Farmer groups are important socioeconomic safety nets for rural communities in sub-Saharan Africa. They provide mutual support to farmers through collective action to enhance improvement of livelihoods. These groups have been increasing in number in the post liberalization period in East Africa. However, it is not clear how these groups' organizational arrangements have been changing over time, and the contribution of these changes on effectiveness of collective action. This study, therefore investigates the various organizational changes in groups and how these changes influence effectiveness of collective action of farmer groups in East Africa. Data were collected from 195 farmer groups in Kenya and Uganda through a structured questionnaire survey, supplemented by focus group discussions. Results show that changes in leadership and governance structures were more important in influencing effectiveness of groups. These include having a considerable number of leadership positions, introducing clear leadership tenure, having committees in groups, holding frequent committee meetings and putting in place more enforcement mechanisms for regulating group conduct. This study proposes adoption of effective leadership and governance structures by farmer groups to enhance effectiveness of collective action.

Farmers form and participate in groups in order to benefit socially and economically through collective activities. However, membership in groups alone is not enough to facilitate improvement of livelihoods, owing to the fact that only successful groups would be able to fully exploit their potential and meet the interests of their members. Understanding group performance is therefore a pertinent issue among social researchers and development practitioners working with farmers groups. Findings from literature indicate that scholars have measured performance of group differently, and this can be divided into three broad areas: Group performance measured by level of cohesion/group characteristics, group performance measured by outputs/benefits and group performance measured by both level of cohesion/group characteristics and outputs/benefits. The measurement of performance of groups engaging in the same activities has been much easier, however; for the groups that engage in diverse activities the measurement of their performance becomes even more complicated. This study concludes that group performance can be measured in various ways depending with what researchers and development practitioners want to investigate and achieve.