Towards a more gender-responsive bean breeding program: lessons from East Africa

In a world of changing climate, growing population pressure and urbanization, a deeper understanding of the traits that users such as farmers and processors prefer, and how they are changing, is critical for achieving impacts through breeding solutions.

Participatory variety selection methods (PVS) that were introduced in late 1990s has been the only method used in bean improvement for learning about farmers preferred traits. Through PVS processes, breeders gained knowledge about farmers’ preferred bean traits, which they use when setting breeding priorities in the breeding cycle.

For breeders to meet end-users’ needs, they must understand the priorities that women and men assign to genetically determined traits. Photo: G.Smith/CIAT

Over the years, however, the information from PVS could not reveal the economic value farmers derive from each trait nor provide actionable insights with regard to new areas of value that breeding programs could to tap into and create new market space for beans. One example that would pop up during discussion with farmers was the need to reduce cooking time. However, breeders had not given it consideration as a priority trait for genetic improvement. Yet, as urbanization in sub-Saharan Africa grows, such unmet preferences can drive future demand away from bean consumption.

In 2012, we decided to employ a choice experiment (CE) method and analyze preferences for bean variety traits in detail in order to understand how farmers value each trait, segment the respondents into homogenous subgroups according to their preferences, estimate the relative segment size and compare the needs of each segment with the bean breeding goals.

We implemented the study in eastern and western parts of Kenya on a sample of 452 bean growing households where male and female respondents participated in the CE and provided additional information on their demographic characteristics. Production (tolerance to stresses, yield, and maturity time), and consumption traits (taste and cooking time) were given varying levels by experts from national agricultural research systems and the International Center for Tropical Agriculture (CIAT) and combined into different variety options, each associated with a “monetary price” using computer-aided methods. Data from each province was analyzed using a latent class model.

The case study is included in GBI’s new publication ‘State of the Knowledge for Gender in Breeding: Case Studies for Practitioners’, linked below. Photo: CIAT

We found that all traits included were important, but there were surprises. While information from PVS always portrays yield as ranked top most important, CE revealed that farmers derive higher economic value from consumption traits such as taste and reduced cooking time. Another surprise was that though women are responsible for cooking beans, we found that men derived more value from reduced cooking time than women, which we believe emanates from the fact that population pressure has resulted in firewood from community woodlots being replaced by planted woodlots or purchased charcoal, all of which are considered men’s role. In eastern parts of Kenya, 45% of respondents were eager to shift to a hypothetical variety with a faster cooking time and more flavor. These results excited the breeders – igniting discussions to breed for reduced cooking time.

The results brought to light the relevance of knowing economic values from each trait in breeding cycle. Because the identified segments are also characterized, breeders are empowered to match product profiles with needs of each segment when setting priorities. Beyond priority setting, information derived from such research can help to unmask the necessary triggers of variety adoption that seed business managers can use to design seed marketing strategies and maximize their business profits. This is especially useful when analysis is based on data from representative sample of intended users.

This case study was presented and discussed in the Annual Review and Planning meeting organized under two projects and held in April 2018 in Kampala, Uganda. Approximately 100 people comprising of breeders, social scientists and research managers from the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), the International Institute of Tropical Agriculture (IITA), CIAT and their respective national partners from Eastern and Western African countries and donor representatives attended. It was also shared and discussed with participants in the Gender Responsive Legume Breeding course organized in July 2018 jointly by Cornell University and Makerere University. The course was attended by plant breeders and social scientists from national agricultural research scientists across sub-Saharan Africa.

This is one of 10 case studies explored in the CGIAR Gender and Breeding Initiative’s new publication ‘State of the Knowledge for Gender in Breeding: Case Studies for Practitioners’. The working paper is part of a series of knowledge products designed to share the Initiative’s collective knowledge more widely across CGIAR and partner breeding programs.


Enid. M. Katungi

Scientist, International Center for Tropical Agriculture (CIAT)Enid. M. Katungi is an agricultural economist with more than 20 years’ of experience in social science research conducted in numerous African countries. Currently, she is a scientist at the International Center for Tropical Agriculture (CIAT), where she coordinates impact assessment for the Bean Research in the Pan African Bean Research Alliance (PABRA) and provides mentorship to NARs. Her current research focuses on: impact evaluation of bean research on poverty, food security and gender outcomes, factors driving and constraining bean technological adoption, gender analysis and integration in bean value chains. She has conducted an ex-post impact assessment of improved bean technology adoption on household welfare in Rwanda, Uganda and Malawi and is now assessing an ex-post impact evaluation of bean research in Tanzania, Ethiopia, Zimbabwe and Burundi. She previously served as a gender focal person in PABRA and a member of the Grain Legumes Program research management committee, coordinating the “Impact assessment, priority setting, knowledge management and gender Organizations” flagship. Enid holds a doctorate from Pretoria University, South Africa in agricultural economics.