Determining the Effectiveness of Multiple Nonresponse Followup Contact Attempts on Response and Data Quality

• Activity Number: 21
• Type: Topic Contributed
• Date/Time: Sunday, August 7, 2005 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Jennifer Tancreto*+ and Michael Bentley
• Companies: U.S. Census Bureau and U.S. Census Bureau
• Keywords: Nonresponse Followup ; contact attempts ; Census 2000

Abstract

Nonresponse Followup (NRFU) is one of the most expensive operations in a decennial census. The number of contacts enumerators attempt to collect data for nonresponding households in NRFU is one area that could be considered in an effort to cut costs. Official Census 2000 enumeration procedures specify that field staff make up to six contact attempts to collect data from households that failed to return their census forms. We implemented an ad hoc research study to determine if we could feasibly use fewer than six contact attempts to collect data from these households without impacting response or data quality. The results suggest enumerators rely on the sixth contact to make last-chance efforts to find a respondent, resulting in a significant increase in the interview rate at contact 6 compared to contact 5. In this sense, the sixth contact appears to be critical in achieving high response; however, we expect this gain could be achieved at an earlier contact set as a limit. Hence, we recommend designing a split-panel experiment, where randomly grouped sets of enumerators are given different limits on the number of contact attempts permitted.