Administrative Data Facility Enables Cross-State Data Sharing
Today, many students transfer programs across state lines or move to another state for work after graduation. Many workers also have a job in a state different from where they live. Because most states only have data from within their own borders, it is difficult to truly understand how our nation’s education and workforce training programs are preparing people to find employment. Moreover, many localities cannot access state data, nor can states access local data. In order to make data sharing easier for state and local agencies, WDQC wants to share a new option for cross-agency data sharing, the Administrative Data Research Facility (ADRF). Julia Lane, the system’s creator, has detailed it below.
The New York University (NYU) Administrative Data Research Facility (ADRF) offers a new approach to sharing data across agencies and across state lines. It enables multiple local, state, and federal agencies to keep data in separate secure areas, and link data for approved projects.
The ADRF allows agencies within the same state or different states to agree to share their data in a common area in the cloud for specific approved projects. If approved, staff from multiple agencies can jointly access the common area, so that they can work together to develop new integrated datasets, share information about coding differences or similarities, and develop common measures. No personally identifiable information is stored in the cloud – the data are hashed and deidentified (with a common hash algorithm) -- before being transferred to the secure area. Moreover, data stewardship modules can be deployed so that agencies can track use and work output. Over 175 agency staff, from about 50 agencies, have already accessed and used the ADRF.
The NYU ADRF was set up by the Census Bureau to inform the decision making of the Commission on Evidence based Policy (and was highlighted in their final report). It was designed to build on the lessons learned from establishing the Census Bureau’s Local Employment Dynamics (and Longitudinal Employer-Household Dynamics) program – mainly that state and local access to and use of linked data was essential to the creation of high value products. Access to the ADRF was made possible by the Laura and John Arnold Foundation and the Overdeck Family Foundation through a set of scholarships to government agency staff.
During a course enabling staff from state agencies to get a feel for the ADRF, agency staff used exit and admission data on ex-offenders from the Illinois Department of Corrections, welfare recipient data from the Illinois Department of Human Services, and unemployment insurance wage records from the Illinois Department of Employment Security to answer questions including “what is the effect of neighborhood characteristics and transportation on the earnings and employment outcomes of ex-offenders and welfare recipients and their subsequent recidivism or exit from welfare?”
Moving forward, agency staff will be able to work with one another to determine their own research projects. More information is available on the Coleridge Initiative website. You can also email Julia Lane at email@example.com.
Julia Lane is a professor at New York University and was one of the founders of the U.S. Census Bureau's Longitudinal Employer Household Dynamics program.