Benefits and challenges of engaging Majority World children in interdisciplinary, multi-qualitative-method, mental health research

ABSTRACT

The qualitative community embraces transparent dialogue through sharing knowledge to improve rigour and develop new initiatives. In this paper, we recognise there are many complexities within qualitative research, leading to important debates. We explore the benefits and challenges, as well as the practicalities and technicalities, of conducting research that 1) employs multiple methods within the qualitative paradigm, 2) is interdisciplinary, and 3) examines a sensitive research topic 4) with a vulnerable group of participants. Specifically, when research is also 5) cross-cultural and 6) utilises participatory techniques. Through our discussion, we draw upon an existing project and examine the complexity of designing and completing intra-paradigmatic mixed methods research with children from different sociocultural contexts, underpinned by interdisciplinary perspectives, in a complex area like mental health.

KEYWORDS:

• Qualitative methods
• children
• vulnerability
• majority world
• intra-paradigmatic
• mental health

Introduction

Cross-cultural qualitative mental health research with seldom heard groups like disadvantaged children and youth is crucial to advance our understanding of issues that inform mental health promotion, prevention, and intervention. Undertaking this type of work, however, can be complex, with challenges arising from working with a sensitive topic, with vulnerable populations. It is therefore important to use innovative and creative qualitative methods to address complicated research questions. Interdisciplinary knowledge providing dialectical epistemic value, multiple methods that strengthen the cohesion of the qualitative approach, and developmentally appropriate techniques that encourage collaboration and coproduction with young populations has benefit for applied research but also raises additional complexities.

In practice, this kind of work is imbued with epistemological and methodological tensions, where there exist arguments about the congruence across different theoretical perspectives and in that sense, have implications for quality. Many tensions are beyond the scope of this paper, such as the different perspectives on the existence of reality (e.g. realism vs. relativism), or the location of construction within the cognitive level of the individual (constructivism) against social interaction (constructionism). Yet, researchers must reconcile incongruence at this theoretical level within a study for credibility and transparency. Our focus in this paper is to explore the methodological challenges in relation to mixing qualitative methods, taking an interdisciplinary approach, including vulnerable populations from disadvantaged communities and different cultures, and in the sensitive topic of mental health.

Mixing-methods and multi-disciplinary knowledge

Quality indicators for qualitative research promote congruence to assure coherence, credibility, and transparency (Lester & O’Reilly, 2021). It is necessary for researchers to anchor their decisions to a methodological framework and assure that quality parameters are fully adhered to. For applied researchers seeking to ensure a holistic perspective, it may be necessary to use multiple methods to address their research question. The typical conceptualisation of mixing methods denotes the combination of quantitative and qualitative, that is, inter-paradigm research, whereas mixing within the qualitative approach is referred to as intra-paradigm research (O’Reilly et al., 2020).

Inter-paradigmatic mixing has been subject to critique, yet it is generally agreed that combining approaches allows researchers to compensate any weaknesses inherent in a single design (Bryman, 2008) and capitalise on the strengths of each approach (O’Cathain & Thomas, 2006). For intra-paradigmatic mixing, there has been a misconception that this is less problematic because the work is wholly qualitative. Because the multiple methodologies or methods are within a single paradigm, it is believed that there are no concerns or issues that warrant attention, but the heterogeneity of the approaches and theories mean this is not the case (Barbour, 1998).

In research aiming to address complex research problems, intra-paradigmatic mixing may be preferred and beneficial, especially when the research team represents different disciplinary knowledge, methodological preferences, and vocabularies. This interdisciplinarity is significant, because mono-discipline approaches are often insufficient and unable to account for complexity (Tobi & Kampen, 2018). Increasingly researchers are recognizing that, to comprehensively address critical questions, disciplinary fields need to be linked, and skills, knowledge and perspectives from multiple disciplines integrated (Fischer et al., 2011; Lakhani et al., 2012).

Children’s voice and mental health

A diverse skill set from different perspectives can be especially useful when managing a complex project with vulnerable children and youth. This is because of the necessity to manage ethical issues such as promoting agency, respecting rights to make decisions, and how to balance consent with safeguarding. This is important when researching sensitive topics like mental health. In the context of ethics, children have frequently been subjected to an over-protectionist position. Historically, they were excluded from research because of concerns about their competencies to contribute, and fears about the risks of harm (James & Prout, 2015). In contemporary research, children are now conceptualised differently and are viewed as having agency (James & Prout, 2015). This understanding of children has been influenced by the global treaty, the UN Convention of the Rights of the Child (1989) advocating children’s engagement in decisions affecting their lives. Thus, it is necessary that research is well-planned, with policies and programmes being responsive to the concerns and needs of this population (Boyden & Ennew, 1997).

In qualitative research, a child-centred approach is favoured through genuine engagement and active listening. In so doing, researchers must balance empowering and promoting children’s voices, with protecting them from harm. In mental health, the importance of capturing children’s voices on different life aspects that influence their wellbeing is increasingly acknowledged by researchers, policy makers and service developers (Faithfull et al., 2019; Larsson et al., 2018; Prout & Tisdall, 2006). Their inclusion in research has, therefore, necessitated innovation and creativity, to account for their developmental competencies and variation in conceptualisation of mental health, as well as to promote research participation as meaningful and engaging.

Participatory approaches are viewed as a mechanism to promote inclusion of seldom-heard groups. Participatory techniques present opportunities for collaboration that allow young participants to be involved in the research process from conception to dissemination (Higginbottom & Liamputtong, 2015) and favour inclusivity over researcher expertise (Cargo & Mercer, 2008). However, participatory research with children is challenging and researchers need to be mindful of any difficulties those children face in participating and help them to overcome any barriers to inclusion (Boyden & Ennew, 1997). It is necessary for any usage of techniques to be genuine and ensure their participation is not tokenistic (Montreuil et al., 2021). While the asymmetrical research relationship may never be fully eliminated, giving voice to children is fundamental for qualitative scholars (Haffejee & Theron, 2022), which is especially pertinent in cases where those children are disadvantaged or marginalised.

Child mental health research in the majority world

Most child mental health research originates from minority (western) contexts. Recently there has been a drive to establish the mental health needs of children in majority world countries (MWC), particularly commonalities that can inform intervention and service models that transcend cultures (World Health Organization, 2020). Such evidence necessitates methodological approaches that actively involve children across different sociocultural settings. Child mental health research in MWC is faced with challenges such as positioning children in certain adult-centric societies, stigma of mental health, and poverty barriers (Huus et al., 2021). Nevertheless, there are strengths when researchers relate to the wealth of community participation in MWC contexts, narration of experiences, and creative approaches in sharing those (Larsson et al., 2018).

To this effect, there is increasing use of participatory research in MWC. These have been predominantly used to address important social issues, like the impact of war or gender-based violence (Abdullah et al., 2021) and building resilience (Cameron et al., 2013). To date, there has been less attention to multi-method participatory approaches in MWC child mental health research, and this has usually focused on single sites. Participatory multi-methods research approaches can generate multi-cultural child-centric evidence to inform international policy, and therefore these need to be implemented in multiple sociocultural contexts. This research gap hence informed the rationale for this research and paper.

Aims of the paper

We aim to consider some of the core challenges associated with undertaking projects that contain multiple challenges and provide potential solutions by using empirical examples from an existing research project. Our intention is to provide a pedagogical narrative addressing the question ‘how might qualitative researchers resolve some of the methodological challenges in multi-method research with majority world children in mental health?’. This focuses on the benefits and challenges of undertaking complex qualitative projects that embrace methodological arguments shaping and directing the work. Not only does this add methodological value for the qualitative community in terms of showcasing how it is possible to address some complicated qualitative tensions, but also illuminates children’s voices for promoting pedagogical messages.

Considering there is limited evidence from children in the majority world on mental health, our inclusion of data examples is prominent in terms of transferring those messages. In this way, we demonstrate how to avoid tokenism, embrace modern ideological recommendations for co-production, and ensure genuineness for working with children beyond the traditional minority world perspectives.

Method

Consistent with the exploratory scope of our question, we adopted a thematic analysis approach, allowing us to identify the salient issues at stake and explore the value of different methods. This promoted a person-centred approach, where children’s voices were central to informing the project and dissemination.

Theory

Our person-centred focus aligned with our macro-social-constructionism position. This reflects the view that children and childhood are constructed and change over time (James & Prout, 2015). It accounts for ways in which children transform their experiences through their personal accounts and narratives, using various tools, including verbal, textual and visual. This recognises that children’s voices must be heard to inform policy debates that relate to their best interests and wellbeing (Livingstone, 2013). A macro-social-constructionist epistemology is appropriate when researchers seek to explore children’s views of the social world through focus on meaning and language (Fraser et al., 2004).

Context and recruitment

With support from local non-governmental organisations (NGOs), we recruited children aged 8–10 and youth aged 13–17 years, to include two developmental stages, in five MWC: Kenya, South Africa, Turkey, Pakistan, and Brazil. We use the broad term ‘children’ throughout, except for specifying age groups in data. We developed partnerships with NGOs and invited children and youth via those local networks by engaging with families in each of specific location and matching our age criterion. Information was widely distributed through the NGO and those expressing interest were invited to a briefing session. All children within our age categories who consented were given the opportunity to participate. Sampling adequacy was assured within and across countries and groups (Hancock et al., 2016), which was necessary for a thematic design (O’Reilly & Parker, 2013).

While each country is situated in the majority world and experiences socio-economic challenges, there are structural and cultural differences. In Kenya fieldwork was conducted in Nakuru city, which has high levels of poverty in terms of basic needs like food, shelter, and clothing. The city lacks basic social amenities like adequate housing and access to digital media. In South Africa fieldwork was conducted in Ekurhuleni Metropolitan Municipality, in the Gauteng Province of Johannesburg. Large numbers are unemployed and many live in poverty, although there is some established infrastructure like electricity, water, and a sewerage system. In Turkey, Karatay and Selcuklu in Konya are low-resource areas with high crime and poverty, as well as many refugees. In Pakistan, Manzoor Colony Mehmoodabad is a neighbourhood in Karachi East. This is an underprivileged area, with high rates of crime, domestic violence, and cultural conflict. In Brazil, we included Rocinha in Rio de Janeiro, a large favela with overcrowding, and lack of quality housing and sanitation.

In each country we were mindful of the cultural values and country specific norms. To facilitate this, we were guided by the partner NGOs in how to approach and engage the children. To promote inclusion, we involved peer researchers in each country to facilitate, reassure and comply with local norms. We were able to adapt and be flexible with our data collection styles to suit the local context, the culture of the children, and their developmental abilities. For example, we had a schedule of questions, but the way in which these were delivered differed, such as being more concrete and directive in Pakistan than in South Africa. This was also facilitated by the interdisciplinarity of the core team, and the wider network of partners whose disciplinary and methodological knowledge enhanced data collection, but also helped with the linguistic and developmental issues.

Data collection

Our data were collected through three methods, focus groups, diaries, and drawings, thus representing verbal, textual and visual information:

a. Verbal data were collected via focus groups, as these provided a mechanism for collaboration, discussion and sharing ideas. Within each country, one senior and one peer research moderator provided space to allow participants to engage with the contributions of others and reflect on ideas (Willig, 2008). We ran 20 focus groups: four focus groups in each country, two per age group and we met with each group twice. Focus groups were conducted face-to-face, except in Brazil, where they were conducted remotely on Zoom, because of COVID-19 safety guidelines. The schedule was designed to be broad, and person-centred. The schedule reflected three areas of questions on how children:

i. Conceptualised their wellbeing along different life socioecological domains (individual, family, and community).

ii. Experienced their wellbeing being influenced by disadvantage and related adversities.

iii. Formulated informal and formal supports to address their needs.

Traditional style interviews or questioning can be daunting (Hore et al., 2021), particularly for children, whose experience and/or understanding of group discussion, may be pressurised situations such as debates or tests in class, where there is a desire to perform well and in an appropriate way. Therefore, visual data collection techniques were also adopted, in line with our child-centred approach, as they supported a ‘less directive line of questioning’ (Heath et al., 2009, p. 125), in the form of participatory drawing.

b. Participatory drawing has gained momentum in child research, by minimising distance between researcher and participants (Haffejee & Theron, 2022); as a way of moving beyond only valuing oral communication, and of promoting ‘non-textual ways of knowing’ (Singhal & Rattine-Flaherty, 2006). Drawing is a familiar activity and means of expression for many children, so it helped dilute the formality of the setting, and encourage authentic discussions. More than just a useful tool in instigating conversation in focus group settings, these visual representations were helpful in deepening engagement, sparking memories, and foster more detailed, reflective explanations of adverse experiences, especially as the study was conducted during the COVID-19 pandemic. Their use was valuable, as the children represented different educational and literacy levels and this method promoted inclusion, as they could write or draw or both. These techniques also offer insight into different way of seeing and understanding subtleties or nuances in experience, which might otherwise have been overlooked (Hore et al., 2021). The primary function of these methods was to engage and include children, recognising that the research process can be intimidating. While they were not formally analysed using a specific methodological frame, they were mapped against the written and verbal data to strengthen thematic analytic points and were included in the coding process of the thematic analysis.

c. Diary entries operated as a task-centred method to enable children to exert more control over their contributions, as they were completed daily, in their own time, over one month, and without interference from the research team – other than occasional prompting by the local researcher (Heath et al., 2009). Diaries allowed participants an opportunity to reflect on their thoughts and feelings before offering the final collection of their ideas, but also to contribute views they may have felt less comfortable raising in a group setting and/or that occurred to them later (Gauntlett, 2007).

Analysis

We utilised thematic analysis to attend to the data, because of the emphasis on language and meaning from the perspective of those central to the research (Braun & Clarke, 2006). We engaged with a codebook form of thematic analysis to allow for the conflation of inductive and deductive coding processes and ensured coder agreement through a multiple coding process (see, Bentley et al., 2019). Two coders independently coded all data to create a synthesis of the core issues identified. The different epistemic disciplinarity of the different team members strengthened analysis and via a thematic mapping consultation process. This allowed the multiple codebooks to be mapped and conceptual categories agreed, while accounting for the disciplinary language of the different members. By doing so, we were able to bring to the fore interesting issues that ensured the holistic approach accounted for the range of expertise and topical knowledge of those members.

Ethics

The research was approved by the University of Leicester ethics committee. Written consent was provided by all adult parties and verbal assent from children (with parental consent). All quotations represented are anonymous. Ethics jurisdictions varied in the five countries, also depending on the recruitment process (for example, recruitment through a non-statutory organisation did not always fall under the current ethics governance). For this reason, each recruiting NGO acted as facilitator, while safeguarding children’s rights, using local protection processes.

Findings

In attending to the tensions and issues of quality for research, we identified three dominant (albeit overlapping) methodological challenges arising from our iterative process of data collection, analysis, and dissemination. These reflect the wider qualitative challenges of mixing methods of data collection, interdisciplinary working, sensitive topics, vulnerable populations, utilising participatory techniques and our cross-cultural inclusion. Each of the challenges is considered below, followed by our approach in addressing it, arising benefits, and empirical supporting data.

Challenge one: Researcher-participant dynamics

Issue: A primary challenge was to minimise the asymmetry between researchers and participants, accounting for cultural nuances, whereby children have respect for adults and power may normatively operate within other child-adult settings like schools. The team was aware that childhood is not a universal concept, and while the UNCRC treaty was adopted in all five countries, its interpretation, meaning, and implementation differed, especially in relation to characteristics like gender and the meaning of mental health.

Solution: We provided different spaces for children to express themselves without adult supervision, and involved peer researchers with lived experiences, who had previously conducted multi-country research with the same research group (Faithfull et al., 2019). Sharing experiences removed myths about mental health and helped break barriers, both in terms of age and mental health. To facilitate data collection, we engaged researchers with cultural and topical knowledge, complemented by the interdisciplinary expertise of the team. These country-specific researchers listened to children’s voices, and had good knowledge of local sociocultural needs, experience of research and/or clinical practice with children.

Benefit: The range, depth and diversity of data, and the significant volume of expression from children in different modalities was of real applied benefit to our work. The inclusion of local and peer researchers meant that we enhanced children’s confidence in expressing views by providing platforms for them to voice their perspectives in different ways.

Empirical: In practice, power dynamics are omnipresent and shape data collection. Inevitably, some children will talk more in focus groups and interrupt others, and thus effective moderation is important. However, quieter children had other ways to express their views during the focus groups as well as outside of them. We accounted for power dynamics between researcher and participant, noting how this takes different forms. To encourage peer discussion, we utilised participatory techniques which enabled children to generate and share ideas (Baum & Smith, 2006), encouraging equal participation, as well as engaging local peer researchers to facilitate the process. In the focus groups this was achieved by encouraging children to discuss the topic without asking permission from adults to speak, ask questions about the research, and contribute ideas as active agents, by providing mechanisms for them to be creative so as to reduce tokenism and commit to genuine engagement.

‘Where did this research start, where is it coming from?’

Child, Brazil (focus group)

“You can speak and answer without my permission, as you can think this as a discussion group. Let’s try not to interrupt each other too much, but you don’t need to ask me to speak, we’ll talk together.”

Researcher, Turkey (focus group)

During the focus groups, children were encouraged to participate ‘equally’ and to feel heard by combining different strategies. These involved interaction with the researchers and peers, paired and group activities (e.g. by pulling together their ideas into group posters – Figure 1). Children were encouraged to ‘brainstorm’ together before completing an activity, rather than responding to an adult-led task or question.

Figure 1. Peer group activity and poster on future recommendations, Pakistan.

Similar principles of participation and co-production were applied to researchers’ verbal interactions with the children, for example, by researchers sharing their own experiences.

‘The place where I live, is also an apartment’

Researcher, Turkey (focus group)

Researchers were mindful of not identifying with and reproducing cultural authority stereotypes such as hierarchy, question-answer, or permission to talk by the authority figure such as the ‘teacher’, which would block children from opening up. As seen in the quote below, the researchers were sometimes treated as figures of authority, and this was managed with reassurance of neutrality and a position of curiosity and was facilitated by the presence of peer researchers. This was not a single reassurance occurrence, however, and did need to be reiterated at intervals.

Researcher: “Do you have a suggestion?”

Child: “No, my teacher.”

Child, Turkey (focus group)

Similar risks could exist within the peer group because of cognitive, social or language differences, and were addressed by spreading questions, prompts and activity leads across the group. Children were encouraged to develop their own ideas in response to broad topics rather than questions. Examples of adverse experiences thus emerged more spontaneously during focus groups discussions, rather than at pre-set time points.

To reduce the asymmetry between researchers and children, and to build researcher/child relationships, diary completion was protracted for one month. This longevity helped build a relationship and ‘pace’ the data collection process, instead of condensing all questions in one contact (O’Reilly & Dogra, 2016). Asking children to complete diaries prior to each focus group, initially without specific requests (i.e. describe what was important for them each day) gave children a sense of control in co-defining the research agenda. In addition, they were encouraged to choose whether they want to write down or draw their experiences and ideas. Overall, this process of employing multiple participatory strategies and the supporting evidence indicates that children were actively involved throughout the study, rather than had a tokenistic presence.

Challenge two: Transcending developmental and language barriers

Issue: In traditional research, usually led by an adult researcher, children are viewed as subjects to be studied and the methods used may be overwhelming, less engaging, and less fun (Milne et al., 2011). Methods that rely fully on verbal exchanges, may exclude children who lack language skills, depriving them of opportunities to be heard.

Solution: Participatory methods that incorporate visual methods, like drawing, transcend such literacy, language, and developmental barriers. Drawings offer children an alternate means of self-expression, in ways that may be more comfortable for them (Callaghen et al., 2016). Participatory methods are also recognized as a means by which children may access difficult experiences and engage, in a safe space, on topics that may be difficult to talk about (Mahadev, 2015). Participatory methods promote children’s agency and make room for varying levels of involvement. This egalitarian approach suggests that the extent of participation in the research process is dependent on the participant, and the process is seen as collaborative rather than dictatorial.

Evidently, using multiple methods is a useful way to address the challenges posed, so children who cannot, or do not want to, engage with a certain modality are likely to be more comfortable in one of the other methods. Furthermore, children were given greater control through the participatory techniques, as some chose to use stickers, colours, cut out pictures and add them and so forth, whereas others chose not to do this. The use of participatory methods allowed us to access both visual and verbal commentaries and insights into children’s experiences and insights.

Benefit: Adoption of a participatory approach that encouraged both visual and verbal responses allowed us to surpass cultural and language barriers. The use of multiple methods at different time points and of participatory techniques within the focus groups, provided a space for children to draw their experiences, and write diary entries away from adults. The use of diaries provided greater control and agency over the process, which was child-led, as there were no minimum or maximum entries required. The use of dairies also provided further opportunities for reflection, as there were no time pressures to produce a drawing or share an experience as in the case of the focus group discussions; here children had opportunities to think about what they wanted to record and capture and could do so when they felt inspired or interested. The data elicited were rich and nuanced, and the drawings offered glimpses into children’s inner world. The study further benefited from the inclusion of researchers from different cultural backgrounds in the data collection and analysis phases, thus ensuring cultural relativity.

Empirical: The use of participatory methods can be a fun and accessible medium for engagement. In our study, this was observed in the excitement children expressed when writing in and using the diaries; they thus made the diaries their own, decorating and deciding what they wanted to include.

Child 1: ‘I’ll put a sticker on now, to make it look cute.’

Child 2: ‘I added a little heart ribbon to it.’

Child 4: ‘I really like to draw; it makes me feel very good too.’

Children, Brazil (diaries)

In the excerpt below, we see how a drawing led to a more detailed discussion around a child’s difficulties at school.

“My picture was a book. I like reading story books and newspapers or magazines but not schoolwork. I love stories because they make me happy, but schoolwork makes me think and scratch my head. Stories make me laugh, as well know the news of what is going on in other places.”

Child, Kenya (diary)

The use of diaries also provided a safe space for children to write about the difficulties they had been experiencing and to express themselves in their language of choice.

Challenge three: Congruence of the intra-paradigmatic approach

Issue: Founded in theoretical debates referred to as the ‘paradigm wars’ (Bryman, 2008), mixing methods has been imbued with tensions and scholarly debate for decades. While intra-paradigmatic mixing has been subject to less concern, the combining of multiple qualitative methods is nonetheless plagued with similar theoretical tensions as inter-paradigmatic mixing. Thus, while using multiple methods was necessary to address the imbalance of power, and to manage the challenge of language, in so doing, another challenge was created.

Solution: It was for this reason we opted to undertake a mixed qualitative methods approach, rather than a synthesised methodologies one (O’Reilly & Kiyimba, 2015). From the outset of planning, and recognising the interdisciplinary nature of the research team, there was transparency in epistemological positioning, with all broadly committed to social constructionism, a child rights approach, and a thematic framework. This meant that, despite our different disciplinary backgrounds, we were able to maintain a singular approach with multiple facets, allowing for integration and coherence.

Benefit: Providing there is epistemological congruence and a singular methodological anchor guiding decisions, data collection and analysis, then using multiple qualitative methods can be beneficial. Nonetheless, it remains challenging to integrate and analyse several datasets, from different countries, that are borne from different methods, and are in different formats. When integrated well, subjected to a single analytical approach, and disseminated collectively, the core messages from a project can be considerably strengthened.

Empirical: The integration of verbal, textual and visual data collection techniques assured that participants were offered choices, promoted inclusion, and accounted for any developmental concerns or power issues (as illustrated in challenge one), as well as accounting for linguistic and cultural issues (as illustrated in challenge two). By treating the three forms of data as one overarching dataset for thematic coding, mapping, and verification, as underpinned by a single epistemological position, we ensured that the data were analysed holistically. While the verbal, textual and visual data collected were not necessarily consistent in depth and breadth across all five countries, or across participants, they did all contribute to the generation of codes and themes in the final codebook.

Furthermore, the integration and thematising of the complete data corpus, allowed for further observations, regarding children’s expressions in relation to their mental health. For example, typically, drawings were more often positive in expression, whereas focus groups delved deeper into the feelings of the participants during the pandemic, and diary accounts were more personal. In this way, the data collection methods reflected different snapshots in time, whilst providing mechanisms for reflection, and anticipation of future directions. Many drawings depicted positive activities, or at least, they demonstrated ways in which children were actively coping with adversity such as the COVID-19 pandemic (Figure 2).

Figure 2. Eating pizza.

“Today I ate pizza with my family.”

Child, Brazil (diary entry and drawing)

“Something that is very helpful is reading a book, trying to relax, because we cannot be nervous right now.”

Child, Brazil (diary entry)

Importantly, of course, some of the narratives derived from the focus groups also considered coping strategies, and some were also expressed verbally.

“Read something, read stories.”

Youth, Turkey (focus group)

Integrating the data collectively through thematic analysis and in a single codebook, provided a rigorous way for the team to identify common threads across countries, and across data collection modalities. The use of multiple methods also allowed children to express their mental health responses (for example, to COVID-19 related risks), and the public health measures designed to protect them. While some participants could articulate those verbally, others preferred to write in their diary or display visually. The combination of methods allowed us to achieve an understanding of the spectrum of emotions, expressed in different ways to collate together under specific thematic headings.

“I have a friend that is suffering, at their house they don’t have food. I really don’t like that.”

Youth, South Africa (focus group)

“I was angry, because I could not go out to buy a new drawing book.”

Child, Kenya (diary)

“I am sad, because we were closing school.”

Child, Kenya (focus group)

Discussion

In this paper, we have reflected on the development and implementation of a multi-method, interdisciplinary participatory approach with children from five majority world countries and living in disadvantage about mental health and wellbeing. The objective of the wider research was to elicit children’s lived experiences of adversity and influence on their mental health. Although most arising issues apply to all child-centric research, we particularly focused on the potential benefits of the approach in cross-cultural child mental health and its applications in informing global policy and guidelines, while pedagogically addressing several of the complexities of doing this kind of qualitative work.

Children living in MWC are at risk for a multitude of difficulties, including poor health, poverty, and poor access to resources and services. Yet they, like children everywhere, are seldom included in decisions impacting their wellbeing. There is a tendency to over-emphasise their vulnerability, despite arguments they should be actively included in research. Multicultural, cross-country research involving children from disadvantaged MWC backgrounds, therefore, requires significant reflexivity, methodological, and ethical consideration. By over-emphasising the vulnerability of young participants, this conceptualisation can conflict with a child-centred ideology and an overly protectionist position can inhibit seldom-heard groups from having the platform to project their perspectives on an important issue such their mental health. Vulnerability is not static, and neither is it absolute, rather it is fluid and contextual (Nordentoft & Kappel, 2011). Thus, for researchers, there must be a balance between empowerment, child-centeredness, and children’s rights, with harm-reduction, protectionism, and risk management within and across MWC sociocultural ‘norms’ and research ethics governance. Researchers must account for the heterogeneity within the population they study, recognise developmental, cultural and language barriers, and provide safe ways for children to have their say. Indeed, empowering children to have a voice through research is essential as they are often relatively powerless and may not fully understand the adult world (Boyden & Ennew, 1997).

In addressing these issues through a participatory multi-method approach with children across five MWC, we encountered three major challenges. First, was the challenge of researcher-participant dynamics, especially with younger children. This is reflected in the literature, as even qualitative studies based on interviews typically engage adolescent participants, because they can articulate more easily their mental health needs. Nevertheless, when interviewing techniques have been adapted for pre-adolescent children, they can provide brief, albeit eloquent accounts of their inner world (O’Reilly & Dogra, 2016). We therefore worked to resolve the challenge of engaging children in our research by using co-production strategies, involving researchers with cultural insights, multiple modalities of data collection and participatory techniques. This had the benefit of engagement, rapport, and cultural relevance. By using a participatory approach, we worked to dilute hierarchies of power, making the process more egalitarian and giving voice to participants (Haffejee & Theron, 2022). This was accomplished by inviting active participation from our participants and being flexible, respectful, and making the process enjoyable (Flewitt et al., 2018).

Second was the challenge of transcending cultural and language barriers, which we worked to resolve by using multiple methods at different time points and participatory techniques. We achieved this by providing space for children to draw their experiences, and write diary entries away from adults, as well as benefitting from researchers from different cultural backgrounds in data collection and analysis. Thus, by involving children as ‘community partners’ through participatory techniques, we encouraged them to articulate their stories (Thompson et al., 2019) and be in control of their contributions.

Third was the challenge of congruence and integration of using a multi-method intra-paradigmatic approach, which we worked to resolve with an overarching epistemological framework that centred on children’s agency and voice, connecting with the evidence-base, and using verbal, textual, and visual data. This had the benefit of offering participants choices, promoting inclusion at different levels, and accounting for developmental concerns. The different disciplinary experience and expertise of the research team provided a foundation for open dialogue and resolution of analytical tensions across the data sets in a way that provided for a more holistic approach to the integration of the data. Furthermore, the combination of methods recognised that there are two possible ways to conduct intra-paradigmatic work (O’Reilly & Kiyimba, 2015). First, is a synthesised methodologies approach, whereby two different studies are integrated at the end, and there are epistemological tensions in that integration. Second is a mixed qualitative methods study, whereby there is a single methodology, but multiple methods of data collection and the mixing occurs throughout the process and there is theoretical congruence (O’Reilly & Kiyimba, 2015). We used multiple methods of data collection anchored to one thematic approach and thus maintained the integrity of our epistemology, methodology and methods across the project.

For qualitative researchers engaging in sensitive topics, there is a need to be considerably mindful of the ethicality of the project in terms of the participants, but also the possible emotional repercussions for the research team (Dickson-Swift et al., 2008; 2016). At an ethical level, there are a range of possible difficulties that require attention, including the assurance of privacy and confidentiality (Bourne & Robson, 2015), with the need for breach boundaries and safeguarding firmly in place. These challenges are compounded in several MWC without fully developed research ethics jurisdictions. Children in MWC vary in their formulation of beliefs, which are commonly influenced by stigmatising attitudes of peers and adults within their community (Tamburrino et al., 2020). Resulting fears and anxieties can hinder further their engagement and articulation of mental health needs, if solely relying on traditionally verbal approaches such as individual interviews and focus groups. The design and implementation of the research, therefore, needs to ensure that there is a minimisation of harm, that participants are well-supported and directed toward sources of support, and that the work is conducted with integrity. This is essential for demonstrating to partners and gatekeepers the value of the work, to ensure they do not take an overly protectionist position and deny participants the opportunity to speak about their lives (Biddle et al., 2013). Co-production of the research process with partners in MWC (such as the NGOs in this study) and training of local researchers can help overcome some of these ethics concerns. In multi-country research, joint training and forums can ensure consistency and share ethics lessons.

In interpreting lessons from this study certain limitations should be acknowledged. Selected sites and young participants were not necessarily representative of other communities of the included countries, or MWC globally, hence the findings are not automatically generalisable to other resource-constrained settings. A multi-informant mixed methods approach would complement children’s views with those of adults, as well as with quantitative measures of risk and psychosocial functioning. This is important, as it is well established that children and youth in MWC and globally often process differently mental health needs and supports from those of adults such as their parents and professionals (Getanda et al., 2017; Haffejee et al., 2022). Furthermore, it was not our intention to undertake any meaningful methodological comparisons across our sites, rather we dealt with the broader qualitative arguments about this kind of research and the general themes of issues identified. Future work would benefit from exploring the value of different methods and designs with different cultural groups and countries to explore how these are perceived and engaged with. Nonetheless, there were some commonalities across the groups (see, Haffejee et al., in press for detail). Despite these conceptual, ethical, methodological, and organisational challenges, cross-cultural research with children is increasingly important in understanding communalities as well as context-specific differences, hence informing global policy and guidelines in the design and implementation of preventive and responsive interventions (Tamburrino et al., 2020).

In conclusion, we argue that complex qualitative research that integrates ideas from the methodological literature has a benefit. Doing research into sensitive topics, with vulnerable groups, and across cultures has value in practice. When conducted using various qualitative methods and combining expertise of team members from different disciplinary backgrounds, it promotes innovation, creativity, and new ways of thinking about research problems. By breaking down barriers to interdisciplinary working, such as varying researcher skills and competencies, paradigmatic differences, and tensions in collaboration (Fischer et al., 2011), a study can provide translatable messages for application (Kvarnström, 2008). This is especially pertinent for research that addresses complex or sensitive topics, and/or engages populations who are constructed as ‘vulnerable’ such as children. Thus, we argue there is a need for further development, implementation, and validation of child-centric participatory multi-methods across different sociocultural contexts, as this can be an important driver for a global approach to evidence-based policy that can maximise limited resources and impact on vulnerable children and their families.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Michelle O’Reilly

Michelle O’Reilly is an Associate Professor of Communication in Mental Health and Chartered Psychologist in Health for the University of Leicester, and a Research Consultant and Quality Improvement Advisor for Leicestershire Partnership NHS Trust.

Sadiyya Haffejee

Sadiyya Haffejee is a practicing psychologist and a Senior Researcher at the Centre for Social Development, University of Johannesburg. She enjoys working at the interface of research, practice and policy.

Seyda Eruyar

Seyda Eruyar is an Assistant Professor in the Department of Psychology, Necmettin Erbakan University, Konya, Turkey.

Grace Sykes

Grace Sykes is a Lecturer in Sociology, in the School of Media, Communication and Sociology, University of Leicester.

Panos Vostanis

Panos Vostanis is a professor of Mental Health, and founded the World Awareness for Children in Trauma programme (www.wacit.org), affiliated with the University of Leicester, and the University of Johannesburg, South Africa.