Federal Comment on Data Sharing in Support of Evidence Building

  • Relationship barriers — Data flow at the speed of trust, and trust can be difficult to establish across agencies interested in building evidence through data sharing. Too often agencies lack a common language, are focused on their own distinct priorities, and are operating under time and resource constraints. We have found that frequent, sustained, and collaborative cross-agency data governance is the key to overcoming these inevitable challenges.
  • Real or perceived legal barriers — While navigating the disconnected web of privacy laws, it is difficult to get beyond the default “no” stance many agencies take towards data sharing. Agency counsel are conditioned to be risk averse. However, when agency leadership clearly articulates the benefits of sharing and empowers lawyers to “work towards yes,” with guidance on best practices, legal counsel can support routine processes to ensure data sharing meets all legal and privacy requirements.
  • Data quality standards and documentation — Data linkage and use for evidence-based policymaking is made much more difficult by lack of standardization (e.g., common fields in different datasets defined differently) and lack of good metadata to help assess data quality. Spending time and resources on these activities, as well as on upstream improvements to data collection, can make a difference at the local level, but strong federal guidance is needed.
  • Secure and flexible technical approaches — There is much work to be done in advancing best practices and standards for data sharing technology, both technology that facilitates the linkage of data extracts from across agencies and technology that facilitates access to the linked datasets for analysis. In our experience, the focus should be on ensuring that purpose drives design and data remain a public asset. Too often, agencies each procure their own custom solutions only to find that data are “held hostage” by expensive and inflexible third-party technology providers. While these technical challenges are common, we find they are usually avoidable or surmountable if relational, legal, and data issues have been properly approached and resourced.
  • Understand the complex needs of individuals and families
  • Allocate resources where they’re needed most to improve services
  • Measure long-term and interconnected impacts of policies and programs
  • Better address systemic racism and promote equity
  • Documenting the economic and social impacts of the COVID-19 pandemic
  • Addressing school readiness and educational achievement gaps driven by out-of-school factors, such as housing segregation and access to early education
  • Supporting better health outcomes for populations with complex disadvantages, such as those experiencing homelessness or those who were formerly incarcerated
  • Combatting the opioid epidemic and addressing the ripple effects of substance use on children and families
  • Provide leadership on responsible data stewardship and meaningful stakeholder engagement in order to increase public trust in data efforts at every level of government. This will require transparency and plain language communication about what data are being collected, for what specific purpose, and how they will be stored, shared, and safeguarded.
  • Support alignment of learning agendas across federal agencies and facilitate dialogue about shared or intersecting data needs. This is critical if the federal government is to create a coherent approach to complex challenges, such as pandemic recovery, economic mobility, and systemic racism.
  • Require that agency learning agendas include a discussion of equity considerations in their data collection and evaluation strategies. This discussion should include strategies for improving the accuracy and consistency of demographic categories, which are currently captured differently between state agencies and even across different programs funded by the same federal agency. While this may require support for states to update data collection, it could also leverage linked administrative data to generate aggregate statistics on racial disparities without the additional burden and privacy concerns associated with new data collections.
  • Promote standards for algorithmic fairness, accountability, and transparency in agency decision-making to address legitimate concerns arising around the use of predictive analytics and machine learning tools, particularly when those tools are proprietary.
  • Ensure that Chief Data Officers have expertise in the relational, legal, and technical aspects of data sharing and work to cultivate staff at all federal agencies with relevant expertise.
  • Encourage clear guidance to states and local government from federal agencies regarding the permissibility of data sharing, especially regarding data sets that are often tightly restricted such as Unemployment Insurance records and birth records.
  • Expand access to the proposed National Secure Data Service beyond the elite research community and create data products to meet state and local information needs. This process could be informed by a pilot program to promote cooperation among federal agencies and integrated data systems at the state level, testing methods for users to upload their data and link with federal sources for analysis.
  • Incentivize states to develop and sustain data linkage capacity by increasing administrative set-aside dollars in federal programs that can be used for developing shared data infrastructure, including staffing (past examples of federal funding streams that supported state data capacity in this way include Preschool Development Block Grants, Race to the Top Funds (US DoEd), and Medicaid’s Mechanized Claims Processing and Information Retrieval Systems 90/10 rule).
  • Refrain from issuing new technical requirements (e.g., data security or IT standards) that are unattainable given current resource constraints without also offering flexible support for both federal and state agencies to meet them.
  • Create a special grant program to fund innovation and capacity-building in state and local data sharing, with an emphasis on staffing — not technical solutions.
  • A special grant program should:

--

--

--

At AISP, we help state and local governments collaborate and responsibly use data to improve lives.

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

Republicans and States’ Rights

Through the Looking Glass

New Law & Educational Requirements for Notaries Public Takes Effect July

The Full Extent of Trump’s Damage

Federal immigration rule affecting Texans’ health?

Making Tragedy a Campaign Cry

Careful, 45…Your Racism Is Showing

Why Brett Kavanaugh is a disaster for the environment

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Actionable Intelligence for Social Policy (AISP)

Actionable Intelligence for Social Policy (AISP)

At AISP, we help state and local governments collaborate and responsibly use data to improve lives.

More from Medium

Understanding Eigenvalues Centrality Score

A project for the long run

The Growing Need for Predictive Maintenance in IoT

Learning 2 be Flexible