Disappearing, altered federal websites are a problem for everyone

The federal Centers for Disease Control’s Behavioral Risk Factor Surveillance System website, which includes a range of data sets and tools long used to map public health trends, includes this note: 

“CDC’s website is being modified to comply with President Trump’s Executive Orders.” 

What exactly has been modified? The website doesn’t specify. Methodology and data glossaries are no longer readily available. 

The CDC isn’t the only federal agency during President Donald Trump’s second term to alter public information online. A litany of federal websites vanished following Trump’s Jan. 20 inauguration only to reappear later with little to no information about any changes made. Some websites, like the Department of Justice’s database of defendants charged in the Jan. 6 storming of the Capitol, are gone altogether. 

For a data journalist, this removal and possible manipulation of federal data are concerning and frustrating because it limits the information we can use to make sense of our world. 

What exactly is data journalism? The term might confuse some people. To me it means using numbers to investigate inequity and injustice and find patterns and anomalies in an otherwise anecdotal world. 

Credible and accessible federal, state and local data make such investigations possible, allowing us to identify solutions to challenges that affect Wisconsin communities. Journalists are hardly the only people to rely on such data. Federal data sets are used by researchers, public officials and students across the world to understand our communities. 

Certain changes to government websites under a new president are relatively common, as illustrated by the End of Term web archive. The archive has, since 2008, preserved information from government websites at the end of presidential terms — collecting terabytes of information. The difference this time? The Trump administration has sought to tear apart full data sets to remove information it doesn’t like, particularly data related to diversity, equity, inclusion and accessibility. 

A stark example is the Youth Risk Behavior Survey — a nationally representative study that “measures health-related behaviors and experiences that can lead to death and disability among youth and adults.” The survey produced volumes of data, which could help communities understand how race, mental health, gender identity and sexual orientation shape health-related behaviors. The data was temporarily taken offline until a court order required the Department of Health and Human Services  to restore the website. 

A note on the website now says, in part: “This page does not reflect biological reality and therefore the Administration and this Department rejects it.” 

Still, like with other restored websites, we don’t know whether information has been scrubbed or changed to conform with the Trump administration’s worldview. We don’t know whether other data or information will change without notice. 

Thankfully, journalists and coding experts are archiving all the data they can get their hands on. Big Local News, Library Innovation Lab, Internet Archive and Data Rescue Project are among organizations making sure the public has access to such information, as is our right.

But these archivists can save only what is already available. They can’t tell us what is being removed or manipulated before data reaches the public. They can’t tell us what information is being kept secret. Americans have long disagreed on politics, and that’s OK. Partisan debate is healthy and necessary in a democracy. But partisanship is now sowing mistrust in the data we rely on to tell the American story. 

And right now? We need concrete facts more than ever.

This article first appeared on Wisconsin Watch and is republished here under a Creative Commons license.