720px-Census_Bureau_seal.svg

[Image courtesy of wikimedia]

Not long after the EAC released its treasure trove of data from the 2014 election, the Census Bureau is following suit with the release of its biennial Voting and Registration Report.

Here’s the release (all emphasis mine):

The 2014 congressional election turnout rate of 41.9 percent was the lowest since the U.S. Census Bureau first began asking Americans about voting and citizenship status in 1978. The 2014 voting rate was 7.0 percentage points lower than in 1978 and down from the 45.5 percent that reported voting in the 2010 congressional election.

These statistics come from Who Votes? Congressional Elections and the American Electorate: 1978-2014, which uses data collected by the Census Bureau’s Current Population Survey. This report provides a detailed historical portrait of voters in congressional elections, and it examines voting patterns by age, race and Hispanic origin and includes a look at early and absentee voting.

The voting rates of every age group between 18 and 64 also dropped between 1978 and 2014, while the voting rate in 2014 for those over 65 was not statistically different from 1978. In 2014, the voting rate for the 65-and-older age group was 59.4 per­cent compared with 23.1 percent of 18- to 34-year-olds. For more information on voting trends by age, see Figure 4 in the report.

“In recent congressional elections, we’ve seen low levels of engagement among young people and the opposite for older Americans,” said Thom File, a Census Bureau sociologist and the report’s author. “These age differences cut across racial and ethnic groups as well. Regardless of whether we’re looking at non-Hispanic whites, non-Hispanic blacks or Hispanics, voting rates tend to increase significantly with age.”          

The report also explores early and absentee voting. In the most recent congressional election of 2014, nearly a third (31.2 percent) of all voters reported either voting early, voting by mail or using some other form of voting. This was about a threefold increase from 1996, when only 10.5 percent of voters reported voting by alternative methods.

In addition to the report, the release also includes a detailed table package. Highlights from these tables include:  

  • Voting rates for non-Hispanic whites — the largest portion of the electorate — have declined from 50.6 percent in 1978 to 45.8 percent in 2014. The group voted at a rate higher than their eligibility by 6.4 percentage points in 2014.
  • Voting rates for Hispanics declined from 1978 to 2014, dropping to a rate lower than their eligibility by 4.1 percentage points in the 2014 election. 
  • Being married with a spouse living in the household corresponded to higher voting rates (50.9 percent) in 2014, particularly in comparison with those who reported having never been married (25.9 percent).
  • In 2014, reported voting rates were also high among those with advanced degrees (62.0 percent) and those who had lived in their current home for five years or longer (57.2 percent). The top tier of the voting rate distribution also included government workers (56.5 percent) and military veterans (54.2 percent). The voting rate of government workers was not statistically different from the voting rates of those who had lived in their home for five years or longer. 

It’s important to remember (as Florida’s Michael McDonald pointed out repeatedly yesterday on Twitter) that these turnout numbers are self-reported by voters and thus tend to be higher than reported state and local results. Here’s what the Bureau has to say on page 1(!) of the report (again, emphasis mine):

The data in this report are based on responses to the November CPS Voting and Registration Supplements, which survey the civilian noninstitutionalized population in the United States. Voting estimates from the CPS and other sample surveys have historically differed from those based on administrative data, such as the official results reported by each state and disseminated collectively by the Clerk of the U.S. House of Representatives and the Federal Election Commission (FEC). In general, voting rates from the sample surveys such as the CPS are higher than official results (Bauman and Julian, 2010; De Bell, et al., 2015). Potential explanations for these differences include misreporting, problems with memory or knowledge of others’ behavior, and methodological issues related to question wording, method of survey administration, and nonresponse. Despite these issues, the Census Bureau’s November supplement to the CPS remains the most comprehensive data source available for examining the social and demographic composition of the electorate in federal elections, particularly when examining broad historical trends for subpopulations.

This Census data will continue to be examined very closely by scholars and practitioners alike for trends and insights – and if you’re looking for a single place for that kind of analysis, you can’t go wrong with McDonald’s United States Elections Project.

The summer between federal election years is always exciting because of data releases like this – the information (from both the EAC and the Census) comes while the memory of the last election is still fresh but the prospect of the next election looms larger than ever. I look forward to seeing what McDonald and others find in these files, given that it’s likely to drive not just the political narrative but also discussions about the future of the nation’s election administration system.