The Committee on National Statistics (CNSTAT) is a standing committee of the National Academies. It was established in 1972 at the recommendation of the President's Commission on Federal Statistics to fill a critical gap as an integrative force for the federal government's highly decentralized statistical system. The Committee works to improve the data and analytic methods that contribute to public policy research and decision-making. During this grant period, the Committee will hold public seminars and develop studies in areas of increasing challenge for federal statistical agencies, including: the tension between protecting confidentiality and privacy and providing research access to data; stemming the decline in response to government surveys; measuring the changing economy; and pursuing innovation and coordination in preference to the status quo. The Committee also will respond to requests for particular studies from sponsor agencies. Some current and expected projects include: measurement of food insecurity and hunger; residence rules and coverage evaluation for the 2010 census; estimation issues for the new American Community Survey; the adequacy of federal business statistics on small business dynamics and innovation; the design of national health accounts; and an update of the Committee's white paper, Principles and Practices for a Federal Statistical Agency. Joint projects with other units in the National Academies include the feasibility of a national ballistics imaging database and applications of data mining and information fusion for public health, terrorism prevention, and other governmental functions.
Committee on National Statistics projects are designed to have broad application beyond the specific issues covered. Many studies help improve government policy and operations. Almost all contribute to the social, behavioral, and economic sciences in such ways as: improving federally sponsored data collections on which much social science research depends; furthering the development and use of statistical methods for policy and methodological research; providing forums for intellectual exchange and consideration of system-wide issues, such as human research participant protection for social science research; and advancing the social, behavioral, and economic sciences within the National Academies and through the Committee's relationships with Congress, OMB, and federal research and program agencies. Many projects also contribute to better public understanding by improving key national indicators (e.g., the poverty measure, the Consumer Price Index) and the quality, relevance, and accessibility of key data sets.
The Committee on National Statistics (CNSTAT)—established at the National Academy of Sciences as a standing committee in 1972—continued its work during the 2005-2012 grant period to evaluate and help integrate the federal government’s highly decentralized statistical system, to improve the data and analytic methods for informing policy-relevant research and public policy decisions, and to foster the development and accessibility of data and measures for basic and applied research in the social, behavioral, and economic sciences. The Committee’s portfolio covered not only aggregate social, economic, and other statistics, such as counts from the decennial census and American Community Survey, but also detailed microdata for individuals, households, firms, and other entities required for innovative analysis of the dynamics of human and social behavior. It conducted its work through consensus studies, workshops, public seminars, and continued updating of its influential white paper, Principles and Practices for a Federal Statistical Agency, for which the third edition was issued in 2005 and the fourth edition in 2009. Interacting with NSF, federal statistical and research agencies, and the social, behavioral, and economic sciences research community, the Committee studied problem areas in which statistical information and methods are important for research and decision-making, including persistent problems that could not be resolved by one-time efforts and problems that span the interests of more than one agency. Its work focused on significant challenges to federal statistics, including the need to integrate household and business surveys, confront declining survey response, use administrative records to improve data quality, enlarge the pipeline of technically qualified agency personnel, and bolster innovation in federal statistics. It studied issues of balancing confidentiality protection with research access to data, measuring the changing economy, enhancing the quality and cost-effectiveness of major censuses and surveys, and refining and applying statistical methods, such as experimental design and data mining, to important national issues. The intellectual merit of the Committee’s work is evident in all of its activities, which bring together leading scientists from a variety of perspectives and disciplines to identify not only state-of-the-art methods for statistical data collection, estimation, and analysis, but also needed improvements in methods. Committee studies addressed such methodological challenges as improving household and business survey response, compensating for missing data in cross-sectional and longitudinal surveys, measuring cost-of-living and nonmarket activities, validating analytical and policy models, providing model-based small-area estimates for research and policy use, improving the design of experiments for censuses and surveys, and many other aspects of the data and methods that support behavioral, social, and economic sciences research and policy analysis. The Committee’s work is expected to have broader impacts, directly and indirectly. Its recommendations for using statistics in federal fund allocation formulas directly affect billions of dollars in aid for programs in education, nutrition, and other public assistance. Its recommendations for improving major surveys and censuses result in more accurate statistics and data sets for research, which in turn supports better policy analysis, which, in turn, can support more informed public decision making. Its recommendations on principles and practices for statistical agencies assist them to maintain objectivity and credibility and work unhindered to develop the most accurate, relevant, and timely data for research and policy use and broad public understanding. Landmark publications from CNSTAT studies during this grant period include: Expanding Access to Research Data: Reconciling Risks and Opportunities; Food Insecurity and Hunger in the United States: An Assessment of the Measure; Gender Differences at Critical Transitions in the Careers of Science, Engineering, and Mathematics Faculty; Measuring Research and Development Expenditures in the U.S. Economy; The Prevention and Treatment of Missing Data in Clinical Trials; State and Local Government Statistics at a Crossroads; and Understanding Business Dynamics: An Integrated System for America’s Future.
