In many East African countries, women and men have different levels of access to formal markets for agricultural inputs, including seed, reflecting a combination of gender norms and resource constraints. As a result, women and men may have different levels of participation in—and reliance upon—informal seed systems for sourcing preferred planting material and accessing new crop varieties over time. We use network analysis to explore differences in seed networks accessed by women and men for three major food security crops—beans, finger millet, and sorghum—in Kenya, Tanzania, and Uganda. Drawing on data from an original survey of 1001 rural farm households across five study sites, we find that women, on average, have fewer connections to experts and farmers’ groups than men but are relatively better connected in farmer-to-farmer social networks across different farming systems. We further find women’s and men’s networks are clustered by gender (i.e., women’s networks include more women, and men’s networks include more men)—and that men’s networks are more likely to exchange improved seed. Women’s networks, though sometimes larger, are less likely to exchange improved varieties that might help farmers adapt to climate change. Women farmers across contexts may also be more reliant on farmer-to-farmer networks than men due to their relative isolation from other seed and information sources. Findings emphasize the need for careful attention to the different implications of seed policies, market interventions, and other seed system reforms to support gender-equitable food security options for women and men in sub-Saharan Africa.

With the increase in the severity of natural disasters linked to climate change, the role religious congregations play in response is not well known, despite their substantial involvement in general charitable activity. Most disaster response research is event-based, ex post, and not focused on religious institutions or donor capacity considerations. Data from the National Study of Congregational Economic Practices fill this gap. The findings indicate that in 2017, a substantial percentage of US congregations participated in disaster-related charitable giving and volunteering. However, the profiles of participating congregations are nuanced and do not always follow prior theory. A congregation's religious tradition is less predictive than some might expect in the context of disasters. Consistent with open systems theory, we find that congregations' interorganizational networks and their proximity to disaster areas are better predictors of participating in disaster relief efforts. Based on our finding of robust congregational participation in disasters, we call for more integration of religious institutions in nonprofit disaster response research.

No abstract