Doing narrative feminist research: Intersections and challenges

Four project examples

The four projects we use to illustrate intersectional narrative feminist research in action were all conducted through universities, received ethical clearance, and involved the de-identification of participants, including the use of pseudonyms. The first project example is the In the Name of Love, Women’s Narratives of Love and Abuse which first started in the late 90s and provided a fine-grained, line-by-line analysis of stories about adult sexual love relationships produced by 10 Australian women living in Melbourne who had been abused as children and had spent time away from their birth families. Individual narrative feminist interviews were conducted with another 74 women in Canada in 2003–2004 about their ideas and experiences of sexuality, romance, and intimate love relations (Fraser, 2005, 2008). A central finding was that while love and abuse should not coexist in women’s lives, they often do. Yet, to appreciate the diversity among women, the intersections of gender, class, sexuality, and age came into view.

Between 2010 and 2013 The Helping Alliances with Drug Treatment Clients project involved 20 individual interviews with 5 treating practitioners and 15 clients, of whom 10 were women and 5 men. Interviews were conducted at two sites, in the clinics located in impoverished, working class suburbs, where the public welfare clients received pharmacotherapy (Fraser and Taylor, 2016). Significantly, all 15 clients had chronic exposure to poverty, social disadvantage, and had been diagnosed as comorbid, which in Australia is predominantly used to describe the co-existence of chronic mental health and substance-use problems (Degenhardt and Hall, 2003). From the clients we interviewed—many of whom were heavily medicated with Methadone and/or anti-psychotic and anti-anxiety pharmaceuticals at the time—we heard how helping alliances with treating practitioners could give them hope that they would not always be seen by authorities as “hopeless,” “anti-social,” or “personality disordered,” “junkies;” as people inherently weak and automatically suspect. Helping alliances refer to the nurturing, insight-oriented and collaborative bonds sometimes developed between health/welfare therapists/practitioners and their clients. Analyzing the personal and political aspects of the stories they told about abuse, impoverishment, trauma and stigma was part of the process (Fraser and Taylor, 2016).

The third project, In Good Company (2013–2014), used three focus groups (audio-recorded and transcribed) to narratively examine the roles companion animals play in women’s lives. Thirty-two women living in Australia recounted how they were comforted through their bonds with animals. Through an interest in the stories women told about their dogs, cats, chickens, and lizards, we noticed not only the animals’ therapeutic input into the lives of women (particularly those suffering anxiety and depression) but also their everyday playfulness and fun (Fraser and Taylor, in press). The project explored women’s love relationships beyond the conventional parameters of human relationships, heterosexual partnership, and motherhood. Women reported loving their companion animals differently but as much as they did their partners and children. We noticed the energy the women derived from simply discussing their “pets” (showing photos, telling stories) and discussed how this energy may be a potentially important source of inspiration that social workers may wish to harness in their day-to-day work with women, particularly women classified as “hard to reach” (Duvnjak and Fraser, 2014).

The fourth project used for the purposes of illustration is called Distress in Childbirth, A Social Work Perspective and is currently in progress. It uses individual narrative interviews with a cross section of Canadian women to explore their birthing experiences. This project is designed to use birthing mothers’ stories to map the potential distress, not just the joy and hope the women may experience in childbirth, unsettling the stereotypical assumption that childbirth is universally experienced as a happy event. Using an intersectional approach, this study considers birthing distress not in isolated individual terms but in relation to the women’s ethnicity, age, class, and ability. Attention is being given to the problematic dominant view of distress in childbirth, which tends to pathologize those who express non-idealized birthing accounts. Because relatively little is known about women’s distressing experiences of birthing, especially from a social work perspective, this project is purposively sampling women who want to talk about distress in childbirth. Privilege and oppression experienced in relation to gender, Indigeneity, sexuality, age, and ability are key parts of this project. Projects such as the Distress in Childbirth recognize that without contributions from women living in chronic hardship our knowledge of childbirth is likely to be severely limited.

Narrative feminist research

We use narrative feminist research because of its potential to challenge androcentrism, which recognizes the historical tendency to marginalize if not ignore women’s experiences and accounts of experiences (Anderson et al., 2004; Hesse-Biber, 2008). Yet, this does not have to mean that all projects have to focus exclusively on women. For instance in the project, In Good Company, gender was not specified as a criterion for participation (Fraser and Taylor, in press). It became a women’s project after only women took up the invitation to participate in the focus groups. In the project Distress in Childbirth (MacDougall, in process), the gender focus was already set by the nature of the topic, that is, those currently able to give birth mostly identify as women. And the analytical focus for the In the Name of Love (Fraser, 2008) was pre-set by exploring women’s experiences of love and abuse.

Participants in narrative and/or feminist qualitative studies are often encouraged to be active agents, exploring subjective meanings and experiences that have been ignored or silenced in previous research, allowing for new insights and awareness into social processes (Holstein and Gubrium, 2011; Riessman, 2013). All four projects allowed participants to express ideas about love, abuse, companion animals, birthing, and drug treatment in their own ways, even if this meant longer interviews, contradictory stories told about the same experience and/or stories told that did not at first glance appear to be relevant. Participants’ questions such as, “You know what I mean?” may be treated as rhetorical or a sign of something difficult to articulate, sometimes due to the inability of language to reflect subtleties or nuances of experiences (DeVault, 1990).

Similar to other narrative projects, narrative feminist research projects may be concerned with a single event, an experience, or an entire life, and within the project the analysis may focus on plots, characters, roles, themes, structures, props, functions and linguistic turns, and the interplay between the story and the storytelling (Chase, 2005; Fraser, 2004; Holstein and Gubrium, 2011; Riessman, 2013). While some will focus more on the plots, characters, settings and themes, others will concentrate more on the structure, function, and form of language used in the storytelling process (Holstein and Gubrium, 2011; Riessman, 2008).

Attention may be paid to how the stories are told (Andrews, 2013) as well as the content of talk. For example, narratives may be linear (with a beginning, middle, and resolved ending), or circular (ending back where it began without a sense of resolution), and they may be thematically organized rather than temporarily organized (Fraser, 2004, 2008; Riessman, 2013). Stories may also seem to be rehearsed and familiar, or seem unsure and novel, appearing to develop in an uncertain way through the telling (Clandinin and Connelly, 2000; Fraser, 2004, 2008). Because the Distress in Childbirth project steps into so many discourses about how women are meant to experience birthing, participants are being given the space and freedom to articulate non-idealized experiences without feeling shame, blame, or inadequacy.

Narrative researchers often embark on multiple, iterative, and overlapping phases (Fraser, 2004, 2008). A common dilemma can be figuring out where a story begins and ends. Some stories do not have clear boundaries, with obvious beginnings and endings (Blix et al., 2013; Riessman, 2013). Care needs to be taken not to, “prune out moment of excess, ambiguity and multivocality within people’s stories” (Chadwick 2014: 49; see also Riessman, 2013). For the Helping Alliances project, this was particularly important as many participants told dramatic stories of abuse, assaults, imprisonment, and suicide attempts. The palpable despair that so many participants faced—men and women—was important to appreciate as well as their expressions of delight and respect for those they considered allies (see Fraser and Taylor, 2016).

Doing narrative feminist research means taking care to link the personal with political and understanding the effects of social problems in ways that do not hyper-individualize, denigrate, and/or pathologize the people who experience them (Fraser, 2004, 2008). Participants often make these links themselves, links that researchers need to pick up on. For instance, one of the practitioners in the Helping Alliance project said, Robyn: What it [helping alliance] comes down to is that you can feel someone else’s misfortune—their systemic disadvantage … You’ve got some empathy [for clients], ‘This wasn’t a life you chose”…

Narrative feminist researchers view storytelling as a process of co-construction. In this co-construction, stories are woven within the interaction of telling, listening, and conversation, rather than representing a single truth through the recounting of a specific time or event (Clandinin, 2013; Clandinin and Connelly, 2000; Davis, 2002/2014; Riessman, 2008). Attention is given to how researchers’ stories intersect with those told by participants (Kimpson, 2005; Padgett, 2008). This means accepting that many plausible interpretations can be made; that acts of interpretation need to be judged against criteria for trustworthiness rather than whether it conforms to a single fact or correct answer (Fraser, 2004). Emotion and curiosity work together with researchers paying attention to gaps, silences, and anomalies, as well as what is said (Fraser, 2004).

Narrative feminist interviewing and focus group facilitation require researchers to be aware of their role in the co-construction of the narratives that unfold (Huisman, 2008; Hydén, 2013). How we ask questions, listen for responses, and interact with participants is of central importance (Polkinghorne, 2007). For instance, after participants described their preferred practitioner styles and approaches, we (the researchers) took months to find a term that adequately described what we believe they were getting at. By referring to helping alliances, a term that they had not used, we helped to co-construct their stories.

Irrespective of social status, narrators adjudicate (consciously and otherwise), which stories they will tell, how they will tell it, and in the process, what they will stress or leave out. Our fieldwork has taught us that these decisions are made so quickly and routinely, and with high degrees of sensitivity, that they are not immediately apparent to tellers or listeners, nor easy to discuss. For instance, it took courage and trust for the In the Name of Love participants to disclose to an unknown researcher incidents of child abuse, sexual assault, and domestic violence. During the interviews most participants must have assessed their feelings of safety and made decisions in the moment about what they would share.

Narrative and/or feminist researchers often analyze how language can be used to normalize (or resist) gender stereotypes and express (dis)approval. We may examine clichés, adages, and conventional turns of phrase to learn how people interact with others and different aspects of themselves, and we may reflect on what was not said—what was left out, stumbled over, or was difficult to articulate (Alvesson and Sköldberg, 2009: 29; Best, 2003; Brown, 2012; Hydén, 2013). For instance, “yeah—nah” is a peculiarly Australian expression reflective of a wish to disagree without confrontation. In all three Australian projects participants often used “yeah-nah” to gently express disagreement with researchers.

Hesitations and confirming questions can act as a sign that something important is being said and/or that participants are trying to develop a shared understanding (Best, 2003). For instance, participants from the In Good Company project collectively discussed how easily they could be dismissed as “overly-emotional women” because of their love of companion animals. And they did so while appearing to interrupt each other, often talking at once or finishing each other’s sentences, even though most were not known to each other prior to the group. Rather than read these utterances as impolite, we could see the collaborative stories being built, reflecting shared knowledge and experience of having an important form of love denigrated or dismissed by others (see Fraser and Taylor, in press).

Narratives with similar plots, events, and/or characters may be clustered together for analysis. From these clusters, a shortlist of narratives may help deepen the analysis. The ways that power operates in participants’ lives may show up in the plots of the stories or the roles characters are expected to play. Participants’ stories may show agreement with and use of dominant discourses and/or resistance to and rejection of dominant discourses (Chadwick, 2014; Frank, 2008; Fraser, 2004, 2008; Riessman, 2013; Shuman, 2005). Care is taken to understand the effects of social problems in ways that do not hyper-individualize, denigrate, and/or pathologize the people who experience them.

Narrative feminist researchers pay attention to what happened in the stories but also how the participants felt about it; the emotions they experienced at a particular time, and those elicited in the retelling (Anderson et al., 2004). Emotions are important sources of embodied knowledge pointing to areas of information and possible future conversation and analysis (Anderson, et al., 2004; Bold, 2012; Davis, 2002/2014; Harrison et al, 2001; Josselson, 2007). For instance, during the In the Name of Love interviews participants often laughed, sometimes humorously, other times cynically or with self-deprecation, at what they had been prepared to do for love. Some women cried, other women got angry, sad or a (different) mix of emotions. Deciphering what these emotions meant to the narrator (not just the researcher) was part of the challenge. We expect a similar pattern of mixed emotions will be expressed during the Distress in Childbirth project.

Silences, breaks, laughter, humor, pauses, false starts, and uncertainties in the narratives may represent the emergence of mulitvocality and counter-narratives (Chadwick, 2014; Clandinin, 2013; McKenzie-Mohr and Lafrance, 2014; Riessman and Quinney, 2005). Revisiting transcriptions of stories several times might be needed to see them from different vantage points. In the Love and Abuse project, this was needed to wrestle with the complexities associated with love and abuse co-existing in intimate sexual relationships—a finding that at the time had no precedence in either social work or women’s services.

Responding to questions about credibility

For many narrative feminist researchers political, philosophical, and pragmatic considerations converge in complicated ways when thinking about data as evidence. Beyond neoliberalism or traditional notions of objective science, we might reflect on the meanings we can draw from our studies. We might reflect on the number of participants involved and the insights we might draw from their testimonies. Why are some stories taken up, and in what ways? What makes a testimony convincing and relevant? Which stories resonate with others, how and with what effects? Using our own criteria of honesty, transparency and accountability we can assess whether they are trustworthy (also see Fraser and Taylor, 2016).

The goal of narrative feminist research is not to find universally generalisable themes and understandings of experience but to offer insights, glimpses into others’ worlds and ways of seeing the world. We see stories as the medium through which individual participants provide insight and understanding about their experiences (Bold, 2012; Chadwick, 2014; Clandinin and Caine, 2013; Clandinin and Connelly, 2000; Riessman, 2002). Among narratives, patterns often surface that are relevant to others and deserve further exploration (Clandinin and Connelly, 2002, Mishler, 1995; Polkinghorne, 2005).

All social work research, especially qualitative forms that are used to promote social justice and sexual equality, face several challenges to its credibility. While stories can be used to poignantly illustrate the material, psychological, and social realities of privilege and oppression (Fraser, 2004, 2008; McCance et al., 2001), the influence of neoliberalism can mean that narrow definitions of evidence are adopted with quantitative findings preferred, particularly statistics. Statistics can be more easily represented as objective than [mere] “personal anecdote,” an expression often used to subordinate women’s views. Through this lens, narrative feminist research may be devalued, if not dismissed, as too emotional and partisan. From our perspective, such a position is myopic, falsely dichotomizing quantitative research from qualitative research, both of which can work together to build knowledge.

We take the position that evidence-based social work needs to be defined broadly to include stories and other forms of personal testimonies provided through the co-construction process outlined above. In our experience with the methodology, which spans more than a decade, narrative feminist research has the potential to generate rich data to which social workers across fields and modes of practice are likely to relate (see Fraser, 2010). Here, we are referring to data that has depth, candor, insight, and potential relevance, or resonance for others. Sometimes this involves working with peoples’ experiences, hopes, and suffering not captured or adequately captured by existing knowledge or theories (Fraser, 2003, 2008, 2010). When approached from an intersectional perspective, and using a systematic and articulated approach to collecting and analyzing stories, we can show how and why specified groups may face much higher risks of some social problems compared to others, not just in large numbers but in specifically gendered ways. Domestic violence is a good example.

For instance, in the project, In the Name of Love, women’s hopes for love were often shown to merge with the cultural injunction that they should love even if it meant repeatedly getting hurt. Many women told stories about feeling compelled to prioritize the love of partners and children over themselves and their needs. From an analysis of the stories told by women who reported abuse in their love relationships, it was evident that the injunction that women should love increased their vulnerability to intimate partner abuse (see Fraser, 2003, 2008). The project showed how love and abuse could coexist, culturally, structurally, interpersonally and intrapsychically, findings that were not anticipated. The project produced evidence that has contributed to a more expansive understanding of domestic and family violence influencing the conceptual and practical orientation of many staff training sessions, and some organizational policies operating in counseling centers and shelters, in Australia, Canada, and England (see Fraser, 2003, 2008, 2010). For example, many domestic violence workers, counselors, and family therapists now talk with women about their hopes for love, not just experiences of abuse, with some women’s services even producing pamphlets called “In the Name of Love.”

As feminist researchers, we remain mindful that gendered conventions can influence the norms for the construction of evidence. We often question who gets to speak, on which questions, for how long, and with what kinds of responses. To paraphrase Fine (2011), we consider whose evidence counts, whose evidence is collected, and how it is interpreted. Attention is given to these possibilities and not just through the choice of focus (for instance on women’s stories) and subsequent recruitment strategies but also through the ways “the evidence” is presented, whether it is in writing, in person, over the airwaves or online.

From an Indigenous perspective, reciprocity in research—the act of giving back to the participants and communities involved—justifies it taking place (NWAC, 2013; Smith, 1999). This includes sharing the knowledge gained through the research in meaningful ways (Harrison et al., 2001; Huisman, 2008; NWAC, 2013) and may involve providing access to the written academic work, as well as creating and providing other means for sharing the knowledge gained, as requested by the participants and communities with which researchers are involved.

Social work’s commitment to social change also requires researchers to share the information from the research project within the broader social work community, and with agencies, organizations, and policy makers in positions to create material changes (CASW 2005a, 2005b), which may be accomplished by publishing the findings and presenting broadly in the academy and through policy and practice networks. Researchers may also decide in collaboration with participants to write non-academic accounts of the work, such as a slide show or other visual or spoken presentations to share with those who are interested (NWAC, 2013). A slide show was used to represent main findings of the Helping Alliances project, back at the clinics where the interviews took place. Mainstream and alternative media may be useful channels of data dissemination. For instance, local news media produced a story about dogs being (some) women’s best friends (see Broadstock, 2015). Social media discussions followed and when another related project started up, some who had read this story stepped forward to get involved. Social media websites, blogs and other electronic forms of communication can be equally important methods of publicizing findings and stimulating discussion about how findings might be used as evidence for the purposes of social change.