparked research into the resettlement of African refugees in Australia, but it is a new field and research is subsequently limited. This article reports ethical issues involved in research with adult South Sudanese students, specifically the impact of participants' dependency and the researchers' partiality. A qualitative approach, taken mainly due to ethics considerations, informed the generation of research questions. Far from inhibiting the research, we argue that an acceptance of these issues enabled the researchers to ask the 'right' questions.

Ethical issues in refugee research

forced migration research is 'partisan', rather than neutral, to the plight of the subjects of its investigation: studying the experiences, causes and consequences of displacement is done with the implicit or explicit intent to influence the development of better policies and programmes on the part of governments, non-governmental and inter-governmental agencies and refugee community organizations (Dona, 2007, p. 210).

A number of ethical issues arise when researching vulnerable populations such as refugees (see Bailes, Minas & Klimidis and responses, 2006; special issue on research methods, Journal of Refugee Studies, 2007). These include power differentials, consent, representation, research fatigue, dependency relationships, confidentiality, trust, risk, social justice, partiality, autonomy and agency, and cultural difference (Mackenzie et al., 2007). We focus on how issues of dependency/autonomy and researcher partiality/student positionality shaped the direction of the research, especially through the generation of research questions, and how treatment of these issues enhanced rather than compromised academic rigour.

Background to the study

Considering these challenges, education is a significant issue in refugee resettlement, especially since participation in educational opportunities and institutions is known to be central to the 'cultural transition process' (Vedder & Horenczyk, 2006). Language learning and education generally are vital for refugees, in terms of both employment and everyday life. The Australian Government provides educational services for migrants and refugees (DIAC, 2007), and there are also a number of informal services on offer. The three main learning environments available to adult South Sudanese are those investigated in the present study: Technical and Further Education (TAFE) colleges, universities, and community groups. The Adult Migrant English Program (AMEP), frequently offered through TAFE colleges, is a government funded English language service provided to some migrants and all refugees. Community groups are voluntary groups often initiated specifically to assist refugees settle in Australia. Finally, students also have access to mainstream learning environments such as universities, often through 'access and equity programs', provided that they meet the entrance requirements. There is little research about the relative benefits of each option, nor the challenges involved (see Brown, Miller & Mitchell, 2006).

Anecdotal evidence indicates that mainstream service providers, not trained to address the cultural, educational and health issues of African refugees, see them as particularly problematic among the refugee population, due to low levels of literacy, significant cultural difference, and psycho-social issues relating to displacement experiences. Settlement difficulties experienced by South Sudanese, including problems associated with education, were used by the former Minister for Immigration, Kevin Andrews, as justification for decreasing the numbers of African refugees accepted for resettlement in Australia (Andrews, 2007). The current research was undertaken on the assumption that the South Sudanese group, already highly visible, were likely to become marginalised if they were not equipped with the education needed to navigate a highly literate Australian society.

Designing the study

Next, social research methods needed to be selected to ensure both high levels of ethical and academic standards. It is well known that participant observation and semi-structured interviews facilitate an interpretive research process (Liamputtong & Ezzy, 2005; Lincoln & Guba, 1985), enabling the study of human experiences within particular social contexts (e.g., Corti & Thompson, 2004; Marshall & Rossman, 1989). Certainly, ethnography enhances the likelihood of being able to understand reality from the participants' perspective (Acker, Barry & Esseveld, 1991; Filstead, 1970; Spradley, 1979). Ethnography also allows for an iterative process of question development, and ultimately, theory development. The methods have the potential to produce "thick" or detailed description (Geertz, 1973) generating rich, high quality data of use to communities as well as to academics.

To determine the best setting for the research, various sites of learning were explored. There were a number of false starts as we attempted to determine the key problems which needed addressing and the best sites to collect information. In the early stages, questions about learning issues for primary school children from a range of African refugee backgrounds, and for refugees settling in regional areas, led to the exploration of options for investigation in these contexts. Finally, due to accessibility and the seriousness of issues in these environments, it was agreed that adult English language learning would be the focus. Three learning environments, with significant similarities and differences in terms of levels of formality, opportunities for teacher-student interaction, and function, were selected. The three researched groups included a church-based women's community group for Sudanese women which addressed perceived acculturation/socialisation needs, a mixed-gender mixed-ethnicity TAFE group studying a structured curriculum of English literacy and English conversation, and an all-male first-year university group of South Sudanese students receiving generalised English support for their discipline-based studies.

Group	Observations	Focus Groups	Interviews*
University Group	4 weeks informal participant observation in alternative entry program to university 13 1-hour sessions of formal participant observation in English support unit 2 1-hour sessions of observation in English support unit Informal participant observation: providing academic assistance to students in English support unit outside class	1 focus group conducted with 11 students	7 interviews conducted 4 students interviewed individually 2 students interviewed together 2 pairs (4) of teachers interviewed
TAFE Group	3 weeks participant observation in pre-literacy class 3 days observation of 3 different literacy classes	No focus groups conducted	4 interviews conducted 3 groups of three (9) students interviewed** 4 teachers interviewed in group
Women's Community Group	25 2-hour sessions of participant observation and observation	No focus groups conducted	4 interviews conducted 3 pairs (6) of students interviewed* 3 volunteer teachers interviewed in group
*All interviews semi-structured **Interpreter required

The first author had access to both the TAFE and university sites due to her role as a teacher. An opportunity to participate in the women's community group arose through a colleague, enabling the third site for comparison. Participant observation in the three sites involved over 40 South Sudanese students. Additionally 21 students and 11 teachers participated in semi-structured in-depth 'active' interviews (see Holstein & Gubrium, 1995). The fieldwork took place over a nine month period in 2006-07. See Table 1 for a summary of the data collection.

We particularly sought to gather evidence that could be used to enhance teachers' awareness of how student participation in learning is influenced by past experiences in a different cultural environment, and by student-teacher and student-student interactions in the Australian classroom. As such, the research had a clear policy focus. A systemic model of student participation in cross-cultural learning environments was developed (adapted from McCaslin, 2009), exploring the relationship between three related sources of influence - cross-cultural, cultural and social. This model provides a theoretical framework for training teachers how to facilitate student participation in a cross-cultural learning environment (Turner, 2009). The research questions formulated during the course of the research are shown in Table 2.

Cross-cultural source of influence	What are the expectations of South Sudanese students in Australia regarding how teaching and learning should take place, and how do they compare and contrast with the expectations of their teachers? How does the degree of congruence between students' and teachers' expectations influence the students' participation in their learning environment?
Cultural source of influence	To what degree do the South Sudanese students demonstrate an interdependent self construal through a focus on relationships? How does this focus on relationships influence the students' participation in their learning environment?
Social source of influence	How do teachers and students act on opportunities generated in the cross-cultural learning environments to modify teaching and learning practices? How do (un)modified teaching and learning practices influence the students' participation in their learning environment?
Relationships among the sources of influence	How do the three sources of influence interrelate to influence student participation in the learning environment?

The ways these questions were generated arose partly out of engagement with the ethical issues of dependency and partiality which drove our focus on expectations, identity, and the role of interpersonal relationships. These issues are now outlined, and their influence on the research questions discussed.

Dependency and consent

For example, Mackenzie et al. (2007) argue that the giving of informed consent provides protection only when three conditions are satisfied: when there is a balance in power between the researcher and researched; when participants understand what it is to which they are consenting; and in circumstances where participants are capable of making their own decisions. These conditions are extremely difficult to achieve in refugee research (see Dona, 2007, p. 216-217). We acknowledge Dona's application of Foucaultian notions of power to the refugee/researcher relationship where it is pointed out that power is never a zero sum construct, and that refugees are agentive to some extent. However, we also recognise that, in the present study, there was a clear power imbalance due not only to the researcher's position as researcher, but also as teacher or assistant in the learning environments. This influenced the approach taken when obtaining consent. Because the students participating in the research did not always understand the implications of giving consent, and were not in a relatively equal position of power, finding a way to obtain informed consent was an extremely important aspect of the research design and implementation.

The approach taken was based on the notion of iterative consent, and was possible due to the ethnographic nature of the study. Iterative models of consent 'start from the assumption that ethical agreements can best be secured through a process of negotiation' (Mackenzie et al., 2007, p. 306). Therefore, obtaining consent goes beyond form-filling. Consent becomes apparent through the processes, styles and outcomes of interaction in the research situation. Participants, especially in the women's community group and the university group, were supportive of the research, and aware of its possible positive impact on learning opportunities for refugees. Participants who better understood the objectives of the research were able to explain it to other participants so that they could make a more informed decision about whether or not to participate. The participants who spoke English fluently also provided advice to the researcher on how to conduct the interviews, and acted as interpreters to ensure those without adequate language could be heard.

However, this developing relationship with the researcher may have made it difficult for participants to decline involvement. Levinson (1998) notes that it was difficult for participants in his Mexican study to refuse to be interviewed once they had already engaged with him, despite being told that they were under no obligation. This may have been true with many of the participants in the present study. Once trust and rapport had developed, it was difficult to know whether the rapport was the reason participants had agreed to be involved. Even so, participants exercised their autonomy by indicating their need for reciprocity from the researcher, such as by asking for help with their studies and to provide other information that they required. This allowed two-way exchanges of information (Klein 1986, p. 14) and a sense that the research was providing an immediate benefit for participants.

Ethics, reciprocity and the generation of research questions

Not all students were found to be visiting in order to develop a strategic relationship with the teacher, however. It became apparent that some students were seeking a sympathetic ear. This finding also became an important part of the study, and was taken as another way to address the ethics issue of dependency. Even though ethical issues, such as 'the complex moral relationship of the observer to the observed, [and] of the relevance of the observed person's situation to the situation of the observer's own society [...] are never resolved in any ethnography' (Markus, 1998, p. 75), listening to participants when they need to talk is a way to ameliorate exploitation. Patai (1991) found that the participants of her study enjoyed the undivided attention the interviews gave them to explore and make sense of their own lives, and Levinson also notes that, in his study of Mexican schoolgirls: 'Interviews and chats [...] became an enormously reflexive and cathartic exercise' (1998, p. 345).

Of course, one issue arising from the participants' desire to talk may be the expectation that the researcher has the power to change the circumstances of the researched rather than simply to report on these circumstances and advocate (Patai, 1991; Tilbury, 2006). In the current study, the prolonged reciprocal interaction with the participants in the three learning environments enabled the clear message to be given that the research would be of use to government and policy makers, and would hopefully impact on people's lives positively, but that it was not in the researchers' capacity to make structural changes to the learning environments generally or immediately, or effect change of personal benefit to participants.

Finally, due to the increased exposure arising from the first author's opportunity to reciprocate by assisting in teaching and advising the South Sudanese participants of the study, she was able to observe students' strong cultural focus on community relationships in an educational setting. In the women's community group and the university group in particular, the students often spoke in a collective way. They used the term 'we' rather than 'I' when asking for help, or referred to other Sudanese students as their 'sisters' in the case of the community group. The students were further found to demonstrate a strong focus on relationships in class. The theme of collectivity and group identity was then explored with the students in interviews, and generated questions about the relevance of community at different levels for both sense of identity and for social support, and about the effect of the focus on community on learning. Findings on this aspect of the study are discussed elsewhere (Turner & Fozdar, in press).

Partiality, generalisability and 'giving voice'

Certainly, the study was designed to give participants as much voice as possible, and to provide opportunities to build relationships and informally observe and reflect on learning environments before developing specific research questions. This is particularly important when researching vulnerable groups. Alcoff (1994) maintains that we should resist the temptation to speak on others' behalf, but if the need arises, we should critically address the relevance of the study to the community, respect and listen to criticism, and consider the possible effects of the research. The current study was considered important in terms of potential positive outcomes for the community, most specifically in terms of greater understanding of learning styles and strategies, which could inform individual teaching approaches. Every attempt was made not to speak in place of the South Sudanese, for example through extensive use of direct quotations from transcripts (see Turner, 2008), in order to voice to the participants.

Closeness to the subject and constant comparison across research sites (Patton, 2002; Charmaz, 2006) were attempts to ensure depth of understanding and to avoid bias, since these were considered to provide an authenticity that a more distanced approach could not. However, the pragmatics of the researcher's different relationships with participants may have affected partiality. For example, the types of opportunities available to conduct fieldwork across the three groups resulted in different opportunities to develop rapport with individual participants. The women's group and the university students were more accessible, enabling longer term connections to be established. Developing this rapport increased the credibility of the data collected in these settings, since participants were more likely to feel comfortable and speak freely (see Oakley, 1981). The study was also influenced by the tailoring of the interview situation to participants (Reinharz, 1992, p. 208). Non-hierarchical methods and a personalised approach meant that each interview was slightly different, since rapport was found to make it difficult to collect the same data in the same way from all participants.

Partiality was a further threat due to differences in education levels and English fluency among, and also within, the three groups, since it is tempting to give greater weight to more sophisticated and articulate opinions, and advocate on behalf of these opinions. The university participants in particular were able to reflect on their own learning as a result of their high level of education. Very few of the students in the women's community group and TAFE technical college group were able to reflect on their learning to the same degree. This skewed the analysis somewhat, in that the three groups and learning environments were not equally represented in the data. In order to counteract this during data analysis, an attempt was made to focus on each learning environment to the same extent. For example, the capacity of the university participants to provide reflective comments on their learning was used to improve the researchers' understanding of learning issues generally, and not to position these participants as the main focus of the study. Indeed, the rapport developed meant being able to talk to participants in one environment about issues arising within any of the three learning environments. The feedback provided then helped to inform the interview questions, and also subsequent observations.

Agency and positioning

All research questions related to aspects of positionality, but particularly the questions on the ways students and teachers co-constructed meaning. Two questions focussed on cross-cultural sources of influence on student participation, the first question addressing the expectations of students about how teaching and learning should take place, and how these expectations compared with those of their teachers. The second question then addressed the degree of congruence between the two sets of expectations, and the influence of this congruence on the students' participation in their learning environment. These questions arose directly from participant observation, becoming conspicuous as a result of friction apparently caused by differences in expectations. In interviews the research questions were broken down. Students were asked about their expectations regarding teaching and learning, and then provided with a number of scenarios of student/student or student/teacher interactions, such as students asking teachers questions, or asking another student for help during class, or trying to stop another student from doing or saying something that would disrupt the class. The participants were then invited to comment upon these scenarios. Furthermore, since a greater degree of friction was observed in the university context than in the women's or TAFE groups, this was also explored in interviews though specific allusion to actual experiences of friction, such as: 'In the first oral presentations that we did, there seemed to be a level of frustration in the class. What do you think about that?'

Next, an acceptance of the research as partial was imperative, given that some students' level of education and proficiency in English provided them with a greater role in reflecting on the research, and in interpreting and explaining other students' opinions. The design of the research questions took this aspect of positionality into account. For example, the questions related to the social source of influence on student participation focused on opportunities generated in the learning environments, and how these were acted on to modify teaching and learning practices. In addition, changes were made to specific interview questions related to the social source of influence during interviews as a result of students' differing abilities to reflect on their own learning. In order to explore the opportunity theme, a set of cards was created from which students could select preferred learning activities (written on different cards) which were then discussed. Participants with a lower level of education took so long to select preferences that a modification was made to the design, allowing them to simply discuss the different activities. This was one instance where the ongoing necessity to adapt questions to particular groups was highlighted.

An illustration of the need for more personalised interviews also derives from issues related to students' differential abilities to speak English. It was necessary to use an interpreter in the TAFE and women's community group, but not in the university group. Furthermore, different interpreters were used for each group. There is some controversy over whether a change in interpreters is beneficial or problematic for the interview process (see Kapborg & Berterš, 2002; Twinn, 1997) but, ultimately, the interpreters were chosen for both their acceptability to the participants, and their own involvement in the research. A case in point is the university participant who volunteered his services as an interpreter for the TAFE group. He was of a similar age and ethnicity to most participants and, since he and the researcher shared a rapport, the interviewees treated the researcher as less of an outsider. On the long drive back to the university, it was also possible to discuss the themes arising from the interviews, and this student's reactions provided an added dimension to the analysis. In the women's community group it was clearly more appropriate for a woman to act as interpreter; therefore, a member of the group with a high level of English was approached.

The choices made in terms of the interviews can be considered ethically appropriate since acceptable interpreters were used (see Mackenzie et al, 2007, p. 304). However, although these choices generated rich data in each learning environment, the credibility of the data was dependent on the questions asked. We would argue, with others (see Bailes et al., 2006; Journal of Refugee Studies, 2007.) that, in this type of ethnographic refugee research, ethical considerations need to be constantly negotiated not only through methodological choices, but also through asking the 'right' questions. This may at first appear to compromise the academic rigor of the study. However, in order to enhance the 'trustworthiness' of the findings (Lincoln & Guba, 1985), particularly given the ethics issues of dependence and partiality, we have attempted to focus on transparency. An acceptance of the issues shaped the direction of the research, widening the view of student participation from an initial focus on the student to a more holistic focus on the interactions between the student and the new cross-cultural learning environment. Although not foolproof, this new focus enhanced the role of researcher reflexivity.

Conclusion

As ethnographic research, the study necessarily involved representation of 'others' (Dona, 2007, p. 220), which included both 'speaking of' in terms of constructing accounts of South Sudanese refugees' learning practices, and 'speaking for', since the project advocates for the needs of this community and proposes ways to mediate these needs. This dual function can only occur in situations where reciprocal benefits for refugee participants or communities underpin the research (Mackenzie et al., 2007). The approach also requires recognition of the need to promote the autonomy of participants and their communities.

Finally, Jacobson and Landau's (2003b) call for more robust refugee research - research that meets the standards of 'scientific' social research more generally and makes its methods explicit - has met with some criticism, including that of Rodgers (2004) who suggests that this approach ignores the link between knowledge and power, and silences the voices of those most in need of being heard. We hope to have shown that ethnographic methods which allow a constant review of ethics issues may lead to asking questions which resonate with a community about which little is known, and also produce rigorous results. This, in turn, has the potential to help generate theory which can be applied, in policy and practice, to the area of adult learning.

References

Alcoff, L. (1994). The problem of speaking for others. In S. O. Weisser & J. Fleischner (Eds.), Feminist nightmares - women at odds: Feminism and the problem of sisterhood. New York: New York University Press.

Andrews, K. (2007). Australia's refugee and humanitarian programme, and skilled migration. Retrieved 9/10/2007, from http://www.minister.immi.gov.au/media/media-releases/2007/, verified 6 July 2010 at http://parlinfo.aph.gov.au/parlInfo/download/media/pressrel/9RIO6/upload_binary/9rio61.pdf;
fileType=application/pdf#search=%22MINISTER%20FOR%20IMMIGRATION%20AND%20
CITIZENSHIP%20no%20andrews,%20kevin,%20mp%22

Bailes, M., Minas, I.H. & Klimidis, S. (2006). Symposium on ethics, multiculturalism and health research: Mental health research ethics and multiculturalism. Monash Bioethics Review, 23(4), 53-63.

Brown, J., Miller, J. & Mitchell, J. (2006). Interrupted schooling and the acquisition of literacy: Experiences of Sudanese refugees in Victorian secondary schools. Australian Journal of Language and Literacy, 29(2), 150-162.

Charmaz, K. (2006). Constructing grounded theory: a practical guide through qualitative analysis, London: Sage.

Clifford, J. (1986). Introduction: Partial truths. In J. Clifford & G. E. Marcus (Eds.), Writing culture: The poetics and politics of ethnography (pp. 1-26). Berkley: University of California Press.

Corti, L., & Thompson, P. (2004). Secondary analysis of archived data. In C. Seale, G. Gobo, J. F. Gubrium & D. Silverman (Eds.), Qualitative research practice. London: Sage.

Department of Immigration and Citizenship (DIAC). (2007). Sudanese community profile. Retrieved (25/04/2010), from http://www.immi.gov.au/living-in-australia/delivering-assistance/government-programs/settlement-planning/_pdf/community-profile-sudan.pdf

Department of Immigration and Citizenship (DIAC). (2009). Australian immigration fact sheet 60: Australia's refugee and humanitarian programme. Retrieved 28/04/2010, from http://www.immi.gov.au/media/fact-sheets/60refugee.htm

Dona, G (2007). The microphysics of participation in refugee research. Journal of Refugee Studies, 20(2), 210-229.

Dyregrov, K., Dyregrov, A., & Raundalen, M. (2000). Refugee families' experience of research participation. Journal of Traumatic Stress, 13(3), 413-426.

Filstead, W. J. (Ed.) (1970). Qualitative methodology. Chicago: Markham.

Geertz, C. (1973). The interpretation of cultures. New York: Basic Books Inc.

Greeno, J. G. (2006). Learning in activity. In R. K. Sawyer (Ed.), The Cambridge Handbook of the Learning Sciences (pp. 79-96). New York: Cambridge University Press.

Holstein, J., & Gubrium, J. (1995). The active interview. Thousand Oaks: Sage.

Ingleby, D. (2005). Forced migration and mental health: Rethinking the care of refugees and displaced persons. New York: Springer Science and Business Media.

Jacobsen, K., & Landau, L. (2003a). Researching refugees: Some methodological and ethical considerations in social science and forced migration. Geneva: UNHCR.

Jacobsen, K., & Landau, L. (2003b). The dual imperative in refugee research: Some methodological and ethical considerations in social science research on forced migration. Disasters, 27(3), 185-206.

Journal of Refugee Studies. (2007). Special issue: Refugee research methodologies, 20(2).

Kapborg, I., & Berterš, C. (2002). Using an interpreter in qualitative interviews: Does it threaten validity? Nursing Enquiry, 9, 52-56.

Kirk, J., & Miller, M. L. (1986). Reliability and validity in qualitative research. Beverly Hills: Sage.

Klein, R. (1986). The dynamics of women's studies: Its international ideas and practices in higher education. London: University of London.

Levinson, B. A. (1998). (How) can a man do feminist ethnography of education? Qualitative Inquiry, 4(3), 337-369.

Liamputtong, P (2007). Researching the Vulnerable: A guide to sensitive research methods. London:Sage.

Liamputtong, P., & Ezzy, D. (2005). Qualitative research methods (2nd ed.). Oxford: Oxford University Press.

Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Newbury Park, CA: Sage.

Mackenzie, C., McDowell, C. & Pittaway, E. (2007). Beyond 'do no harm': The challenge of constructing ethical relationships in refugee research. Journal of Refugee Studies, 20(2), 299-319.

Markus, G. E. (1998). Ethnography through thick and thin. Princeton. New Jersey: Princeton University Press.

Markus, H. R., & Kitayama, S. (2003). 'Models of agency: Sociocultural diversity in the construction of action', in V. Murphy-Berman & J. J. Berman (Eds.), Cross-cultural differences in perspectives on the self (pp. 1-57). Lincoln: University of Nebraska Press.

Marshall, C., & Rossman, G. B. (1989). Designing qualitative research. London: Sage.

McCaslin, M. (2009). Co-regulation of student motivation and emergent identity. Educational Psychologist, 44(2), 137-146.

Morita, N. (2004). Negotiating participation and identity in second language academic communities. TESOL Quarterly, 38(4), 573-603.

Morita, N. (2009). Language, culture, gender, and academic socialization. Language and Education, 23(5), 443-460.

National Health and Medical Research Council (NHMRC). (1999). National statement on ethical conduct in research involving humans. Retrieved 4/4/2005, from http://www.nhmrc.gov.au/publications/. [verified 6 June 2010 at http://www.universitiesaustralia.edu.au/documents/publications/policy/statements/e35.pdf]

Oakley, A. (1981). Interviewing women: A contradiction in terms. In H. Roberts (Ed.), Doing feminist research (pp. 30-61). London: Routledge.

Patai, D. (1991). Is ethical research possible? In S. B. Gluck & D. Patai (Eds.), Women's words: The feminist practice of oral history (pp. 137-154). New York: Routledge.

Patton, M. Q. (2002). Qualitative evaluation and research methods (3rd ed.). Thousand Oaks, CA: Sage.

Refugee Council of Australia (2008). Australia's refugee and humanitarian program: Community views on current challenges and future directions. Retrieved 15/11/2009, from http://www.refugeecouncil.org.au/docs/resources/submissions/2009-10_intakesub.pdf

Reinharz, S. (1992). Feminist methods in social research. Oxford: Oxford University Press.

Rodgers, G. (2004). 'Hanging out' with forced migrants: Methodological and ethical challenges. Forced Migration Review, 21, 48-49.

Spradley, J. P. (1979). The ethnographic interview. New York: Holt, Rinehart and Winston.

Tilbury, F. (2007). "I feel I am a bird without wings": Discourses of sadness and loss among East Africans in Western Australia. Identities: Global Studies in Culture and Power, 14(4), 433-458.

Tilbury, F (2006). Ethical dilemmas: Principles and practice in research with African refugees - a response to Bailes et al. Monash Bioethics Review, 23(4), 75-84.

Turner, M. (2008). Adult South Sudanese students in Australia: A systemic approach to the investigation of participation in cross-cultural learning. Unpublished doctoral dissertation, Murdoch University, Western Australia. http://wwwlib.murdoch.edu.au/adt/browse/view/adt-MU20090902.11321

Turner, M. & Fozdar, F. (in press). Negotiating 'community' in educational settings: Adult South Sudanese students in Australia. Journal of Intercultural Studies.

Twinn, S. (1997). An exploratory study examining the influence of translation on the validity and reliability of qualitative data in nursing research. Journal of Advanced Nursing, 26, 418-423.

Usher, R. (1997). Telling a story about research and research as story-telling: Postmodern approaches to social research. In G. McKenzie, J. Powell & R. Usher (Eds.), Understanding social research: Perspectives on methodology and practice (pp. 27-41). London: Falmer Press.

Van Maanen, J. (1995). An end to innocence: The ethnography of ethnography. In J. Van Maanen (Ed.), Representation in ethnography (pp. 1-35). Thousand Oaks, CA: Sage.

Vedder, P. H., & Horenczyk, G. (2006). Acculturation and the school. In D. L. Sam & J. W. Berry (Eds.), The Cambridge handbook of acculturation psychology (pp. 419-438).

Voutira, E. & Dona, G. (2007). Refugee research methodologies: Consolidation and transformation of a field. Journal of Refugee Studies, 20(2), 163-171.

Westoby, P. (2008). Developing a community-development approach through engaging resettling southern Sudanese refugees in Australia. Community Development Journal, 43(4), 483-495.

Authors: Marianne Turner has recently completed an interdisciplinary PhD in Education and Sociology at Murdoch University in Western Australia. She has extensive experience teaching English as a second language (ESL) in Australia, UK, Japan, and Spain, including eight years at tertiary level. Email: marianne.turner@murdoch.edu.au Farida Fozdar is Associate Professor in Sociology and Community Development at Murdoch University. She publishes in the areas of race relations, nationalism, migration and discourse analysis, and is co-author of Race and Ethnic Relations (2009), published by OUP. Please cite as: Turner, M. & Fozdar, F. (2010). Dependency, partiality and the generation of research questions in refugee education. Issues In Educational Research, 20(2), 183-197. http://www.iier.org.au/iier20/turner.html