Since August 1, 1933-

"...As the major figures from the American Statistical Association (ASA), Social Science Research Council, and new Roosevelt academic advisors discussed the statistical needs of the nation in the spring of 1933, it became clear that the new programs-in particular the National Recovery Administration-would require substantial amounts of data and coordination among statistical programs. Thus in June of 1933, the ASA and the Social Science Research Council officially created the Committee on Government Statistics and Information Services (COGSIS) to serve the statistical needs of the Agriculture, Commerce, Labor, and Interior departments...COGSIS set...goals in the field of federal statistics... (It) wanted new statistical programs-for example, to measure unemployment and address the needs of the unemployed â- (It) wanted a coordinating agency to oversee all statistical programs, and (it) wanted to see statistical research and experimentation organized within the federal government... In August 1933 Stuart A. Rice, President of the ASA and acting chair of COGSIS,... (became) assistant director of the (Census) Bureau. Joseph Hill (who had been at the Census Bureau since 1900 and who provided the concepts and early theory for what is now the methodology for apportioning the seats in the U.S. House of Representatives)... became the head of the new Division of Statistical Research... Hill could use his considerable expertise to achieve (a) COGSIS goal: the creation of a research arm within the Bureau..."

Source: Anderson, M. (1988), The American Census: A Social History, New Haven: Yale University Press.

Among others and since August 1, 1933, the Statistical Research Division has been a key catalyst for improvements in census taking and sample survey methodology through research at the U.S. Census Bureau. The introduction of major themes for some of this methodological research and development, where staff of the Statistical Research Division¹ played significant roles, began roughly as noted-

• Early Years (1933-1960s): sampling (measurement of unemployment and 1940 Census); probability sampling theory; nonsampling error research; computing; and data capture.
• 1960s-1980s: self-enumeration; social and behavioral sciences (questionnaire design, measurement error, interviewer selection and training, nonresponse, etc.); undercount measurement, especially at small levels of geography; time series; and seasonal adjustment.
• 1980s-Early 1990s: undercount measurement and adjustment; ethnography; record linkage; and confidentiality and disclosure avoidance.
• Mid 1990s-Present: small area estimation; missing data and imputation; usability (human-computer interaction); and linguistics, languages, and translations.

At the beginning of FY 2011, most of the Statistical Research Division became known as the Center for Statistical Research and Methodology. In particular, with the establishment of the Research and Methodology Directorate, the Center for Survey Measurement and the Center for Disclosure Avoidance Research were separated from the Statistical Research Division, and the remaining unit's name became the Center for Statistical Research and Methodology.