The Great Federal SOGI Data Rescue

During the Trump administration’s first year, sexual orientation and gender identity (SOGI) measures were widely removed from national surveys, programmatic data, and administrative systems. These changes often happened with little transparency, accountability, or evidence-based decision-making. While the removals have largely impacted data collection and dissemination beginning in 2025, there is a risk that currently available data will be modified to remove SOGI information. Thus, the Williams Institute, ICPSR/DataLumos, Movement Advancement Project, and Data Rescue Project, have joined forces to systematically track and publicly preserve every federal dataset containing SOGI measures.

Since late 2025, the Williams Institute has been documenting the questionnaire text that asked about SOGI in each federal survey, including the year the questions were introduced. During a lively hackathon in May 2026, over 100 attendees from across the country took the following actions to preserve SOGI data:

• Confirmed whether known SOGI questions were present in current versions of datasets on agency websites
• Preserved copies of datasets in DataLumos where none were found, to safeguard the information in the event of retroactive dataset modifications

In the span of an hour, hackathon participants verified the availability of SOGI data in over 100 federal datasets on agency websites, identified approximately 40 datasets in need of rescue, and archived the missing datasets in DataLumos! A few examples of datasets that were reviewed during the hackathon include the National Ambulatory Medical Care Survey (NAMCS), the Health Information National Trends Survey (HINTS), and the National Survey of Children’s Health (NSCH). You can check out the slides from the May hackathon.

Still, many more datasets need to be verified and rescued. If you’re interested in keeping the momentum going, please join us for our next hackathon on Thursday, August 13th at 12:00-1:30 pm ET.

Register for the August 13th webinar!

No advanced technical experience is required, and attendees will receive guidance on how to locate, review, and document relevant materials. Comfort with internet research, navigating websites, and/or familiarity with downloading data is welcomed. 

Much work will be needed to restore the scientific integrity of the U.S. data ecosystem, including rebuilding trust with those most at risk of harm. However, many brilliant, dedicated people are already working together and will carry this work forward into a future we can all be proud of.  

Take care,

Lauren Bouton: boutonl@law.ucla.edu

Erin Meyer: meyerin@umich.edu