A new study finds that America’s statistical data is under threat, not from cyber criminals but from bureaucratic mismanagement.
The nation’s data is at risk, according to a new report from the American Statistical Association published in partnership with George Mason University and supported by the Sloan Foundation. The 15,000-member organization, made up of statistics professionals from academia, business, government, and research, harshly criticizes governments’ degrading ability to capture and disseminate essential statistical information.
The Value of Federal Data
Objective information is crucial to any decision-making process. What if we didn’t know how many children are falling behind academically in the wake of Covid-19 lockdowns? What if we couldn’t track high lead levels in human blood? The American Statistical Association (ASA) warns that these things may not be far off. They say that the tide can be reversed, though, and have provided a list of 15 specific recommendations for ensuring the continued production of high-quality data for public and private use well into the future.
Federal agencies track the numbers on crime, education, health, employment, income, and many other areas. However, this information does not remain siloed within the federal government. Businesses and other private organizations rely on this impartial information to make informed decisions. Our public health, infrastructure, and economy require a steady stream of unbiased information for effective management.
A Troubling Assessment of Government Operations
Thirteen federal agencies primarily exist to create this mountain of data, which the Chief Statistician of the United States oversees. Karin Orvis currently holds that position within the Office of Management and Budget (OMB). The study evaluated each agency on five criteria: professional autonomy, support from their parent agency, budget and staffing, data quality and innovation, and user engagement. The authors stated that “similar to how engineers regularly assess America’s transportation infrastructure…the goal was to identify capabilities, threats, opportunities, and to recommend needed action.”
Their final determination? “Federal statistics are at risk.”
ASA identified three common issues across the 13 agencies, with each agency experiencing at least one of them:
- A lack of autonomy threatens the process from collection to dissemination due to the potential for partisan interference or even the inability to meaningfully deal with data quality issues.
- The parent agencies that house these data-focused agencies aren’t prioritizing support for the data mission, leaving these agencies without sufficient autonomy, staffing, or budget to do the work effectively.
- Real dollar investments in these agencies have declined in the last 15 years, exacerbated by inflation. This leads to tighter resources for the same old tasks and limits agencies’ ability to test new methods for adjusting to a rapidly changing information landscape.
The consequences of these failures manifest themselves in a challenging information environment. As forms of communication and our overall way of life changes, data collection methods for topics like the unemployment rate can become “outdated in content and methods.” Agencies are also only considering their internal uses for the data, cutting programs that outsiders like businesses or Congress have long relied upon to make decisions.
With current privacy concerns, agencies are finding it more practical to cut access to data rather than invest in stronger data protections. Finally, without the autonomy and support needed, these agencies are becoming increasingly vulnerable to politicization and biased data collection or reporting.
ASA outlines potential opportunities and provides examples of what agencies could do to mitigate these issues. One idea would be to move toward a blending strategy instead of relying solely on surveys. As Americans become increasingly averse to answering surveys, difficult questions could be answered by administrative and commercial data. Another option is increased public education on the need for information and how it can directly benefit citizens.
A Path Forward for Government Data Agencies
After producing an intensive survey of where these agencies are at and comparing it to the changing data landscape, 15 recommendations have been developed. These recommendations are aimed at the agencies and a variety of stakeholders. Grouped by agency, the recommendations target Congress, the parent agencies, the agencies themselves, and the OMB.
Recommendations for Congress include enacting legislation explicitly outlining the autonomy of data collection agencies, increasing the ability to share information between agencies, adequate funding, balancing access for all with privacy, and more direct annual monitoring. Parent agencies are again encouraged to push for autonomy, increase the timeliness of shared services, sufficiently budget and staff the agencies, and regularly meet with the agencies to get eyes on opportunities and challenges. The OMB has a role that needs to be fulfilled in regulating and disambiguating federal information agencies. The office needs to formalize the responsibilities of the agencies and their parents, develop a strategic plan across agencies, and ensure the Chief Statistician’s office has everything it needs to do the job.
The data agencies themselves aren’t off the hook, however. They need to communicate more effectively with Congress and parent agencies, stakeholders, and users. Overall, the study emphasizes that bilateral dialogue needs to be established to understand user needs and ensure communication with stakeholders across the political spectrum.
Ryan Whalen is a writer based in New York. He has served in the Army National Guard and holds a BA in History and a Master of Library and Information Science with a certificate in Data Science. He is currently finishing an MA in Public History and working with the Harbor Defense Museum at Fort Hamilton, Brooklyn.