Key Considerations for Managing and Analyzing Longitudinal Photovoice Research Data

Context/Justification: Photovoice is a qualitative, participatory, action research method that originated in the health sciences over 30 years ago. With theoretical roots in Paulo Freire's approaches to empowerment education, feminist theories, and documentary photography, it has been lauded as a research method with potential to both empower participants and generate rich, qualitative data in partnership with research populations that are often marginalized. However, longitudinal applications of Photovoice are exceedingly rare in the literature-health science or otherwise. To the author's knowledge, no publications exist to guide researchers on how to proceed or what to avoid when managing and analyzing Photovoice data longitudinally.  Goals/Methods: This paper draws on the author's experiences from nearly a decade as the Principal Investigator or Co-Principal Investigator of three longitudinal Photovoice studies. These studies examined 1) health and integration experiences of refugee women in the United States,  2) health and integration experiences of urban refugee women in Kampala, Uganda, and 3) the learning and career impact of study abroad to South Africa for U.S. university public health students. The aim of this paper is to systematically describe lessons learned and considerations for other researchers when managing and analyzing longitudinal Photovoice research data. Results: Two key areas of the author's experiences with longitudinal Photovoice will be described including, 1) how high volumes of data from multiple sources over several years (ranging from 2 to 8) were organized and managed; 2) the evolving approaches to analysis including recurrent cross-sectional and trajectory approaches. Conclusions: Longitudinal Photovoice is a novel application of a research method, so there is a gap in the literature regarding management and analysis of Photovoice data longitudinally. While the central focus of this paper is on longitudinal Photovoice data, lessons described may be of interest to readers/audience members engaged with other forms of longitudinal qualitative data.