Principles and Practices for a Federal Statistical Agency: Eighth Edition (2025)
Chapter: Summary
Summary
Highlights
- Public trust in federal statistics and in the agencies that provide them is essential to a shared understanding of our economy and society.
- Articulating agreed-upon principles and practices provides a common foundation to producers and users of federal statistics.
- In 1992, the board members of the Committee on National Statistics were first tasked to issue guidance for the federal statistical system as whole.
- Five principles guide federal statistical agency programs.
- Ten practices support achievement of these principles.
- These principles and practices are largely unchanged from prior editions.
- Modest revisions in this eighth edition reflect evolution in data ecosystems and significant federal statistical policy changes observed or anticipated since the passage of the Evidence Act of 2018.
THE VALUE OF NATIONAL STATISTICS
Statistics are essential not only for policymakers and program administrators at all governmental levels, but also for individuals, households, businesses, and other organizations to make informed decisions and for scientists to add to knowledge. Even more broadly, the effective operation of a democratic system of government depends on the unhindered provision of impartial, scientifically based statistical information to its public on a wide range of issues, including employment, growth in the economy, the cost of living, crime victimization, family structure, physical and mental health, educational attainment, energy use, and the environment. Furthermore, this flow of trustworthy information is critical to shaping citizens’ understanding when confronted with today’s avalanche of statistics in traditional and social media.
A DECENTRALIZED SYSTEM
In the United States, there are more than a dozen federal statistical agencies1 whose principal function is to collect, compile, analyze, and disseminate information for such statistical uses as monitoring key economic and societal indicators, informing policy, allocating legislative seats and government funds, evaluating programs, and conducting scientific research. Federal statistical agencies often rely on data provided by other federal agencies and by state, local, and tribal governments to produce official statistics.
Statistical agencies have unique responsibilities. Although statistical agencies provide objective and impartial information that informs policymakers, they should not advocate policies or take partisan positions that would undercut public trust and the credibility of the statistics they produce. In addition, recent statutory and regulatory changes have expanded the responsibilities of federal statistical agencies and other federal agencies to include advising departments on the use of statistical data for evidence-based policymaking.
A COMMON FOUNDATION
Communicating the unique responsibilities of federal statistical agencies—and best practices on achieving them—remains essential to a
___________________
1 The United States has a decentralized statistical system. There are thirteen recognized statistical agencies and three recognized statistical units. In addition, using a threshold definition as $3 million in estimated or direct funding for statistical activities in the forthcoming or either of the past two fiscal years, there were more than 109 other agencies that conducted statistical activities in 2022 (Office of Management and Budget [OMB], 2023b, 2024c).
well-functioning statistical system. To this end, the Committee on National Statistics (CNSTAT) of the National Academies of Sciences, Engineering, and Medicine articulates five principles and 10 practices as part of its mission to provide an independent review of federal statistical activities.
From its first edition in 1992 to this eighth edition, Principles and Practices for a Federal Statistical Agency (P&P)2 is intended to support the invaluable role of relevant, credible, trusted, independent, and innovative government statistics. With modest refinement over the years, the principles and practices have endured. They have influenced federal statistical policy in substantial and lasting ways. Indeed, acts of Congress as well as regulations and statistical policy directives issued by the OMB have codified many of these principles and practices. The volume is used across federal statistical agencies and an ever-expanding community of producers and users of national statistics to introduce and support their staff in a shared culture.
Five Principles
Federal statistical agencies are coordinated by OMB and are subject to government regulations and guidance, but their missions and contributions to the public good are best seen as resting on five well-established and fundamental principles. In brief, these principles are relevance, credibility, trust, independence, and innovation. These are coequal, not ranked, and are intended to reinforce one another in practice. They are summarized below, listed in Box S-1, and described in greater detail in Chapter 3.
Principle 1: Relevance to Policy Issues and Society
Federal statistical agencies must provide objective, accurate, and timely information that is relevant to important public policy issues. They must also help ensure that these products are on the table when decisions are made. To develop relevant statistics needed by policymakers in Congress, the executive branch, and other entities, statistical agencies must have a solid understanding of the public policy issues, federal programs, and information needs in their domains. To ensure that they are providing relevant information, statistical agencies need to reach out to a wide range of their data users, including staff in their own departments and other federal departments who use their data, members of Congress and their staffs, state and local government agencies, academic researchers, businesses, and members of the general public, including local organizations. To facilitate best use of their products, statistical agencies must educate users and adapt dissemination to their users’ needs. However, statistical agencies should be
___________________
2 We refer to this report as P&P throughout the remainder of the text.
careful not to become involved with policy development or implementation, as those activities could affect their ability (or the perception of their ability) to conduct impartial and objective statistical activities.
Principle 2: Credibility Among Data Users and Stakeholders
Federal statistical agencies must have credibility with those who use their data and information. The value of statistical agencies rests fundamentally on the accuracy of their data products. Because few data users have the resources to verify the accuracy of statistical information, users rely on an agency’s reputation for disseminating high-quality, objective, and useful statistics in an impartial manner. Agencies build and maintain credibility through clear public commitments to professional practice and transparency in all that they do, including informing users of the strengths and weaknesses of their data. Agencies should communicate the “fitness for use” of their statistical products as meeting the degree of accuracy required for a particular, responsible use.
As alternative data sources are used to supplement (or sometimes replace) survey data, federal statistical agencies must continually strive for verifiable improvements in accuracy and timeliness in the estimates they produce. This can include reduction of known bias (such as from the use of administrative data blended with survey data) measured by the reduction in measures of total error (National Academies of Sciences, Engineering, and Medicine [NASEM], 2023c).
Principle 3: Trust Among the Public and Data Subjects
Federal statistical agencies must have the trust of those whose information they obtain. Because virtually every person, household, business, state or local government, and organization is the subject of some federal statistics, public trust is essential for the continued effectiveness of federal statistical agencies. Individuals and entities providing data directly or indirectly to federal statistical agencies must trust that the agency is collecting information that serves a public purpose, and that the agency will appropriately handle and protect their information. Federal statistical agencies not only have legal and ethical obligations that require them to fulfill these expectations, but they also have the obligation to effectively communicate the value of the data they collect and the methods they use for obtaining and protecting them. An effective statistical agency has policies and practices to instill the highest possible commitment to professional ethics among its staff and builds a culture of protecting the confidentiality of its data and engendering respect for those who provide data.
Principle 4: Independence from Political and Other Undue External Influence
Federal statistical agencies must be independent from political and other undue external influence in developing, producing, and disseminating statistics. Statistical agencies must be impartial and execute their missions without being subject to pressures to advance any political or personal agenda. They must avoid even the appearance that their collection, analysis, and reporting processes might be manipulated for political or other purposes or that individually identifiable data might be obtainable for nonstatistical purposes. Only in this way can statistical agencies serve as trustworthy sources of objective, relevant, accurate, and timely information. Protection from undue outside influences requires that statistical agencies have the authority to make professional decisions concerning their programs, including authority over the selection and promotion of staff; the processing, secure storage, and maintenance of data; and the timing and content of data releases, accompanying press releases, and documentation.
Principle 5: Continual Improvement and Innovation
Federal statistical agencies must continually seek to improve and innovate their processes, methods, and statistical products to better measure an ever-changing world. Federal statistical agencies and programs cannot be static but must continually work to create reliable information on new policy questions, adopt improvements in all aspects of their operations, and respond to user demands for more timely and granular information. An effective statistical agency not only seeks out and evaluates potential new data sources that could provide useful information, but also tests and implements new methods to enhance the accuracy and cost-effectiveness of its data collection, processing, and dissemination processes. It also seeks ways to reduce burden on data subjects. It works closely with data users to identify potential new and useful statistical products.
Ten Practices
To fulfill these five principles, 10 practices are essential for statistical agencies to adopt. These practices represent the ways and means of making the basic principles operational and facilitating an agency’s adherence to them. Practices 1 to 4 pertain to an agency’s operations, internally and within the federal government; practices 5 to 7 bridge internal operations and external relations with the professional statistical and research communities; and practices 8 to 10 focus externally on an agency’s key
BOX S-1
Principles and Practices
Five principles guide federal statistical agency programs:
- Relevance to Policy Issues and Society
- Credibility Among Data Users and Stakeholders
- Trust Among the Public and Data Subjects
- Independence from Political and Other Undue External Influence
- Continual Improvement and Innovation
Ten practices support achievement of these principles:
- A Clearly Defined and Well-Accepted Mission
- Necessary Authority and Procedures to Protect Independence
- Commitment to Quality and Professional Standards of Practice
- Professional Advancement of Staff
- An Active Research Program
- Strong Internal and External Evaluation Processes for an Agency’s Statistical Programs
- Coordination and Collaboration with Other Agencies
- Respect for Data Subjects and Data Holders and Protection of Their Data
- Dissemination of Statistical Products That Meet Users’ Needs
- Openness About Sources and Limitations of the Data Provided
constituents: data users, data subjects, and data holders.3 The 10 practices are listed in Box S-1 and detailed in Chapter 4.
THE CONTRIBUTION OF THIS REPORT
The preparation of this report is unique among other National Academies reports. Because P&P is intended to reflect and uphold the most salient guidance for the federal statistical system as a whole, the authoring committee is CNSTAT itself, due to its role as an independent, objective, and centralizing influence on the system. CNSTAT members include experts in statistical and computational methods, survey research, economic, social and demographic measurement, administrative law, and other relevant fields. Additionally, unlike many other National Academies reports, this report does not include recommendations. Instead, this report provides
___________________
3 The terms data subject and data holder in this edition have replaced the term data provider used in the prior edition to reflect greater precision when describing entities contributing or facilitating access to data. See glossary.
principles and practices, which serve as guidelines and not prescriptions. By adhering to the principles and following the practices, a federal statistical agency will be well positioned to provide the relevant, credible, trusted, independent, and innovative statistical information that the public requires.
Since 2001, CNSTAT has issued a new edition of P&P to coincide with the start of each presidential term. This serves two key functions: to communicate with a new audience these principles and practices, and to account for the expected, ongoing changes in society and the federal statistical system. The principles and practices described in this edition remain unchanged since the prior edition. The updates made to this volume instead consider how changes in the federal statistical system intersect with these longstanding guidelines. Throughout, the volume accounts for the wide-ranging observed and anticipated impacts of the Foundations for Evidence-Based Policymaking Act of 2018 (2019), which expanded the role of heads of the recognized statistical agencies and units in their departments; expanded the responsibilities of the Chief Statistician in coordinating expanded data sharing across agencies and access to confidential data for evidence building; and prescribed an enlarged role for federal surveys and administrative records to be used in support of sound policymaking. This edition has added a glossary and tightened terminology to improve accessibility to an ever-widening audience, and its appendices have been extensively updated and rearranged for ease of reference.
CNSTAT intends for these principles and practices to assist statistical agencies and units, the research organizations supporting them, as well as other government agencies nationwide engaged in statistical activities, to carry out their responsibilities to provide accurate, timely, relevant, and objective information for public and policy use. It also intends this report to inform decision makers, data users, and others about the characteristics of statistical agencies that enable them to serve the public good.
This page intentionally left blank.
