Voting and Registration data have been collected biennially in the November Current Population Survey (CPS) since 1964. The statistics presented are based on replies to survey inquiries about whether individuals were registered and/or voted in specific national elections. For the purpose of these estimates, election types are considered to be either Congressional (e.g. 2002, 2006, etc.) or Presidential (e.g. 2004, 2008, etc.).
People who are not United States citizens are not eligible to vote. The voting-age population also includes a considerable number of people who cannot register to vote despite meeting citizen and age requirements. Some people are not permitted to vote because they have been committed to the penal system, mental hospitals, or other institutions, or because they fail to meet state and local resident requirements for various reasons. The eligibility to register is governed by state laws legalities that differ from one another in many respects.
Registration is the act of qualifying to vote by formally enrolling on an official list of voters. People who have moved to another election district must take steps to have their names placed on the voting rolls in their new place of residence. The state of North Dakota has no formal registration requirement voters merely present themselves at the polling place on election day with proof that they are of age and have met the appropriate residence requirements. Therefore, in North Dakota, people who are citizens and of voting age (and who meet the residence requirement), are automatically considered registered.
Over the years, changes have been made to the Voting and Registration supplement. The only constant is that in all iterations of the survey a separate question has been included regarding both voting and registration. In some years (1982, 84, 86, 90) these were the only questions asked. In other years, additional questions were included. For example, in 1984, respondents were asked a total of six questions, two of which concerned the time of day that a respondent voted. Similarly, in 1988 respondents were asked if they had registered for that election specifically.
The Voting and Registration supplement has remained relatively consistent since 1996. In that year, respondents were asked specifically whether or not they registered to vote after January 1st, 1995. This allowed analysts to directly assess the influence of the National Voter Registration Act (or The Motor Voter Act). In 2004, the Census Bureau stopped asking specifically about this date, but the supplement continues to gather information on whether respondents were registered to vote and by what means this registration occurred.
In recent years, voter-participation data were derived from replies to the following questions. Voting age citizens were asked:
"In any election some people are not able to vote because they are sick or busy, or have some other reason, and others do not want to vote. Did (this person) vote in the election held on November (date varies)?"
Respondents were classified as either "voted" or "did not vote." In most tables, this "did not vote" category includes those who reported "did not vote" or "do not know," as well as noncitizens and non-respondents. The data on registration were obtained by asking the following question to those who reported they did not vote:
"Was (this person) registered to vote in the November (date varies) election?"
It was assumed that those who reported voting were also registered. Therefore, the total registered population was derived by combining the number of people who voted with the number of non-voters who reported being registered.