The National Academy of Sciences at 150.

of the United States and began to compile data annually to provide the record of all sorts of domains of American life. Data on population, finance, commerce, agriculture, exports, mining, railroads, telegraphs, immigration, education, and public lands were all published and developed into time series collections thereafter. The leaders of the Federal Statistical System became major leaders in American science as well. For example, Academy member Francis Amasa Walker was census director in the 1870s and 1880s, president of MIT, president of the American Statistical Association, and the founding president of the American Economic Association. The member biographies of the Academy in the 19th century include many people who crossed disciplinary lines in this more amorphous period in science. Efforts at Centralization By the early 20th century, the US government produced a wide variety of high-quality statistics, both from the census and periodic surveys and from administrative statistics. Presidents from Theodore Roosevelt to Franklin Roosevelt proposed administrative streamlining and consolidation of the activities of the myriad agencies that produced official statistics. These efforts failed as Congress or supporters of particular agencies resisted administrative reform. Franklin Roosevelt finally achieved a modicum of central coordination in the early 1940s with the creation of the position of Chief Statistician in the Bureau of the Budget, Executive Office of the President. Individual statistical agencies would not be brought under the authority of the Chief Statistician, but as head of the Office of Statistical Standards, he had the authority to approve all data collection forms within the federal government. That power provided the capacity to eliminate duplication, standardize procedures and classification systems across agencies, and encourage technical innovations in data collections. In the 1950s and 1960s, statistical agencies introduced computer processing and management of statistical data, starting with the Census Bureau’s use of the Universal Automatic Computer (UNIVAC) for the 1950 census. At the time, the individual agencies each built their own systems. Further efforts at centralization were proposed in the 1960s, when the potential for saving money and fostering research by consolidating the myriad computer databases attracted the interest of the Budget Bureau and academic social scientists. Congress, however, did not see the value of building such a “national data center” and raised significant privacy concerns. The proposal was quietly withdrawn. However, Congress and the President did agree to create the President’s Commission on Federal Statistics of Wallis Commission, which recommended the establishment of the Committee on National Statistics (CNSTAT) in the National Academy of Sciences. At the same time, Phillip Handler was becoming the Academy’s new president, and one of his initiatives was to increase the presence of the social sciences in the Academy. Since its founding, CNSTAT has produced 242 reports, 181 of them in the last 20 years. If one looks at some of the major technical innovations in the federal statistical system in the last generation, it is hard not to see the stamp of CNSTAT reports on the work. The federal statistical system remains decentralized, but the creation of CNSTAT has provided the venue for the leadership of statistical sciences to work with and support the further development of the scientific innovation and offer policy guidance. The Committee on Population Robert Hauser, National Research Council In the early years of the Committee on Population (CPOP), much of its activity focused on population growth, especially on fertility control and the relationships between population growth and economic development. However, despite the continuation of rapid population growth in some parts of the world, and especially in Sub-Saharan Africa, there was a sharp decline in demand for NRC work in this area. Some of the factors in this decline were the success of family planning programs in some parts of the world, a replacement of concerns about population growth and family planning per se with those of women’s empowerment and economic growth, and changes in the research agenda of the Agency for International Development (AID) and the major foundations that once supported research on population growth and policy. For example, despite projections of continued growth in the world’s population to 10 billion, and of increasingly deleterious effects of global climate change, the Division of Behavioral and Social Sciences and Education has been unable to win support for any large-scale study of sustainability and economic conditions in the context of these projections. Instead, with guidance from Richard Suzman, director of behavioral and social research at the National Institute on Aging, CPOP has increasingly and productively focused on issues related to the causes and the consequences of population aging. This work began with the seminal edited volume The Demography of Aging, which was published in 1995. Since then, CPOP has produced a mix of consensus reports, edited volumes, and workshop reports, now numbering almost three dozen. The Science of Population Aging Four themes of the series have had enormous influence on the science of population aging. First, CPOP’s work has led to increasing interdisciplinary studies of population aging and, in particular, to the integration of biomedical, genomic, economic, social, and psychological research. The first contribution to this theme was an edited volume called From Zeus to Solomon published in 1997, which dealt with such diverse topics as the role of the elderly in other species and among human societies past and present, the contribution of evolutionary theory to our understanding of humans, and the potential for collecting genetic material in household surveys. This work and its sequels are now complemented by current study of new developments in biodemography. Second, CPOP’s contributions have provided a series of motivating, illustrative, and instructive reports on the inclusion of biological measures in social surveys. The lead contribution was a volume called Cells and Surveys: Should Biological Measures Be Included in Social Science Research? published in 2001, and most recently, Conducting Biosocial Surveys: Collecting, Storing, Accessing, and Protecting Biospecimens and Biodata, published in 2010. Third, with leadership from the National Institute on Aging, CPOP has contributed to the development of an international cohort of longitudinal biosocial surveys of aging and health. More than two dozen such surveys around the world are located mainly in and around Europe as well as the United States. In the recent past, CPOP has conducted workshops and produced reports that encouraged the development of such surveys in Asia, and a project now under way at DBASSE will undertake a similar effort, focusing on aging in Latin America. Fourth, CPOP has addressed the factors in US longevity and mortality. Here the two key reports are Explaining Divergent Levels of Longevity in High-Income Countries in 2011, and in 2013, US Health in International Perspective: Shorter Lives, Poorer Health. At ages above 50, not only does life expectancy in the United States lag behind that in many other developed nations, but Olson PNAS | June 24, 2014 | vol. 111 | suppl. 2 | 9355 SA CK LE R CO LL O Q U IA