Methodological reflections on the use of photovoice in school health research

Introduction

There is overwhelming evidence to support the empowering potential of photovoice, a process whereby research participants take photographs to present their views on a significant aspect of their lives. First introduced in 1994 as Photo Novella by Wang and Burris (1997), photovoice has since proved to be a powerful research strategy that continues to be used across the social sciences, including in health, health promotion and health education (Liebenberg, 2018; Wang, 2020). The work of Wang and Burris suggests photovoice methodology is political in that it derives from critical theory, which centralises issues of power and social justice. Thus, photovoice seeks to create space for people to critically engage with and actively determine their health and well-being, both as individuals and as groups. Indeed, one of the typical strengths of photovoice methodology is its participatory nature and its resultant ability to facilitate community change, often through knowledge sharing (Switzer and Flicker, 2021). As an empowerment-driven strategy, photovoice is well suitable for studying immigrant children and youth, who face the unique challenge of navigating unfamiliar settings and school systems.

While it is befitting for all citizens of this world to ‘voice’ their concerns on issues affecting their health, using photovoice can be challenging due to its methodological peculiarities. First, unlike other photographic research approaches, photovoice typically uses participant-generated pictures and an analytical interview protocol to prompt discussions that seek to promote self-awareness, advocacy and the emergence of novel ideas (Cooper and Yarbrough, 2010; Esau et al., 2016; Henry et al., 2013). Second, photovoice aligns with collaborative analysis, even though photovoice researchers tend to use conventional qualitative approaches to data analysis (Wang, 2020). Third, photovoice demands close attention to ethics, especially concerning the taking of photos and dissemination of research findings (Creighton et al., 2018). Fourth, photovoice researchers often need to provide incentives to encourage participation, such as money, gifts cards, art supplies and cameras (Creighton et al., 2018; Adekeye et al. 2014; Shannon, 2013). Fifth, participant–researcher relationships are particularly important to photovoice research considering the usually prolonged engagement between participants and the researcher.

Although photovoice is a participatory action research method in which community engagement is central, it is important to also consider the role of photovoice researchers who can significantly influence data collection, analysis and representation (Cooper and White, 2012). This article builds on recently published work (Nyika, 2022) and reports on the application of photovoice in an effort to advance conversation concerning reflexivity and positioning of the photovoice researcher. In particular, it seeks to provide practical insights and recommendations on the use and relevance of photovoice to help prepare beginning researchers for the complexities of working with the technique. This is important because research reports often do not describe the challenges and tensions involved in data collection activities, hence beginning researchers may not expect the many complexities they will encounter in the field.

Development of this methodological reflection was influenced by my wondering how my experiences using photovoice as a research tool might contribute to the future use of photovoice methodology. To my knowledge, this is the first known photovoice methodological reflection involving African students in health promoting schools (HPS) or comprehensive/coordinated school health. HPS take the form of integrated school settings designed to collectively optimise stakeholders’ health and broader processes of schooling (Lee et al., 2014; Macnab, 2013; Simovska, 2004). Informing methodological reflection here is Archer’s (2003, 2007, 2012) theory of reflexivity or self-talk, which proposes that an individual makes decisions and acts in accordance with their personal cognition and social environment. The article begins by describing the context of the study on which my methodological reflection is based. This is followed by discussion of my thoughts, feelings and observations using photovoice to examine African immigrant students’ school experiences. I conclude by making recommendations for future use by beginner researchers.

Research context

Given that HPS are cultural institutions developed to attain health and educational goals (Nyika and Murray-Orr, 2017), participation within them is likely to be contested, especially for African immigrant students, who may bring different, even contrasting, understandings of what schooling and health should be. As a health and education professional of African descent, on relocation to Canada, I had interest in the health and well-being of African students who were participating in Canadian HPS. In particular, I sought to understand how African immigrant students imagined, felt and thought about themselves in relation to HPS, as Nova Scotia school stakeholders.

The study was framed by a philosophical bricolage (Lévi and Strauss, 1966) informed by critical race theory and social constructivism. Study participants (eight from Africa and seven from the Caribbean region) were purposefully selected from four Nova Scotia high schools with the largest numbers of immigrant students of African ancestry. Participating schools (given the pseudonyms A, B, C and D) were considered to be HPS because they offered health-related programmes (e.g. breakfast programmes, youth health centres and student support services) in partnership with governmental, non-governmental and community organisations (McIsaac et al., 2012). Some of the study participants knew each other before hand, but others like myself did not know any of the participants ahead of time. Like the participants, I, the researcher was Black and had recently migrated to Canada, which made me something of an insider.

Participants used the following pseudonyms (with actual ages in parentheses): Alex (18), Bottle (18), Breanna (20), Emma (12), Furtuna (21), Henrietta (18), James 1 (17), James 2 (19), Lisa (18), Michelle (18), Natasha (16), Rashida (18), Riley (16), Seth (20) and Timothy (16). Each received a Canadian $50 gift card as a gesture of appreciation on completion of the research project. The research ethics process as detailed elsewhere (Nyika, 2018) involved seeking approval from my home university research ethics board and a responsible school board. The research question to be pursued was as follows: How do African immigrant secondary school students perceive their participation in HPS in Nova Scotia?

Reflections

Entering the research field, I was mindful of the power and privilege of being a researcher (Cooper and White, 2012), yet as an inexperienced photovoice researcher, I was uncertain about my role in the aforementioned study. Thus, I thought preliminary interviews would give participants and me, the researcher, an opportunity to get to know one another and for me to begin building research relationships (Call-Cummings et al., 2019; Solheim et al., 2021). Preliminary interviews enabled me to capture detailed accounts of participants’ cultural backgrounds, including their family situation, migration experience, and health beliefs and practices. The interviews were conversational because from time to time, I shared my own stories and experiences, which appeared to make participants feel at ease. Interestingly, preliminary interviews are seldom used in photovoice research and would be a valuable addition in future studies.

Insofar as photovoice has particular methodological requirements as indicated by the need for informed consent, training photography and the use of open-ended questions, I adopted Warne et al.’s (2013) photovoice design as a guide to the study’s photovoice data collection. However, since both the content and context of a picture are central to photovoice (Bendell and Sylvestre, 2017; Liebenberg, 2018), I incorporated other research methods to provide multiple perspectives and enrich the research findings, such as focus groups, individual interviews, observations and fieldnotes.

A series of photovoice workshops marked the beginning of data collection and focused on photovoice ethics training, how to use disposable film cameras and participant conduct during the photovoice process. It is likely that most participants knew how to take pictures, but perhaps less so using age-old, disposable film cameras. As with Warne et al.’s (2013) photovoice protocol, I use disposable film cameras in the hope that the cameras would encourage participants to carefully select their photovoice artefacts since disposable film pictures cannot be altered after they have been taken. I discussed this aspect of the study with participants during the preliminary workshops as well as the sometimes fortuitous and inconvenient nature of photography (Wang and Redwood-Jones, 2001).

It was not possible to hold one photovoice workshop for all participants as initially planned because some participants had to attend to urgent school and personal matters. As a result, I held sessions with three, two and sometimes one participant. I combined the last four preliminary individual interviews with the photovoice workshops. During the interview-cum-workshops, I found it helpful to encourage participants to take a few pictures each day aiming to use the whole film of 27 exposures by the end of 1–2 weeks.

Some research participants had problems taking pictures using disposable film cameras. One participant had to be given a replacement camera after taking pictures with the camera in reverse, essentially shooting ‘selfies’ the whole time. Other participants’ photographs looked blurred, perhaps due to poor lighting. Yet, other participants were uncertain whether it was them or their friends who were to take the photovoice pictures. Lisa, for example, took most of her photovoice pictures collaboratively with Michelle and a friend from school. Others did the same despite being given what I thought was clear guidance for each participant to take pictures individually. The inability of participants to adhere to photovoice instructions and/or use cameras effectively is evidenced in some prior research (Boonekamp et al., 2020; Labbé et al., 2021), signalling the need to check in with and support participants as they take pictures. Despite this, it is possible that as an inexperienced photovoice researcher, the photovoice workshops I conducted may have inadequately addressed the topic of photography.

The local school board required the research project to cause minimum disruption to participants’ school life, especially learning activities but, this proved difficult to achieve due to unforeseen challenges related to some participants working part-time. It was especially challenging to organise data collection activities in line with the limitations imposed by African Nova Scotian student support workers being available only 3 days a week instead of 5, as were the teachers and students. African Nova Scotian student support workers played a key role in the research project by acting as contact persons for the recruitment of participants and the coordination of research activities. In the end, I was forced to hold one focus group and several preliminary and photovoice individual interviews during school hours.

As noted previously by Call-Cummings et al. (2019), school-based photovoice research is strongly influenced by the schools’ stakeholders and their gatekeeping strategies. Early on, I wrote the following notes about photovoice group interviews:

I have realised today that it’s not easy to conduct focus group discussions. There is a lot of issues with them, everybody is busy in this Canadian dream. [The study participants have] got to work, they got to navigate (like everybody else) the various determinants of health/life. Each time I go to school A, students and everybody, the African student support worker, all look like they are always in a hurry. (Fieldnotes, 29 April 2016)

With so much going on in the lives of participants, gatekeepers and myself, it proved necessary to make adjustments to the photovoice data collection plan. Like Boonekamp et al. (2020), the unpredictabile of research field caused me to make changes and adaptations to the photovoice methods. I learned to be a negotiator and narrative researcher (Clandinin and Connelly, 2000) during the 4 months I was involved in data collection as everything else became of secondary importance as I worked hard to negotiate research schedules and events with the various stakeholders and keep everyone updated electronically and/or in person. It also took me a while to adjust to the challenge of (re)scheduling both individual and group interviews because I first had to learn to live with the uncertainties of the research field. For example, one participant’s pictures could not be processed in time for the photovoice discussions since she returned her camera the last day of data collection. Another participant failed to return his camera altogether and efforts to contact him afterwards proved fruitless. A third participant’s pictures were excluded from the study as he could not attend the accompanying photovoice interviews for personal reasons.

Photovoice provided opportunities for participants to consider their role in comprehensive school health through analytical conversations. During the first photovoice group discussion, I felt slightly unsure about the research agenda as participants and I struggled to cope with Wallerstein and Bernstein’s (1988) rather academic interview technique known as SHOWeD. What was I really looking for? Wallerstein and Bernstein’s framework, which draws on Freire’s (1973) germinal work on critical education, requires the following questions to be asked: (a) What do we See here? (b) What is actually Happening here? (c) How does this concern Our lives? (d) Why does this affect our lives? (e) What can we Do about it? As an inexperienced photovoice researcher, I began to wonder how productive this kind of group discussion would be and what the focus group data would look like. The group interview format also reminded me of my early high school teaching experiences when I had to manage class and group discussions. My control over participants’ responses and the direction of the group interview was inefficient, in contrast to the individual interviews where I felt more certain. Subsequently, I decided to modify the SHOWeD framework to align more closely with the study’s research question. It proved useful to include the following questions that spoke to participants individually: what does the picture you selected mean to you, and what made you want to take the picture? These questions, which helped kick-start photovoice conversations and focus in on the context of the pictures, could be important additions to future research. In addition, during the photovoice interviews, I initially struggled to synchronise the picture slide show and audio recording. However, with time, I was able to more efficiently manage the group conversations, positioning myself alternately as a researcher, moderator and participant.

With time, during photovoice discussions, participants looked comfortable and engaged, suggesting a deepened research relationship. Compared to individual interviews, which felt a little formal, in focus group discussions, there was more laughter, sighing and grimacing. Some participants even made conversation by asking me personal questions, such as the name of my university and my academic qualifications. I wonder if that move signified progress toward stronger research relationships, greater authenticity with respect to my insider status, or just plain curiosity. One participant, Michelle, was particularly interested in the focus of the research project. She said, ‘I expressed my ideas as open[ly] as I could, and it was nice to know that you are at [a Canadian] university representing [African immigrant people]’ (Michelle, photovoice group interview). Her comment highlighted the importance of establishing authentic participant–researcher relationships, with the ability to positively impact the research process (Nyika and Murray-Orr, 2014; Solheim et al., 2021).

Unlike during the preliminary interviews, many of the participants provided more engaging narratives during photovoice interviews due to more established research ties and a sense of ownership of the pictures. It is possible that participants and I became closer, even to a degree falling in love (Clandinin and Connelly, 2000) over the course of the individual interviews, photovoice workshops, and later occasional verbal and/or electronic communication.

Limitations

It is important to recognise that the methodological reflections here reflect the views of one school health researcher of African heritage, with accompanying biases, assumptions and limitations. Moreover, the reflections shared relate to the study of a relatively small group of African immigrant youth (from four particular schools) who volunteered to participate in a local research project. Despite this, a number of suggestions can be made which, if implemented, may strengthen future research.

Conclusion

This article contributes to ongoing conversation on qualitative research methods by drawing attention to the role of the researcher in photovoice research. As evidenced here, photovoice brought interesting dynamics to data collection in this study, changing the tone, the feel and the energy of my interactions with research participants. Preliminary individual interviews which had been conducted earlier in the study gave participants and the researcher the opportunity to familiarise themselves with each other, and for the researcher to start to build participant–research relationships. The preliminary individual interviews helped me, as a researcher, to better contextualise participants’ photovoice narratives. A major lesson learned from the work was the value of making changes to photovoice data collection in ways informed by initial photovoice workshops and group interviews. To help enhance the value of the photovoice interviews, I modified Wallerstein and Bernstein’s (1988) interview protocol by adding questions that specifically addressed my own research agenda and participants’ personal circumstances and feelings.