Attention: You are using an outdated browser, device or you do not have the latest version of JavaScript downloaded and so this website may not work as expected. Please download the latest software or switch device to avoid further issues.
| 24 Jun 2026 | |
| Blogs |
By Annique Garnier, Senior Analyst with the Center for Data Policy at the Data Foundation
This month, the federal statistical system reached a milestone in making confidential datasets more accessible for researchers. The new Standard Application Process (SAP) Portal is now live at sap.nsf.gov. The new portal replaced the legacy ResearchDataGov.org, which closed permanently on June 16 of this year. The data catalog, the participating agencies, and the application itself are all unchanged. What changed is the platform underneath. A streamlined portal establishes a solid foundation for future improvements to accessibility and ease of use for researchers. The address changed too, and that matters just as much.
It would be easy to file a launch like this under routine IT housekeeping. That would miss what the SAP actually represents. It is one of the most concrete, usable deliverables of the Foundations for Evidence-Based Policymaking Act of 2018, better known as the Evidence Act. And its move to a .gov domain says something quietly important: secure, single-point access to the nation's confidential statistical data is now permanent public infrastructure, meant to be official, trusted, and durable.
A recommendation that became a requirement
To appreciate the moment, it helps to go back to September 2017. That is when the U.S. Commission on Evidence-Based Policymaking, a 15-member bipartisan body created by Speaker Paul Ryan and Senator Patty Murray, released its final report, The Promise of Evidence-Based Policymaking, with 22 unanimous recommendations.
One of the Evidence Commission's central findings pointed to a simple but stubborn problem. A researcher who wanted access to confidential federal data faced a different application, a different set of rules, and a different process at nearly every agency. That fragmentation was itself a barrier to evidence. Researchers spent time finding and trying to unlock sixteen front doors instead of doing the work the data could support.
The Evidence Commission's answer was a uniform process: one front door through which qualified researchers could apply for access to restricted data, rather than navigating a maze that changed agency by agency. Congress took that idea and wrote it into law. Title III of the Evidence Act, the Confidential Information Protection and Statistical Efficiency Act of 2018 (CIPSEA 2018), mandated the establishment of the SAP
What the Evidence Act requires
The new SAP Portal is not just a technical upgrade. It advances a statutorily required process. Under CIPSEA 2018, codified at 44 U.S.C. § 3583, each recognized federal statistical agency and unit must establish an identical application process. The law is specific about what "identical" has to include:
Two companion pieces complete the picture. The first is a public metadata catalog, so researchers can discover what confidential data assets exist and where to apply. You cannot request what you cannot find. The second is the tiered-access framework under 44 U.S.C. § 3582, which asks agencies to categorize the sensitivity of each data asset and match it to an appropriate level of access. Together, these provisions turn an abstract promise, "expand secure access to data for evidence," into a working system with rules, timelines, and recourse. OMB formally established the SAP through guidance (M-23-04) in 2022, and the Interagency Council on Statistical Policy has coordinated the effort ever since.
What it looks like in practice
The requirements come to life the moment you open the catalog. Every dataset is described with the same standardized metadata, which means a researcher can compare very different datasets from very different agencies on equal terms.
Take one example: the NAHMS Feedlot Study from USDA's National Agricultural Statistics Service. A researcher browsing the catalog can see the study's description and purpose, the years of data available (1994 through 2021), the smallest geographic unit offered, and the way access is provided, in this case through USDA's Virtual Data Enclave. The same entry shows the sensitivity tier, listed as Trust Level 2, whether a public-use version exists, any fees that apply, and a detail that matters a great deal to the research community: whether non-U.S. citizens may apply, and under what conditions.
That is the same structured information a researcher would find for a Census Bureau survey or a Medicaid dataset from CMS. The standardization is the real value here. It is what allows a single front door to work across sixteen recognized statistical agencies and units, each with its own history, rules, and data.
An Evidence Act milestone worth celebrating
Good data infrastructure is mostly invisible when it works. It rarely makes headlines. But it is quietly essential to the evidence that makes the government smarter, and to the researchers, evaluators, and analysts whose work depends on responsible, secure access to data the government already collects.
Congratulations to NSF, the Interagency Council on Statistical Policy, and every federal statistical agency and unit that carried this across the finish line. For those of us who follow federal data policy closely, it is satisfying to watch a recommendation from a 2017 report grow into permanent public infrastructure. This is what delivering on the Evidence Act looks like, and it is worth pausing to recognize how far the work has come.