Agencies Are Making ‘Uneven’ Progress Meeting Evidence Act Mandates, GAO Says

The Foundations for Evidence-Based Policymaking Act passed in 2018 with a series of mandates for federal agencies to use hard data in policy decisions, such as determining whether a certain program is successful and deserving of more funding.

The Trump administration’s Office of Management and Budget put out the first set of guidance a few months after the law was signed, with more published less than a year later. The Biden team followed suit, releasing a presidential memo in the early days of the new administration and new OMB guidance superseding what was issued by Trump.

However, despite all that guidance, there are “longstanding weaknesses across the federal government regarding agencies’ capacity to build and use different types of evidence,” according to a new report from the Government Accountability Office.

GAO auditors conducted an online survey in 2020 of nearly 4,000 federal managers at 24 agencies.

“Results from GAO’s 2020 survey of federal managers showed that nearly all managers—an estimated 95%—reported having at least one type of evidence for their programs,” GAO wrote, noting OMB defines several types of evidence, including “performance information, program evaluations and other types of data, research and analysis.”

Going deeper, 71% of respondents reported using two or more types of evidence to evaluate their programs, and 35% used all three types noted above. Half to two-thirds of respondents told GAO that “when they had evidence,” it was used in the decision-making process.

However, far fewer federal managers said they had or were able to identify people in their organization with the data collection, management and analysis skills to fully implement the Evidence Act.

“For example, between 50% and 60% of managers reported that their programs had staff with skills needed to collect, analyze and use different types of evidence,” the survey found. However, that drops to 45% to 47% when those same managers were asked about skilled staff agencywide.

GAO auditors noted that data collection is meaningless unless officials can “understand what conclusions can and cannot be drawn from them.”

Forty-three percent of federal managers said they had requisite staff with the “knowledge and skills to integrate and compare findings from different types of evidence.” The survey garnered similar results when managers were asked about access to data analytic tools, with “about half of managers report[ing] that both their programs and agencies had tools to collect, analyze and use performance information.”

Managers also reported difficulty using data to show the public how a certain program was performing; only “one-third of managers reported taking that action,” the report states.

But these findings were “uneven” across the government.

“When we disaggregated the survey results, we found that the reported presence of different aspects of capacity varied widely across the 24 agencies,” the report states.

• Across the survey questions related to existing capacity, agency-level results ranged from an estimated 23% to 80%.
• Similarly, across survey questions related to actions that can enhance capacity, agency-level results ranged from an estimated 16% to 74%.
• When looking at the difference between the highest and lowest estimated agency results on each individual question, that difference ranged from 23 to 46 percentage points.

GAO auditors said they hope the findings gathered in the report will inform ongoing efforts by OMB, the Office of Personnel Management and the Chief Data Officers Council to improve data for decision-making across government.

“This could include using our survey results to help identify promising practices at certain agencies and address challenges at others,” they wrote. “Results could also be used to help identify cross-cutting capacity issues affecting multiple agencies, and prioritize actions to address them.”