Improving Coverage of an Address-Based Sampling Frame for the National Children's Study, Los Angeles County

Researchers have shown that address-based sampling (ABS) frames for in-person surveys suffer from undercoverage of the household population (Iannacchione 2011). To bridge this coverage gap, RTI International developed the Check for Housing Units Missed, or CHUM, (McMichael et. al. 2008) frame supplementation procedure for in-person surveys that use ABS as a sampling frame. Similar to the half-open interval (HOI) often used for field enumerated frames, the CHUM is susceptible to errors in implementation by field staff. A recent evaluation of the HOI found field staff largely failed to implement the HOI field procedures properly (Eckman & O'Muircheartaigh 2011), suggesting that the performance of field staff could impact the CHUM process as well. As part of its sampling activities, the National Children's Study used CHUM to assess ABS coverage in Los Angeles County. This CHUM process, which required specific field staff training, provided several key lessons learned. For example, we found that several factors influenced how the CHUM was implemented including: • Type and nature of training; • Type of field staff used; and • Field support provided. This paper describes these findings, along with the methods used to effectively implement the CHUM process. We also discuss the results of the CHUM process, including the improvement in sample coverage.