Comprehensive sexuality education (CSE) is recognized as a critical tool for addressing sexuality and reproductive health challenges among adolescents. However, little is known about the broader impacts of CSE on populations beyond adolescents, such as schools, families, and communities. This study explores multi-level impacts of an innovative CSE program in Madagascar, which employs young adult CSE educators to teach a three-year curriculum in government middle schools across the country. The two-phased study embraced a participatory approach and qualitative Human-centered Design (HCD) methods. In phase 1, 90 school principals and administrators representing 45 schools participated in HCD workshops, which were held in six regional cities. Phase 2 took place one year later, which included 50 principals from partner schools, and focused on expanding and validating findings from phase 1. From the perspective of school principals and administrators, the results indicate several areas in which CSE programming is having spill-over effects, beyond direct adolescent student sexuality knowledge and behaviors. In the case of this youth-led model in Madagascar, the program has impacted the lives of students (e.g., increased academic motivation and confidence), their parents (e.g., strengthened family relationships and increased parental involvement in schools), their
schools (e.g., increased perceived value of schools and teacher effectiveness), their communities (e.g., increased community connections), and impacted broader structural issues (e.g., improved equity and access to resources such as menstrual pads). While not all impacts of the CSE program were perceived as positive (e.g., students start experimenting with sex and love), the findings uncovered opportunities for targeting investments and refining CSE programming to maximize positive impacts at family, school, and community levels.

Data were collected in June 2021 during seven workshops held in six regional cities in Madagascar. Principals and one other member of the school were invited from each of of the youth-led, non-governmental organization's 45 partner schools during the 2020-2021 school year. Principals had the discretion to invite second representatives, which was often the person who worked most closely with or alongside the CSE Educator. If the principal could not attend, then they assigned someone to attend in their place. In total, 90 school administrators participated, which included principals, vice principals, and school monitors representing 45 schools, both urban and rural.

Workshops were facilitated by the organization's Monitoring and Evaluation Manager and lasted 3-4 hours. Workshops were held in the organization's regional offices or rented working space (i.e., not in schools), and discussions were facilitated in the local language, Malagasy. Additional organization staff members in attendance took detailed notes which they then cross-referenced with the audio recordings of the discussions as needed to accurately fill a standardized data entry sheet in Microsoft Excel file. The final data entry sheet was translated into English for qualitative analysis.

At the workshops, participants responded to two research prompts:

1) What are the different effects CSE programming has on students’ lives?

2) What other effects does the CSE programming have in the school or in the community?

In brief, participants brainstormed responses to the prompts individually on post-it notes, then participated in an affinity clustering exercise in which they grouped their ideas based on perceived similarities and differences.

In phase 2, data were collected in May 2022 during the non-governmental organization's end of the school year symposium for principals of partner schools to gather feedback on the cluster analysis conducted in phase 1, validate findings, and vote on priority areas. Fifty-one principals from the partner schools from the 2021-2022 school year were invited to participate, and a total of 50 representatives from the schools attended the workshops, representing three regions of Madagascar.

Two members of the research team consolidated key findings across the seven workshops conducted in phase 1, paying particular attention to key themes that were raised in multiple different workshops using qualitative content analysis. The team used grounded theory to analyze the data using an inductive approach, in which initial cluster themes were compared across all workshops to draw broader comparisons, and the relationships between clusters were studied to propose a final conceptual model of the CSE program impacts. The final cluster labels were created by the team based on a review of the items within the clusters.

The tallied votes from the phase 2 “Visualize the Vote” exercise were collated in a standardized Microsoft excel spreadsheet. Basic counts were calculated for each cluster to reflect the principal’s observations of areas impacted by the CSE programming and the principal’s vote for the areas of impact most important to them.  Translations of the Phase 2 principal feedback were reviewed by the coding team and select quotes were used as illustrative examples to contextualize the key clusters.