Online Program

Use of Paid Media to Encourage 2010 Census Participation among the Hard-to-Count

Doug Evans, George Washington University 
*A. Rupa Datta, NORC at the University of Chicago 
Ting Yan, NORC at the University of Chicago 

Keywords:

TheU.S. Bureau of the Census employed a paid advertising effort as part of its communication campaign to encourage participation in the 2010 Decennial Census. Paid advertising was targeted at a range of hard-to-count groups, including those identified by race or ethnicity, as well groups identified by age and other demographic characteristics. The 2010 Census Integrated Communications Program Evaluation (CICPE) was sponsored by the Census Bureau and conducted by NORC at the University of Chicago to evaluate the full communications effort for the Census, including the paid media campaign. The evaluation included best practices from public health and marketing evaluations of media campaigns, such as: eliciting confirmed awareness of advertisements, capturing receptivity to advertising messages, tracking changes in attitudes and beliefs addressed in advertisements, and the use of advertisement metrics such as gross ratings points to independently estimate exposure to the paid media campaign. Other relevant features include use of actual Census participation data rather than self-reported mailback status, and the availability of data from experimental variation in paid media exposure in a sub-set of sites. This paper describes the 2010 Census paid media campaign and summarizes key results regarding the overall effectiveness of the campaign in increasing mail return rates to the Census. The authors then extend the prior literature by examining the extent to which any effects of the paid media campaign appear to be driven by: - Differential effectiveness across people’s initial dispositions towardthe Census, for example, those who were ill-disposed, indifferent, or inclined to return, - Inducing changes in attitudes or beliefs targeted in advertising content, or - General receptivity to advertisements. These results can help in understanding how advertising might be helpful in boosting participation among hard-to-count groups in other survey contexts, as well as when advertising may be a less effective tool at encouraging survey participation.