Chapter one: Rationale for the research and design of the study
This Chapter outlines the rationale for the methodological approach of the study. It examines some of the challenges in carrying out research in this field and how they were addressed in the design of the study. The aim of the research was to elicit a rich corpus of material which reflected the experiences of Pakistani women who might seek refugee protection in the UK against domestic violence1. Women from many countries seek asylum in the UK on the grounds of domestic violence, but Pakistan was specifically chosen as a focus for the research because of the landmark 1999 case of ‘Shah and Islam’. This asylum case set a precedent for the consideration of women from Pakistan as a ‘Particular Social Group’, one which experienced persecution on the grounds of membership of the group and could be considered for asylum on that basis. This ruling seemed to indicate that many women from Pakistan would be well positioned to make a claim for asylum on the grounds of domestic violence. Yet, case work in the UK highlighted a range of issues that recurred across asylum cases and drew attention to apparent flaws in the process of assessment and adjudication, often resulting in the rejection of women’s claims for asylum. In a similar way, women who had come to the UK as sponsored spouses, and had experienced domestic violence, often found themselves with insecure immigration status. By adopting a case study approach, focusing on a country where women had been granted the status of a ‘Particular Social Group’, we aimed to draw out these issues and to create a systematic analytical resource for asylum practitioners and service providers in dealing with cases that involved domestic violence.
The researchers for the project had diverse skills and experiences relevant to the themes of the study and which intersected with many of the service provision needs of women seeking protection against domestic violence. Indeed, South Manchester Law Centre as the host organisation has over 30 years experience of providing specialist legal advice and support to Pakistani women.
1.1 Methodological rationale – the need for a new evidence base
The number of Pakistani asylum seekers entering the UK is well documented by the Home Office (Bennett, Heath et al. 2007). Their Research Development and Statistics department provides comprehensive figures for the gender of those asylum seekers, and also indicates where Pakistani asylum seekers have been dispersed within the UK. However, the figures for dispersal are (1) not broken down by gender and so also (2) they do not give us a clear picture of where women from Pakistan are being dispersed to, or (3) their experience after dispersal. Equally, whilst Home Office country reports give an outline of the context of an asylum seeker’s country of origin, there are several significant flaws and omissions. The Advisory Panel on Country Information (APCI), which has been given responsibility for reviewing the Country of Information Service (COIS) reports produced by the Home Office, provided critical commentary on COIS’s April 2006 report on Pakistan. In that commentary APCI recognised that these reports rely on a small set of secondary sources and only recently incorporated references to certain established international and national NGO sources such as the Human Rights Commission of Pakistan. Moreover the commentary warned of the dangers of over-using secondary sources because of the risks associated with what they call information ‘roundtripping’, a situation which, upon careful scrutiny, might identify COIS itself as the original, therefore 1 Whilst asylum and immigration law in the UK is applicable in England, Wales, Scotland and Northern Ireland (unlike other types of domestic legislation), interviews were concentrated in England and Wales for this study. The decision to conduct interviews in these two countries was based on the project team’s view that the material gathered would be sufficiently and reliably robust to reflect the aims and objectives of the project.
The information in these reports is presented without analysis and it is explicitly claimed that no judgment is made as to the quality of the information, or how it might best be interpreted. These issues highlight a need for original, empirical research which offers both the detail and depth provided by fieldwork.
Equally, in designing and developing the study the research team drew attention to a vastly wider set of documentary sources beyond those relied upon by COIS, and which could be critical to inform wider debates on asylum and domestic violence. Specifically designed fieldwork enabled the analytical rigour and explanation required to examine the processes which both generated flight, whether in the UK or Pakistan, and determined the outcome of a claim for asylum. This was particularly the case where asylum was claimed on the grounds of domestic violence, where issues of gender, race and class intersect with those of asylum in both the country of origin and the country of refuge.
This need, for a more systematic, rigorous and complex understanding of the processes which both precipitated flight and decided its outcome, determined the selection of a qualitative research approach for design of the study (see Parker, 2004). The study was designed to identify and draw on relevant participants from both Pakistan and the UK. Qualitative interviewing was chosen as one method which could highlight and examine the range of journeys undertaken by women seeking asylum in the UK, exploring the factors which informed and constrained their choices at key chronological and geographical points in that journey. It also correspondingly enabled the generation and critical evaluation of in-depth accounts of the contexts (at a service based and wider level) in both Pakistan and the UK (Byrne 2002; Mason 2002). The strength of this approach is its focus on specificity, as also consistent with the epistemological framework guiding the study (see also section 2.7 of this report, see Wilkinson and Kiztinger,1996), so avoiding the production of a homogenized picture of the experience of Pakistani women and generating the detailed context-specific information needed to inform asylum case work. Interviewing was used alongside a range of supplementary methods including: individual (legal) case studies in the UK, a scoping survey in the UK, what we describe as ‘familiarisation’ visits to a range of service provision contexts in Pakistan, group discussions and the collation of a range of documentary sources across both countries. By using a range of sources and research methods the study aimed to address gaps in knowledge in the field and illuminate the ways in which women’s journeys might be shaped.
1.2 A note on terminology in the report
Discussions within the qualitative research methodological literature highlight the need to use appropriate terminology to reflect the model of research and analysis (Parker, 2005). While we would claim to have generated original ‘evidence’ that both supplements and comments upon available sources, rather than calling this ‘data’ or ‘findings’ (and so implying an untenable separation between the research process and its product), we have called this ‘material’, from which we have elaborated ‘analyses’. By this, we are highlighting the interpretive processes inevitably involved in the generation and evaluation of qualitative research. While such research cannot lay claims to being exhaustive (and the selection of samples and geographical arenas is thus oriented to be indicative), it is systematic and generative in its specificity and – as already outlined above – offers a complexity and depth of analysis missing from previous accounts (Somekh and Lewin, 2005). Similarly, this interpretive approach means that we avoid terms like ‘triangulation’ to describe the relationships between the different research methods adopted in conducting the study, or even between the different parties whose accounts the research documents, as this would imply claims to a spurious (and naïve) realism. Rather we take a perspectival approach that seeks to generate, explore and interpret diverse accounts. (And we use the term ‘account’ here to highlight how we can not, or rather can only indirectly, infer feelings, beliefs and experiences, through the reports or verbal accounts provided by participants to the researcher(s) (Burman, 1997; Alldred and Burman, 2005; Parker, 2005 ). This sensitivity to and acknowledgement of the contextually-shaped character of qualitative ‘evidence’ (which is also true of, but less usually acknowledged in quantitative research), is central to claims made for the ‘quality’ of qualitative research. Far from undermining its rigour and systematicity, such attention to specificity is understood to guarantee these (Denzin and Lincoln, 2000; Parker, 2004).
Further, in the field of gender and migration, language is often a highly charged and contested arena. Quite deliberate choices had to be made as to the way in which service providers and Pakistani women who took part in the research would be described, both in the course of conducting the research and in presenting this analysis. All those who were interviewed are usually referred to in this report as ‘participants’, rather than ‘respondents’. While the latter formulation may appear to be a standard social science term, it was recognised that this could (for some readers) confuse the research context with legal process (so also highlighting also the diverse audiences for this report). Moreover, in some cases, the nature of ‘participants’ involvement went far beyond the passive response indicated by the description of ‘respondent’, for some of the service provider participants actively facilitated contact with other organisations or women victims/survivors, provided advice, assistance and support to the research and additional written and secondary source information.
There is equivalent debate and discussion regarding how to describe the position of women who have experienced domestic violence (often alongside other kinds of abuse see for example Reavey and Warner, 2003). It has been argued that calling them ‘victims’ can reinforce stereotypical notions of passivity, which – significantly – are often associated both with Asian women and with victims/survivors of domestic violence. On the other hand, the more recent term ‘survivor’ – which is prevalent within the Anglo-US literature – has attracted criticism for overstating and presuming (rather than as is part of the remit of this study, exploring) the extent to which women may have overcome or recovered from the impacts of experiences of abuse. Hence, where it seems appropriate we move between alternative formulations. The women survivors/victims who took part in the research were diverse in their backgrounds, interests and in their responses to the experience of violence. Their contributions were formative in the development of the structure and arguments discussed in this report.
The term ‘service provider’ is also used both in the UK and Pakistan context. While this term is used for the sake of convenience, it is important to note the diverse range of organisations and individuals that this category covers; from the Home Office to a small domestic violence unit in a locality. It is equally important to note that in the UK and Pakistan context the term ‘service provider’ may describe organisations which are fundamentally different, both philosophically and structurally. In terms of the Pakistan material, we offer a profile of the services researched and at times identify, or otherwise characterise, the specific service from which an account was generated in order to facilitate more adequate evaluation of the significance of the source. It should of course be noted that no participant was identified without securing permission to name them.
Following practice in the two national contexts under investigation here, we write of ‘refuges’ in the UK context and ‘shelters’ in the Pakistani context.
Finally, a key set of terms in this report surround notions of ‘flight’ and ‘relocation’. While western domestic violence researchers often prefer to discuss ‘escape’ rather than ‘flight’, in the context of this study’s focus on national and trans-national contexts for seeking protection from violence and the often forced and protracted journeys undertaken, this criticism seems less relevant. Where possible, and in due respect to the often very significant temporal and geographical distances traversed (whether intra- or trans-nationally) as well as to offer a less loaded formulation, we discuss women’s ‘journeys’. There are of course different meanings associated with notions of ‘relocation’, ‘internal’ vs. ‘trans-national flight’, alongside key substantive issues at stake for this study in precisely what is understood by ‘safety’, ‘protection’ and ‘return’. We deal with these as analytical topics in the analysis and final discussion.
1.3 Overview of the research
In order to gain a coherent and comprehensive view of the complexities of (intra-country or transnational) flight and relocation involved in seeking protection against domestic violence, the research drew on a wide range of participants in Pakistan and the UK who would be working with women’s cases at various stages in the journey. This included service providers, policy makers and women themselves. The effects of domestic violence, asylum and insecure immigration on children are discussed in the report, as they were raised as issues during the field work – including the ways children impact on women’s decisions, the limits of service provisions to women with children, and in some cases (of child brides and forced marriage for example) as direct participants within the research (taking the definition of ‘child’ as under 18 years, as specified by the United Nations Convention on the Rights of the Child). However children did not directly participate in the research in the sense of being interviewed separately, (with one exception – see section 2.2.1) or their specific perspectives taken as the topic of study. We would therefore emphasise that there is an urgent need for a specific, dedicated study addressing this gap.
The study used semi-structured schedules in face to face individual and telephone interviews but, in recognition that interviews alone were unlikely to generate the depth of information we required for the study (Silverman 2003), the research also employed group interviews and case studies in the UK and in Pakistan. There was also an extensive survey conducted in the UK which drew on professional networks to identify service providers working with this client group. Lastly, the researchers carried out an initial literature review of the field which was used to contextualise the conduct of the research and inform the development of the interview schedules. This review was supplemented during the course of the study as the fieldwork also generated access to further, previously unknown sources, and a range of source material was collated and used to inform this final report, including government documents in Pakistan as well as material from NGOs and local service providers.
In using mixed methods the research aimed to generate material from a variety of sources to ensure thorough evaluation of relevant material, including its remit, strengths and limitations, to inform coding and analysis (Barbour, Rosaline et al. 2001). Thus this range of sources enabled the researchers to attend to diverse perspectives and evaluate the contribution of different sources of evidence in producing different perspectives and interpretations (see also Stronach, 1997).
Overall the research consisted of –
Pakistan based research in the geographical regions of Punjab, Sindh and the North West Frontier Province:
UK based research in England and Wales:
In both countries the research took part in two phases. The first phase consisted of a scoping or piloting stage, where key informants and organisations were identified. In the UK this comprised a postal survey, and in Pakistan it consisted of a field trip where ‘familiarisation’ interviews with relevant stakeholders in the research took place. The familiarisation interviews were intended to identify potential participants and give a broad understanding of the practices and systems in Pakistan concerning domestic violence. This phase was used to inform the topics and areas of questions structuring the schedules, and to pilot schedules with a small sample of potential participants. It should be noted that this initial phase was conducted differently in the UK and Pakistan, in recognition of the need for different methods in the two contexts. The rationale for the two approaches is described in the following sections. The second phase consisted of field research in both the UK and Pakistan, carrying out individual interviews, observing key contexts of practice and facilitating group discussions.
1.4 Methodological challenges
The aim of the WASP research project was to generate a more informed understanding of the issues faced by Pakistani women in attempting to escape domestic violence and claim asylum in the UK. There were several key substantive research questions which needed to be addressed by the study in order to realize that aim. These questions emerged from case work at South Manchester Law Centre and from an initial literature review of research. The questions were refined and developed with the collection of further material during the familiarisation interviews in Pakistan and the scoping exercise in the UK. These questions were used to frame and focus the research in order to ensure that the results could contribute to the knowledge base used by case workers, service providers and decision-makers:
These research questions informed the design of the research and guided initial discussions with participants. Moreover, the researchers were also concerned to document diverse perspectives and to elicit a broad range of responses. This was reflected in the choice of a range of methods and participants.
1.4.1 Issues in trans-national research
Carrying out trans-national research in this way presented several methodological challenges. The differing contexts in the UK and Pakistan demanded different approaches to the research design, particularly in the first phase of the work.
In Pakistan, services for women who had experienced domestic violence were largely acknowledged to be sparsely provided and poorly resourced. Such services often relied on local NGOs, local activists, or even philanthropic individuals, for their survival and were found to be very diverse in their structure and focus (analysis of the Pakistan context is presented substantively in Chapters four to seven). In some services provision involved male workers and organisations were sometimes led by men as this was perceived to provide some degree of protection to the women service users and workers. Equally, services which provided support to women victims/survivors in Pakistan did not necessarily have domestic violence as their main focus of work. Many organisations that took part in the research focused on supporting families, and saw their role in terms of mediation and reconciliation within the family. It was important in generating and interpreting the research to be clear about the very different philosophical approaches towards domestic violence service provision operating within the two countries. Yet, to achieve a balanced and accurate interpretation of the emerging trans-national nature of barriers to women, it was also necessary to highlight apparent parallels between the UK and Pakistan in limitations to service provision (these issues are discussed in depth in Chapters eight to ten).
In Pakistan the professional networks of such services and regulation within the sector are also in their infancy. We note these issues here to clarify the asymmetrical character of domestic violence service provision in the UK and Pakistan, and to indicate the challenges and limits to conducting trans-national research on this topic. This is reflected in the availability of information, as monitoring and evaluation data was not as easily available in Pakistan as in the UK. Much information was sourced from newspapers for example (and here it is significant to note that both government and non-government agencies often appeared to rely on press coverage of cases to assemble their statistical profiles of domestic violence). Furthermore, most of the information was held within informal networks and knowledge about the prevalence, and issues surrounding, domestic violence was available from individual, rather than formal, sources.
In the UK services for women who had experienced domestic violence, or were seeking asylum, were formalised, networked and regulated (in the sense of being set out in national or publicly available policy documents, with agreed standards). They were usually run by paid workers, operating within a professionalised structure, and levels of service provision were comparatively much better resourced. Resources were also usually provided through more formal institutions and service providers often had to formally report to funding bodies and make information about the service available in the public domain.
Such differing contexts demanded different methods to identify research participants. As is described later in the Chapter, in the UK participants were identified through a survey which used formal networks, while in Pakistan the researchers used informal networks, ‘familiarisation’ visits and ‘snowballing’ to contact potential participants. Despite these different methods, the sampling in both countries ensured a final group of research participants which included a broad range of service providers and women victims/survivors.
It is also important to highlight the physical and geographical difficulties of carrying out research in Pakistan. Whilst the research focused on specific regions, and largely urban areas, there were still logistical challenges in travelling across great distances whilst ensuring the safety of the researchers. Most of the organisations who were supporting women victims/survivors had experienced threatened or actual violence and most employed security guards at their premises. It was decided that for travel within urban areas the researchers would use the services of a paid driver, who would be available during the research period. The safety of the researchers was also considered in booking hotel accommodation and interregional aircraft journeys and, as part of the ethics scrutiny process (see section 1.4.2), the researchers were required to undertake a risk assessment before undertaking the work to ensure that neither they, nor any participant would be put at significant risk as a result of the work.
The focus of the research was also necessarily different in the two contexts. Questions around asylum and migration were of little interest or relevance to service providers and policy makers in Pakistan. The focus there was on domestic violence and its relationship with women’s rights within the country. In the UK respondents were interested in the asylum system as it affected their clients, but often had little awareness of the context in Pakistan and the difficulties that clients had experienced in their countries of origin in getting support or accessing services. The differences in the interests and knowledge of the two samples were reflected in the construction of the schedules. Despite this differing focus, wherever possible, the schedules used in the UK and Pakistan mirrored each other and addressed equivalent areas. This was the case for both service providers, policy makers and women victims/survivors and was made possible by using questions which were framed around the chronological journey of the woman, from her familial environment in Pakistan to asylum in the UK.
1.4.2 Ethical considerations in the research
The research plan was subjected to review by the Departmental and University Ethics Committees at Manchester Metropolitan University, the academic partner in the research. Ethics approval was gained prior to embarking on fieldwork. However, there were additional ethical issues which were raised and addressed during the course of the fieldwork which it is relevant to note here.
1.4.2.1 Women victims/survivors, anonymity and confidentiality
In Pakistan the most acute ethical dilemmas encountered were those relating to the needs of women victims/survivors. An indicative example is how, whilst carrying out research in shelters and crisis centres, the researchers found that women sometimes disclosed abuse or mistreatment that they had experienced within the service. The women who raised these issues asked the researcher not to act on the information or to share it with any other party. The women wanted to be able to discuss their experiences, but were concerned that any complaint might lose them their accommodation or lead to them being further discriminated against. That such concerns were expressed already indicates key issues for the evaluation of available service provision. In terms of the research process, it was recognised that even if the women had been willing to share the information more widely, it would have been difficult for the researcher to intervene as there were rarely any formal procedures that could be followed to report such issues and there were few, if any, alternatives for the women should they lose their accommodation. After discussion with the supervisor and with those providing support in Pakistan it was decided not to act on this information both in order to respect their requests and maintain the women’s access to the (albeit) limited services. This example does highlight ethical difficulties of carrying out research within an unregulated sector. We have however included some such responses in our analyses but in such a way as to prevent identification of the specific service as well as the participant.
The impact of the unregulated nature of the sector was also confirmed when in one shelter the researcher encountered two unaccompanied children, girls who were aged seven and thirteen, who had been sold into prostitution and marriage. The researcher was able to interview the child aged thirteen for the study as the particular circumstances surrounding her residence at the shelter appeared to be consistent with the accounts given by women of abuse from an early age. The shelter was an inappropriate place for such young children, but there was no other more suitable accommodation for them and no regulations concerning appropriate service provision for children. The researcher also met, although did not interview, two unaccompanied girls in the advice and resource centre for Christian women run by CLAAS2 who were later taken to their shelter.
While participants were all guaranteed anonymity, some women were living ‘in hiding’ because of the particular nature of the violence they were escaping, notably the threat of honour killings. Additional efforts have been made to prevent specificity of detail from rendering women who were living in such circumstances identifiable. This has meant that, in publicly available accounts of the specific ‘case histories’ or long accounts of women’s stories that we have provided, we have sometimes changed some specific biographical information. Obviously these changes had to be made without compromising the substantive analytical points to be interpreted from the material, for example substituting the name of a city for another in the same region. Given the extremity of some of the circumstances described in these cases, such amendments were deemed a necessary precaution particularly as the report will be widely disseminated in Pakistan.
1.4.2.2. Support needs of women victims/survivors
Women victims/survivors also disclosed issues such as self-harm and during the interviews and discussed painful life events such as rape, or physical violence. The difficult nature of these experiences sometimes resulted in the woman becoming acutely distressed and the researcher had to use judgement and sensitivity to know how far to take the questioning. This sometimes limited what information could be gained from the interview, but it was important to allow the women to have some control over the interview process and feel free to end the discussion at any time. In most cases the interviews were completed but over a longer period, sometimes involving two or three separate visits.
In addition the presence of children at the interviews often interrupted and prolonged the process. The research team were aware of the ethical dilemmas this presented; whether to continue interviews and potentially expose children to distressing information as well as to visibly distressed parents or to negotiate alternative times to complete interviews. In situations where this arose women were often unwilling to be parted from their children and were keen for the interviews to continue. The researchers observed and noted such responses; indeed, the presence and behaviour of children at interviews is elaborated on in later Chapters.
Many women had also been forced to leave their children behind, or had had their children taken away from them. Speaking about such difficulties could also result in distress. Again, it reflected the unregulated and poorly resourced nature of the sector that little or no support was available in the shelters after the interview. Indeed although shelters in Pakistan do have counsellors, it was often unclear how often residents were able to access their services, and indeed it emerged (see also Chapter six) that little or no longer term support was offered. The research team were very concerned about the lack of physical and emotional support available to women within shelters, and were aware that the interviews could raise difficult and painful emotions when there was no adequate aftercare in place. Some women victims/survivors were aware of this lack of support and commented that they had been interviewed before but did not feel they had benefited from the process.
2 CLAAS – the Centre for Legal Aid Assistance and Settlement is an interdenominational organisation working for Christians who are being persecuted because of their faith in Pakistan. It has offices in the UK and Pakistan.
As a result, the team became aware of a need to provide some, albeit token, recompense for the women victims/survivors’ time and commitment to the research. There has been much debate within the research literature about the role of the researcher in qualitative research in terms of feedback, benefit and accountability (Finch, 1984; Ribbens and Edwards, 1998). Hence alongside maintaining an analytical approach to the work of soliciting and documenting participant accounts, provision of additional advice after the woman’s participation in the research was seen as delivering on a necessary ethical commitment. This advice typically drew on the researcher’s wider knowledge of services in the region. Whilst it was considered inappropriate to arrange payment to an individual woman who was resident within a shelter for her participation, payment in kind was made on an equal basis for the shelters which facilitated access to their residents and helped in the research. This contribution amounted to approximately £50 per shelter and was made in kind, in the form of food (types identified as lacking in the residents’ diet – including chicken and fruit), plus in some cases purchase of educational materials for the children. In the UK, individual women who participated in group discussions through their involvement with support groups were offered payments of £20.
1.4.2.3 Consent to participate
With such a vulnerable group of women, who were often living in insecure accommodation, it was important to stress the notion of informed consent and to ensure that the women were informed about the research and what their participation would mean. However, many of the women survivors were not literate and could not read any material about the project, so consequently were unable to give written consent. Instead the research was thoroughly discussed with each woman and her verbal consent gained. Although the researcher did speak Urdu and Punjabi (two widely spoken languages in Pakistan) some women spoke a local or regional dialect as their first language and the first task of the interview was sometimes to agree a language which both the woman and the researcher could use. This meant that whilst the woman might be very competent in the language used in the interview, it was not necessarily her first language. To ensure that the woman understood what was being asked of her, and gave her formal consent, the researcher was particularly thorough in explaining the research and checked with the woman at each point to ensure that she understood. However, this was an inevitably uneven process, and whilst the researcher did not pursue any women she felt were unsure about the research or hesitant about taking part, the research was largely reliant on the judgement of the researcher to ensure that consent was fully informed and fully given.
Despite the significant logistical and socio-political difficulties of conducting research in Pakistan, ultimately the numbers of research participants were higher. This perhaps can be understood in terms of the eagerness with which women victims/survivors and service providers sought to participate in the research. Another factor which may provide an explanation is that the unregulated nature of services meant that service providers sometimes allowed open access to participants with little attention given by them to confidentiality/consent issues. This allowed access to women other than those who had been previously selected by the service providers to be invited to participate in the study. Indeed this posed a further ethical dilemma for the researcher, when women who had not been identified as willing participants by the service providers who were facilitating the research process subsequently approached the researcher and asked to take part. Where possible the researcher conducted interviews with these women and in doing so this can be regarded as compensating for or correcting any systematic distortions introduced by workers’ selection of women residents. Again the researcher followed the process of obtaining informed consent as described above.
1.4.2.4 Supporting the researchers
In addition to the support needs of the participants, it should be mentioned that the distressing character of participants’ accounts, alongside the often very limited and impoverished circumstances in which residents at shelters and other facilities were living, posed ethical issues regarding support for the researchers. This was addressed during the second field trip by setting up additional support in Pakistan with Farida Shaheed, Director of Shirkat Gah as a relevant and experienced senior researcher specialising in the field of violence against women in Pakistan. We note this matter here as the impact on and duty of care for researchers, as well as researched, is too often overlooked and if unaddressed can also work to compromise the quality and sustainability of conducting research in this area. It is also clearly a relevant consideration for subsequent studies that we hope will take further some of the research questions and issues identified by this study.
1.4.3 Responses to the research
Response rates to the research varied widely both across and within the two countries. In the UK a survey was used to identify service providers who were working with Pakistani women fleeing domestic violence. The rate of response was initially low in this exercise, for a range of reasons. The survey questionnaire was deliberately sent to a large number and range of organisations in order to identify those who were working with Pakistani women, including those who might not usually be recognised for their involvement in such work. Those who did not work with these service users did not see the questionnaire as relevant to their work and did not return. This was expected as the net had to be cast widely in order to identify service providers who had cases or worked with women in this situation. However, in following up the survey questionnaires it was clear that there was a widespread sense of ‘research fatigue’. Many of the workers, particularly those working in asylum and immigration, felt inundated with work and were resentful of incursions into their limited time. For workers who were also funded through the work they performed, such as solicitors and some legal case workers, there was also a sense that time was money. As one case worker pointed out - “We have no financial support, even though these questionnaires are very important we have no financial assistance to cover the cost.”
Service providers also responded differently in the UK and in Pakistan. The often informal nature of provision in Pakistan made it very difficult to make initial contact with providers. Making initial contact, and pursuing those contacts to arrange a meeting, demanded a level of persistence that the UK research did not. However, as the research progressed within Pakistan, and participants began to invest in the research, the researchers found that the participants facilitated access to other service providers and to women victims/survivors who were using services. This facilitation was carried out despite a lack of resources and the fact that services in Pakistan were often responding to crisis.
In the UK service providers were easily identifiable and largely responsive once initial contact had been made by letter. However the formal nature of service provision meant that despite an initially welcoming discussion, progression beyond that was at times difficult. Service providers were concerned to protect vulnerable clients and were conscious of issues around data protection and the legal requirements they had to meet. Services in the UK were also often concerned with the possible ramifications in terms of funding, or the requirements they had to meet in terms of provision, which could at times make them defensive or unresponsive to the research. This was illustrated in the response of the Home Office to the research. Several individuals with responsibility for gender, asylum and/or domestic violence policy and practice had initially agreed to be interviewed, but later declined and instead submitted a joint written response.
Similarly NASS3 were invited to take part in the research but stated that, as they had no knowledge of where the asylum seekers they supported came from, they could not be aware of the reason for their claim. When asked if they would be interested in commenting on the other aspect of the research, that of Pakistani women who experience domestic violence in the UK, who might be involved in a joint asylum claim with a partner, they declined and referred the researchers to the NASS policy around domestic violence. It is difficult to make any specific interpretation of the reason for these responses but, especially as in some cases there were implications regarding resources linked to targets, it is hard not to surmise that the topic of the research was such that some service providers and policy makers seemed uncomfortable in being open or transparent about their own systems and practices.
1.4.4 Design of the interview schedules
The design of the interview schedules presented three distinct challenges for the research team. Firstly, the schedules had to be specific enough to draw out responses which addressed the research questions, yet also allow room for emergent themes to appear and to explore the particular perspective of the individual participant. Secondly, they had to be designed to focus on the specific interests of each ‘type’ of participant, whether an NGO worker a local police officer or national policy
3 During the course of the study, and after the interviews had been completed in March 2007, NASS (the National Asylum Support Service) ceased its operations as the Home Office agency responsible for the asylum dispersal and support system. These functions are now undertaken by the Home Office’s new Border and Immigration Agency (BIA), which is now responsible for all issues connected to asylum and immigration control in the UK.
maker, whilst still addressing common themes which would allow for analytic comparison. Lastly, the schedules had to be responsive to the specific contexts in each country, whilst again exploring equivalent narratives and issues to allow for comparison across the two contexts.
In consideration of these issues, the schedules were designed chronologically to identify the key issues posed for a woman victim/survivor at each point in her journey, from the decision to leave her family of origin, or marry outside Pakistan, to her eventual status in the UK. Using this chronological approach allowed the researchers to explore similar points in the victims/survivors’ experience from multiple perspectives, and attended both to similarity in themes and specificity in an individual participant’s perspective. Where necessary questions were added for specific participants, such as the police, or for policy makers (see Appendix One for examples of the schedules) but the construction of the schedules around a chronological journey enabled the research team to build coherence into the organisation of questioning and the exploration of themes. Equally, despite the differences in the focus of the research in the UK and Pakistani contexts, following a chronological journey allowed the questions in the two sets of schedules to reflect each other and track key stages in that journey.
The actual topics explored in the schedules were informed by the knowledge of the team based in the Law Centre and as framed by the initial research aims. The topics and specific questions were refined by the feedback from the first stage of the research; the familiarisation interviews in Pakistan and the scoping exercise / survey in the UK. The feedback from the first phase identified specific points in the journey of a woman victim/survivor where her choices were constrained by gendered legal, social and economic processes. Feedback from participants in the first phase of the research also confirmed that the initial research questions were appropriate and focused on the areas where there were either conflicts in interpretation or a lack of clear evidence.
In Pakistan the schedules were presented for comment to the following key partners in Pakistan for discussion; the AGHS Legal Aid Cell4, PANNAH shelter5, the Aurat Foundation6 (Peshawar), CLAAS7 advice and resource centre and DASTAK shelter, whose feedback provided relevant ‘member checks’. The schedules were amended before being used in the second phase of the research in Pakistan.
In the UK, the schedules were piloted with three workers from South Manchester Law Centre who were working with Pakistani women asylum seekers. Feedback from the pilot interviews was used to adapt the schedules so they would be more applicable to specific participants.
1.4.5 Interpretation, recording and transcription
In both Pakistan and the UK non-English speaking respondents were provided with female interviewers who could speak several languages. In this way many of the women victim/survivors who were interviewed were able to respond in their first language. However, this was not possible in all cases, many of the women spoke local dialects and a common language for the interview had to be agreed at the outset any discussion took place. Interviews with service providers in Pakistan were sometimes conducted in English, depending on the preference of the participant.
It was anticipated that all the interviews would be taped and then transcribed for detailed analysis after the field work. However initial work in Pakistan indicated that participants were often reluctant to be audio-taped. Both government officials and local service providers were often more comfortable with note taking and participants such as the police refused to have interviews taped. Equally, women were often interviewed in busy shelters where noise levels made audio recording impractical.
As a result the majority of the interviews in Pakistan were recorded using notes taken during the interview, and completed following each interview with additional notes and observations made about the interview and the context. In order to mirror this method the majority of interviews in the UK were recorded in the same way: interviews were recorded using notes, which were written up with additional observations soon after the interview had taken place. Some interviews with women
4 AGHS Legal Aid Cell is a relatively small but important NGO in Pakistan. It was set up in 1980 to provide free legal aid to women.
5 A women’s shelter based in Karachi.
6 The Aurat Foundation is a women’s rights organisation based in Islamabad, Pakistan.
7 CLAAS – the Centre for Legal Aid Assistance and Settlement. ibid.victim/survivors in the UK were audio-taped, but there were similar difficulties with the audio quality of the tape as the interviews took place in busy settings and young children were sometimes present. The recorded interviews in Pakistan were transcribed and translated in Pakistan and the transcriptions were checked by native speakers for errors or omissions as well as by the interviewer. Similarly, any UK recorded interviews were (where relevant) translated, transcribed and checked for errors by the relevant UK interviewer. Citations used in the report are taken from both the recorded and transcribed interviews and from detailed notes taken during the interviews.
1.5 Research in Pakistan
This section outlines how the research was carried out in Pakistan and the rationale for the choice of methods. There were two fieldwork periods in the Pakistan research: November to February 2005-6 and May to July 2006.
In Pakistan it was not possible to identify a sample using formal survey or sampling methods. The scattered nature of provision, and the lack of formalised networks, meant that the field workers had to use less formal methods to contact potential participants.
In Pakistan the research was largely carried out by a sole researcher (Nadia Siddiqui) who could speak Urdu and Punjabi. The researcher had previous experience of carrying out research on related topics in Pakistan (a short field trip in December 2003 and a further trip from May to July 2004 funded by the Winston Churchill Memorial Trust) and had an existing network of contacts around domestic violence and services for women in some of the cities focused on in the research. This researcher carried out all the interviews with women victims/survivors and with the majority of the service provider interviews in Pakistan. Her work was aided by a research consultant from the project team (Yvonne Prendergast) who joined her in Pakistan for a one month period during the second phase of the research to assist with and conduct interviews.
1.5.1 Identifying the research participants
The research in Pakistan involved using systematic non-probabilistic sampling (May, Nicholas et al. 1995; Barbour, Rosaline et al. 2001). This method involved identifying target organisations with specific characteristics; they were working with survivors of domestic violence, or working in this policy area, they had knowledge of the issues surrounding domestic violence and they were part of informal networks working around domestic violence at a regional or national level. The first field trip took place from November 2005 – February 2006 and was used to carry out familiarisation interviews with this initial group of individuals and organisations. The aim of these familiarisation interviews was to –
These interviews enabled the researcher to ‘snowball’ contacts and develop a network of relevant partners which could be used to identify further services and to make contact with women victims/ survivors of domestic abuse. A second field trip was made in June – July of 2006 where more formal interviews were carried out with both service providers and women who had experienced domestic violence.
The geographical regions covered within the study are Punjab, Sindh, and the North West Frontier Province (NWFP). The geographical diversity and size of Pakistan made it necessary for the research to be limited to specific regions. The experience of South Manchester Law Centre in working with this client group indicated that a sizeable proportion of cases in Manchester originated from the above regions. Equally, although women came from a wide range of backgrounds, including both urban and rural settings, women who left abusive relationships were known to gravitate to cities where they were more likely to access services with a degree of anonymity. As a result it was decided to focus largely on the following key cities: Peshawar in the North West Frontier Province, Lahore, Faisalabad, Islamabad, Rawalpindi, Karachi and Hyderabad.
Table 1 Participants in the Pakistan research – Phase One ‘Familiarisation interviews’ November – February 2006
Punjab
Sindh
North West Frontier
Totals across Pakistan
INGOs
3
-
-
3
NGOs
7
6
2
15
Shelters / Crisis Centres
5
3
-
8
Police
-
1
-
1
Other key Informants
3
3
-
6
Number in each area
18
13
2
33 familiarisation interviews with service providers
Table 2 Participants in the Pakistan research – Phase Two, Field Work May – July 2006
Punjab
Sindh
North West Frontier
Totals across Pakistan
INGOs
3
-
2
5
NGOs
10
7
2
19
Shelters / Crisis Centres
6
3
-
9
Police
2
1
-
3
Government Officials / Government Minister
2
1
2
5
Other key Informants
5
-
-
5
Number in each area
28
12
6
46 individual interviews with service providers
1.5.2 Selecting the sample for the research
The familiarisation interviews enabled the field workers to identify a viable list of over seventy organisations who had experience in working with this group of women. Potential organisations and individuals were then identified on the basis of –
Semi-structured individual interviews were carried out with these participants. The interviews focused on three broad topics; the legal and social context in Pakistan around issues of gender and domestic violence, the barriers that women faced when trying to flee or prosecute violence and the issues that organisations face in providing services to women. Appropriate schedules were devised for the different respondents (See Appendix One) and the final balance of interviews, and spread across the regions, is represented in Tables 1 and 2.
A total of 33 familiarisation interviews were carried out during the first field trip (See Table 1). The time which elapsed between the first and second field trips allowed the research team to reflect on the research and make any necessary methodological adjustments to the study. On the second field trip a further 46 interviews were carried out with organisations and individuals across the three regions, representing participation from a wide range of organisations including refuges, legal representatives, NGOs, INGOs and government officials. The material generated from both field trips has been used in the analysis. In carrying out the fieldwork with service providers we also identified a sample of women who had experienced domestic violence and were willing to be interviewed. The women were drawn from a range of regions and 25 interviews were eventually carried out, the breakdown of the sample is detailed in Table 2.
1.6 Research in the UK
In the UK the research was carried out by Nadia Siddiqui and Sajida Ismail, the two researchers on the project, with the assistance of a UK based research consultant who joined the project team (Meg Allen) and two researchers from Roehampton University, (Aisha Gill and Aarunima Bhatnagar) both of whom who were Urdu speakers. The research consultant and one of the project researchers (Nadia Siddiqui) carried out interviews with service providers, while one project researcher (Nadia Siddiqui) carried out the focus group discussion with women victims/survivors. The researchers based at Roehampton University were commissioned to carry out interviews with women victims/survivors, using the contacts that had been made through the exercise which identified potential participants (see the description of the scoping exercise in section 4.1).
In the UK the sample was identified using a formal survey questionnaire and by the addition of other key participants. The material gathered from this scoping exercise was analysed by the research consultant in consultation with the project researchers.
1.6.1 Identifying the research participants
In order to identify participants for the research in the UK a wide-ranging scoping exercise was carried out. The aim of this exercise was to –
The exercise involved a postal/e-mail questionnaire that was sent to 971 organisations. The experience of the workers at the law centre indicated that women would be using a specific range of services and the service respondents were selected accordingly –
Citizen’s Advice Bureaux were selected, as they have been funded nationally to provide specific services to refugees and asylum seekers across the UK. National CAB provided the team with a list of 361 area bureaux who were sent a postal questionnaire and then followed up by an e-mail reminder.
All Women’s Aid affiliates, who are listed on the national website, were e-mailed the questionnaire and were followed up by e-mail. Local Refugee Councils and Regional Refugee Actions were asked to provide lists of local organisations that gave support or advice to refugees. Solicitors were selected from the Law Society’s list of accredited immigration and asylum practitioners and were sent a postal survey. Some were then followed up by telephone interviews.
All law centres in England and Wales were sent a postal survey, which was then followed up by e-mail and telephone.
Local race equality organisations were selected from a list of CRE (Commission for Racial Equality – now merged into the Equality and Human Rights Commission) funded organisations and sent a postal questionnaire. These were followed up by an e-mail reminder.
• Specific Asian women’s organisations were also targeted and either sent a postal or e-mail questionnaire. This was then followed up by telephone contact. The questionnaires were sent to a broad range of organisations, many of which might not be working with this client group. However, it was important to cast the net as widely as possible in order to contact those groups or individuals who were and whose contributions might have been missed by previous research. This inevitably made the initial response rate poor, but this was not an issue for the research as the purpose of the questionnaire was to identify particular services rather than provide a systematic review of provision. Initial analysis showed a poor return from specific areas and from certain organisations. In order to improve the return rate, organisations were selected from areas with poor returns; London, Wales and Yorkshire & Humberside. These organisations were contacted and completed the questionnaire by telephone. Several interviews were also carried out with targeted organisations in the North West. These supplementary interviews gave a final return rate of 16.3%. Many of those who did not want to take part in the research nevertheless gave valuable feedback which has contributed to the report.
The scoping exercise generated a viable list of contacts for the consultation. We identified over thirty organisations that were working with Pakistani women asylum seekers and an additional twenty who had no contact but were interested in talking part in the research. The aim of the scoping exercise was also to enable us to develop a database of contacts for organisations and individuals who were working in this area. These contacts could be used for later networking and for the dissemination of the results of the research.
1.6.2 Selecting the sample for the research
In the research plan we had also identified several key areas where we wanted to carry out the research – the North West, the Midlands, London and Wales. These areas did not include all the areas targeted in the scoping exercise as described in paragraph above. The North West, the Midlands and London were selected because they had large Asian populations and high levels of asylum seeking populations. The high numbers of returns from these areas in the scoping exercise confirmed our initial selection. Wales was additionally chosen as an area where there were low percentages of ethnic minority populations, and where dispersal was a relatively recent phenomenon. The return rates for Wales in the scoping exercise reflected these low numbers, and we were concerned to explore how services were delivering in an area where the Asian population was low, and expertise around asylum and immigration less available.
Table 3 Participants in the UK research – Phase Two Field Work October 2006 – April 2007
North West
Midlands
London
Wales
Totals across the UK
Home Office
-
-
1 (and one written response)
-
2
NASS officials
-
-
1
-
1
Lawyers / Barristers
3
1
2
-
6
Refuge / Domestic violence services
3
1
2
1
7
Refugee Organisations
1
1
2
1
5
Police
1
-
-
-
1
Legal Support
2
-
1
-
3
Other key Informants
-
-
1
1
2
Number in each area
10
3
11
3
27 Individual interviews with service providers
Women asylum seekers
-
6
1
1
8 Individual interviews with women asylum seekers
From the scoping exercise we identified thirty-six organisations from these four areas which had worked on cases of this kind and were prepared to take further part in the research. We also identified several organisations which would be key informants such as the Home Office, NASS and the police, who had not been included in the original scoping. Participants were then selected according to one or more of the following criteria –
Semi-structured face-to-face interviews were carried out with these respondents. The interviews focused on three broad topics: the issues that workers faced in dealing with such cases and the systems and methods that were used to assess cases; the culture and situation in Pakistan; and the perceived difficulties of return and the issues for the women themselves in such cases. Appropriate schedules were devised for the different participants and the final balance of interviews and spread across the regions is represented in Table 3.
A total of 28 organisations and individuals were interviewed across the areas, representing feedback from a wide range of organisations including refuges, legal representatives, asylum and refugee organisations and the Home Office. One face-to-face interview was completed with the Home Office, but other participants there consented to a joint written response. This response has been included in the analysis. In carrying out the field work with service providers we also identified a sample of Pakistani women who were claiming asylum on the grounds of domestic violence. The women were drawn from a range of regions and the breakdown of the sample is detailed in Table 3.
1.7 Individual case studies
This study was precipitated by the work of South Manchester Law Centre where specific cases had informed early discussion and shaped the research questions. Four of these UK cases were selected to give a sense within the report of the circumstances which have led to women and their children seeking refugee protection in the UK and how these cases fared within the asylum process. In particular, the UK cases illustrate the kinds of considerations that appeared to be taken into account in arriving at asylum determinations, and the logic or lack of logic, of those decisions.
The case studies from Pakistan were formulated from research interview records as part of the research aim to document a woman’s account of seeking safety from domestic violence, including attention to the sources of support she did and did not seek out, and the kinds of responses she encountered.
Beyond this, both sets of cases figure within the report as a vivid record of the impacts and meanings of familial or domestic violence and its implications for a particular woman’s current situation. Apart from the obvious difference of national location, the two sets of cases are asymmetrical in the sense that the UK cases were generated as ongoing legal cases while the Pakistan cases are generated from research interviews (which are of course different from legally-focused asylum interviews). Nevertheless they both convey, and work as a powerful reminder of, the position of a particular Pakistani woman. The specificity of a woman’s position, and her account of her experiences, can often be lost within a wider thematic analysis of the kind we have undertaken in this study (see Shacklock and Thorp, 2005; Beverley, 2000). We therefore frame our substantive Chapters with these longer narrative accounts. This attention to narrative chronology not only exposes specific issues that emerge as relevant in particular cases, but it also works as a useful corrective to the cross-sectional analysis undertaken as the primary mode of presentation of the material (i.e. juxtaposing the accounts of different participants) and brings back into focus the perspective of the women around whose circumstances and provision this study was formulated to investigate.
1.8 Group interviews / discussions
The group interviews were conducted in part to give additional depth to the individual interviews and provide opportunity for discussion in a group setting (Fontana and Frey 2003). Group interviews are particularly useful for exploring people’s knowledge and experiences and were used to generate wider comments on the participants’ views of the research questions (Kitzinger 1995). Group discussion of this kind is typically more wide ranging and generative than one to one interviews, as participants associate with, and develop, each others’ responses. They are useful to check out emerging interpretations, counter sampling biases (of those individuals willing to be interviewed) generate a wider pool of participants and highlight areas of consensus and debate amongst the participants’ accounts. In this sense the group interviews were used both as an exploratory and phenomenological tool (Sinagub, Shay Schumm et al. 1996). The use of groups in a natural field, allowing freedom for group members to develop the discussion, enabled the researchers to make juxtapositions with material from the interviews with service providers and women victims/survivors and ensure coherence and systematic evaluation for the thematic analysis. Moreover, by enabling group participants to develop the discussion, the process also enabled women to safely withdraw from certain stages in the discussion, for example, when the focus turned to more personal elements of women’s accounts of violence. Indeed, in one group discussion which was originally composed of thirty women, approximately half the participants decided to not participate once the discussion turned to individual women’s accounts. A topic guide was devised which explored similar broad themes to those of the individual interviews (see Appendix One).
In Pakistan seven group interviews / discussions were held. Two were facilitated by NGOs in Lahore in which the participants were women from a Christian shelter, women from the community and local workers. The third was held with NGOs working in Hyderabad and Interior Sindh areas. Three were held at shelters in Lahore and Karachi where the participants were residents and ex-residents at the shelters. The final group was facilitated by Kwendor Kor (an NGO) with the researcher, where the participants were a group of local women in Peshawar.
A group interview was also carried out in the UK in the North West. The group was enthusiastic and the discussion engaged, therefore, it was extended to take place over two sessions a week apart and was hosted by an organisation providing support to asylum seekers. Women who attended the group were Pakistani women asylum seekers and victims of domestic violence and two students on placement at the project also attended.
1.9 Analysis
The aim of the research was to elicit a rich corpus of material which could reflect the asylum journey of a woman fleeing domestic violence in Pakistan. At the beginning of the project the experience of the research team had informed the structure and focus of the research, and initial research questions were identified at an early stage. Since the schedules had been constructed chronologically, using the research questions as a starting point across all the participants and both countries, it was decided that the research material would be analysed systematically using a thematic approach. Burman (1998) defines a ‘thematic analysis’ as –
‘A coherent way of organising or reading some interview material in relation to specific research questions. These readings are organised under thematic headings in ways that attempted to do justice both to the elements of the research question and to the preoccupations of the interviewees.’ (p.49)
Using a thematic analysis presents a range of challenges. Whilst a distinction is often made between either generating themes purely from the material itself or from pre-determined research questions (Braun and Clarke 2006), it was important in the context of this study that the analysis was a balanced representation of the research questions (and the topics and issues these covered) and other relevant issues raised by the participants during the fieldwork. The research team were concerned to ensure that the analysis answered the questions raised by case work, but also paid attention to convergence and divergence across the research themes and highlighted diversity and variation in responses as well as commonality.
In terms of paying sufficient attention to each specific context, the material from each national context was analysed separately and then juxtaposed with the other, to highlight commonalities and divergences. In order to address both of these needs the analysis began at an early stage in the field work. In the first phase, whilst carrying out the field work, the researchers regularly reviewed their interview tapes and notes in order to familiarise themselves with and consolidate the material. This process involved forming initial ideas as to the emerging themes, those that both addressed the research questions and presented new issues. The field workers wrote regular reports, drawing on their interviews and field notes, which were discussed with the research supervisors in Pakistan and the UK and developed during the course of the work. This process was ongoing and lasted until the completion of all fieldwork in late 2006, by which point a provisional collection of themes and subthemes had been identified for each corpus of material.
After the completion of the field work each of the researchers worked with the supervisory team to analyse the material. This corpus of material was reviewed and relevant features or issues were coded systematically and collated into themes (see also Parker, 2005). These themes were compared against those which had been identified during the field work and were reviewed and refined through discussion within the team. This was an iterative process, including successive consultations and discussions, with an attention to convergent and divergent perspectives and to minor as well as major themes. This was considered by the team to be especially important as the earlier case work had indicated a need to pay due attention to the diversity of issues within cases, as well as the commonality of issues. Despite this concern to identify and document specificity great consensus across the participant accounts emerged as to the key issues and the dominant themes within the research area. A wide range of participants, with diverse perspectives and positions, echoed very similar views and identified similar issues. The Home Office response differed from those of the rest of the UK participants and these differences are discussed in subsequent Chapters.
1.10 Structure of the report
The structure of the report mirrors the design of the research in that it follows, chronologically, the journey of a Pakistani woman who seeks asylum in the UK on the grounds of domestic violence. The research in the UK and Pakistan is presented in separate Chapters of the report in order to provide the reader with ease of access to the information. However the continuity of the women’s journeys, and the integrity of their accounts, is retained through the insertion of specific and detailed case studies between individual Chapters. These case studies highlight the need to retain a consciousness of the individuality and specificity of each woman’s history.
The background and context to the research is detailed in Chapters two to four. Chapter two gives an overview of the international context, highlighting the situation of women globally and current international law as it affects women asylum seekers. This Chapter also describes the ethos of the research project and contextualises the way in which the research was approached and carried out. Chapter three provides additional legal and contextual information, but focuses on the UK context, providing an overview of UK laws governing domestic violence and asylum and the impact on women victims/survivors. Chapter four gives an overview of Pakistani government and legal structures and the law as it impacts on women seeking protection from violence in Pakistan.
Chapters five to seven report on the research and feedback from participants in Pakistan. Chapter five examines the situation of women in relation to domestic violence in Pakistan, what protection is available to them, how the law is used in domestic violence cases and the response of the police and judicial agencies. Chapter six examines the status of service provision in Pakistan, looking specifically at access to shelters and exploring the nature and limitations of provision. Chapter seven tracks the journey of a woman who is forced to flee violence and explores broader issues which influence the choices available to her, such as class, ethnicity, customary practices and the notion of ‘honour’. The Chapter also describes governmental responses to the issue of domestic violence and response of women’s groups and human rights groups within Pakistan.
Chapters eight to ten report on the research which was carried out in the UK context. Chapter eight explores perceptions of Pakistan from the UK context and examines the UK based respondents understanding of domestic violence as it affects Pakistani women and the extent and limitations to women’s choices in fleeing violence in that context. Chapter nine echoes the content of Chapter six in that it examines the status of service provision for Pakistani women fleeing domestic violence in the UK, but it also goes further in exploring the issues that are perceived to have an impact on the assessment and outcome of asylum cases in the UK. Chapter ten describes the impact of the asylum system of women and children and the effects of the journey on their physical and mental health.
In Chapter eleven we summarise the information and develop some of the themes which have been highlighted in the research.