The Stanford Institute for Research in the Social Sciences (IRiSS)’s Secure Data Center hosts one of the nation’s 29 Federal Statistical Research Data Centers, providing access to restricted data for Stanford researchers and those from nearby universities.
What data is available in Federal Statistical Research Data Centers?
The Federal Statistical Research Data Centers (RDCs) located across the country—managed and administered by the U.S. Census Bureau—make two types of restricted data accessible to approved researchers: survey and census data, and administrative data.
The former is largely comprised of data collected by the U.S. Census Bureau in course of its surveys and censuses, such as the decennial census and the American Community Survey. As a whole, the content of these Census Bureau RDC data can be broadly categorized as demographic (e.g.., decennial census, American Community Survey, Current Population Survey, American Housing Survey, National Crime Victimization Survey) and economic (e.g.., Longitudinal Business Database, economic censuses, Annual Survey of Manufactures, Survey of Business Owners, Longitudinal Firm Trade Transactions Database).
By contrast, the collection of administrative data is in some ways a byproduct of a functioning organization or government bureaucracy, as it is data that federal agencies, state and local governments, and other organizations collect in course of running their programs, such as unemployment and disability claims data. Administrative data from the Social Security Administration, state-level public assistance programs, and private companies such as CoreLogic are some of the datasets available to researchers through the FSRDC network.
Another important type of Census dataset uses administrative data linked to economic and demographic data to examine the connection between employers and employees: the Longitudinal Employer-Household Dynamics (LEHD) and the Innovation Measurement Initiative’s (IMI) Universities: Measuring the Impacts of Research on Innovation, Competitiveness, and Science (UMETRICS) data.
Some agencies are partnering with the Census Bureau and also providing survey data that can be accessed in the RDC. The main agencies are the the National Center for Health Statistics (NCHS), the Agency for Healthcare Research and Quality (AHRQ), the Bureau of Labor Statistics, and the Bureau of Economic Analysis. The full list of participating agencies can be found on the Census Bureau website.
What is restricted microdata?
Microdata refers to the individual-level responses and data that are collected; from people, with regard to demographic data, and from businesses, with regard to economic data. Census provides survey and census results aggregated at various geographic levels, such as states, counties, and cities, online to the public, but researchers may find that using microdata allows for more customized aggregations. Microdata also allows researchers to conduct sophisticated statistical analysis and may be necessary for certain research questions.
Census and its partnering agencies often make some microdata available to the public online. For example, researchers can download public microdata for the American Community Survey and the decennial census (few economic surveys have publicly available microdata, in contrast). However, these public microdata have limited geographical detail, may be missing some variables, have censored values, and may include only a sample of respondents to protect privacy.
However, within the RDCs, researchers can access much richer microdata files that have all respondents, much greater geographical specificity, and less or even no censorship of values. These microdata are never released to the public due to privacy considerations, but if their projects are approved and if they obtain the necessary security clearance, researchers may view the data within the RDC, conduct their statistical analysis, and release the results of their statistical models. The RDC is especially useful for economic research because there is little microdata on firms and establishments available to the public. The valuable datasets available only through the RDC can unlock a trove of research insights for social scientists across multiple fields including economics, business, sociology, political science, and health research.
What types of studies do RDC microdata make possible?
A unique benefit of working with different types of restricted microdata is the ability to link data from different sources together to weave together a dataset that answers questions beyond what looking at single survey or source of administrative data can address. Census datasets often include firm, establishment, or person identifiers, allowing economic datasets to be linked to each other and some demographic datasets to be linked to each other. In some cases, economic and demographic datasets can be linked together too, for example, the Longitudinal Employer Household Dynamics (LEHD) dataset combines administrative data from states' unemployment insurance records with Census Bureau demographic data and economic data, revealing relationships between trends in unemployment rates, worker demographics, employer history, and industry health.
Examples of research projects that have or are using restricted-access microdata through the RDCs can include:
· Working papers that use Census Bureau data:
· Research featured in the Census Bureau Center for Economic for Economic Studies annual reports
· Projects using NCHS data
For more information about the FSRDC program and available data, please see the Census's FSRDC webpage.
Who can access the RDC?
To gain access to the RDC, qualified researchers in social science disciplines must submit proposals to the Census Bureau. The proposed research projects must, provide benefits to Census Bureau programs, demonstrate scientific merit, require non-public data, be feasible given the data, and pose no risk of disclosure. In addition to having proposal approved, researchers must obtain Census Bureau Special Sworn Status, which includes a background check. Learn more on the Census Bureau website. In order to access the data that affiliate organizations make available in the RDC - National Center for Health Statistics, Bureau of Economic Analysis, Bureau of Labor Statistics, Agency for Healthcare Research and Quality - researchers must apply directly to each organization directly.