Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: Internet surveys can be designed so that a respondent can simply click on a link to indicate that they do not want to fill out the survey. This link can be embedded in an e-mail invite or, in online invite surveys, it can be included on the invite page. The decline option is appropriate for those respondents who are not actually the end-user and, thus cannot answer most of the questions. This option can potentially improve response rate estimates as well as provide additional information about respondents. However, there is a concern that the decline option provides an easy opt-out for legitimate respondents.

The Urban Institute’s Assessing the New Federalism (ANF) project periodically invites its public use data users to complete a customer satisfaction survey. The survey is an Internet survey that is accessed and tracked using an e-mail invite. In the past we have not offered the decline option. However, we estimate that about half of the people registered to use the ANF public use data, actually do not end up using the data. For the next round of this survey we plan to conduct an experiment offering a random half of our respondents the decline option. We plan to analyze the effect the decline option will have on the response rate and compare the results of the two groups’ survey responses. The findings from our analysis will be the core of our APPOR paper and presentation.