Across settings, it has been shown that the co-residential household is an insufficient measure of family structure and support. However, it continues to be the primary means of population data collection. To address this problem, we have developed a new instrument – Kinship Support Tree (KST) – to collect kinship structure and support data on residential and non-residential kin and tested it on a sample of 462 single mothers and their children in a slum community in Nairobi, Kenya. This instrument is unique in four important ways: 1) it is not limited to the co-residential household; 2) it distinguishes potential from functional kin; 3) it incorporates multiple geospatial measures; and 4) it collects data on kin relationships from the perspective of children. In this paper, we describe the KST instrument, assess the data collected in comparison to data from household rosters, and consider the challenges and feasibility of administration of the KST.