Coleridge Initiative

Administrative Data Research Facility


The Administrative Data Research Facility (ADRF) is a secure cloud-based computing platform designed to promote collaboration, facilitate documentation, and provide information about data use to the agencies that own the data. The ADRF was established by the Census Bureau with funding from the Office of Management and Budget to inform the decision-making of the Commission on Evidence-Based Policy.

The ADRF has provided secure access to over 100 confidential government datasets from 50 different agencies at all levels of government. The ADRF is FedRAMP Certified and is listed on thFedRAMP Marketplace, it has received Authorization to Operate from the Census Bureau, the US Department of Agriculture, and the National Science Foundation, and it won a 2018 Government Innovation Award.

ADRF System Components

The ADRF is a FedRAMP certified environment that currently hosts de-identified data (often generated from confidential data) and is operated as a secure remote access data facility within the Amazon Web Services GovCloud. It provides secure access to data sets that historically have not been available for analysis by authorized researchers allowing them to develop rich metadata and access search features to find and use federal datasets. Balancing security and usability requirements, this system provides a comprehensive facility for administrative data analytics to meet evidence-based policy goals. 

The platform’s data processing and management capabilities include facilities for data ingestion, data documentation, data analytic tools, and data stewardship.  All access is exclusively via web browser over HTTPS connections, and it leverages cloud-based services for efficiency, reusability and cost transparency. The desktop environment is based on Microsoft Windows 10 with secure access to databases, file systems and, if required, select externally accessible websites and/or services.

system architecture

Five Safes Data Strategy

The ADRF platform implements the Five Safes Framework to ensure the safe and responsible use of data. To learn more about each of the five safes, click on the appropriate button below.


Safe Projects

Safe People

Safe Settings

Safe Data

Safe Exports