Dissonance,Disagreement,Difference: Challenging Thematic Consensus to Decolonise Grounded Theory

Abstract

Over the past two decades researchers have been exploring new hybrid methodologies to decolonise Indigenous mental health research. Grounded Theory with Community Participatory Action Research (CPAR), often using Indigenist methods, is the most common. Grounded Theory’s claim to rigour is its transparent, organised process of sifting and theme finding, while CPAR respects Indigenous self-determination and Indigenous ways of knowing doing and being, involving Indigenous research participants in all stages of the research from data collection to analysis. On the surface it would seem to be the ideal methodological approach to navigate the cultural divide. However, this article will argue that Grounded Theory’s inherent weakness is in the process of thematic analysis, which uses consensus during the analysis phase to find dominant themes. Drawing on the social systems theory of Niklas Luhmann and the political theory of Chantal Mouffe, this article will argue for a group process of “agonistic pluralism” instead. Searching for shared truths has a tendency to smooth out differences. The article proposes an approach for configuring a research team and conducting team analysis that struggles with, and accounts for, dissonances, disagreements and differences. Furthermore, it argues these differences should be recorded as important findings along with agreed themes. The approach has been developed to explore community perspectives on the relationship between culture and health, and in turn, to develop culturally appropriate mental health therapies for First Nations young people within a Western paediatric hospital.

Background

The traumatic effects of colonisation have had far-reaching impacts on the mental health and wellbeing of First Nations peoples around the world. Australia, the only colonised nation to have been settled by an invader without a Treaty with its First Peoples, has a particularly poor record. Atkinson (2002) described the effects of post-colonisation trauma to be substantial, enduring and multi-generational. Moreover, Deferio (2019) found mental illness, in turn, is an independent risk factor for worse general health status and premature death. Australia’s First Nations youth mental health statistics are appalling: more than twice as many First Nations than non-Indigenous young people report high or very high levels of psychological distress (33% vs. 13%) and almost twice as many report a long term mental health condition (29% vs. 16%). Twice as many are hospitalized for intentional self-harm (5% vs. 2%) and four times as many die from intentional self-harm or suicide (29% vs. 7%) (AIHW, 2018). Overall, Aboriginal and Torres Strait Islander people carry a 2.3 times higher burden of disease than non-Indigenous people yet health expenditure on First Nations people is only three-quarters of that spent on non-Indigenous people per capita (https://humanrights.gov.au/our-work/aboriginal-and-torres-strait-islander-social-justice/publications/close-gap-10-year-review#:∼:text=This.review.assesses.the.most,(health).equality.by.2030.) ApologiesAzzopardi et al., (2018) note that trends of poorer mental health for First Nations peoples are evident across a range of international databases. Disconnection from families and kin, and adverse events are the most reliable risk factors for predicting poor mental health outcomes (Young et al., 2017).

Although there has been evidence of disparities in health and wellbeing since invasion, efforts to address them did not begin in earnest in Australia until fifteen years ago. At the Close the Gap: Indigenous Health Equality Summit in 2008 – a meeting of the Council of Heads of Australian Government (COHAG) and a number of representatives of First Nations peak bodies for health and social justice – participants signed a Statement of Intent that committed to “achieve equality of health status and life expectancy between Aboriginal and Torres Strait Islander peoples and non-Indigenous Australians by 2030” and to ensure “primary health care services and health infrastructure for Aboriginal and Torres Strait Islander peoples are capable of bridging the gap in health standards by 2018” (https://humanrights.gov.au/our-work/close-gap-indigenous-health-equality-summit-statement-intent). The strategies that flowed from this, however, have had minimal impact. The Close the Gap: ten year review (2018) found that despite the stated commitment to involve First Nations communities in developing and implementing the Strategy, there was neither appropriate consultation nor sufficient concrete action to achieve real traction. Consequently, many of the ambitious 2018 targets have not been met (https://humanrights.gov.au/our-work/aboriginal-and-torres-strait-islander-social-justice/publications/close-gap-10-year-review#:∼:text=This.review.assesses.the.most,(health).equality.by.2030.) A subsequent National Strategic Framework for Aboriginal and Torres Strait Islander Peoples’ Mental Health and Social and Emotional Wellbeing 2017–2023, had neither an implementation plan nor clear funding streams. The ten-year review (2018) demanded a ‘reset’.

Most policies draw on large scale epidemiological and clinical nomothetic research to establish need (ABS, 2018-19; ABS 2023) supplemented by qualitative research that includes idiographic case studies of lived experiences to animate group nomothetic statistics with insights into individual’s thoughts, beliefs, attitudes, and feelings (Cawte, 1988; Eley et al., 2007; Hunter, 1991; Petchkovsky & San Roque, 2002). While this kind of research is ostensibly reliable, repeatable and reproducible, there are inherent Western biases in these methodological approaches. As Indigenous scholars, Maggie Walter (Palawa) and Stephanie Russo Carroll (Ahtna-Native Village of Kluti-Kaah) have argued (2019) there are hidden politics at play with the way quantitative data are collected, analysed and interpreted:

Statistics are human artefacts and in colonizing nation states such numbers applied to Indigenous Peoples have a raced reality … It is dominant society questions that are hidden behind the cover of claims of objective methodology. Within this, the Indigene remains the object, caught in a numbered bind, viewed through the straitjacketing lens of deficit

Qualitative methods are not immune to bias either. Paine et al. (2019) have noted that when research is stripped of its relational context it can easily be misinterpreted. Further, Marley, (2020) has cautioned there are ethical problems associated with any universalising practices, such as secondary analyses by other researchers. Knowledge in Indigenous communities is particular to people and place and non-consensual sharing can both violate rights of the participants and lead to false interpretations. First Nations mental health research, therefore, requires research methods that are able to respect cultural difference and navigate the divide between Indigenous and Western epistemologies and ontologies. Policies and programs, meanwhile, need to be informed by such research to ensure they focus on matters that are important to communities and draw on the strengths and resilience of Indigenous peoples to develop mental health interventions that better meet the needs of First Nations people.

Surveying the Research Frameworks for Indigenous Health

Before considering what these frameworks might look like, it is important to understand the broad range of current research in the field of Indigenous mental health, and interrogate their methodological strengths and weaknesses.

Western Nomothetic and Idiographic Research

Western nomothetic and idiographic research are the dominant methodologies enlisted by Western trained scholars. Nomothetic research seeks generalisable truths while idiographic research focuses on data that are unique to individuals. Quantitative research is wholly nomothetic, while qualitative research spans both, depending on the degree to which it adopts a positivist lens (searching for generalisable themes) or an interpretivist approach (White & Cooper, 2022). Key characteristics of good quantitative nomothetic research in the field of mental health include a psychometric definition of a dependant variable under investigation and definition of the independent and control variables. It presumes that groups can be objectified if the measures used have proven adequate psychometric validity and reliability. In this approach many factors that form part of the bio-psycho-social ecosystem (Engel, 1977) in which the groups exist are rendered invisible.

Qualitative researchers, by contrast, work with idiographic data. Unlike quantitative nomothetic methods, in which data is numerical, idiographic data is verbal, textual, visual or sensual, capturing subjective thoughts, beliefs and feelings. Qualitative researchers who adopt a positivist orientation (Glaser & Strauss, 1967) nevertheless adopt a process of analysis that involves finding generalisable thematic patterns shared amongst individuals within a group. Interpretivists, on the other hand, reject the notion of generalisability altogether, believing research to be always context-bound (Charmaz & Thornberg, 2021).

Psychiatry and Psychology were historically committed to idiographic research but are now dominated by ‘evidence-based’ nomothetic research. The Social Work and Nursing fields, on the other hand, are built on qualitative research. Good clinical practice, however, is always informed by both nomothetic research and the individual experiences and challenges facing the patient (Bloch, 1997; Windelbrand 1998 [1904]). Health policies broadly, and mental health specifically, have consequently tended to draw on both too (Australian Government Department of Health 2017, 2021; AIHW & NIAA, 2020; Close the Gap Campaign Steering Committee for Indigenous Health Equity, 2018; Council of Australian Governments, 2008; DPM&C, 2017; Dudgeon et al., 2014; Dwyer et al., 2011; Eckermann et al., 2010; Human Rights and Equal Opportunity Commission, 1997; Swan & Raphael, 1995).

Both nomothetic and idiographic research have their epistemological and ontological foundations in the Enlightenment, a period arising in the 17th century when human intellectualism in the fields of science and philosophy challenged religion, pagan tradition and spirituality as sources of knowledge; the public sphere emerged as a place of debate and commerce; and individualism began to shape cultural mores (Bristow, 2011; Habermas, 1989). The same forces drove global exploration and colonization of Indigenous cultures. While researchers in the social sciences debate the extent to which knowledge is objective and generalisable, versus subjective and particular, researchers from all traditions are concerned about scholarly rigour. Moreover, Western research from all fields continues to believe that knowledge should be disseminated universally. That is, knowledge may be particular to different cohorts depending on how they are situated, geographically, socially and historically, but everyone has a right to know about it. Indeed academic inquiry relies on international peer review, publications and presentations. Promotion is built on impact, and citations are a key line of evidence. Open-source journals are becoming increasingly common to ensure the widest possible readership. The universalist epistemology and ontology of the Enlightenment may have been successfully critiqued by interpretivist qualitative theories, but global neoliberal capitalism is the inescapable epistemic foundation of Western knowledge exchange (Marginson, 2010).

First Nations communities have a deep distrust of these forms of research given their complicity in cultural dispossession, destruction and marginalisation. Throughout the settler-colonial era, Western researchers were voracious in their collection, examination and dissemination of diverse Indigenous knowledges, including secret knowledge about healing practices, sacred knowledges of ceremonies, and details about bush medicines, and artefacts that were part of daily social lives (Eckerman et al., 2010). Atefacts and knowledge were exchanged with no shared understanding of the protocols for re-sharing with others. Much of what was gathered is still accessible today in museum collections, libraries and online. Indigenous knowledge ‘theft’ and dissemination by Western scholars have been significant contributors to trans-generational Indigenous trauma. On the one hand, the concerns are cultural and spiritual: the removal of artefacts that were an intrinsic part of ceremonial practices have disrupted initiation rites, amongst other things, while removal of human skeletal remains for research interfered with burial practices and beliefs about the passage from physical to Spirit worlds. On the other hand, the concerns have been socio-economic: while Western researchers and collecting institutions have profited from the Indigenous knowledges they have published and displayed, First Nations peoples themselves have been impoverished through the dispossession. Over the past two decades, Indigenous scholars have played an important role in illuminating and challenging these practices. As Maori anthropologist, Smith (2021: 1) said:

From the vantage point of the colonized, a position from which I write, and choose to privilege, the term ‘research’ is inextricably linked to European imperialism and colonialism. The word itself, ‘research’, is probably one of the dirtiest words in the Indigenous world’s vocabulary. When mentioned in many Indigenous contexts, it stirs up silence, it conjures up bad memories, it raises a smile that is knowing and distrustful. It is so powerful that Indigenous people even write poetry about research.

To make a difference within communities, a number of researchers have felt it important to develop new methodologies for research that are not regarded with suspicion or ignored by the communities they intend to serve.

Indigenist Frameworks and Method

Indigenism was a term coined by a number of Indigenous researchers in the late 1990s early 2000s to describe scholarship that centralised Indigenous worldviews (Garroutte, 2005; Lavallée, 2009; Martin & Mirraboopa, 2003; Rigney, 1999; Smith, 1999; Wilson, 2009; Wilson et al., 2022). These scholars came from a range of disciplinary backgrounds – education (Karen Martin-Booran Mirraboopa, Margaret Kovach, Linda Tuhiwai Smith, Lester-Irabinna Rigney, Miriam Rose Ungunmerr Baumann); sociology (Eva Marie Garoutte); Indigenous and feminist studies (Aileen Moreton Robinson); law (Irene Watson); and health (Lynn Lavallée, Bronwyn Fredericks, Denise Wilson, Alayne Mikahere-Hall and Juanita Sherwood). Indigenist research always begins from a principle of self-determination and is underpinned by “ceremony, respect, authentic engagement, critical reflexivity, reciprocity, relationality and responsibility” (Quinn, 2022, p. 2). Indigenist research has an entirely different epistemological and ontological foundation to all Western research traditions. Knowledge is not universal: it has been passed on orally over millennia and is particular to people, their stage of development through the human life cycle, place, season and time. Nor is it individual: it is always relational, reciprocal and embedded in community. While this may seem to have an affiliation with idiographic research traditions, which affirm the particularity of people and context, First Nations peoples have strict rules about what can be shared, and with whom; and where and when knowledge can be passed on. The context itself has an important role in telling stories as kinship is more-than-human. Country – a term that denotes Ancestral lands – is believed to be alive with both living ecologies and numinous Spirits. In many traditional communities, knowledge particular to place must be shared in that place (Benterrak et al., 1984).

Indigenist research frameworks for First Nations health have multiplied over the past 25 years. They are preferred by First Nations communities, who run community-controlled health initiatives, because they honour Indigenous ways of knowing, doing and being, decentre power, and maintain First Nations data sovereignty (Kukutai & Taylor, 2016). An expansive systematic review of black and grey literature from 1990-2012 identified 305 Aboriginal and Torres Strait Islander healing programs around the world, 118 of which were disseminated in peer reviewed literature (McKendrick et al., 2013). Overarching frameworks include Kaupapa Māori from Aotearoa/New Zealand (Cooper, 2012; Keegan, 2012; Royal, 2012; Stevenson, et al., 2018), Tobacco Ties (Lavallée, 2009), and Sikolohiyang Pilipino (Enriquez, 1975; Pe-Pua & Protacio-Marcelino, 2000). Within these frameworks there are also a variety of Indigenist data collection methods. These have been developed from knowledge sharing practices in particular communities. They include, amongst others, Dadirri – a Ngan’gikurunggurr and Ngen’giwumirri concept from the Aboriginal peoples of the Daly River region in Northern Australia that means a “deep contemplative process of ‘listening to one another’ in reciprocal relationships” (Ungunmerr-Baumann et al., 2022, p. 94), yarning (Bessarab and Ng’andu, 2010; Walker et al., 2014; Geia et al., 2013; Barlo et al., 2021; Hughes & Barlo, 2021; Kennedy et al., 2022), storying (Murrup-Stewart et al., 2022; Phillips & Bunda, 2018), story-work (Kovach, 2009) and sharing circles (Lavallée, 2009). These are all cultural forms of communicating that differ from structured or semi-structured interviews and standardised questionnaires. These methods are not only culturally familiar to the participants, they take place within the context of established relationships of care. Yarning, for example, is generally undertaken by insider-researchers with their own communities. They differ from interviews in that they are non-hierarchical, circular, multi-layered and intimate, building community through the process rather than merely extracting knowledge for the benefit of the researcher.

While Indigenist mental health research has been important for communities, it has been slow to impact policy in Australia. There are two reasons for this: sometimes it is not disseminated outside communities for cultural and plitical reasons; more often it is regarded with suspicion by Western policy writers because it is deemed to lack Western scholarly psychometric objective rigor and is rarely generalisable (McKendrick et al., 2013). Mental health care, as a consequence, has remained siloed. Aboriginal controlled community health organisations work in small community facilities grounded in Indigenous ways of knowing doing and being, while mainstream health is delivered in hospitals and clinics framed through the lens of Western science and social science.

Hybrid Methodologies: Combining Western and Indigenist Methods

Hybrid qualitative research methodologies have been developed to bridge the divide through syncretism of Western and Indigenist research methodologies. Two frameworks for blending Indigenous and Western methods emerge from Indigenous ways of knowing: Etuaptmumk or Two-Eyed Seeing (Marshall et al., 2015; Wright et al., 2019) and Ganma (Ryder et al., 2020; Yunggirringa & Garnggulkpuy, 2007). Etuaptmumk is a Mi’kmaw term from the Northeastern woodlands of Canada coined by Elder Albert Marshall that means ‘learning to see from one eye with the strengths of Indigenous knowledges and ways of knowing, and from the other eye with the strengths of mainstream knowledges and ways of knowing, and to use both these eyes together, for the benefit of all’ (Reid et al., 2021, p. 244). Ganma is a Yolηu word from north-east Arnhem Land in Australia describing the place where salt water and fresh water meet, and churn to form new matter—foam. It is used metaphorically to depict the way new knowledge is formed from Western and Indigenous origins, leading to a deeper understanding. Others modify Western methodologies with Indigenist data collection methods and/or First Nations governance, but retain structural and analytical approaches drawn from Western methodologies. These include Community-led Participatory Action Research, or CPAR (Cahill, 2007; Sharmil et al., 2021), and CPAR with Grounded Theory analysis (Murrup-Stewart et al., 2021; Priest et al., 2017). McKendrick et al. (2013) found CPAR alone and Grounded Theory with CPAR to be the most commonly used methodologies.

Grounded Theory is the preferred method where little is ‘known’ about a phenomenon. It develops theory from the ground up, finding common, generalisable, themes through immersion in idiographic data. Its key contribution is the development of a rigorous process of analysis. Analysis begins immediately through an iterative, reflexive, inductive process of interpretation. There is a careful structured approach to data coding that moves from the general to the abstract, supported by memo writing throughout the process. This ends when ‘saturation’ is reached. That is, when any new data support the themes that have already been identified and support theory that has developed from them. (Charmaz & Thornberg, 2021; White & Cooper, 2022). CPAR, on the other hand, focuses on the dissolving boundaries between researcher and participants, observer and subject. It is based on Participatory Action Research (PAR), which “assumes knowledge is rooted in social relations and most powerful when produced collaboratively through action” (Fine et al., 2003, p. 173). In the context of a hybrid methodology, First Nations communities participate in a range of ways. They are usually closely involved in governance; have oversight over data collection, often employing Indigenist methods; and will also participate in identifying the themes that emerge from the data. When paired, CPAR provides an ethical framework of engagement while Grounded Theory provides a methodological rigour to the analysis. On the surface, Grounded Theory with CPAR seems one of the most promising methodologies for navigating the cultural divide between Western and Indigenous epistemologies. It combines a methodology from the qualitative research arsenal accepted for its rigorous approach while simultaneously honouring Indigenous cultural mores and data sovereignty.

Grounded Theory has had many evolutions since it was first developed in 1967 by Glaser and Strauss (White & Cooper, 2022). Key to all variations is that data are used to produce theory rather than test a preconceived hypothesis. Rigour is developed through close observation of reality. While Glaser & Strauss (1967) believed that the researcher had a capacity to be objective, Charmaz & Thornberg (2021) recognised that researchers bring their own diverse perspectives to bear on the research process and are shaped by the epistemologies in which they have been immersed and the theories in which they have been educated. Their interpretivist model includes a process of transparent critical reflexivity, where researchers position themselves and consider the biases they might bring to bear on the interpretations they make (Charmaz & Thornberg, 2021). In this way, there are clear synergies with Indigenism. As Lavallée (2009) notes:

Indigenous research is not objective, nor does it see itself as unbiased. Following from the belief in interconnectedness … research cannot possibly be completely objective. Individuals conducting the research are necessarily connected to the individuals being researched, and all concerned are connected to all other living things. Emotions are connected to all mental processes. Every time we think, use reason, and figure, emotion is tied to that process; therefore, it is impossible to be free of emotion and subjectivity in research (p. 23).

Cooper & Burnett (2006) argue that reflexivity during the analysis phase of Grounded Theory is not enough. Interviewers are equally instrumental in shaping the direction of a conversation during the data collection. They promote a discursive reflexivity between the interviewer and interviewee to co-construct the narrative that is recorded. Grounded Theory with CPAR offers an opportunity for discursive reflexivity between researchers and participants throughout other stages of a research project as well. When Indigenist methods based on cultural forms of communication are employed, data are intrinsically circular and relational and discursive reflexivity often becomes part of the analysis too.

Grounded Theory and CPAR with Indigenist Methods in Action: Decolonising Research Practice

Hybrid frameworks incorporating Grounded Theory, CPAR and First Nations specific data collection methods appear to be the best methodology for serving community interests while simultaneously speaking to the Western policymakers. Recent studies of note include a co-designed Western PAR methodology, informed by critical theory, with Dadirri and Ganma health and wellbeing approaches, developed by Sharmil et al. (2021), Priest et al. (2017), and Murrup-Stewart et al. (2021). Sharmil et al.’s (2021) study to improve services for Indigenous people with mental health and alcohol and other drug comorbidities was co-led by Elders, with Dadirri governing the process by which participants communicated, and Ganma ensuring everyone remained mindful of each person’s contribution, recognising that the work composed together was not claimed as ‘mine’ or ‘yours’, but as ‘ours’ (Sharmil et al., 2021, p.6). Priest et al.‘s study (2017), which sought to better understand the concepts of health and wellbeing from the perspective of Indigenous urban children in an Australian city, was developed with community support, including a formal partnership with two communities, an Indigenous advisory group, and an Indigenous research assistant with close community connections, and a community exhibition and feedback session at the end. They used Indigenist data collections methods (yarning) as well as recognised Western approaches (photovoice) and adopted constructivist Grounded Theory (Charmaz, 2006, 2014) to analyse the results. The team took care to ensure that they undertook a discursive process with the project advisory group after the project to check themes (Priest et al., 2017). Murrup-Stewart et al.‘s study (2021) explored the relationship between culture and wellbeing, also focussing on young First Nations adults in an urban Australian context. They used Indigenist data collection (yarning), and thematic analysis derived from Grounded Theory. Specifically, the researchers used NVivo for initial coding and two researchers independently coded the first 20% of transcribed yarns, and the larger team met to agree on sub-themes based on Aboriginal ways of knowing, being and doing (Martin & Mirraboopa, 2003) before proceeding to code the rest. The process was place-based, relational, holistic, ancestral, connective, and collectivist (Murrup-Stewart et al., 2021).

A key strength of CPAR in combination with Grounded Theory, which each of these studies explores, is its collective process. As social systems theorist, Niklas Luhmann, 1986 has argued, groups are only transformed when there is a feedback loop of a “critical outside” (Wolfe, 1998, p. 64–70). These studies explore different methods for engaging communities in the research analysis so they help make sense of the data through cultural perspectives and lived experience. The group process ensures that there is an opportunity for interpretations of researchers to be challenged by the community participants and “insider-researchers”, and for discoveries to be debated. Western health research must learn to listen to Indigenous people if it is to address the failures of previous policies and better meet the needs of Indigenous communities.

But each of these research teams also recognised implicit challenges in marrying Western and Indigenist methods. Most identified power to be an ever-present phenomenon in this binary pair, despite the promise that hierarchies would be flattened, making adjustments to compensate. For example, one group abandoned the approach, believing at the time that the epistemological and ontological differences were irreconcilable. Kennedy et al., 2020; Sharmil et al., 2021 used critical theory to remain mindful of the risk that hegemonic power paradigms might be reinforced. The lead author navigated her position as an ‘outsider’ to the participatory community by spending considerable time developing relationships and deferring leadership to community Elders, but she is mindful that there is rarely the money and time in research projects to allow for true reciprocity. Priest et al. (2017) attempted to straddle both worlds by including both Indigenous and Western scholars and cross-checking themes with their participatory community in multiple instances before coming to a consensus about the themes. As community ‘insiders’ alert to the risk of Western bias, Murrup-Stewart et al. (2021) decided to modify the rules of Grounded Theory, and carefully filtered themes to ensure they represented an Indigenous worldview rather than allow themes to emerge spontaneously from the data. In a later collaboration (Murrup-Stewart et al., 2022) Murrup-Stewart further explores the tendency of Western epistemologies to be privileged through an imagined yarn with a research participant called ‘Settler-Colonisation’. The paper concludes that settler-colonisation remains unfinished business, finding new guises even within methodologies that seem to promote First Nations worldviews. Kennedy et al. (2020) abandoned their planned thematic analysis partway through their research believing it impossible to work across such wholly different epistemological and ontological frameworks. They switched to an Indigenist relational methodology instead because they were anxious about further entrenching Western ontologies, but reflected afterwards that involving First Nations participants in the study design and validation of findings would have been better for addressing power imbalances. They write: “Such lessons are powerful reminders that decolonization is an evolving, self-implicated process that requires ongoing critical reflection and corresponding action” (Kennedy et al., 2020, pp. 10–11). While reflexivity and reciprocity are essential parts of CPAR praxis, there are often limits to how far they can be taken within the constraints of a research project.

Towards ‘Agonistic Pluralism’: Difference, Dissonance and Disagreement as Findings

There is, however, a further challenge that has not been identified. One of the intrinsic features of Grounded Theory with and without CPAR is that the pathway to rigour is configured around reaching consensus about the themes during analysis. A key step in the process of developing theory from the ground up is progressing from open coding to categorisation into subthemes and principle themes. Categorisation, which unfolds continuously from first coding through to final theory development, is a process of finding the generalisable within the particular. This is done by researchers, iteratively, who address the potential of bias in their findings through self-reflexivity, and in the case of hybrid models employing CPAR, discursive reflexivity with either insider-researchers or community participants. These themes are finally condensed into a theory. Research teams using these hybrid methodologies generally try to identify between 3 and 8 overarching themes by sifting through the many stories they have gathered (Busija et al., 2020; Murrup-Stewart et al., 2021; Priest et al., 2017). Sometimes these themes are verified through an additional process of testing against the literature – a form of triangulation (Quinn, 2022). Whether the themes are discerned by a cross cultural research team or through a participatory process with community, a key assumption is made: that themes reflecting the majority are the most appropriate representation of the concerns of the group.

Consensus-seeking invariably smooths out important differences amongst diverse participants. As political theorist Chantal Mouffe (1999) contends, debating until an agreement is reached reinforces a dominant discourse. She writes: Before every consensus exists as a temporary result of a provisional hegemony, as a stabilisation of power, and that always entails some form of exclusion” (p. 756). Deliberative democratic processes are underpinned by several assumptions: that they are governed by social norms of equality and symmetry; that everyone has the right to question and have an opinion; and everyone has the right to initiate a reflexive argument (Mouffe, 1999, p. 747). But, Mouffe writes, power is rarely symmetrical, and dissonances will always exist within pluralist contexts. Mouffe (1999) argues instead for a process of “agonistic pluralism” (p. 754) – a struggling with difference across divergent worldviews. In the context of Grounded Theory, research analysis would not focus on shared themes alone, but also include the dissonances, disagreements and differences expressed across the cohort. Where research budgets are insufficient, or participants are dispersed, the analysis ought to be completed by a multi-perspectival team carefully configured around social difference. Again, the dissonances, disagreements and differences across the team should be included as key findings along with any agreed themes.

‘Agonistic Pluralism’ in Action

The hybrid Grounded Theory with CPAR research method presented here has been developed and tested in a study exploring Indigenous culture’s role in First Nations young people’s mental health in south-east Australia. The community participation took the form of a governance board of Elders, an advisory group of First Nations health professionals, First Nations leadership of the research team (Author AV) and a First Nations research assistant who was a member of the community of participants. Author (JM), a non-Indigenous researcher, also co-leads the project. Participants included 46 First Nations community members (2 subsequently withdrew) who were known to the First Nations research assistant and recognised by Elders as having knowledge that would be valuable to the project. The participants deliberately included people with diverse ages, lived experiences, tribal affiliations and perspectives. There were a range of age groups from young adults through to respected Elders; urban and rural dwellers; men and women; people with varying exposure to culture, including some who were born on Aboriginal missions, some who had been removed from their First Nations families as infants, and some who had mixed settler-First Nations heritage; and people with varying educational backgrounds including those who had not completed secondary school through to people working as academics.

The Indigenist method of yarning was used to gather data. Originally planned to be convened in community groups around the state, COVID-19 lockdowns forced a last-minute change to the process and the research assistant ended up conducting yarns via Zoom, one-on-one with participants. This was a time of considerable stress and uncertainty in the social context in which the team was working with six lockdowns totalling 262 days over 2020-2021 (Pocket, 2021). While in hindsight it was conceivable that these yarns could have been group processes, coordination would have been more complex, and once we realised the benefits for deeper, more reflective yarns and the ease of recording them in Zoom, we abandoned the group process. For many, these yarns were a source of solace and connection in an otherwise isolating, lonely time.

Each yarn went for between 30-90 minutes. They were recorded and transcribed. The team decided not to use NVivo to process the results, concerned that word-sorting with artificial intelligence would not account for local lingo used by many participants, more likely in the context of a yarn between community members. To avoid the bias and blind spots human ‘coders’ inevitably bring, depending on their social, cultural, educational and disciplinary experiences, a multi-perspectival, discursive approach was adopted. Four coders reviewed all the yarn footage and/or transcripts. Each of the team members had a distinct social, cultural, educational and disciplinary perspective: two had Aboriginal heritage and two were non-Indigenous; one was a former Aboriginal Health Liaison Officer (AHLO) working in the hospital with secondary education to year 11, one had an undergraduate degree, and two had doctorates; those with tertiary training straddled diverse disciplines too: mental health (psychiatry), social sciences (anthropology), and the spatial disciplines (architecture and cultural geography). One of the coders was a participant-researcher – the research assistant who conducted the Zoom-yarns – others were observers. The authors wanted all the emergent perspectives of the yarns to be valued and heard, believing this would best inform the subsequent stage of the project so initial coding was completed independently.

Memos were written by each of the ‘coders’ to define, explore and question their own pre-existing ideas and assumptions. This self-reflexivity ensured the coder remained mindful of their positionality. Once the individual coding and thematic analysis was complete the team met, again in Zoom, over four extended sessions. Each coder took it in turns to share what they had ‘discovered’ through each yarn, a different person leading each time. A process of deliberation followed. This enabled the research team to engage in a process of discursive reflexivity (Cooper and Burnett, 2006). We committed, overtly, to record not only shared perspectives, but the differences. Sometimes the group was united in what they believed had emerged from the data. But, at times the dissonance between perspectives was the key finding.

On close investigation of the coding and themes each of the members of the research team had identified, there were clear dissonances. Each had focussed, or at least prioritised and explored more deeply, areas that were shaped by their own disciplinary bias and cultural interests. The ethnographer teased apart what the word culture meant to people, the Aboriginal psychiatrist (author AV) was interested in the ‘old ways’ of healing, while the architect (author JM) was especially attuned to discussions of place and context. Lived experience was also a shaper. Author (JM) had coded barriers to health but had not even identified it as a key theme. It was appended at the end of the coding matrix. The former AHLO by contrast had identified multiple themes related to barriers to health-seeking including observations of racism, lack of cultural safety, trauma, pressures of living across two worlds, and service barriers. While there was no intrinsic disagreement that these were evident in the yarns, there was a powerful dissonance in the extent to which team members recognised them as important themes. The team resisted pursuing consensus, allowing these differences to be important findings that shaped the second stage of the project: the development of an adjuvant mental health therapy based on cultural practices within a Western hospital for young people with mental health difficulties. This has been discussed in more detail in another publication (Vance et al., 2023).

Difference, dissonance and disagreement were present within the participant group too. While the team agreed that there was consensus amongst participants that culture was essential for health, exactly what ‘culture’ meant varied widely. It was variously described in terms of practices (smoking ceremonies, craft, music, dance) knowledges (language, how to prepare bush medicines), relationships (family, kinship networks, animals and Country), and a way of being (feelings, sensations, connection with Ancestral Spirits). Most participants agreed Western health overlooked the importance of culture to health. But they did not agree about where to go if sick. Many had experienced racism and alienation in Western hospitals and general practices, while others had reservations about going to community-controlled health organisations because of lack of privacy. There was also a lack of consensus about whether culture can be integrated into Western care or whether it is wholly different and must stand alone. Consequently, the team has published an overview that presents key shared findings (Vance et al., 2023), but a number of other papers that tease out these variation findings are in press or development.

Limitations/Future Directions

Research at the interface of Indigenous and Western health is still in its infancy, particularly in the field of young people’s mental health. A global systematic review by Jongen et al. (2023) of peer refereed and grey literature describing programs that address mental health and wellbeing for First Nations young people specifically found only 14 intervention studies, five of which were in Australia located in primary health care, education, juvenile justice and Indigenous communities. Four key approaches were identified, all of which focused on enhancing compliance with Western healthcare systems. Meanwhile the systematic review by Saini (2012) of studies that had attempted to validate hybrid Western and Indigenist research designs found none of the 68 papers that met the inclusion criteria had cross validated the overarching hybrid Western and Indigenist research designs. In each case the researchers had focused on validating the Indigenist research methods alone. The validation of Western health quantitative and qualitative research methods for Indigenous peoples was just not countenanced. This is concerning, given Tseng’s (2006) detailed outline of the influence of culture on the formation of a mental disorder; influencing the coping patterns to deal with stress of a disorder; the clinical manifestation of a disorder; the severity and frequency of occurrence of a disorder; and the management of a clinical condition. Saini (2012) hypothesises, with reference to Witt (2007) that Indigenist research methodologies are being actively validated by First Nations researchers “searching for rational justification to defend [their] cherished worldviews against the attack by those who constantly wish to denigrate them” (Witt, 2007, p. 23 in Saini 2012, p. 5). But Western methods are accepted to be already validated, even though they rarely take cultural difference into account. While Saini (2012) concludes that lack of cross validation is an inherent weakness for hybrid methodologies, we conclude otherwise that Western and Indigenist research methods within hybrid methodologies need to be critiqued separately, given their wholly different epistemological and ontological basis.

Conclusion

This paper has offered a comparative exploration of the three overarching research methodologies – Western nomothetic and idiographic, Indigenist, and hybrid Western and Indigenist – employed to study the mental health challenges faced by First Nations communities. The epistemological and ontological underpinnings of each is a key contributor to their essential difference. First Nations scholars have been important contributors to Indigenous mental health research and policy development, working within Western quantitative and qualitative research methodologies (see Dudgeon et al., 2022; Gone, 2023; McGaw et al., 2022a; Milroy et al., 2023; Thai et al., 2023; Vance et al., 2017). The reasons for the use of Western methodologies and development of hybrid frameworks are twofold. The first is strategic: in order to speak to, and be heard by, those with the power to make a difference, it has been necessary to adopt the accepted quantitative and qualitative research methodologies of the hegemony (Foucault, 1972; McGaw, Vance, & Herb Patten, 2022). The second is socio-cultural: Indigenous identity is complex in the contemporary globalised world and First Nations people have been assimilated to varying degrees into Western epistemologies and social mores and trained within particular disciplinary modes of practice (DeLanda, 2006; McGaw & Pieris, 2014). Meanwhile the use of Indigenist epistemologies is motivated by a desire for the academy to recognise other knowledge systems and to support communities in their aim of self-determination. This paper contends that a hybrid methodology that uses Grounded Theory with Indigenist data collection methods and community participation seems most promising. It also advocates for a multi-perspectival team for analysis. We caution against the usual practice in Grounded Theory of working toward consensus around themes. As Mouffe has argued, consensus seeking privileges dominant voices and majority views, concealing important dissonances and disagreements likely within groups constituted around difference. This paper does not argue that an agonistic Grounded Theory model should replace all others. Rather, it proposes that the field needs more. Where groups are complex and diverse, the findings ought not to be limited by thematic consensus alone. The field needs to be prepared to struggle with the tension of disagreement.