Have You Really Seen This Ad? Using Confirmed Awareness to Correct Estimates of Self-Reported Exposure

• Activity Number: 61
• Type: Topic Contributed
• Date/Time: Sunday, July 31, 2011 : 4:00 PM to 5:50 PM
• Sponsor: Section on Government Statistics
• Author(s): Rupa Datta*+ and Peter Hepburn and Ting Yan and Doug Evans
• Companies: NORC at the The University of Chicago and NORC at the The University of Chicago and NORC at the The University of Chicago and The George Washington University
• Keywords: 2010 Census ; measurement error ; campaign evaluation ; paid media ; self-report

Abstract

The 2010 Census Integrated Communications Program Evaluation (CICPE) is conducted by NORC at the University of Chicago to evaluate the effectiveness of the communications campaign to promote participation in the 2010 Decennial Census. Critical to the evaluation of the campaign is public awareness of and exposure to the campaign. Survey respondents were asked whether they had seen or heard a particular Census advertisement. Such self-reported exposure is subject to measurement error: respondents may report exposure because of social desirability or selective attention bias or because they confused the target ad with other ads. To investigate possible measurement error, we employ two validation techniques: 1) confirmed awareness, through which respondents' self-reported awareness can be confirmed as either true or false awareness; and 2)mode-specific cueing, which compares recognition of the ad to the self-reported exposure (which is based on recall). This paper uses these two techniques to correct our estimates of self-reported exposure. This paper investigates the use of these two techniques to correct our estimates of self-reported exposure.