Experts slam Trump’s unusual new census report as a ‘hijack by the White House’

Census Bureau’s Unsigned Report on Noncitizen Voting Draws Sharp Criticism from Statisticians and Former Officials

Earthguardiansonline.com – A document posted to the U.S. Census Bureau’s website this week — one bearing no author names, no contributor credits, and no standard editorial review trail — has ignited a firestorm among statisticians, former agency leaders, and voting-rights advocates. The report asserts that more than 24,000 noncitizens cast ballots in the 2020 presidential election, a figure the White House has seized upon to revive its long-running narrative of systemic electoral fraud.

President Donald Trump amplified the finding on Truth Social on Tuesday, declaring in all caps:

“I WON THE ELECTION!”

He framed the bureau’s work as definitive proof that widespread illegality cost him the 2020 race, and he predicted the tally would “explode” once analysts finished processing the remaining 32 million voter records not yet included in the initial sweep.

Scale and Statistical Context

To grasp the magnitude of the claim, consider the numbers. Roughly 160 million ballots were cast nationwide in November 2020. Twenty-four thousand votes represent less than 0.02 percent of that total. Joe Biden carried the national popular vote by more than seven million votes — a margin roughly 290 times larger than the figure in question. The report’s findings also fall well short of Biden’s victory margins in the most tightly contested states, meaning the alleged noncitizen ballots, even taken at face value, would not alter the outcome in any state.

Moreover, because American ballots are secret, the document cannot — and structurally cannot — identify which candidate each matched voter supported. The analysis establishes only that certain names in a commercial voter file correspond to records flagged as noncitizens in federal databases; it says nothing about how those individuals voted, if they voted at all.

How the Analysis Was Conducted

The methodology, as described in the document released Tuesday, involved cross-referencing a commercial voter-registration database against federal records held by the Internal Revenue Service, the Social Security Administration, the State Department, and the Department of Homeland Security. The commercial data originated with Data-Clear, a Virginia-based company that received a $649,500 federal contract in 2024 to procure voter-registration information. Analysts then ran the Census Bureau’s record-linking system to flag individual voters whose names matched noncitizen entries in the government databases.

The report itself is labeled a “beginning analysis,” suggesting further rounds of matching are planned. Commerce Secretary Howard Lutnick, whose department supervises the Census Bureau, posted on social media:

“It’s just the beginning,”

adding that the next phase “will likely show tens of thousands of additional illegal voters.”

Why the Unsigned Format Raises Alarm

The Census Bureau has historically been governed by career statisticians and has operated at arm’s length from the White House. Publishing a statistical document without named authors or contributors is, by the agency’s own norms, highly irregular. Former senior officials say the format could signal that career staff were bypassed or sidelined in the production of the analysis.

Earlier this year, the administration placed a cohort of partisan analysts drawn from the Trump-aligned America First Policy Institute inside the Census Bureau, according to three former bureau officials who had served under Trump. AFPI did not immediately respond to a request for comment. The Census Bureau, the Commerce Department, and the White House likewise did not respond to comment requests.

Expert Reactions

John M. Abowd, who served as the bureau’s chief scientist from 2016 to 2022, told reporters that the statistical signal is too small to separate from the background noise of routine matching errors. Record-linking systems, even well-designed ones, generate false positives. The 24,000 matches constitute approximately 0.019 percent of the 128 million records examined — a rate consistent with ordinary data-quality noise.

“None of the professional researchers that I knew or worked with at the Census Bureau would make such a claim from that study,” Abowd said. “Professional researchers would have reported that result and given it some context.”

Robert Santos, who led the Census Bureau from 2022 until resigning in the opening days of Trump’s second term, was more pointed. He described the publication as an obvious “hijack by the White House.”

“I was shocked and dismayed that a document of this sort – and I refuse to call it a report – would appear on a Census Bureau website,” Santos said. “It is clearly a product of White House and Department of Commerce analysts.”

Santos identified what he called a “fatal flaw” in the underlying assumption: the analysis treats federal, state, and local voter records as perfectly accurate. In practice, voter files routinely contain outdated addresses, duplicate entries, and stale registrations. Those data-quality problems were exacerbated during the pandemic, when election offices faced shutdowns, staffing shortages, and surges of mail-in registrations that outpaced verification capacity.

Broader Political Context

The report lands amid an intensifying administration push to assemble documentary support for Trump’s longstanding assertion that the 2020 election was compromised by widespread fraud. In a primetime address last month, the president cited a trove of newly declassified documents to argue that the U.S. election system harbors gaping vulnerabilities that leave future contests exposed to foreign interference and noncitizen participation. The Census Bureau’s “beginning analysis” appears to be the latest installment in that effort — a statistical artifact deployed to lend empirical weight to a political narrative that independent audits, state-by-state recounts, and court rulings have repeatedly failed to substantiate.

For readers tracking the story, the critical questions going forward are straightforward: Will the remaining 32 million records yield the dramatic surge Trump has promised, or will they confirm what statisticians already expect — that the match rate stays within the range of ordinary data noise? And will the agency eventually publish a fully attributed, peer-reviewed version of the analysis, or will the unsigned “beginning” remain the definitive statement?