The application of qualitative research in public health has produced very important insights, especially in the field of social determinants of health. The most common qualitative health research techniques are semi-structured interview and focus group (FG).1 Otherwise, different approaches using visual methods like photo-elicitation or photovoice have been successful in studying health issues.2 Considering these advances in the application of qualitative techniques and their scope, we propose a comparison of two group techniques such as FG and photovoice in public health research.

FG was originated in sociology by Merton and Kendall.3 It consists of a formal meeting of 5-10 people selected for being representative of a population group with similar characteristics or experiences. It is used to deeply understand the point of view of the subjects involved in a phenomenon of study and how the collective discourses related to a particular theme, are socially shared and co-constructed.4

FG research has reported relevant results in the study of factors influencing food habits in diverse population groups5 as well as in other urban health topics, like air pollution or alcohol environment.6

On the other hand, photovoice is a relatively new technique. It uses photography to study participants’ perspectives considering their opinions as the main actors of a particular problem. Recently, the inclusion of the perspective of people participating in qualitative studies is taking steps towards their participation not only as subjects-objects of study, but also considering their capacity of originating changes. This approach is also classified within citizen science7 and is usually framed under the umbrella of participatory action research.8 It was first formulated by Wang and Burris,9 who developed it for studying life and health conditions of women in rural China. It consists in putting cameras in participants’ hands and using the photographs they take with their narratives, to evaluate community resources, while stimulating critical reflection and social action.

The objective of this article was to compare, from an empirical point of view, the use of FG and photovoice as techniques for health research. Our goal was to identify differences, similarities, potentialities, and limitations of each technique through the analysis of two studies with the same object of study.

We show key differences and similarities between the two techniques in Table 5, as well as their potentialities and limitations in Table 6.

The aspects included in Table 5 and Table 6 correspond precisely to characteristics previously described in literature that define and differentiate each of the techniques. Related to the differences, we describe two of them; the facilitation and the data analysis.

The facilitation model used for FG or for a photovoice group is quite different. Conducting FG requires moderation tasks only the day of the meeting. In photovoice, facilitators had to monitor/accompany participants for at least one month and a half. The facilitator in photovoice usually gets personally involved in the problem addressed by the group, while in FG the role of the research team in relation to participants is more traditional.

The process of data analysis is different. In FG, researchers work together to interpret the transcripts and in photovoice the analysis is mainly done by participants. In the last session participants also summarize their results by presenting a map of categories and photos (see online Appendix B) as an overview of the findings “constructed” over the five sessions

Considering the purpose of the study was to compare from an empirical perspective, the use of FGs and photovoice as we conducted two studies on food environment in neighbourhoods with different socio-economic profiles, both techniques proved to be very fruitful, providing novel and useful results.5 We cannot state that one technique is better than the other for the study of urban health. In fact, in addition to our work, several studies have used these techniques with equally remarkable results.14 Some of the differences showed in the results, had already been described in literature. For example, our approach to moderating FG was framed within the Qualitative School of Madrid15 and other authors,16 in which moderator intervenes little and concentrates on listening and observing. There are no clear rules on how to facilitate photovoice groups.17 In practice, we carried out some of the facilitator functions or skills used in action research,18 such as balancing interventions, giving meaning or encouraging group meaning-making, among others. Related to results, from FG these are drawn from the readings and analyses done by different researchers19 of the literal transcriptions of the FG. We used the qualitative sequential discourse method.12 In photovoice, as Wang and Burris9 propose, participants themselves analyse their photos and debates. Results would be analysed as if it were a qualitative description. Qualitative descriptive studies produced findings closer to the data as given.20 In our photovoice study, several participants in Villaverde wrote proposals for policy recommendations, explained those to decision makers and actively collaborated in the writing of a book and scientific articles.11

Given all of above, we found advantages and difficulties with each technique, which are explained bellow.

The importance of the discourse could be a noteworthy reason to use FG in a study. Discourses gathered through FG allowed us to use different theories for analysis. For example, in our qualitative study, we used grounded theory to analyse the perceived effects of a comprehensive smoke-free law according to the socioeconomic status of neighbourhoods,21 and the qualitative sequential discourse method to understand how the residents’ perceptions of their food environment affect their dietary habits.5 Despite the only analytical purpose in photovoice is giving participants the ability to voice their analysis of the situation, we also found some theoretical analysis conducted after the whole participative and action process, based on the transcripts, photos and/or the processes experienced.22

We consider several key elements when choosing the photovoice technique instead of FG: increasing the critical awareness of participants, involving local or regional decision makers before starting the study, the atypical role of the research team and the implications of using photography. In photovoice, social intervention was not sought, although it was not disdained either. An attempt was made to empower participants so that they were aware of the reality of their neighbourhood and acted accordingly. In this sense, we highlight the relevance that participatory studies with transformative vocation and social intervention could have in the general framework of public health.

FG does not intend to increase the critical awareness or knowledge of the study population. However, according to some authors the mere fact of participating in FG makes participants reflect and acquire new ideas on the research topic. FG are not simply a mean of eliciting knowledge from participants but are often reported to be significantly creative and transformative experiences for them.23

Despite the possibility of increasing critical awareness through FG, the difference with photovoice in this sense is enormous. Our study gave us the opportunity to see how the fact that participants met five times raised awareness about the problem,24 as other photovoice studies have shown.25

Furthermore, sometimes in classic FG, participants do not agree with how researchers have analysed their discourses. In photovoice the results are validated simultaneously during the analysis of participants. In qualitative research, the validation of results, or member checking, is a fundamental step that does not normally occur.26 We consider this type of extra feedback session necessary to listen to participants, making sure they feel represented by the researchers’ interpretations.27

In addition, within FG based research, post-project evaluation is not usually implemented either. We found very few publications regarding the design of specific interventions supported by FG findings.28

Considering the application of both techniques in several evidence-based decision-making studies, like changes in drug prescriptions by general practitioners after participating in a qualitative study28 or generate policy recommendations on alcohol urban environment using photovoice,29 we highlight the potentialities of these techniques as a great input in citizen science.30

As mentioned previously, in photovoice, at the end of the five sessions, results are ready: the concept map, the photos and the SHOWED forms. No further analysis by researchers is required. According to the objectives of the technique, the only thing left to do would be presenting the results publicly informing policy/decision makers. That is, after a month and a half of work, there would be results. While transcribing every FG and then conducting a triangulate analysis can be very time consuming, photovoice requires more dedication during and between sessions.

Another characteristic of the role the research team will play in photovoice is the involvement with local or regional policy makers and the close collaboration with other organizations within the community (non-governmental organizations, neighbourhood associations, community centres, etc.). This collaboration demands great efforts synchronizing agendas, languages, and priorities (for example, the time a researcher concerned with scientific papers is different as those of a social worker focused on the needs of the community).

Photography is a tool only included in photovoice. Although there are FG studies supported by images,31 we cannot make an empirical comparison with the way we designed and conducted our FG study.

Regarding potentialities, using FG we get a deep knowledge of the study population, by extracting the type or expected discourse of that population.32,33 In photovoice, participants provided new knowledge, and also gained social recognition that transformed their self-perception.25 Results could be transformed into recommendations for policymakers related to food.

Related to ethical considerations, informed consent, and release of images (both by the participants and by the people who may appear in the photographs) are essential, but not enough. Special care must be taken with sensitive or marginalized populations.34

This article should serve as a guide for other qualitative studies planning to use the techniques of photovoice and/or FG.

Depending on the research objectives, one technique may be more beneficial than the other. Both techniques are complementary and may even be used within the same study.

If the research objective is to increase or produce new scientific knowledge within a specific topic, FG would be the most appropriate technique. If the objective includes involving the population to generate changes like acquiring new knowledge and developing awareness of certain troublesome and/or approaching political decision spaces, photovoice would be more suitable.

We found that photovoice is a powerful methodology in the study of physical aspects connected to health or social issues requiring extra effort from researchers and participants. In addition, photovoice requires special care with related ethical considerations.

We found that the effect of participating in FG studies has not been sufficiently assessed. Analyses of possible changes in awareness, self-consciousness and/or behaviour should be included in the design of qualitative research in general, at least when it comes to communities or population groups that are vulnerable or susceptible to such changes.

FG may be the most appropriate technique for producing new scientific knowledge within a specific topic. Photovoice is a participatory action research technique focused on social change that necessarily increases participants’ engagement and awareness, and involves decision makers to channel proposals. The atypical role of photovoice researchers and ethical implications of photography must be considered.

Data made available to people who request it by contacting the corresponding author.

Jorge Marcos Marcos.

The corresponding author on behalf of the other authors guarantee the accuracy, transparency and honesty of the data and information contained in the study, that no relevant information has been omitted and that all discrepancies between authors have been adequately resolved and described.

Conceptualization: P. Conde. Investigation: P. Conde, J. Rivera-Navarro, M. Gutiérrez-Sastre, I. González-Salgado, M. Franco and M. Sandín Vázquez. Formal analysis: P. Conde, J. Rivera-Navarro and M. Sandín Vázquez. Funding acquisition: J. Rivera-Navarro, M. Franco. Supervision: J. Rivera-Navarro and M. Sandín Vázquez. Writing original draft: P. Conde, J. Rivera-Navarro and M. Sandín Vázquez. Writing, review and editing: P. Conde, J. Rivera-Navarro, M. Gutiérrez-Sastre, I. González-Salgado, M. Franco and M. Sandín Vázquez.

The authors would like to thank Jorge Barriuso and Ulises Díaz-Ropero for their linguistic assistance on this article, and every person (residents and health workers) who participated and assisted in any of the focus groups or photovoice sessions and meetings.

This work was funded by Spanish Ministry of Economy and Competitiveness [CSO 2016-77257-P] and by the European Research Council under the European Union's Seventh Framework Programme [FP7/2007-2013/ERC] Starting Grant Heart Healthy Hoods agreement no. 336893.

None.