Voter Turnout in American Elections Since 2000

A healthy democracy depends on widespread participation in elections. When more citizens vote, our elected representatives are more likely to truly reflect the people they’re meant to serve. In contrast, low voter turnout can skew our democracy toward the interests of a small segment of the population.

All sorts of factors influence voter turnout. It changes over time and varies by state. Turnout rises and falls depending on the type of election, the candidates and offices on the ballot, and the issues in play at any given time. And turnout can be measured in more than one way.

Use this first-of-its-kind resource to explore voter turnout in every state since 2000, for both primary and general elections.

In our decentralized system, elections are run by the states. And there are many variables that create distinct voting behaviors across the states. State-specific voting laws can influence participation in all elections. In primaries, some states allow all voters to participate, whereas others allow only registered party members. In general elections, turnout rates are often dependent on whether federal, state, and local elections are scheduled on the same day.

Demographic characteristics such as the age, income, education level, and racial makeup of the population also contribute to varying turnout rates. So does the political culture of a state—how deeply ingrained voting and civic participation are in the state’s identity.

Turnout in the United States has historically been lower than in other advanced democracies. However, it’s picked up since 2000 in both presidential and midterm elections. Despite occasional dips in turnout, this upward trend reflects growing engagement by the electorate in American elections and democracy.

In presidential election years since 2000, national primary voter turnout has varied between 18 and 29 percent, and general election turnout has fluctuated between 56 and 68 percent. In midterm years, primary turnout is about the same, but tends to be lower for the general.

In U.S. elections since 2000, average turnout for general elections is more than twice as high as primary turnout.

American voters may see general elections as more significant, but primary elections are a key part of the democratic process, enabling engaged voters to choose which candidates they want to send to the general election. Low participation in primaries means a smaller subset of voters plays an outsized role in shaping our elections—and they tend to be the most passionate party supporters.

Participation tends to fall off as voters move down the ballot. This is called roll-off: Some voters make a choice for president or governor — often the highest profile races in any given year — then leave other sections blank. And positions further down the ballot hold considerable influence over elections and government, especially local government.

Completing the entire ballot gives voters a say in all levels of government. Voting in all races helps maintain a balanced and effective government at every level.

To calculate voter turnout rates in this report, we relied on data from a variety of government sources. A majority of the data came from the websites and databases of secretaries of state — in most states, the secretary is the chief election official. State government websites archive extensive election-related data, including results, voter turnout, and registration figures. When historical data were missing from state sites, we obtained it by contacting the states’ election data administrators.

Occasionally, states providing multiple sources of data from different points in the counting and certification process led to discrepancies. For example, some states made public multiple documents – some that reported certified vote totals, while others reported preliminary or recount totals. In such cases, we used the totals from the most recent document. When historical data was unavailable from a state source, we used information from the U.S. Election Assistance Commission (EAC), election statistics from the U.S. House of Representatives, and the Book of States. In some cases, the required data was not available in any source. Similarly, states that hold caucuses do not report turnout.

The methodology for reporting voter turnout varies nationwide. Most states calculate it as a percentage of registered voters, while others use the voting-age population as the denominator. To facilitate cross-state and over-time comparisons, we collected four different bases for comparison. We collect these measures annually to account for demographic changes, such as people moving, dying, and registering to vote, ensuring our data remains accurate and up to date. Users can choose among them to better understand how different bases affect reported turnout rates:

1. Total number of registered voters: This figure includes all individuals listed as active and inactive voters on the state’s registration rolls.
2. Voting Age Population (VAP): This is an estimate made by the Census Bureau of all the individuals in a state who are 18 or older, thus legally old enough to vote. VAP does not account for age-eligible people who are disqualified for other reasons. It is the default measure used on the site.
3. Voting Eligible Population (VEP): Another estimate, developed by Michael McDonald of the University of Florida, refines the VAP by excluding those who are not eligible to vote, such as non-citizens or those disenfranchised for legal reasons.
4. Citizen Voting Age Population (CVAP): The CVAP is a five-year estimate of the number of individuals within a state who are eligible to vote, accounting for citizenship status. The Census Bureau calculates and updates the CVAP annually.

Sources for these figures include secretary of state websites, the EAC, the U.S. Elections Project led by Michael McDonald at the University of Florida, and the U.S. Census Bureau.