The Impact of Prompting on Response Rates: Experience with Touchtone Reporting in the CES Program

• Activity Number: 314
• Type: Contributed
• Date/Time: Wednesday, August 14, 2002 : 10:30 AM to 12:20 PM
• Sponsor: Section on Survey Research Methods*
• Author(s): Richard Rosen*+ and Tony Gomes and Ralf Hertwig
• Affiliation(s): U.S. Bureau of Labor Statistics and U.S. Bureau of Labor Statistics and U.S. Bureau of Labor Statistics
• Keywords: Nonresponse ; Delinquency control ; Response rates

Abstract

Achieving desired response rates in a timely and cost-effective manner is a top priority in the Current Employment Statistics (CES) program. CES is a monthly survey of 250,000 business establishments that provides estimates of employment, payroll, and hours for the nation, states, and metropolitan areas. CES uses a number of collection methods. Touchtone Data Entry (TDE) is the predominant collection method, with over 60% of the sample. Under TDE, delinquent respondents receive one or two reminder messages each month. This paper measures the impact of these messages on response rates. Independent test and control samples were selected from the TDE respondent population. In one test sample, the advance notice message was withheld for a five-month period. In the other test, the nonresponse message was withheld for a five-month period. Both groups had a parallel control group for comparison. The results show that prompting respondents twice each month achieves the highest response rates. Witholding either message reduces the response rate by 13 percentage points. The paper compares response rates over the five-month period and explores mode effects associated with different prompting methods.