Over the last year exercises termed participatory poverty assessments have been carried out as part of the process of preparing World Bank Country Poverty Assessments in a number of countries. In Ghana, Zambia and Kenya such exercises have been carried out in rural and urban areas using methods based on the RRA/PRA 'family'. The article questions some of the assumptions underlying the methods, drawing on experiences in Ghana and Zambia. It argues that assumptions of community, mutual knowledge and homogeneity in livelihood patterns derive from the rural-based traditions of the RRA/PRA approach and are not relevant to an urban context.
These observations will be of interest to researchers and others using participatory methods in an urban context.