Evidence-based policymaking gains traction in states
- By Matt Leonard
- Jul 25, 2018
Lawmakers want to pass laws that have a positive impact, but often "they simply don’t have access to the information that helps them do that," said Sara Dube, director of the Pew-MacArthur Results First Initiative at the Pew Charitable Trusts. "Even if they do have that information available somewhere, they don’t know how to get to it and they don’t know how to interpret it.” She made the comments at a July 24 event hosted by the Bipartisan Policy Center.
Evidence-based policymaking is gaining traction in states, but statewide systems for supporting it and making it a priority are less common, according to a report from the Pew Charitable Trusts.
The report defines evidence-based policymaking as "the systematic use of findings from program evaluations and outcome analyses (‘evidence’) to guide government policy and funding decisions." If governments can focus their limited resources on programs that have demonstrated positive results, they can "expand their investments in more cost-effective options, consider reducing funding for ineffective programs and improve the outcomes of services funded by taxpayer dollars."
Pew’s report categorizes all 50 states and Washington, D.C., based on their implementation of evidence-based policymaking as either leading, established, modest or trailing. Four states were listed as leading: Washington, Oregon, Utah and Connecticut.
Washington's move into evidence-based policymaking began when the Washington State Institute for Public Policy the efficacy of programs to reduce recidivism among juveniles in the justice system, said Eric Trupin, director of the Division of Public Behavioral Health and Justice Policy at the University of Washington.
The organization concluded that investing in community-based programs would be more effective at reducing recidivism and helping the economy than increasing the number of beds in juvenile facilities, Trupin said.
That "built an appetite" for legislators to increasingly say, “'Let’s look at evidence before we make some decisions about where we’re going in terms of policy,’” he added.
The state has built a data infrastructure that allows it to track the impact of juvenile justice legislation. A shared database within the Department of Social and Health Services' Office of Juvenile Justice allows officials to examine outcomes based on considerations such as the use of Medicaid services and an individual’s previous involvement with the justice system.
“There is an integrated database that allows us to really ascertain outcomes in various domains, so that is a very, very fortunate circumstance,” Trupin said.
The road to evidence-based policymaking does have bumps. Some people in government can be hesitant to share data for policy research because it is often laced with personally identifiable information, said Jessica Corvinus, research and evidence-based policy manager at the Colorado Office of State Planning and Budgeting.
To address privacy concerns, Colorado officials are focusing on data linking rather than the sharing of PII-filled data with other agencies and expect researchers to protect that information. Linking aggregates the data to eliminate PII, Corvinus said.
David Yokum, director of The Lab @ DC, said it's understandable that government employees are wary of sharing administrative data.
“Internally, administrative data has, typically, a tremendous quantity of quirks” that could easily be misinterpreted, he said. The datasets weren't built for research, and data should be cleaned and documentation written before it is handed off to researchers.
An important part of evidence-based policymaking is creating a strong data infrastructure that features "automated, high-quality and real-time reporting systems" as well as interoperable systems and knowledgeable staff, according to the Pew report.
The technology can be tricky when it comes to contracting and funding, Yokum said, but actually making the technology work is more straightforward. From a technical standpoint, "we know how to do this,” he added.
But Yokum cautioned against “focusing too tightly on just investing in a data system for a data system’s sake. Other than scientists and data scientists, no one cares about data per se. What they care about is some problem” data can help solve.
Matt Leonard is a former reporter for GCN.