Comparing household listing techniques in a rural midwestern Vanguard Center of the National Children's Study

The National Children's Study (NCS) is a longitudinal study that will examine the influence of environmental and social factors on the health and development of 100,000 children, following them from before birth until age 21. Proposed participant recruitment methods call for locating and listing all dwelling units (DUs) located within randomly selected segments within the 105 NCS sites. One of seven Vanguard Centers of the NCS includes four rural counties that span approximately 2,500 square miles. The size of this sampling area presents unique geographic challenges. In order to determine the most efficient method for listing DUs within this large area, a study was undertaken to investigate the differences in the percent of DUs identified and the cost of four different approaches. It compared the on-site listing method of physically identifying each DU with three other methods: plat maps, postal listings, and Geographic Information Systems (GIS)/satellite imagery techniques. The on-site method had the strongest kappa (.85) in terms of identifying true DUs. There was a moderate agreement (.59) with the plat map method, fair agreement (.34) with the postal method, and only a slight agreement (.14) with the GIS/satellite imagery method. The plat map, postal listing, and GIS/satellite methods were less time-consuming than the on-site method.