Skip Main Navigation Skip To Navigation Content

Information Quality

Skip top of page navigation
Census.gov Information Quality Main Statistical Quality Standards › Preface

Preface

1. Introduction

Purpose

This document specifies the statistical quality standards for the U.S. Census Bureau. As the largest statistical agency of the federal government, the Census Bureau strives to serve as the leading source of quality data about the nationís people and economy. The Census Bureau has developed these standards to promote quality in its information products and the processes that generate them. These standards provide a means to ensure consistency in the processes of all the Census Bureauís program areas, from planning through dissemination. By following these standards, the Census Bureauís employees and contractors will ensure the utility, objectivity, and integrity of the statistical information provided by the Census Bureau to Congress, to federal policy makers, to sponsors, and to the public.

Background

In 2002, the United States Office of Management and Budget (OMB) issued Information Quality Guidelines (OMB, Guidelines for Ensuring and Maximizing the Quality, Objectivity, Utility, and Integrity of Information Disseminated by Federal Agencies, February 22, 2002, 67 FR 8452-8460), directing all federal agencies to develop their own information quality guidelines. In October 2002, the Census Bureau issued its information quality guidelines (U.S. Census Bureau, U.S. Census Bureau Section 515 Information Quality Guidelines, 2002). These guidelines established a standard of quality for the Census Bureau and incorporated the information quality guidelines of the OMB and the Department of Commerce, the Census Bureau’s parent agency.

Following the OMB’s information quality guidelines, the Census Bureau defines information quality as an encompassing term comprising utility, objectivity, and integrity. Our definition of information quality is the foundation for these standards.

Utility refers to the usefulness of the information for its intended users. We assess the usefulness of our information products from the perspective of policy makers, subject matter users, researchers, and the public. We achieve utility by continual assessment of customers’ information needs, anticipation of emerging requirements, and development of new products and services.

  • The statistical quality standards related to utility include: Planning a Data Program (A1), Developing Data Collection Instruments and Supporting Materials (A2), Developing and Implementing a Sample Design (A3), Acquiring and Using Administrative Records (B2), Reviewing Information Products (E3), Releasing Information Products (F1), and Providing Documentation to Support Transparency in Information Products (F2).

Objectivity focuses on whether information is accurate, reliable, and unbiased, and is presented in an accurate, clear, complete, and unbiased manner.  Objectivity involves both the content of the information and the presentation of the information.  It requires complete, accurate, and easily understood documentation of the sources of the information, with a description of the sources of errors that may affect the quality of the data, when appropriate.

  • The statistical quality standards related to objectivity include: Developing Data Collection Instruments and Supporting Materials (A2), Developing and Implementing a Sample Design (A3), Establishing and Implementing Data Collection Methods (B1), Acquiring and Using Administrative Records (B2), Capturing Data (C1), Editing and Imputing Data (C2), Coding Data (C3), Linking Data from Multiple Sources (C4), Producing Direct Estimates from Samples (D1), Producing Estimates from Models (D2), Producing Measures and Indicators of Nonsampling Error (D3), Analyzing Data (E1), Reporting Results (E2), Reviewing Information Products (E3), Releasing Information Products (F1), Providing Documentation to Support Transparency in Information Products (F2), Addressing Information Quality Complaints (F3), and Managing Data and Documents (S2).

Integrity refers to the security of information – protection of the information from unauthorized access or revision, to ensure that the information is not compromised through corruption or falsification. Several federal statutes and Census Bureau policies govern the protection of information, most notably Title 13 and Title 26.

  • Protecting Confidentiality (S1) directly addresses issues concerning the integrity of the data. All the statistical quality standards contain requirements for protecting information from unauthorized access or release.

In September 2006, the OMB issued Standards and Guidelines for Statistical Surveys, which specify requirements for federal statistical agencies to ensure that their information products satisfy the information quality guidelines. The OMB standards are not intended to describe all the efforts that an agency may undertake to ensure the quality of its information. These Census Bureau statistical quality standards provide additional guidance that focuses on the Census Bureau’s statistical programs and activities and that addresses the Census Bureau’s unique methodological and operational issues.

2. Scope

The Census Bureau’s statistical quality standards apply to all information products released by the Census Bureau and the activities that generate those products, including products released to the public, sponsors, joint partners, or other customers.  All Census Bureau employees and Special Sworn Status individuals must comply with these standards; this includes contractors and other individuals who receive Census Bureau funding to develop and release Census Bureau information products.

The Census Bureau often conducts data collections and performs associated work for sponsoring agencies on a reimbursable basis. The work performed by the Census Bureau under such contracts is in the scope of these statistical quality standards, whether performed under Title 13, Title 15, or another authorization. If a sponsor’s requirements or funding constraints result in noncompliance with these standards, the Census Bureau’s manager for the program must obtain a waiver, except where noted in the standards.

For the purposes of these standards, information products include printed, electronic, or digital formats (e.g., Web, CD, DVD, and tape) of: news releases; Census Bureau publications; working papers (including technical papers or reports); professional papers (including journal articles, book chapters, conference papers, poster sessions, and written discussant comments); abstracts; research reports used to guide decisions about Census Bureau programs; public presentations at external events (e.g., seminars or conferences); handouts for presentations; tabulations and custom tabulations; public-use data files; statistical graphs, figures, and maps; and the documentation disseminated with these information products.

Exclusions to the Scope

None of the following exclusions apply to Statistical Quality Standard S1, Protecting Confidentiality, or the requirements for protecting confidentiality in the individual standards.

These standards do not apply to:

  • Information products intended for internal Census Bureau use that are not intended for public dissemination.
  • Information products delivered to agencies within the Department of Commerce for their internal use.
  • Internal procedural or policy manuals prepared for the management of the Census Bureau and the Department of Commerce that are not intended for public dissemination.
  • Information products that result from the Census Bureau’s administrative or management processes.
  • Information products released in response to a Freedom of Information Act request.
  • Documents intended only for communications between agencies, within agencies, or with individuals outside the Census Bureau if the documents contain no data and do not discuss analyses or methodological information.
  • Informal communications between Census Bureau employees and colleagues in other organizations that do not disseminate Census Bureau data or results based on Census Bureau data.
  • Information products delivered to sponsors or oversight agencies, including the Congress, relating to the management of Census Bureau programs.
  • Information products authored by external researchers at the Census Bureau’s Research Data Centers.
  • Information products that use Census Bureau data and are authored by Special Sworn Status individuals employed by other federal agencies or organizations for their agencies (e.g., SSA, GAO, and CBO).
  • Information products generated by other agencies or organizations to which the Census Bureau has given only technical assistance or training.  However, Census Bureau staff providing such assistance should consider these standards as guidelines.
  • Information products developed from surveys intended to measure Census Bureau customers’ or users’ satisfaction with Census Bureau products or to measure Census Bureau employees’ job satisfaction. However, any public release of results of such surveys must explain that they do not meet the Census Bureau’s statistical quality standards because the respondents are self-selected and may not be representative of all customers, all users, or all employees.
  • Communications released via social media. Social media must not be used to disseminate data or statistical analyses not previously cleared for external release. Such communications must follow the Census Bureau’s Policies and Procedures Governing the Use of Social Media.

The scope statements of the individual standards provide additional information to clarify the scope and to list exclusions specific to each standard.

3. Responsibilities

All Census Bureau employees and Special Sworn Status individuals are responsible for following the Census Bureau’s statistical quality standards in their work to develop, deliver, and release information products.

Responsibilities of the Program Areas and the Supporting Directorates and Divisions

Divisions and offices within the Economic Programs, Demographic Programs, and Decennial Census plan, process, analyze, and disseminate data. The Census Bureau’s Center for Statistical Research and Methodology supports all three directorates in areas of statistical, methodological, behavioral, and technological research and development. The Field Operations Directorate and Information Technology Directorate collect, transmit, and process data for demographic household surveys, the Decennial Census, the Economic Census and surveys, and the Government Census and surveys. The Census Bureau’s other directorates and divisions provide various types of administrative, logistical, and strategic support to the program areas.

The responsibilities of the program areas and the supporting directorates and divisions with respect to these statistical quality standards include:

  • Ensuring that the necessary resources are available to comply with the statistical quality standards.
  • Implementing and verifying compliance with the statistical quality standards.
  • Guidance on implementing the standards and verifying compliance can be obtained from the program area’s Methodology and Standards (M&S) Council representative as shown in Table 1.

    Table 1.  M&S Council Representatives

    Program Directorate

    M&S Council Representative

    Decennial Census Directorate

    Demographic Programs Directorate

    Economic Programs Directorate

     

    All other directorates

    Chief, Decennial Statistical Studies Division

    Chief, Demographic Statistical Methods Division

    Chief, Office of Statistical Methods and Research for Economic Programs

    Chief, Center for Statistical Research and Methodology


  • Reporting situations where requirements of the standards might need revision (e.g., a program’s processes or products may have changed so that some requirements of the statistical quality standards may also need to be revised).
  • Following the procedure to obtain a waiver if unable to comply with one or more of the statistical quality standards.

Responsibilities of the Methodology and Standards Council

The Census Bureau’s M&S Council consists of the division and office chiefs of the statistical methodology groups in the various program areas. The Council advises the Census Bureau’s Program Associate Directors on policy and issues affecting research and methodology for Census Bureau programs. The Council also ensures the use of sound statistical methods and practices, and facilitates communication and coordination of statistical methodology and research throughout the Census Bureau and the broader statistical community.

The responsibilities of the M&S Council with respect to these statistical quality standards include:

  • Promoting awareness of and compliance with the Census Bureau’s statistical quality standards.
  • Reviewing waiver requests and forwarding their recommendation for approval or denial of the waiver to the Program Associate Director.
  • Conducting periodic reviews and evaluations of the standards to study how well the standards are working and to identify difficulties in implementation.
  • Maintaining an archive of evaluation findings, waiver requests, and suggestions for improvement to inform future revisions of the Census Bureau’s statistical quality standards.
  • Updating the standards as needed.

The responsibilities of the individual M&S Council members for their directorates (See Table 1.) include:

  • Provide guidance on interpreting the standards to the programs in their directorates and to directorates that participate in conducting and implementing their programs (e.g., the Field Operations Directorate).
  • Provide assistance in implementing and verifying compliance with the standards to the programs in their directorates and to directorates that participate in conducting and implementing their programs (e.g., the Field Operations Directorate).

4. Interpreting and Using the Standards

The complete set of statistical quality standards includes process standards (designated with “A” through “F”) and supporting standards (designated with “S”). The process standards are organized according to the different processes associated with developing and releasing information products. The organizational framework for these process standards is:

  1. Planning and Development
  2. Collecting and Acquiring Data
  3. Capture and Processing Data
  4. Producing Estimates and Measures
  5. Analyzing Data and Reporting Results
  6. Releasing Information

The supporting standards address issues that cut across all the process standards. The two supporting standards are S1, Protecting Confidentiality, and S2, Managing Data and Documents.

The standards are written at a broad level of detail, to apply to all the Census Bureau’s programs and products. They describe what is required and do not delineate procedures for how to satisfy the requirements. Each standard has a list of key terms that are used in the standard. These terms are defined in the glossary to provide clarification on their use in relation to these standards.

To help managers interpret the requirements of the standards, examples are often provided. These examples are intended to aid the program manager in understanding the requirements and to provide guidance on the types of actions that may be useful in satisfying the requirements. It is important to note that the examples listed under a requirement are not all-inclusive; nor will every example apply to every program or product. Finally, there may be more than one acceptable way to comply with a requirement. That is, several equally acceptable actions might be performed to comply with a requirement, rather than only one unique set of actions.

Program managers must use their judgment to determine which actions must be performed for their program to comply with a requirement. The program manager is expected to carry out all the actions needed to comply with a requirement. This may include performing activities not listed in the examples. The expectation is that program managers will balance the importance of the information product and the size of the project with the constraints of budget, schedule, and resources when determining how to comply with the requirements.

If the program manager believes it is not feasible to comply with a requirement, the program manager must request a waiver. The Waiver Procedure provides a standard mechanism to exempt a program from compliance with a statistical quality standard when such an exemption is warranted. The Waiver Procedure also promotes proper management and control in implementing the standards. Finally, the Waiver Procedure ensures that appropriate documentation of exceptions to the standards is generated and maintained to inform future revisions of the statistical quality standards.

5. History of the Development of the Standards

The Census Bureau has a long history of delivering high quality data about the nation’s people and economy.  Technical Paper 32, Standards for Discussion and Presentation of Errors in Data, issued in March 1974, is an example of the Census Bureau’s commitment to promote transparency in the quality of the information and data products it delivers to the public and to its sponsors.1

Over the years, the Census Bureau has developed additional guidance regarding the quality of its products and in 1998 began to formalize its efforts to ensure quality in its products and processes. The Census Bureau began this more formal approach by instituting a quality program based on a foundation of quality principles, standards, and guidelines. The paper, Quality Program at the U.S. Census Bureau, describes the beginnings of the Census Bureau’s Quality Program (Proceedings of the International Conference on Quality in Official Statistics, 2001).

In 2001, the Census Bureau issued the first of eleven new statistical quality standards. Several of these standards updated the content of Technical Paper 32. Over the next four years, ten more standards were developed.

In 2005, after conducting a benchmarking study of the standards of other statistical organizations, the M&S Council initiated a more coordinated approach for developing a comprehensive set of statistical quality standards. While the existing standards were a good start, this approach aimed to improve consistency and cohesion among the standards, as well as to reflect all the requirements of the OMB’s Standards and Guidelines for Statistical Surveys in the context of the Census Bureau’s programs, products, and processes.

The new approach to developing statistical quality standards relied on five key components: 1) a dedicated staff to develop the standards, rather than ad hoc teams; 2) contractor assistance; 3) multiple reviews of draft standards to obtain feedback from the program areas; 4) focus groups to obtain more thoughtful and attentive input from the program areas; and 5) a documented, consistent development process.

The Census Bureau began developing these standards in May 2006. The process was completed in May 2010, when the Census Bureau issued these statistical quality standards.



Footnotes:

  1. Technical Paper 32 is available from the U.S. Government Printing Office, Washington, DC 20401. It was revised in: Gonzalez, M., Ogus, J., Shapiro, G., and Tepping, B. Journal of the American Statistical Association, Vol. 70, No. 351, Part 2: Standards for Discussion and Presentation of Errors in Survey and Census Data (Sep., 1975), pp. 5-23. http://www.jstor.org/stable/2286149.

[PDF] or PDF denotes a file in Adobe’s Portable Document Format. To view the file, you will need the Adobe® Reader® Off Site available free from Adobe. This symbol Off Site indicates a link to a non-government web site. Our linking to these sites does not constitute an endorsement of any products, services or the information found on them. Once you link to another site you are subject to the policies of the new site.
Source: U.S. Census Bureau | Methodology and Standards Council |  Last Revised: August 17, 2011