Federal CIO Suzette Kent said in early December that there was one gift she really wanted for Christmas: the release of the long-awaited final Federal Data Strategy one-year action plan. Now, after several rounds of public comments and a months-long delay, the plan was released Dec. 23, ensuring Kent got at least one gift she wanted this year.
The final action plan — originally slated for release in August — added four new actions for agencies, raising the total from 16 to 20, after the Office of Management and Budget considered the input of stakeholders across industry, academia and government. The 20 actions under the data strategy action plan are grouped across three broad categories: agency actions, community of practice actions and shared solutions actions — all with actions that must be completed within a specific time frame.
Kent told reporters earlier this month that the delay on the data strategy shifted goals and expectations “not at all.”
“Many of the focus areas are foundational and so agencies were working on many of them already," Kent told Federal Times at the Advanced Technology Academic Research Center IT Modernization summit Dec. 11. “What’s new in there is how we’re going to measure success and that required some dialogue with the agencies and across the agencies. And so many of the ones who are going to be held accountable have been part of the dialogue.”
The four new additions in the data strategy spanned all three categories, but one of the most significant additions was the formal launch of the federal chief data officer council — a group mandated under the Evidence Act, a data governance law passed in January this year.
The data officer council will meet in January 2020, the action plan reads, and the council will develop best practices for data, promote data sharing agreements across the government, and identify ways in which agencies can use data for evidence-based policymaking. Right now, the federal agencies’ data sits in silos across their agencies. OMB wants to break down these silos so that agencies owning data that could be useful to other agencies is shared.
For example, “Education and Labor and HHS and the VA all have education programs, and they could have agencies work on a data strategy by mission area. Or you can do something around natural resources with the EPA, or the Department of Interior, or Agriculture,” Dan Chenok, executive director of the IBM Center for the Business of Government, told Federal Times in September.
Another new action mandates that agencies update “comprehensive” data inventories for upload to data.gov by July 2020. Additionally, every 90 days after, all agencies must identify missing or incomplete listings in their data inventories and “ensure metadata is comprehensive for priority data assets.”
Two new action items were added to the shared solutions category, which also had its category name revised after being called the “shared actions” category in previous drafts. The first new action under that section directs the Federal Committee on Statistical Methodology (FCSM) to identify best practices for measuring and reporting on data quality outputs that come from several different data sets.
Along with publishing best practices of measuring data quality, the FCSM must also develop best practices for agencies for “secondary” data likely to be useful in evidence-based policymaking and create a tool to assist agencies use secondary data effectively, all by December 2020.
“Assessing the fitness for specific uses of data is critical to leveraging data effectively,” OMB officials wrote.
The last addition is the creation of the data standards repository by December 2020 “to accelerate the creation and adoption of data standards across agencies." OMB and GSA will work together to create a one-stop shop for information regarding standards for different types of data, communities of practice and standards organizations.
Of course, the federal government’s new focus on its data also means it needs a workforce that’s both fluent in data and trained with the proper skills. Under the action plan, agencies have until July 2020 to identify current data literacy and skill levels within their workforce, and three months after that must have identified skills that the agency needs for its work.
The data strategy authors also added two new data practices to its list of 40, in response to public feedback. The first directs agencies to provide resources to explicitly leverage data assets in response to concerns that agencies wouldn’t dedicate enough resources to effectively use their data. The other new practice told agencies to “align quality with intended use" to ensure data quality.
“Data likely to inform important public policy or private sector decisions must be of appropriate utility, integrity, and objectivity,” OMB officials wrote about the new practice.
Light revisions were pervasive throughout the final strategy action plan, adding clarifications and additional guidance in several areas.
“Federal data has great potential to improve the way government delivers services, to inform research, and to catalyze innovation and grow the economy,” OMB officials wrote. “The views of stakeholders were essential for ensuring that the strategy fully leverages Federal data to provide quality services for the American people, add value for business, and increase government effectiveness and transparency while maintaining data security and preserving privacy and confidentiality.”
The delay in the data strategy wasn’t a concern for GSA CIO David Shive, who said that “doing good data management" in the interim would help agencies be prepared for when the action plan was finalized. Shive defined good data management as establishing a data management framework, good governance and having the right personnel in place to maximize data value.
“If you do those bodies of work ... it is highly likely that there will be massive overlap there and then the work necessary to conform to that strategy is going to be diminutive,” Shive told Federal Times in a December interview before the action plan was released.
The complete list of 2020 data actions is as follows:
- Action 1: Identify Data Needs to Answer Priority Agency Questions
- Action 2: Institutionalize Agency Data Governance
- Action 3: Assess Data and Related Infrastructure Maturity
- Action 4: Identify Opportunities to Increase Staff Data Skills
- Action 5: Identify Priority Data Assets for Agency Open Data Plans
- Action 6: Publish and Update Data Inventories
Community of Practice Actions
- Action 7: Launch a Federal Chief Data Officer Council
- Action 8: Improve Data and Model Resources for AI Research and Development
- Action 9: Improve Financial Management Data Standards
- Action 10: Integrate Geospatial Data Practices into the Federal Data Enterprise
Shared Solution Actions
- Action 11: Develop a Repository of Federal Enterprise Data Resources
- Action 12: Create OMB Federal Data Policy Committee
- Action 13: Develop a Curated Data Skills Catalog
- Action 14: Develop a Data Ethics Framework
- Action 15: Develop a Data Protection Toolkit
- Action 16: Pilot a One-stop Standard Research Application
- Action 17: Pilot an Automated Tool for Information Collection Reviews that Supports Data Inventory Creation and Updates
- Action 18: Pilot Enhanced Data Management Tool for Federal Agencies
- Action 19: Develop Data Quality Measuring and Reporting Guidance
- Action 20: Develop a Data Standards Repository
Andrew Eversden covers all things defense technology for C4ISRNET. He previously reported on federal IT and cybersecurity for Federal Times and Fifth Domain, and worked as a congressional reporting fellow for the Texas Tribune. He was also a Washington intern for the Durango Herald. Andrew is a graduate of American University.