A White House Push for Evidence-Based Decisions

Earlier this year, the White House issued a Memorandum on Restoring Trust in Government Through Scientific Integrity and Evidence-Based Policymaking. This early event indicates the importance of evidence in setting federal policy, and it provides long-term guidance for policymakers, for federal agencies implementing policies, and for state and local governments operating under federal grants directed at implementation.  

To federal agencies, it demands they develop data practices to drive evidence-based policy improvements toward cost efficiency, effectiveness, and oversight duties. 

Meanwhile state and local governments can read into it greater expectations that federal grants, particularly for scientific activities, will come with requirements of data as evidence of performance.

It’s the latest in a string of actions by administrations from both parties to support evidence-based decisions. The increased attention on evidence will have long-term impacts on federal agencies implementing policies, as well as state and local governments operating under federal grants. It also affects policymakers themselves, who must define what constitutes “evidence” and provide or fund mechanisms for collecting it. 

This White House attention on evidence is intended to:

1. Improve cost-effectiveness in updates to existing programs

2. Improve cost-effectiveness in new programs based on evidence from other programs

3. Address shifting risks with the right resources at opportune times, before we hit inflection points of unwanted impacts on environment, human health, national security, and financial stability

4. Re-establish public trust in government with evidence that it is serving the country’s collective interest

What does it mean to be evidence-based?

There are two common interpretations for “evidence-based.” The first focuses on situations where variables may be strictly controlled, like in drug trials. 

The second is oriented to situations that are more complex, nuanced, and difficult to control. That can make it much more difficult to conclude that specific inputs will reliably correlate to outcomes like less school violence or more effective child rearing. (Do children raised in the same household ever turn out identically?) The results of this latter approach are advisory rather than strictly conclusive.

The strict approach may seem more attractive because it is meant to provide certainty: perform X and the result is Y; and the complement, don't perform X and the result is not Y. Of course, that type of analysis isn’t suitable for all situations, if there are tests we do not want to repeat—not just due to complexity but sometimes due to ethical reasons.

Another ethical problem that can arise with the strict approach is the expectation that it is possible to independently verify the results by re-conducting the study. Think of data collected about the impact of industrial disasters or warfare, where it isn’t ethical to re-create circumstances in order to arrive at strict evidence-based outcomes.

When it isn’t feasible, ethical, or even possible to conduct evidence based on strict control, we should recognize our limits and collect evidence to advise policy rather than strictly control it.

The White House Memorandum advises using whichever approach is appropriate for the situation.

A commitment to evidence

While there are naturally those who support the White House memo and those who question it, I recommend reading it for yourself because it is thoughtfully written and lays out specifics for how the federal government can use evidence for the four goals stated above.

This memorandum—and what we may expect to come next via revisions to federal agency policies and proposed laws—is a continuation of the commitments and actions of previous administrations.

A significant legislative advancement for data came under the Obama administration with the DATA Act of 2014. It brought the federal government together around machine-readable data as the data of choice. (See “Understanding Machine-Readability in Modern Data Policy.")

Under the Trump administration, we realized the Foundations for Evidence-Based Policymaking Act of 2018 and the GREAT Act of 2019. Both established Congress’ expectations of machine-readable data as a cornerstone of fiscal management and policy evaluation. There’s also the Federal Data Strategy of 2020.

What this means for government employees

The work begins at the top with an interagency Task Force on Scientific Integrity. It will review and assess scientific integrity practices to ensure “that executive departments and agencies establish and enforce scientific-integrity policies that ban improper political interference in the conduct of scientific research and in the collection of scientific or technological data, and that prevent the suppression or distortion of scientific or technological findings, data, information, conclusions, or technical results.“

Agencies are then required to document scientific-integrity policies that reflect the results of the task force. I read the memorandum as driving collaboration across an agency and down through appropriations and awards to state and local governments, and the sub-recipient companies contracted for services.

After a year of experience with the various policy approaches by the federal government, by states, and even by municipalities to the COVID-19 pandemic, a focus on evidence should come across as a necessary investment. We have a lot to learn from the wide range of adopted policies around the world.

The drive for data will require organizations to have a system and agreed-upon process for collecting, reviewing, approving, and accessing data. 

Collaboration is also key—and an expectation that those involved will possess sufficient data literacy to participate. Whether you’re drafting, implementing, or reviewing policies, whether you’re carrying out research or serving in an operational role, it’s highly likely that you’ll be working with and producing data. 

Documents are important, of course, but the conclusions they contain are increasingly based on data. Data are their essential ingredient, the secret sauce. Thus, data and the information that can be encapsulated as data are an essential part of almost every job in government. For you and your career, data literacy is an essential job skill.

For those who wish to learn more directly from other subject matter experts on this topic, I recommend this Data Foundation webinar recorded February 9, 2021, titled “Prioritizing Scientific Integrity and Evidence-Based Policymaking,” as well as learning about the work of relevant organization focused on making data work for people. At their sites, you’ll find short articles, white papers, recorded and upcoming events, and podcasts (like this one) that will enhance your capabilities.

• Teaching Public Service in the Digital Age
• Data.Gov
• The Data Foundation
• The Open Data Institute 
• The Fintech Open Source Foundation 
• The Data Lodge 
• Results For America