Given the dual purposes of this project—which allows both an in-depth examination of many themes related to civic engagement in the U.S. and a wide-ranging perspective that allows for comparisons to over 20 countries in Europe—the U.S. CID survey has yielded two separate datasets.

The first consists of the full U.S.-only survey, with responses to all of the questions.¹  In order to facilitate and encourage the common use of several key variables, and to help individual users to avoid having to “reinvent the wheel” by creating certain scales and indices, the released dataset also includes the following constructed variables at the end of the variable list:  generalized trust, political action, party identification, participation in voluntary organizations, citizenship norms, the diversity of social networks, racial prejudice/negative stereotypes, national pride, attitudes toward immigrants, and demographic factors.  These additional variables are explained in the Appendix to the Codebook.

The second dataset includes the U.S. alongside the 22 countries that were included in the 2002 ESS, but it only contains the common questions from the U.S. CID and ESS, without the constructed variables.²

Publications based on either of these datasets should include a version of this acknowledgement statement.

Users who are interested in receiving a restricted version of the U.S.-only dataset that includes geo-codes for all respondents can contact the project's director to receive a confidentiality agreement that must be completed before receiving the data.


¹ Note that the weight for the U.S.-only dataset is set to "dweight," but users may want to change the weight to NATWT or NATWTB (see the explanation of weighting options).

² Note that the weight for the merged dataset is not set by default.  Users should weight cases by "dweight" before proceeding with analysis.