Nearly 400 Federal Datasets Deleted or Altered, Guardian Finds

Nearly 400 federal datasets covering public health, education and other government functions have been deleted or altered since the start of President Trump's second term, according to an analysis published by The Guardian. The newspaper's count could not be independently verified against a dataset-by-dataset list, which was not included in the material reviewed for this report.

Nearly 400 US datasets have been deleted or altered since January 2025, touching public health, education and other federal records, The Guardian reported.

How many federal datasets does the analysis identify?

The Guardian's count stands at nearly 400 US government datasets deleted or altered since the Trump administration returned to office, the paper reported. The report does not break the figure down agency by agency in the summary reviewed for this story, and HTT News could not confirm from available material how many datasets were fully deleted versus edited, or which federal agency hosted each one.

What government functions do the affected datasets cover?

The Guardian's reporting ties the changes to public health and education records, among other categories, according to the outlet's own framing of its findings. Federal datasets in those categories typically track measures such as disease surveillance figures, school enrollment counts and other statistics that state health departments, universities and newsrooms draw on for independent analysis. The specific datasets named in the Guardian's reporting were not detailed in the source material available for this article.

What does the Trump administration say about the changes?

No on-the-record response from the Trump administration to The Guardian's findings appears in the material reviewed for this report. HTT News will update this brief if the White House, the Office of Management and Budget or individual federal agencies name the datasets in question or dispute the Guardian's count.

Why do public health and education researchers depend on these datasets?

Federal statistical agencies and health and education departments publish datasets that outside researchers, state governments and journalists use because the federal government is often the only entity with the legal authority and scale to collect certain figures nationwide, such as school-level enrollment data or disease case reports. When a dataset is deleted, removed from a public portal or altered without a public changelog, outside researchers can lose the ability to reproduce prior analysis or compare current numbers against historical baselines.

What is the timeline of the changes?

• January 2025: The second Trump administration takes office.
• 2025–2026: The period over which The Guardian says the nearly 400 dataset deletions or alterations occurred.
• October 2026: The Guardian publishes its tally, the basis for this report.

HTT News does not have access to a month-by-month breakdown of when individual datasets went offline or were edited.

What should researchers and the public watch next?

Readers tracking a specific federal dataset can check whether an agency's data portal lists a last-updated date or a removal notice, and can file a Freedom of Information Act request with the hosting agency if a dataset cannot be located. Journalists and advocacy groups frequently compare current agency web pages against archived snapshots to document when a file changed. HTT News has not identified, in the material available for this report, a single consolidated list naming all roughly 400 datasets The Guardian counted; readers seeking agency-by-agency detail should consult The Guardian's reporting directly.

What does the Guardian's methodology disclose?

The material reviewed for this report does not include a detailed methodology section from The Guardian explaining how reporters identified the nearly 400 datasets, whether they relied on automated web-crawling tools, agency records requests or manual review of federal data portals. HTT News could not confirm whether the count includes datasets that were temporarily taken offline and later restored, or only those still missing or altered as of publication. Readers seeking that level of detail should consult the Guardian's own account of its reporting process.

How does this fit into broader concerns about federal data access?

Public health researchers, universities and state agencies have raised concerns in past years about the durability of federal datasets that underpin disease tracking, education statistics and other recurring reports, though the source material for this story does not name specific prior incidents or compare the current tally to any historical baseline. HTT News has not identified, in the material available, any independent tally from a second news organization, academic group or government watchdog that corroborates or disputes the Guardian's count of nearly 400 datasets. Any such corroboration, if it emerges, would strengthen confidence in the scope of the changes The Guardian describes.