Researching domestic violence and abuse
LISTENING TO LESS-HEARD
Listening to less-heard voices: methodological approaches, considerations and challenges when researching domestic violence and abuse with vulnerable and marginalised women
Sioban O’Brien Green and Sarah Morton
Who are vulnerable and marginalised women, and how do research ethics committees define them? What criteria are used to craft this definition by academics, researchers and research funders, and how does it impact on subsequent research design? Potentially, is a victim/survivor of domestic violence hypothetically always contained within the parameters of the definition of ‘vulnerable’? How does this categorisation within research contribute to increasing safety and supports for women, or potentially — conversely — put them a greater risk, stigmatisation, exclusion and harm? Most importantly, how do women participating in research self-define themselves and convey this to researchers? This chapter aims to address these formidable questions and draw on the experiences and knowledge ot the authors, who have undertaken research that has focussed on listening to women at the periphery of social research due to numerous factors and circumstances, including women experiencing domestic violence.
Rationale and stimuli for undertaking inclusive research on domestic violence
There is an increasing international mandate for conducting research and collating data on domestic violence. Yet any research must be inclusive ot marginalised and vulnerable populations in order to ensure robust and pertinent resultant research data, prevalence rates and statistics, findings and recommendations. Women living in remote, rural settings, women with disabilities, women living in poverty and migrant women all need to be included in research on violence against women, to create data and evidence to effectively combat the issue. By paying attention to diversity in study sampling and recruitment, and by acknowledging the great risks and vulnerabilities that some women experience, the voices of women traditionally excluded from, or on the margins of, social research can be heard and contribute meaningfully to research-focussed and informed policy change and service responses (Beilis,
Hughes, Perkins, & Bennett, 2012; Crenshaw, 1991; Nixon & Humphreys, 2010; Sokoloft & Dupont, 2005).
There is a growing focus on public and patient involvement in research design and process and requests tor more engaged research practice, to better address complex social issues and involve the communities and cohorts most impacted by them (Campus Engage, 2016). All of these stimulants in relation to research require a collaborative, inclusive and ethical approach to research design and processes that are cognisant of the needs and realties ot those participating in the research. Research should seek to measure, understand, and then amend and ameliorate the situations and conditions that allow tor harmful complex social issues and local and global societal challenges, such as domestic violence and abuse, especially regarding vulnerable populations.
The contribution of intersectionality and intersecting circumstances and realities in women’s lives
The contribution that intersectionality makes to a traditional feminist framework on researching violence against women is to explore the role of culture, expand definitions of abuse, look at the structural causes of partner violence, and shift from disempowering presentations of vulnerable and marginalised female victims to resourceful, active survivors ot violence. The social contexts in which women live are formed and maintained by intersecting systems of power and ot oppression, which modify and interact with each other to frame safety from, and responses to, domestic violence, and what this implies tor individual women, their communities and contexts (Sokoloft & Dupont, 2005). Intersectional research is distinguished by the inclusion of less-heard perspectives and making the lives and experiences of marginalised groups visible. Intersectionality is also a theory that examines how individual characteristics and hierarchies of power and oppression interact, construct and reinforce each other. Fundamentally, it is located in praxis and commitment to social justice (Withaeckx, 2017). McKibbin et al. present intersectionality as a feminist discourse that acknowledges diversity and explains how different groups ot women live with, and experience, disadvantage (2015). This form of discourse is helpful to enabling the voices of marginalised women to be heard, and allows for researchers and policymakers to conceptualise women’s diverse range of experiences and develop potential future responses to them.
Addressing intersectionality, marginalisation and vulnerability within research methodology
As with any research field, there are a range of research methodologies relevant to aspects ot domestic violence and women’s experiences, and how the inclusion ot vulnerability and marginalisation needs to be considered throughout. The core assumptions of intersectionality methodology, according to Hankivsky et al. are to recognise that all members of a social group (e.g. the LGBTQI community, migrants, Travellers, ethnic-minority women) do not necessarily have the same experiences; and, therefore, humans cannot be separated into elements, or an essentialist approach taken (2014). It is proposed that researchers undertaking intersectional studies be committed to social change and social justice, and be engaging with a range ot stakeholders (ibid.). This is required both to do the research and to inform and leverage subsequent action and change. Resultant research findings must empower the individuals who participated in the research, and not reproduce or reinforce the inequalities with which they live. The researcher(s) must also reflect on and acknowledge their privileged position as they undertake research with marginalised groups (Withaeckx, 2017).
Quantitative research considerations
Quantitative research design, as a primary data collection and analysis method in relation to domestic violence and abuse, is most often utilised for surveys, to determine the prevalence of the phenomena. Prevalence aims to measure how frequently domestic violence occurs in a specific population, such as women aged between 18 and 65, during a specific time period, such as lifetime/ever, last year, last month, last week, etc. (European Institute for Gender Equality, 2017). Ensuring that surveys include demographic data collection and sufficient options for respondents to self-describe their features and characteristics in order to capture the breadth of a survey sample is vital. This is in order to ascertain it it is reflective of a population and sufficiently captures the diversity contained within populations, especially in relation to migration patterns, which can fluctuate and change at a frequency greater than census or other data sources can capture. In some cases, dynamic survey sampling and the use of purposive booster samples will ensure that sufficient variance is reached, to be representative ot population diversity (Leye et ah, 2017).
Selt-completion surveys or online surveys may not be accessible to women with certain disabilities and/or literacy challenges, and therefore need to be optimised to include these groups ot participants. In all survey approaches, pilot testing of the data collection tools with varied groups representing potential participants is recommended in advance of field research commencing.
Secondary analysis of existing data sets and administrative data collected from the relevant domestic and sexual violence services, police, courts, coroners, etc. may also be important sources of quantitative data, but they often lack sufficient disaggregation in relation to ethnicity, religion, disability, sexual orientation, etc (Goodey, 2017). As a result, they are not comprehensive enough to reflect all of the population. Caution must be also used in relation to categorising ethnicities, nationalities, or other identifying features of a study sample that exist in such low numbers to make identification within a community or population highly likely.
Regular survey data will be required at a country level to monitor progress in relation to the United Nations Sustainable Development Goals and the Council of Europe Convention on preventing and combating violence against women and domestic violence (United Nations, 2019; Walby, 2016). As a result, it is imperative that the data collection categories and sampling frames are as robust as is feasible. This may require utilisation of approaches such as snowball or time-location sampling, to ensure adequate reach into marginalised population groups (Johnson & Chang, 2014). Finally, the data, prevalence and statistics on domestic violence that emerge from quantitative research methods are often the catalyst tor further research or can be used to provide a rationale for research funding and studies that specifically include vulnerable populations or groups.
Qualitative research considerations
Qualitative research approaches are often considered the ideal route to gather data on domestic violence, as they are concerned with meaning, context and lived experiences, and potentially offer private settings to collect data, such as one-to-one interviews. How participants’ experiences are recounted, interpreted and understood by the research participants themselves, together with elements of reflexivity and subjectivity on the part ot the researcher, are essential to qualitative research (Guest, Namey, & Mitchell, 2013).
Some literature suggests that participation in qualitative research studies on domestic violence enables victims to use their experiences to help others, as they may find the interview a cathartic and meaningful experience, and feel valued and listened to as they recount their own personal story, reflections and opinions to a researcher (Downes, Kelly, & West- marland, 2014). Focus groups can also create an environment for peer support, as well as generate dynamic and deep data collection through the group interactions and stimuli (O’Brien Green, 2017). However, given the importance of connection and rapport between interviewer and interviewee in qualitative research methods, it is vital to grapple with the potential tor coercion, pressure and power dynamics to be explored and addressed within qualitative research design, especially concerning vulnerable groups (World Health Organization, 2016).
Action research offers the potential to understand professional judgement, lived experiences, and multiple ways of knowing when seeking to develop effective practice (Gaventa & Cornwall, 2008; Reason & Bradbury, 2012). It is particularly applicable to health and social-care settings, where there is a desire to move beyond measuring outcomes and explore processes of change and practice development based on collaboration, as well as addressing power imbalances (Donnelly & Morton, 2019). Numerous examples exist of the use of action research with regard to domestic violence and marginalisation, across different cultural settings (Sullivan, Bhuyan, Senturia, Shiu-Thornton, & Ciske, 2005).
Mixed-method research approaches
For some research studies on domestic violence and abuse, the most effective approach is a combination of methods, to ensure that the study is as comprehensive and thorough as possible, and to compensate tor when data collection may not be feasible via certain routes to reach all in a sample. In these situations, a mixed-method approach, combining aspects of qualitative and quantitative research, may be most appropriate (Lewis, 2014). For some studies, this may imply in-depth interviews, in addition to a survey questionnaire on the same topic, but to differing study cohorts (Women’s Health Council, 2009). Mixed methods can also be used to verify, explain or seek corroboration of initial quantitative study findings, and therefore act as a form of data triangulation (European Institute for Gender Equality, 2015). Mixed methods may also be utilised when the sample size or population to be researched is potentially very small and hard to reach, so inputs from those working with the population maybe sought (such as domestic violence service staff), in addition to data collection from the core sample, and potentially additional data collection via case file analysis or an online survey (SAFE Ireland, 2016).
While a mixed-method approach can provide a pragmatic solution to researching complex issues with hard-to-reach populations, the study design and sequencing must be rigorous enough to ensure valid research results, and not risk the time and effort that researchers and participants have devoted to the study be dismissed (Lewis, 2014). A mixed-method approach utilising multiple data sources and respondents may be the most appropriate in cases ot intimate- partner femicide, by which the fundamental research participant has been silenced through murder. This approach is in order to enhance risk identification and risk reduction measures via the analysis of data sources available to ultimately reduce the incidence of femicide (Schrottle & Meshkova, 2018).
Research ethics, ethical research design and safety considerations
The absence of research on domestic violence that fully considers the realities that women live with and in, and the potential of research on these intersecting topics to inform professional guidelines, service delivery, screening and legislation, creates an ethical dilemma in relation to policymaking in a research vacuum without adequate, substantial and recent research to guide and inform this process. However, the ethical considerations in relation to researching domestic violence, in the context ot additional vulnerabilities, such as pregnancy, migrant status, drug use, disability, etc., require considerable thought and ongoing reflection in relation to the safety, on multiple levels and contexts, ot the research participants and the researcher. This is because issues can arise before, during and after data collection, and there exists the potential ot retraumatisation and re-victimisation of the interviewee and vicarious trauma tor the interviewer via both the data collection processes and research write-up. There also exists the possibility ot the safety and risk contexts of the research participants changing and deteriorating during the study. Effectively preparing field researchers or interviewers on how to respond to and process what they are hearing in interviews is also necessary, in addition to enacting participant safety and onward-referral protocols, and legal requirements in relation to child protection, if required (Fraga, 2016).
How women are engaged with and accessed to participate in research, are heard and listened to by researchers, have their histories documented, and where the research and reports are published in the context of potentially life-threatening past and future experiences places much responsibility on the researcher in relation to safety, confidentiality, re-victimisation and vicarious trauma (O’Brien Green, 2018). Additionally, the dimensions of pregnancy or a recent birth, current drug use, mental health issues, etc. by participants in a study adds another layer ot potential safety concerns and heightened vulnerability. There exists the possibility of re-victim- isation ot study participants in research on domestic abuse, in particular if women are asked to repeatedly recount incidences of forms ot violence that they have experienced.
Transcription and/or translation utilisation in research studies needs to take into account potential confidentially breaches, especially if the language or dialect spoken by research participants is limited to a very small population subgroup and is particularly distinguishable or identifiable. The European Commission states that the vulnerabilities of research participants can be exacerbated if the confidential information that they share with researchers can, in any way, be linked to them (2018). This is particularity the situation tor research participants who are sex workers, refugees, disabled women, LGBTQI persons and victims ot domestic violence.
The inclusion of women of childbearing potential in research studies and those with other perceived vulnerabilities, such as the use ot substances/drugs or engagement in sex-working, may intensify the sensitivity of research ethics committees to potential harms, and, as a result, they may opt tor additional safety measures or the exclusion ot this population cohort in research studies. Within these committees there may also be a limited understanding of how domestic violence services are structured and operate, and the staffing arrangements and demanding, resource-limited environments in which many of these services function. Yet, overly restrictive or prescriptive study inclusion and exclusion criteria within a study designed to incorporate the most current and comprehensive considerations in relation to ethics and participant safety can result in a situation where the study participation criteria are such that recruitment is exceptionally limited and/or challenging, and/or interviewees are non-existent.
Downes et al. state that the challenges in gaining ethical approval to undertake research on domestic violence may lead to a diminishment of research occurring and policymaking taking place without evidence (2014). This situation may become even more challenging, as researchperforming organisations, such as universities, become more risk aware and averse, and less open to complex research studies with vulnerable populations (Miller & Bell, 2002). However, the challenges and risks associated with researching violence against women, while great, must also be compared with the absence of survivor voices to correctly and comprehensively inform policy, service provision and legislation. Although pregnant and migrant women, women who have experienced domestic violence, women who are victims of crime, and those who have experienced trauma are generally classified as vulnerable populations in research settings, given the prevalence of these experiences in the adult female global population, they cannot be considered rare or infrequent.
In relation to research with women who have experienced domestic violence, both participant and researcher safety gains a heightened importance and emphasis (Ellsberg & Heise, 2005; Fraga, 2016; Sullivan & Cain, 2004). Any study design must be cognisant ot safety issues for interviewees in the following: participant recruitment processes, risk assessment of potential interviewees, researcher contact with potential interviewees prior to and after interviews, location of interviews, and referral routes and supports identified in case of risks or needs changing in the course of the research study or during the interview. The World Health Organization (WHO) states that research interventions with women who have experienced partner violence while pregnant require special ethical considerations and referral routes for interviewees, given the increased risk of pregnancy loss due to violence (2016).
Planning the location ot study interviews and/or tocus groups is a key feature of the safety considerations in any domestic violence study. Sullivan and Cain recognise that the immediate safety needs of women are usually met when research interviews are conducted in domestic violence refuges, but they also acknowledge that study interviews may take place in other locations that women interviewees themselves deem safe and comfortable (2004). Domestic violence refuges and services may have pre-established professional links with local medical services and police, should these services be required during or after a study interview. Nonetheless, there are potential confidentiality issues that may arise when being interviewed in a refuge where a woman has lived or is currently living. As there is an element of mobility attached to seeking safety and protection from a violent partner, and as a woman’s personal situation of risk may escalate and de-escalate rapidly (with a perpetrator being released from prison, or a barring or protection order expiring, for example), undertaking more than one interview with a woman or longitudinal studies, with multiple interviews planned, may not be feasible (Sullivan & Cain, 2004).
At the end of an interview, a researcher can leave, both physically and psychologically, the topics or issues being investigated, and the place of the interview. The researcher may not have the ongoing safety concerns and issues with which their interviewee must continue to live, and, indeed, options such as travel or where to live may not be available to the research participant for safety and child protection/access reasons (Montoya & Rolandsen Agustin, 2013; Womens Health Council, 2009). Being sensitised to this reality and acknowledging that the options and choices that researchers may have at the end of an interview or the research study are not the same as those of their participants is important, both in relation to study safety design and protocols and in terms of researcher reflexivity.
Study documentation and protocols
Documents related to the research study, such as participant and gatekeeper information, must provide clear, comprehensive and accurate information on the study in a format (and, potentially, a language) that is appropriate to the research cohort. An approach to maximising the readability' of the study documents tor participants should be used, and this may imply proofreading documents and checking legibility by people outside of the research team. A concise research study summary may be useful, to generate interest and support and assist with recruitment from the outset. This summary can be reviewed during the consent-gathering process, to ensure that participants are cognisant of the study’s aims, purpose and implications, and the topics and subjects covered in the interviews or survey. An adequate budget tor translation should be part of the study design. To ensure the participation of all study documents — the summary and survey/interview questions, or, at minimum, consent forms — should be available in the languages and/or formats ot relevance to potential participants. This is to ensure that women are not further marginalised and excluded from research participation.
Study documentation should also address data protection and confidentiality, and how the data storage and analysis will respect the confidentiality of study participants. Study documentation must refer to the protocols and policies for dealing with issues that can arise during research on domestic violence. The protocols need to be devised prior to fieldwork and may require legal, child protection or other relevant inputs. Any protocols must be mindful of national mandatory, legal-reporting and disclosure requirements, especially' in the case of an ongoing risk of child abuse or disclosure related to potential self-harm or suicide. In relation to vulnerable populations, protocols must be even more attuned to risk potential and harm mitigation throughout the entire research process, while also providing clarity on the limits of researcher confidentiality as prescribed by' law.
During data collection, unintended and unexpected issues or findings may emerge, and researchers may need to refer interviewees to a wider range of support services than originally anticipated, in particular, in cases ot human trafficking, sexual and labour exploitation, and other criminal activity (European Commission, 2018). Intended or planned forced marriage or female genital mutilation may also emerge during research with particular groups of women, and although they may not necessarily be the primary focus of the study', protocols will need to be in place to respond to these situations (European Institute for Gender Equality, 2015). These unanticipated findings may have harmful current and future implications bey'ond the immediate study' cohort, and researchers have ethical and legal responsibilities to ensure that non-maleficent aspects of their research design are upheld.
Safety and drug-related and/or organised crime
Women may be directly and indirectly' connected with networks of criminal activity, drug- related crime and intimidation which may have implications for the safety of the woman and the researcher, particularly in small communities (Raghavan, Mennerich, Sexton, & James, 2006). For instance, Sandberg and Copes found that researchers needed to take specific safety measures where there was a belief or concern about criminal activity, violence or drug-dealing, including always completing interviews in organisational settings and informing colleagues ot their location and schedule (2013). However, it was noted that having access to such contexts in the course of the research could have both emotional and practical safety implications, and that while researchers may become privy' to information and understandings that are not report- able to police or other relevant agencies, being immersed in such settings was ethically and morally challenging (ibid.). This immersion in, and exposure to, such activity could affect the researcher’s world view and personal and community relationships (Morton & Hohman, 2016). It must be noted that some research with vulnerable populations simply may not be possible due to profound safety concerns, both for the researcher and/or the participant.
While compensation can be seen as incentivising potential interviewees to participate by means ot monetary pressure, and, therefore, possibly biased study samples, it can also mitigate against interviewees’ travel costs, childcare expenses and lost earnings when giving their time to participate in a study (O’Brien Green, 2018). Supporting the reality ot expenses incurred in order to participate in research studies should form part of every research study design and budget. Adequate compensation upon completing a study interview that is offered in a manner that women can utilise discreetly and easily is part of responsible research, yet compensation for study participants can be problematic for research ethics committees and cause concerns about stimulating or over-incentivising research participation (O’Brien Green, 2017). If substantially higher study recruitment is observed amongst specific cohorts ot the study population, a reexamination of compensation on offer and recruitment methods may be required, to ensure that undue inducement is not a factor (World Health Organization, 2016).
For some women, a lack ot extended family and strong informal support networks may mean that childcare options are very limited or completely absent, and the only way that women can participate in research is with their children. This can present challenges tor the researcher, as time will need to be allocated for breaks and noise from the child(ren) can be an issue for the transcription ot interviews (O’Brien Green, 2017). Having a child present can also be a challenge to the participant’s safety. Children should not be present in interviews if they are verbal for any research on domestic violence, tor two main reasons: the interview content (both questions and responses) can be upsetting to the child, and there exists a possibility of children being later interrogated (for example, by the perpetrator of violence or other family or community members) as to what their mother has told others. As a result, having verbal children in interviews or focus groups has the potential to escalate risk and compromise participant safety (Zimmerman & Watts, 2003). When and where interviews are scheduled must try to accommodate women’s childcare needs and should form a part of study design. Providing or funding childcare for women to participate in research studies is also (potentially) an enabling factor for study recruitment and a realistic aspect of robust study design.
Documentary evidence of the voluntary, free and informed consent of research study participants is generally a requirement of research ethics committees and research funders. However, in cases of research with vulnerable and marginalised groups, the process and routes to provide a mutually agreed understanding of consent between researcher and participant, to demonstrate that consent was given, may be more challenging than with other research cohorts. This can be due to language differences between the researcher and the interviewee, literacy, disability impacting on reading, hearing, etc., mental health issues, and women being affected by their substance use or medication. Emphasising the right to withdraw consent at any time without any consequences or impact on accessing services (such those services in which the interview or focus group is being held) is very important (World Health Organization, 2016). It is also imperative to demonstrate that avoiding coercion and any misrepresentation of possible benefits of participation in the study has been addressed fully and clearly with interviewees (Miller & Bell, 2002). Additionally, study participants should be fully aware ot whom is commissioning and/or funding the research and why (European Commission, 2018).
Existing evidence indicates that 50% ot women experiencing domestic violence may have used, or are currently using, substances, usually as a response to their experiences ot abuse (Galvani, 2004; George, Boulay, & Galvani, 2011). Beyond this, women may also be prescribed or misusing medication, the side effects of which can impact on daily living (Romans, Cohen, Forte, Du Mont, & Hyman, 2008). The risks to participants may be resolved by simply not including those who are intoxicated, but given the prevalence of substance use for women experiencing domestic violence, important narratives and data may be excluded. In addition, the opportunity to understand drug use patterns and to inform practice and policy will also be foregone (Barratt, Norman, & Fry, 2007). Rather than engage with fallible methods of assessing intoxication, Aldridge and Charles advocate focussing on the process of ensuring that the information is understood by the research participant during the consent process (2008). This may include, for instance, ensuring that information about the research is understood by the participant and excluding those who are not able to demonstrate such understanding, as well as asking participants directly if they feel that their substance or medication use may be impacting on their ability to provide consent and to fully participate in the data collection process (ibid.). Deciding not to include women who are currently using substances or intoxicated by their use assumes that there will be obvious outward and identifiable indicators of substance or medication use. It also assumes that women using substances are automatically ‘vulnerable’ (Aldridge & Charles, 2008). Given the level of medication use, and sometimes misuse, for women experiencing domestic violence, the effect of medication on the ability of a woman to provide informed consent and to engage in a research process requires special attention (Wuest et ah, 2007).
Finally, confirming that consent is freely given and enabling consent comprehension and affirmation are the responsibilities of the researcher, not the research participant. Enabling women from marginalised and vulnerable contexts to fully consent and participate should be the aim of inclusive and representative research on domestic violence and abuse.
Challenges and overcoming them
Ensuring that, potentially, the most vulnerable and marginalised voices get listened to and included in research in domestic violence is not without challenges. There can be resistance from research ethics committees and research funders, resource and time pressures, and immense logistical and practical challenges to overcome. As community-engaged research and utilisation of approaches that facilitate the inclusion of less-heard voices are receiving greater focus within the field, challenges have been identified with regard to both, especially as they are underpinned by principles of power-sharing and collaboration. For instance, community-engaged research can be complex, requiring the development of processes for participation and the co-creation of knowledge and action, with the aim of transforming how knowledge is produced (Stoecker, 2012). While community organisations, such as domestic violence support agencies, may be ideally placed to facilitate participation and advocate for social change, questions remain as to whether community-engaged research addresses power dynamics or lacks the ability to support societal-level change that will truly address social and health inequities (Belle-Isle, Benoit, & Pauly, 2014; Gustavsen, 2014). There is a growing focus on the meaningful involvement of including seldom-heard voices within research processes, but radical shifts in both research processes and practice are required, to ensure that this is effective (Ni She et ah, 2019). Yet community-engaged research and practitioner partnership approaches to addressing research challenges, while not compromising the confidentiality of participants, can remain complex to negotiate.
Participant (de)selection by gatekeepers
When gatekeepers from domestic violence services are utilised to access a study population and promote participation in a research study, the potential exists for power and control dynamics being exerted and subsequent sample selection bias (O’Brien Green, 2018). There may be efforts to frame how domestic violence supports and services are perceived, and, as a result, only particular clients are informed about the research by the gatekeepers, and then subsequently opt to take part (Vearey, Barter, Hynes, & McGinn, 2017). Recognising that there is potentially a level of study self-selection by participants and then potentially an additional layer of selective participant recruitment by gatekeepers, which ultimately may influence a study’s findings, despite as broad a recruitment effort as possible, is unavoidable. Women perceived as too vulnerable to participate in a research study by gatekeepers may not be offered the possibility of having their voices heard therein (Miller & Bell, 2002). This can further exacerbate the marginalisation of vulnerable women and reduce their access and contributions to research, with an outcome of potentially silencing the voices that most need to be listened to.
The complex realties of women’s lives and experiences of domestic violence are not easy to listen to and process, no matter how much experience a researcher has investigating challenging and complex topics. Various terms have been applied to the potentially traumatic experience of listening to or hearing the violent, abusive or harmful experiences of others (Clemens, 2005; Coles, Astbury, Dartnall, & Limjerwala, 2014; Morton & Hohman, 2016). Vicarious trauma to interviewers/researchers/transcribers can occur during the interview and while revising the data by listening to interview recordings and reviewing interview transcripts and notes repeatedly. Vicarious trauma for the researcher is an often overlooked issue when researching complex and distressing issues with participants who are exceptionally stigmatised and marginalised. Changes in world view, increased sensitisation to the issues of power and control in society, and a re-evaluation of one’s own past and perceptions of violence tend to be common reactions to working with such material (Morton & Hohman, 2016). Building support, debriefing and other processes into the research design, as well as opportunities to discuss study challenges, is crucial to retaining researcher objectivity, completing the study and research study staff retention (Medical Research Council, 2015). Supports for researchers should be in place prior to the data collection commencement, in order to reduce the potential tor vicarious trauma and distress and its subsequent impact on data collection and analysis. These supports need to be built into the study design and researcher supervision processes, including resourcing them from the outset. In addition, debriefing, accessing peer support and having a manageable workload and a supportive work environment can also help reduce the potential for vicarious trauma emergence and impact.
Onward referral of research participants
Given the nature of women’s experiences regarding domestic violence, especially where there are multiple/complex issues or aspects of exclusion or disadvantage, relevant and accessible support services are important. Trauma-informed responses (TIR) are being widely discussed, considered and implemented in different social and community service settings, and specialist services (such as sexual assault services) and exist in most jurisdictions. Yet lack of funding, waiting lists and a ‘silo approach’ to complex issues can persist. In recent research considering women experiencing a range of complex issues, including domestic violence and substance use, the lack of immediacy of service delivery was a key theme, together with women often not meeting basic criteria for access to services, such as stability in their substance use (Morton, MacDonald, & Christophers, 2020). This has implications for any research study, given the importance ot ensuring adequate referral pathways for research participants, should they find that they are emotionally impacted or if re-victimisation occurs by their perpetrator during the research process.
Misuse of research data
Unfortunately, when vulnerable or marginalised groups are identified as part of research cohorts, the possibility exists that the research data may be misconstrued or misused for negative and harmful political or ideological purposes. This potential particularly exists in relation to ethnic- minority, migrant and sex-working study participants, to be misused by racist and anti-migrant campaigns and movements. As a result, the published research findings, as well as the public and media presentation of any research findings, need to be carefully considered and prepared, being aware of ethical concerns and possible ulterior motives for the potential misuse of data (European Commission, 2018; European Institute for Gender Equality, 2013).
Ensuring that women in vulnerable contexts and who may be considered marginalised are core to undertaking research on domestic violence, given the importance of their experiences and their trajectory and reality of living through abuse, is crucial. This is to ensure that all victims/ survivors of domestic violence have equitable opportunities to contribute to the knowledge base, to inform policy, legislation and service development. It is also to ensure that women on the periphery of society, or whose voices are less amplified and heard in research, are centre- staged and acknowledged. Recognising that women can be experiencing and/or have lived through/survived domestic violence, as well as other forms of gender-based violence, in addition to discrimination and specific exploitation due to their vulnerability (disability, mental health, addiction status, pregnancy, etc.) and/or legal status, such as trafficking, etc., is crucial, to reflect the lived realties of women’s lives. Furthermore, women may be participating in research studies related to their circumstances and situations, such as their migration status, pregnancy, substance use, disability, etc. Ensuring that these studies are also cognisant of the potential for the sample to have experienced gender-based violence is important, to guarantee that women receive the supports and referrals they need, but also to ensure visibility' of the intersecting nature of multiple challenges and vulnerabilities with and through which women live.
How research participants are enabled to engage with the research study results, reports and publications needs to be considered, to ensure that a collaborative and appreciative approach to the research design is apparent at the end of the study. Producing and disseminating research study findings in a timely manner and in accessible formats (via open-access publications, research summaries, briefing papers, etc.) is also important, in order to optimally influence the policy and legislative changes that may follow distinctive timetables (such as national/regional parliamentary calendars or European Union directive transposition deadlines). In conclusion, researchers have a vital role to work beyond designations such as vulnerable and marginalised; to reduce power differentials between the academic/researcher and those providing the data for studies; to proactively involve, recognise and compensate all women who agree to take part in research studies while optimising safety, respect and appreciation for those who contribute to the global knowledge base on domestic violence, in order to effectively respond to it, to reduce and, ultimately, end it.
- • By attending to diversity, and acknowledging the great risks and vulnerabilities some women experience, those who are traditionally excluded from social research can be included and integrated into research-informed legal and policy change and service responses.
- • Collaborative, inclusive and ethical approaches to research design are needed, that allow and account for the intersecting systems of power and oppression, vulnerability and exclusion that may exist in women’s lives.
- • Safety for participants, and the researcher, is key throughout the research process, with special attention required to safety during participant recruitment processes, during and after interviews.
- • It is imperative to demonstrate that consent has been provided and that consent is mutually agreed and understood, especially where language issues, disabilities or effects of substance use may be factors.
- • Researchers must work to reduce power differentials between the academic/researcher and those providing the data for studies and proactively involve, recognise and compensate all women who agree to take part in research studies while optimising safety, respect and appreciation for those contributing to the global knowledge base on domestic violence.
Implications for policy, practice and research
- • Ensuring that women in vulnerable contexts and who may be considered marginalised are core to undertaking research on domestic violence, given the importance of their experiences and their trajectories and realities of living through abuse, is crucial. This is to allow all victims/survivors of domestic violence to have equitable opportunities to contribute to the research knowledge base and to inform policy, legislation and service development.
- • It is important to recognise that women can be experiencing and/or have lived through/ survived domestic violence, as well as other forms of gender-based violence, in addition to discrimination and specific exploitation due to their vulnerability (e.g. disability, mental health, addiction, pregnancy, migration/legal status, etc.).
- • Where women are participating in research studies related to their circumstances and situations (e.g. migration status, pregnancy, substance use, disability, etc.) it is important to be cognisant ot the potential for them to have experienced domestic violence and abuse.
- • Ensuring women receive the supports and referrals they need during and after a research study is important, especially where there are complex and intersecting needs and experiences.
- • Research study findings should be produced and disseminated in a timely manner and in accessible formats such as via open-access publications, research summaries, briefing papers. The research findings should also be accessible to the study participants.
Aldridge, J., & Charles, V. (2008). Researching the intoxicated: Informed consent implications for alcohol and drug research. Drug and Alcohol Dependence, 93(3), 191—196.
Barratt, M. J., Norman, J. S., & Fry, C. L. (2007). Positive and negative aspects of participation in illicit drug research: Implications for recruitment and ethical conduct. International Journal of Drug Policy,
Belle-Isle, L., Benoit, C., & Pauly, B. (2014). Addressing health inequities through social inclusion: The role of community organizations. Action Research, 12(2), 177—193.
Beilis, M., A., Hughes, K., Perkins, C., & Bennett, A. (2012). Protecting people promoting health: A public health approach to violence prevention for England. London, UK: Department of Health and National Health Service.
Campus Engage. (2016). Engaged research practice and principles: Society and higher education addressing grand societal challenges together. Dublin: Irish University Association and Irish Research Council.
Clemens, S. (2005). Recognizing vicarious traumatization: A single session group model for trauma workers. Social Work with Groups, 27(2/3), 55—74. doi:10.1300/J009v27n02_05
Coles, J., Astbury, J., Dartnall, E., & Limjerwala, S. (2014). A qualitative exploration of researcher trauma and researchers’ responses to investigating sexual violence. Violence Against Women, 20(1), 95—117.
Crenshaw, K. W. (1991). Mapping the margins: Intersectionality, identity politics, and violence against women of color. Stanford Law Review, 43(6), 1241—1299. doi: 10.2307/1229039
Donnelly, S., & Morton, S. (2019). Creating organisational and practice change through the use of cooperative inquiry groups in healthcare settings. Action Research, 17(4), 451—468. doi: 1476750319855126
Downes, J., Kelly, L., & Westmarland, N. (2014). Ethics in violence and abuse research — A Positive empowerment approach. Sociological Research Online, 19(1), 1—13.
Ellsberg, M., & Heise, L. (2005). Researching violence against women: A practical guide for researchers and activists. Washington, DC: World Health Organization.
European Commission. (2018). Ethics in social science and humanities. Luxembourg: Publications Office of the European Union.
European Institute for Gender Equality. (2013). Female genital mutilation in the European union and Croatia. Vilnius: EIGE.
European Institute for Gender Equality. (2015). Estimation of girls at risk of female genital mutilation in the European Union. Vilnius: EIGE.
European Institute for Gender Equality. (2017). Glossary of definitions of rape, femicide and intimate partner violence. Vilnius: EIGE.
Fraga, S. (2016). Methodological and ethical challenges in violence research. Porto Biomedical Journal, 1(2),
Galvani, S. (2004). Responsible disinhibition: Alcohol, men and violence to women. Addiction Research & Theory, 12(4), 357-371.
Gaventa, J., & Cornwall, A. (2008). Power and knowledge. The SAGE handbook of action research: Participative inquiry and practice (pp. 172—189). London and Thousand Oaks, CA: Sage.
George, S., Boulay, S., & Galvani, S. (2011). Domestic abuse among women who misuse psychoactive substances: An overview for the clinician. Addictive Disorders & Their Treatment, 10(2), 43—49.
Goodey, J. (2017). Violence against women: Placing evidence from a European union-wide survey in a policy context. Journal of Interpersonal Violence, 32(12), 1760—1791.
Guest, G., Namey, E. E., & Mitchell, M. (2013). Collecting qualitative data: A field manual for applied research. London: Sage.
Gustavsen, B. (2014). Social impact and the justification of action research knowledge. Action Research,
Hankivsky, O., Grace, D., Hunting, G., Giesbrecht, M., Fridkin, A., Rudrum, S., & Clark, N. (2014). An intersectionality-based policy analysis framework: Critical reflections on a methodology for advancing equity. International Journal for Equity in Health, /3(1), 119. doi:10.1186/s12939-014-0119-x
Johnson, M., & Chang, S,. (2014). Sampling in quantitative research. In S. Jiro|wong, M. Johnson, & Anthony Welch (Eds.), Research methods in nursing and midwifery: Pathways to evidence-based practice (2nd ed., pp. 165—181). Victoria: Oxford University Press.
Lewis, J. (2014). Mixed methods research. In S. Jirojwong, M. Johnson, & A. Welch (Eds.), Research methods in nursing and midwifery: Pathways to evidence-based practice (2nd ed., pp. 268—286). Victoria: Oxford University Press.
Leye, E., De Schrijver, L., Van Baelen, L., Andro, A., Lesclingand, M., Ortensi, L., & Farina, P. (2017). Estimating FGM prevalence in Europe. Findings of a pilot study. Research report. Ghent: International Centre for Reproductive Health.
McKibbin, G., Duncan, R., Hamilton, 13., Humphreys, C., & Kellett, C. (2015). The intersectional turn in feminist theory: A response to Carbin and Edenheim (2013). European Journal of Women’s Studies,
22(1), 99-103. doi:10.1177/1350506814539445
Medical Research Council. (2015). Guidelines for the prevention and management of vicarious trauma among researchers of sexual and intimate partner violence. Pretoria: Medical Research Council.
Miller, T., & Bell, L. (2002). Consenting to what? Issues of access, gate-keeping and ‘informed’consent. In M. Mauthner, M. Birch, |. Jessop, & T. Miller (Eds.), Ethics in qualitative research (pp. 54—69). London: Sage.
Montoya, C., & Rolandsen Agustin, L. (2013). The othering of domestic violence: The eu and cultural framings of violence against women. Social Politics: International Studies in Gender, State & Society, 20(4), 534-557.
Morton, S., & Hohman, M. (2016). “That’s the weight of knowing”: Practitioner skills and impact when delivering psychoeducational group work for women who have experienced IPV. Social Work until Groups, 39(4), 277-291.
Morton, S., MacDonald, P, & Christophers, L. (2020). Substance use, homelessness and violence: Policy and practice in regard to women with complex needs. Dublin: Merchants Quay Ireland.
Ni She, E., Morton, S., Lambert, V, Ni Cheallaight, C., Lacey, V, Dunn, E., . . . Kroll, T. (2019). Clarifying the mechanisms and resources that enable the reciprocal involvement of seldom heard groups in health and social care research: A collaborative rapid realist review process. Health Expectations, 22(3), 298-306. doi.org/10.1111 /hex. 12865
Nixon, J., & Humphreys, C. (2010). Marshalling the evidence: Using intersectionality in the domestic violence frame. Social Politics: International Studies in Gender, State & Society, 17(2), 137—158. doi:10.1093/ sp/jxq003
O’Brien Green, S. (2017). Organising focus groups: Process and logistics. In E. Leye & G. Coene (Eds.), Researching female genital mutilation/cutting (pp. 85—90). Brussels: Brussels University Press.
O’Brien Green, S. (2018). “I have a story to tell”: Researching migrant women’s experiences of female genital mutilation and gender-based violence in Ireland and Europe. Social Work and Social Sciences
Review, 19(2>), 134-151.
Raghavan, C., Mennerich, A., Sexton, E., & James, S. E. (2006). Community violence and its direct, indirect, and mediating effects on intimate partner violence. Violence Against Women, 12(12), 1132—1149.
Reason, P, & Bradbury, H. (2012). The SAGE handbook of action research: Participative inquiry and practice (2nd ed.). London and Thousand Oaks, CA: Sage.
Romans, S. E., Cohen, M. M., Forte, T, Du Mont, J., & Hyman, I. (2008). Gender and psychotropic medication use: The role of intimate partner violence. Preventive Medicine, 46(b), 615—621.
SAFE Ireland. (2016). SNaP: Specific needs and protection orders. Justice Sought justice lost. Athlone: SAFE Ireland.
Sandberg, S., & Copes, H. (2013). Speaking with ethnographers: The challenges of researching drug dealers and offenders. Journal of Drug Issues, 43(2), 176—197.
Schrottle, M., & Meshkova, K. (2018). Data collection: Challenges and opportunities. In S. Weil, C. Cor- radi, & M. Naudi (Eds.), Femicide across Europe: Theory, research and prevention (pp. 33—52). Bristol, UK: Policy Press.
Sokoloff, N. J., & Dupont, I. (2005). Domestic violence at the intersections of race, class, and gender: Challenges and contributions to understanding violence against marginalized women in diverse communities. Violence Against Women, 1 /(1), 38-64. doi: 10.1177/1077801204271476
Stoecker, R. (2012). Community-based research and the two forms of social chang e. Journal of Rural Social Sciences, 27(2), 83.
Sullivan, C., & Cam, D. (2004). Ethical and safety considerations when obtaining information from or about battered women for research purposes. Journal of Interpersonal Violence, 19(5), 603—618.
Sullivan, M., Bhuyan, R., Senturia, K., Shiu-Thornton, S., & Ciske, S. (2005). Participatory action research in practice: A case study in addressing domestic violence in nine cultural communities. Journal of Interpersonal Violence, 20(8), 977-995. doi:l 0.1177/0886260505277680
United Nations. (2019). Global indicator framework for the sustainable development goals and targets of the 2030 agenda for sustainable development. New York, NY: United Nations.
Vearey, J., Barter, C., Hynes, P, & McGinn, T. (2017). Research ethics in practice: Lessons from studies exploring interpersonal violence in different contexts. Families, Relationships and Societies, 6(2), 273—289.
Walby, S. (2016). Ensuring data collection and research on violence against women and domestic violence: Article 11 of the Istanbul convention. Strasbourg: Council of Europe.
Withaeckx, S. (2017). Researching harmful cultural practices: Values and limits of an mtersectional perspective. In E. Leye & G. Coene (Eds.), Researching female genital mutilation/cutting: Proceedings of the 2nd international academic seminar of MAP-FGM project (pp. 113—118). Brussels: Vubpress Brussels University Press.
Womens Health Council. (2009). Translating pain into action: A study of gender-based violence and minority ethnic women in Ireland. Full report. Dublin: Womens Health Council.
World Health Organization. (2016). Ethical and safety recommendations for intervention research on violence against women. Building on lessons from the WFIO publication Putting women first: Ethical and safety recommendations for research on domestic violence against women. Geneva: World Health Organization.
Wuest, J., Merritt-Gray, M., Lent, B., Varcoe, C., Connors, A. J., & Ford-Gilboe, M. (2007). Patterns of medication use among women survivors of intimate partner violence. Canadian Journal of Public Health, 98(b), 460-464.
Zimmerman, C., & Watts, C. (2003). WHO ethical and safety recommendations for interviewing trafficked women. Geneva: World Health Organization.