WomanVoters

The ballot. The record. The power.

Data

Women voter turnout, and how to read the data

The short answer

The main national source on turnout by sex is a Census Bureau survey supplement, which measures reported voting rather than counted ballots. Read the denominator before comparing any two numbers, because rates based on citizens differ from rates based on all adults.

Go straight to the official source

Official source

Open the Census voting and registration tables

Detailed tables behind the reported turnout estimates, including sex breakdowns.

Open the Census voting and registration tables

Destination: census.gov Published by United States Census Bureau Link checked August 29, 2026

This link leaves WomanVoters and opens the official site. We never ask for the personal details the official system requires.

Official source

Open Census voting and registration data

Official survey estimates of reported registration and voting.

Open Census voting and registration data

Destination: census.gov Published by United States Census Bureau Link checked August 29, 2026

This link leaves WomanVoters and opens the official site. We never ask for the personal details the official system requires.

What the source actually measures

After each federal election, the Census Bureau fields a supplement to the Current Population Survey asking whether people were registered and whether they voted. The results are published as a set of detailed tables covering sex, age, race and ethnicity, educational attainment, income, employment status, disability status, tenure, registration method, voting method, and reported reasons for not voting.

The supplement is a household survey. It reflects what respondents report, and it is weighted to national and state population estimates. It is not a ballot count and it does not identify how anyone voted.

Read these five things before quoting a number

What to check Why it matters
Release and table number Tables are renumbered between releases, and a citation without a table number cannot be verified.
Denominator Total population, citizen population, and registered voters produce different rates from the same behavior.
Population covered Most tables cover the civilian noninstitutionalized population, which excludes some residents.
Margin of error Differences smaller than the published error are not findings, especially at the state level.
Election type Presidential year turnout and midterm turnout are not comparable series.

What the data cannot tell you

  • It cannot tell you how any group voted, because it does not ask for a candidate choice.
  • It cannot confirm an individual’s participation, because it is a sample.
  • It cannot explain a change on its own. A shift can come from a rule change, a competitive race, a survey change, or sampling variation.
  • It cannot support a claim that women vote as a single bloc, and neither can any other source.

How this page will change

We publish estimates only after a person reviews the latest release against the source tables and records the release name, table number, denominator, and margin of error. When a new release arrives, the current figures move to a dated historical page and the newest release becomes the headline, rather than overwriting the older analysis.

The Census voting topic page links every release in the series, including state-level tables and historical time series back to the 1960s. Our methodology page describes how we verify and label any figure we publish, and our sources directory lists the tiers we use.

Official record

What the official record shows

These are survey estimates, not counts

The Current Population Survey Voting and Registration Supplement asks a sample of households whether a person was registered and whether they voted. Election offices count ballots. The two are different measurements and they do not match exactly.

Denominators change the answer

A turnout rate can be expressed as a share of the total voting-age population, of the citizen voting-age population, or of registered voters. The same underlying behavior produces different percentages under each denominator, so a comparison is only valid when both figures use the same one.

Every estimate carries sampling error

Census releases publish standard errors and margins of error. A gap between two groups that is smaller than the margin of error is not a finding, and a state-level estimate from a national sample is far less precise than the national figure.

Perspective

Our perspective

Aggregate turnout is a starting point, not a story

Turnout numbers describe an outcome, not a motive. We use them to ask what changed in the rules, the calendar, and the administration of an election, and we avoid claims that any group votes as a bloc.

This panel states what we believe. It is labeled, it is separate from the record above, and it never changes the voting instructions on this site.

The strongest argument against our view

A reasonable objection is that self-reported voting is known to overstate turnout, since some respondents report voting when they did not, which makes the survey a weak basis for public argument. That is a real limitation. It is also why the survey is best used for comparisons across groups and over time within the same series, rather than as a substitute for official counts.

What remains uncertain

This page publishes no headline percentage. We will add a figure only after a reviewed snapshot of the latest Census release, with the release name, table number, denominator, and margin of error stated on the page.

Sources

  • United States Census Bureau

    Voting and registration detailed tables

    Detailed tables behind the reported turnout estimates, including sex breakdowns.

    Tier 3 source, official dataset. Domain census.gov. Checked August 29, 2026.

  • United States Census Bureau

    Voting and registration topic page

    Official survey estimates of reported registration and voting.

    Tier 3 source, official dataset. Domain census.gov. Checked August 29, 2026.

Next step

Open the official Census voting and registration tables

This link leaves WomanVoters and opens census.gov.

Change history

This page has not been substantively changed since it was published on August 29, 2026.

Found an error? Report it and read the correction log.

Support this project

Help cover hosting, source review, accessibility, maintenance, and independent civic education.

Choose how you want to support

Pick an amount, or choose your own on Stripe's page. Nothing is preselected. Our suggestion, if you want one, is $25.

Choose your own amount

Set the amount on Stripe before you pay.

Choose a custom amount

Every tier funds the same work. A larger amount does not buy anything a smaller one does not, and you can cancel at any time from the billing portal linked in every receipt.

  • Payments go to the independent publisher of this site. Stripe names the receiving entity on the checkout page and on your receipt.
  • Payments are not tax deductible, and do not go to a candidate, campaign, political party, or ballot committee.
  • Support does not influence our factual coverage or editorial conclusions.
  • Voter information on this site is free and stays free.

Checkout is hosted by Stripe. Amount and cadence are shown before you pay, and recurring support is never preselected. See the full support terms and how this publication is funded.