Explorations in Data Innovations - Administrative Data and Research Analysis (ADRA) Project
One of the priorities of the Chief Evaluation Office (CEO) within the U.S. Department of Labor (DOL) is supporting the research community and other stakeholders in accessing and using data. To that end, CEO works with other DOL agencies to conduct Administrative Data Research and Analysis (ADRA) studies. The studies are designed to build evidence on DOL programs and policies and to inform agency priorities by examining administrative data on a wide variety of topics.
In the workforce context, the ADRA studies inform investments in data, methods, and tools that broaden capacity for continuous improvement and evidence-backed decision-making. Since 2018, AIR, as a subcontractor to Westat, has contributed to ADRA studies focused on unemployment insurance, machine learning, and construction safety programs.
AIR Experts Explore Options for Machine Learning
An example of AIR’s work includes the Explorations in Data Innovations report, where AIR experts explored different options for employing machine learning to create and maintain a public-facing, labor-related data catalog for DOL. Between May 2022 and December 2023, the study team worked with experts in machine learning, web scraping, and labor-related data to understand how DOL CEO could use machine learning approaches to automate data collection efforts.
AIR conducted a literature scan, designed automated options to collect and catalog data, and convened a technical working group to solicit external feedback on the options. The report highlighted opportunities to support automated data collection in the future, as artificial intelligence is a rapidly evolving field.
