Linking Mailing Addresses to a Household Sampling Frame Based on Census Geography
Morton, K. B., McMichael, J. P., Cajka, J. C., Curry, R. J., Iannacchione, V., & Cunningham, D. B. (2007, May). Linking Mailing Addresses to a Household Sampling Frame Based on Census Geography. Presented at American Association for Public Opinion Research Conference, Anaheim, CA.
Oftentimes, sampling frames for household surveys are based on Census geography in order to ensure complete geographic coverage of the target population during the first stage of sample selection. Subsequently, counting and listing methods can be used to create household frames within sampled areas for the final stage of sample selection. However, these methods are often cost and time prohibitive. RTI International has conducted research to investigate potential advantages of using residential mailing lists instead of the traditional onsite enumeration methods in an area-based survey. An inherent challenge in this process is that of linking mailing addresses to a corresponding frame. For this research, a Geographic Information System (GIS) has been used to identify addresses located within the sampled block. Other challenges addressed in this paper are those related to inclusion of areas known to have poor mailing list coverage such as rural areas, areas without home delivery of mail, and areas with simplified mailing addresses. Secondary sources are often needed to supplement the residential mailing lists in these areas.