WASHINGTON, DC — Continuing to push assertions that the 2020 presidential election was compromised, Donald Trump recently asserted on social media that the Census Bureau uncovered 24,000 ballots cast by non-citizens during that election cycle.
The assertion, shared alongside a report titled “Noncitizen Voting in the 2020 Election, A Beginning Analysis” and accompanying data visualizations, quickly generated widespread debate and prompted demands for verification online.
### The Claim: 24,000 Non-Citizens Cast Ballots in 2020
Writing on Truth Social, Trump stated that the Census Bureau had initiated a comparison of 2020 voter rolls against official citizenship records.
“On the first 128,000,000 Voters, the Census Bureau proves that over 24,000 Noncitizens voted illegally!” Trump wrote, adding that a subsequent review of another 32 million voters would likely cause that figure to rise significantly. He also reiterated his call for the passage of the SAVE America Act.
Reactions to the post were sharply divided, with numerous supporters endorsing the findings while critics demanded independent fact-checking.
### Why the Claim is Misleading
Despite the presentation of data, experts and records show that Trump’s interpretation is inaccurate. The Census Bureau did not confirm that over 24,000 non-citizens voted unlawfully in 2020. Instead, the process involved matching roughly 24,000 voter records with government databases that suggested an individual might not be a citizen—a match that falls short of definitive proof.
The methodology involved cross-referencing a commercial voter list with immigration and visa databases. When Social Security numbers could not be verified directly, analysts relied on names, dates of birth, and addresses.
Individuals were flagged as potential non-citizen voters if they appeared in immigration files and the bureau lacked subsequent documentation confirming naturalization. However, election and data experts note this limitation stems from incomplete database records rather than actual evidence of unlawful voting; a lack of a citizenship record in a specific database does not mean the person is not a citizen.
David Becker, executive director and founder of the Center for Election Innovation & Research, dismissed the validity of the reported findings.
“This alleged analysis is based on Census data and fails the laugh test,” Becker noted, pointing out that the administration failed to adequately explain its methodology. He added that Census data lacks the unique identifiers necessary for reliable cross-referencing with public voter files, meaning any attempt to do so is prone to generating significant numbers of false positives.
Reference