Authors: Darby Steiger & Chris Jackson, SSRS
Standard Deviations blog posts represent the views of the author/organization, but not necessarily those of The Census Project.
Much of the conversation about the disruptions to the federal statistical system over the past year and a half has focused, understandably, on the impacts on the agencies themselves: staffing levels, budget cuts, methodology changes, and the fate of individual data products. However, the users on the other end of that system, the researchers, advocates, planners, and analysts who rely on federal data to do their jobs, are just as important to be listening to.
In early 2026, we surveyed more than 500 users of federal statistical data, including those in academia, nonprofit and advocacy organizations, government agencies, and the private sector, to ask a direct question, “how have changes at federal statistical agencies affected your ability to do your work?” The results, published in our report Voices from the Data Community, offer a clear look at this question. We found several dominant themes.
The disruption is nearly universal. Ninety-three percent of respondents told us that changes at federal statistical agencies have affected their work, and 54 percent characterized the impact as major. This held across sectors, users in academia, nonprofits, and local government reported similarly high rates of disruption. Respondents described websites going offline without warning, restricted-use data approvals stretching from days to months, and staffing cuts that left them without anyone to call when something broke.
Workarounds exist, but data users are clear that they are not solutions. We asked how people were coping with the changes, and answers ranged from relying on older archived datasets to turning to state and local data to leaning on private-sector sources. But almost universally, respondents told us these substitutes fall short. State and local data doesn’t aggregate cleanly to national estimates. Private data can be inconsistent, incomplete, or costly. Archived data goes stale. As one respondent from an academic institution put it, the existence of gold-standard federal statistics is essential and that private data is helpful only in the context of benchmarks from official statistics.
The workforce crisis is the throughline. When we asked respondents to describe specific examples of disruption, the pattern that emerged most consistently was not about any single dataset or product, it was about the people. Reduced staffing at federal statistical agencies has meant longer approval timelines, fewer answers to technical questions, and a loss of institutional knowledge that respondents worry could take years, if not decades, to rebuild. This is worth underscoring for a policy audience: the health of the federal statistical system is inseparable from the health of its workforce.
Concern about the future is high and broad-based. Looking ahead, nearly 80 percent of respondents told us they are very concerned that federal data collections will simply stop. More than 70 percent are worried about the erosion of public trust in federal statistics, a concern based in reality, given that our companion general-population polling found trust in federal statistics falling from 53 percent to 47 percent between June 2025 and March 2026.(https://ssrs.com/insights/the-publics-views-on-the-value-of-federal-statistics-2026/)
What can be done. These respondents are clear, the most important fix is securing the scientific independence and necessary financial support for the federal statistical agencies. All other solutions are essentially tourniquets on the problem.
But we do identify four short-term, specific recommendations for the civil society and philanthropic community: 1) build a curated, searchable index of existing alternative data tools so users don’t have to discover them by word of mouth; 2) establish a proactive communications function to push information about emerging resources to the dispersed data user community; 3) create a structured effort to document and archive the expertise of departing federal staff before that institutional knowledge disappears entirely; and 4) identify the most consequential data gaps, such as discontinued NSF-supported research or the removal of sexual orientation and gender identity data, where targeted private data collection could help preserve the trendlines over a short period.
However, none of these recommendations are a substitute for a fully funded, fully staffed federal statistical system. These are stopgap solutions, offered in the spirit that this community has approached the moment all along: pragmatic, urgent, and clear-eyed about what’s actually at stake.
The full report, Voices from the Data Community: How 2025 Has Impacted Public Data Users, and the one-page audience snapshots are available at ssrs.com/emerge-initiative.
