At a glance
Election integrity statistics are often discussed in broad terms, but the 2024 data points in the source set are concrete. They show a system with widespread paper records, large-scale voter participation, extensive auditing, and a mix of local procedures that still produce measurable differences worth tracking.
Here is the short version:
- Nearly 65% of the citizen voting age population participated in the 2024 general election, producing more than 158 million counted ballots (EAC 2024 EAVS).
- Over 98% of election jurisdictions used voting equipment that produced a paper ballot or auditable paper record (EAC 2024 EAVS).
- More than 211 million citizens were active registered voters, equal to 86.6% of the citizen voting age population (EAC 2024 EAVS).
- More than 770,000 individuals served as poll workers in the 2024 general election (EAC 2024 EAVS).
- Several state audit reports found very small discrepancies or very high match rates in post-election checks.
Contents
- What the national numbers say
- Registration and turnout
- How ballots were cast
- Paper records and equipment
- Administrative capacity
- What state audits found
- What the audit numbers mean together
What the national numbers say
The strongest pattern in the 2024 election integrity statistics is that the system relied heavily on paper-backed processes and post-election checking rather than a single perfect indicator of accuracy.
The national election administration survey reported that over 98% of election jurisdictions used voting equipment that produced a paper ballot or auditable paper record (EAC 2024 EAVS).
That same source set shows that nearly 93% of states, territories, and DC required voting system testing and certification by statute or rule (EAC 2024 EAVS).
Those two facts matter together. Paper records are only one part of integrity; testing, certification, and audit procedures determine whether those records can actually be checked later.
Fast facts
- Over 72% of voters cast ballots in person at a voting location (EAC 2024 EAVS).
- 35.2% voted in person before Election Day (EAC 2024 EAVS).
- 37.4% voted in person on Election Day (EAC 2024 EAVS).
- Voting by mail accounted for about 30% of total ballots cast in 2024 (EAC 2024 EAVS).
- Nearly 40% of jurisdictions used electronic poll books to assist with voting (EAC 2024 EAVS).
Registration and turnout
The registration base was large, and turnout was high.
More than 211 million citizens were active registered voters, equal to 86.6% of the citizen voting age population (EAC 2024 EAVS). Nearly 65% of the citizen voting age population participated in the 2024 general election, producing more than 158 million counted ballots (EAC 2024 EAVS).
Those two figures are useful because they separate the size of the registered electorate from the number of ballots counted. The gap between the two helps explain why election administration has to manage both access and accuracy at the same time.
Compact comparison
| Measure | 2024 figure | Source label |
|---|---|---|
| Active registered voters | More than 211 million | EAC 2024 EAVS |
| Active registered voters as share of citizen voting age population | 86.6% | EAC 2024 EAVS |
| Participation rate | Nearly 65% | EAC 2024 EAVS |
| Counted ballots | More than 158 million | EAC 2024 EAVS |
State-level records in the source set reinforce the same general picture. North Carolina had more than 7.8 million registered voters, and more than 5.7 million voters cast ballots in its 2024 general election (NCSBE Post-Election Audit Report 2024). North Carolina turnout was about 73.7% of eligible registered voters (NCSBE Post-Election Audit Report 2024).
That is a high-turnout environment, which makes post-election reconciliation and audits more demanding, not less.
How ballots were cast
The ballot-casting mix was diverse, but in-person voting still dominated.
Over 72% of voters cast ballots in person at a voting location (EAC 2024 EAVS). Within that total, 35.2% voted in person before Election Day and 37.4% voted in person on Election Day (EAC 2024 EAVS). About 30% of total ballots were cast by mail (EAC 2024 EAVS).
That distribution matters because each method adds a different operational burden.
- In-person voting depends on polling-place staffing, equipment, and check-in systems.
- Early voting spreads demand across a longer period.
- Mail voting adds ballot tracking, return processing, signature review, and deadline handling.
The source data also shows the scale of military and overseas voting. 1,327,324 UOCAVA ballots were transmitted, 806,743 were returned and submitted for counting, and the return rate was 68.4% (EAC 2024 EAVS). The U.S. total showed 905,343 counted ballots, equal to 96.3% of returned ballots, while 30,401 ballots were rejected nationwide, representing 3.7% of returned ballots (EAC 2024 EAVS).
UOCAVA snapshot
| Metric | Value | Source label |
|---|---|---|
| Ballots transmitted | 1,327,324 | EAC 2024 EAVS |
| Ballots returned and submitted for counting | 806,743 | EAC 2024 EAVS |
| Return rate | 68.4% | EAC 2024 EAVS |
| Counted ballots | 905,343 | EAC 2024 EAVS |
| Rejected ballots | 30,401 | EAC 2024 EAVS |
The same source set notes that 59.9% of overseas citizens received ballots by email, 52.1% of uniformed service members received ballots by postal mail, 27% of returned UOCAVA ballots came back through email, and 51.6% came back via postal mail (EAC 2024 EAVS).
Those differences show that election integrity is not a single workflow. It is a chain of workflows, and the reliability of the chain depends on how each channel is managed.
Paper records and equipment
The paper-trail numbers are among the clearest election integrity statistics in the dataset.
Over 98% of election jurisdictions used voting equipment that produced a paper ballot or auditable paper record (EAC 2024 EAVS). Only 80 jurisdictions in three states reported using systems without a verified paper trail, and just one jurisdiction reported using only voting systems without a verified paper trail (EAC 2024 EAVS).
That is a narrow exception set compared with the overall national footprint. It does not remove the need for oversight, but it does show that paper-based auditability is widespread.
System coverage at a glance
- Over 98% used equipment with a paper ballot or auditable paper record (EAC 2024 EAVS).
- Only 80 jurisdictions in three states reported systems without a verified paper trail (EAC 2024 EAVS).
- Just one jurisdiction reported using only systems without a verified paper trail (EAC 2024 EAVS).
- Nearly 93% required voting system testing and certification by statute or rule (EAC 2024 EAVS).
Another useful operational detail is that states reported sending nearly 40 million confirmation notices to maintain accurate voter rolls (EAC 2024 EAVS).
That number is not flashy, but it is central to election administration. Accurate rolls reduce avoidable errors, duplicate records, and administrative follow-up.
Administrative capacity
Election integrity is not only about machines and audit rules. It also depends on staffing and data-processing capacity.
The source set reports that 47.9% of jurisdictions had significant challenges recruiting and retaining poll workers, while more than 770,000 individuals served as poll workers in the 2024 general election (EAC 2024 EAVS).
Those two numbers can coexist. A large workforce can still be hard to staff if local jurisdictions struggle with retention, scheduling, or training.
State motor vehicle offices accounted for 32.2% of voter-registration transactions processed, and automatic voter registration accounted for 26.4% of voter-registration transactions processed (EAC 2024 EAVS). Those figures show that the registration pipeline is heavily shaped by systems outside the polling place itself.
Administrative workflow share
| Registration process | Share of transactions | Source label |
|---|---|---|
| State motor vehicle offices | 32.2% | EAC 2024 EAVS |
| Automatic voter registration | 26.4% | EAC 2024 EAVS |
The election system also had to manage military and overseas ballots, provisional issues, and equipment administration across a wide variety of jurisdictions. That is why the data points about paper records, confirmation notices, and poll-worker staffing belong in the same conversation.
What state audits found
This is where the statistics become especially useful. Several states reported audit results that allow direct comparison between machine counts, hand counts, and rule-based review processes.
North Carolina
North Carolina’s post-election audit report provides one of the most detailed state-level snapshots in the dataset.
- More than 5.7 million voters cast ballots in the 2024 general election (NCSBE Post-Election Audit Report 2024).
- 93 of North Carolina’s 100 county boards had finalized voter-history audits by the morning of November 26, 2024 (NCSBE Post-Election Audit Report 2024).
- 15 county boards had voter-history totals that matched ballot counts exactly (NCSBE Post-Election Audit Report 2024).
- 56 counties had voter-history differences of 5 ballots or fewer (NCSBE Post-Election Audit Report 2024).
- The combined absolute variance across finalized counties was 865 ballots (NCSBE Post-Election Audit Report 2024).
- The average variance was about 9 ballots per county (NCSBE Post-Election Audit Report 2024).
North Carolina also ran a sample hand-count audit. It selected 200 ballot groupings statewide, including 156 Election Day precincts, 28 early voting sites, and absentee-by-mail ballots in 16 counties (NCSBE Post-Election Audit Report 2024). The audit covered more than 260,800 ballots statewide (NCSBE Post-Election Audit Report 2024).
Out of 200 sampled ballot groups, 177 matched exactly between hand count and machine count, while 23 showed slight machine-versus-hand-count differences (NCSBE Post-Election Audit Report 2024).
Pennsylvania
Pennsylvania counties performed a statutorily required 2% statistical audit after the 2024 general election (Pennsylvania Department of State).
- The statewide risk-limiting audit covered 32 counties (Pennsylvania Department of State).
- It used 55 randomly selected batches of ballots (Pennsylvania Department of State).
- More than 37,000 ballots were reviewed (Pennsylvania Department of State).
- Only 6 vote discrepancies were identified (Pennsylvania Department of State).
- The largest discrepancy resulted in a 2-vote change (Pennsylvania Department of State).
That combination of sample size and small discrepancy count is the kind of result that makes an audit meaningful without implying that every ballot must be manually rechecked.
Virginia
Virginia’s 2024 audit results show two different audit styles in the same source set.
- The ballot-polling risk-limiting audit sampled 1,878 ballots from 125 of 133 localities (Virginia Department of Elections).
- The Senate RLA met the 10% risk limit in the first round (Virginia Department of Elections).
- The batch-comparison RLA for the 1st Congressional District covered 7 of 18 localities and 19 randomly selected batches (Virginia Department of Elections).
- That batch-comparison audit included 137,627 hand-counted ballots (Virginia Department of Elections).
Virginia’s figures are useful because they show that audit design changes with contest type and local structure. The integrity question is not whether one audit format exists, but whether the chosen format meets its statistical target.
Indiana
Indiana’s primary audit results are smaller in scale, but the match rates are straightforward.
- Johnson County sampled 118 ballots across three contests and achieved a 100% match rate on the 106 inspected ballots (Indiana Secretary of State).
- Kosciusko County sampled 126 ballots, inspected 120, and recorded a 100% match rate (Indiana Secretary of State).
- Grant County sampled 77 ballots, inspected 69, and recorded a 100% match rate (Indiana Secretary of State).
These are compact audit results, but they still matter because they show consistent inspection outcomes across different counties and contest sets.
Maryland
Maryland’s audit results add another useful perspective because the rules tied sample size to precinct and ballot categories.
- The 2024 manual audit required at least 2% of election-day precincts and at least 1% of early-voting and mail-in/provisional ballots, with a minimum of 15 mail-in/provisional ballots per county (Maryland State Board of Elections).
- The statewide base was 1,991 election-day precincts, so 2% rounded up to 40 precincts (Maryland State Board of Elections).
- Baltimore City found 3,881 ballots tabulated by the voting system versus 3,880 ballots by manual audit (Maryland State Board of Elections).
- Manual audits validated 72,767 votes cast for U.S. Senator (Maryland State Board of Elections).
- The automated ballot tabulation audit confirmed the accurate count of over 3 million ballots in the 2024 general election (Maryland State Board of Elections).
Ohio
Ohio’s numbers are among the most direct in the dataset.
- The state reported a 99.99% post-election audit accuracy rate for the 2024 general election (Ohio Secretary of State).
- Ohio’s 88 counties reported a 100% accuracy rate for the presidential election audit (Ohio Secretary of State).
Michigan
Michigan’s source data includes both procedural audits and a recount example.
- The 44th State Representative District recount confirmed the original winner and changed the vote differential from 61 votes to 79 votes (Michigan Bureau of Elections).
- Michigan conducted procedural audits in more than 375 precincts after the 2024 November general election (Michigan Bureau of Elections).
- The statewide ballot-audit report said its audit included EV precincts and PreAVCB precincts using the current format for the first time (Michigan Bureau of Elections).
What the audit numbers mean together
The main lesson from these election integrity statistics is not that every jurisdiction reported the same result. It is that the system repeatedly produced measurable, paper-backed, auditable outcomes across large numbers of ballots and across multiple states.
The national survey shows broad use of paper records, testing, certification, and poll-worker staffing (EAC 2024 EAVS). State audits then show what those safeguards look like in practice: exact matches in some sampled groups, small discrepancies in others, and formal risk-limiting or manual audit procedures that can be measured and reported.
That combination is more informative than any single headline percentage. It gives you the scale of participation, the shape of ballot handling, the reach of paper records, and the results of post-election checks, all in one dataset.