Turnout Demographics

Overall voter turnout rates provided on this site cannot reveal who among the electorate voted. For a demographic profile of the electorate, we must turn to surveys.

Among the most widely sited surveys is the Census Bureau’s Current Population Survey, November Voting and Registration Supplement (or CPS for short). The CPS is a very large survey primarily used to calculate the nation’s unemployment rate nationally and among states. Given it’s important nature, the Census Bureau invests significant resources to make sure households respond to the survey. In the November of a federal election year the Census Bureau asks a limited number of questions about voting and registration. When cross-tabulated with the survey’s extensive demographics, the CPS provides a comprehensive snapshot of participation among various demographic groups.

There are differences between the CPS and VEP turnout rates. The CPS surveys non-institutional households, which excludes from the CPS sample military barracks, campus dorms and prisons. The first two categories are more pertinent since felony prisoners are eligible to vote in only three states – D.C., Maine, and Vermont.

Surveys have error as a consequence from random sampling and from other issues, such as who responds to a survey and the truthfulness of their responses. Pollsters have long noted that poll respondents overstate their voting participation, a phenomenon known as over-report bias. Not all people respond to a survey, a phenomenon known as non-response bias. The statistics presented here below correct the CPS for vote over-report and non-response biases by re-weighting the CPS such that state turnout rates are similar to the VEP turnout rates, as described in more detail here.