Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
The People’s Data: Why Federal Data Matters More Than Ever with Nick Hart
359 Plays6 hours ago
In this episode, I talk with Nick Hart, President and CEO of the Data Foundation, about the rapidly changing landscape of federal data, statistical agencies, and evidence-based policymaking. We explore how the Evidence Act reshaped government data infrastructure, why privacy protections and data governance matter more than ever, and what’s been happening behind the scenes over the last year as agencies faced staffing cuts, data removals, and unprecedented political pressure. Nick explains how government data systems actually work, why the U.S. model is both admired and strained, and what a “Data System 2.0” might look like in the future. We also discuss state and local data roles, the risks of politicizing data, and two public-facing initiatives from the Data Foundation: the Evidence Act Hub and the People’s Data 100. This is a wide-ranging conversation about trust, transparency, and why government data quietly underpins far more of our lives than most people realize.
Subscribe to the PolicyViz Podcast wherever you get your podcasts.
Become a patron of the PolicyViz Podcast for as little as a buck a month
Check out the Data Foundation and their People's Data 100 project!
Follow me on Instagram, LinkedIn, Substack, Twitter, Website, YouTube
Email: jon@policyviz.com
Recommended
Transcript
Podcast Introduction
00:00:12
Speaker
Welcome back to the policy of this podcast. I'm your host, John Schwabish. Thanks so much for tuning into the show. I hope you're well, we continue our learning this week about what's going on in the federal data ecosystem. What the administration has been doing to remove data sets, to stop collection of data. And as those of us who are working in the data space, either as visualizing or communicating data as analyzing data is doing research or analysis, or just plain old consuming data, what we can do and what we should be aware of with these actions from the administration.
Introduction to Nick Hart
00:00:45
Speaker
So to help me, i am pleased to be joined by Nick Hart, who is the president and CEO of the Data Foundation based here in Washington, DC. We talk all about what the Data Foundation is doing in this space. We talk about the various efforts that they have underway, including a couple of public survey type efforts, which I will link to in the show notes and you should take a look at those.
Redesigning Data Systems
00:01:06
Speaker
um We also talk about what a federal data system might look like in the future.
00:01:11
Speaker
We have perhaps an opportunity here to rethink how the federal data system and of course state and local data systems are constructed and how they run and how they collect and publish and use data. So I think this is a continuation of a discussion that I've been having with folks over the last six months or so with Rob Santos from the Census Bureau with Bill Beach and Erica Groshen, formerly of the Bureau of Labor Statistics, and most recently with Denise Ross, the former chief data scientist of the US. So if you haven't listened to those episodes, please go back. You can get the show wherever you get your podcasts. Obviously, you're listening to it or you're watching it on YouTube, Spotify, iTunes, wherever you get your podcasts. You can listen to it directly on the policy of this site or you can get it directly from Zencaster as well, where, by the way, I do all the recording. So that's all i've got. That's my intro for you. I'm going to move right on to the interview. So I hope you'll enjoy my conversation on this week's episode of the podcast with President CEO of the Data Foundation, Nick Hart.
Nick Hart's Background
00:02:12
Speaker
Hey, Nick. Good to meet you. Thanks for coming on the show. course. Happy to be here. Excited to chat with you. um I think you know this show is primarily data viz, but I've branched out the last year because a lot has been going on. And so people may not be as familiar with you as they are with some of my other folks. So maybe you could talk about your background a little bit, and then let's talk about what the Data Foundation does.
00:02:34
Speaker
Sure. um Well, hello everyone. Thanks for listening to this this episode. So, Nick Hart, I've been leading the Data Foundation for the last seven years as the president and CEO. Prior to that, I spent about a decade working in the White House's Office Management and Budget, and was also the Policy and Research Director for the Ryan Murray Evidence Commission, or its formal title, the U.S. Commission on Evidence-Based Policymaking.
00:03:00
Speaker
So I've really spent most of my career focused on building policy frameworks and strategies to enable all of us, every American, but businesses, government decision makers to better use information that government's collecting and managing. And as you said, it's been a pretty exciting year in that context that I know we'll talk a little bit
Data Foundation's Mission
00:03:20
Speaker
more about it. So the Data Foundation, this is what we do every single day.
00:03:24
Speaker
We are essentially advocates and champions for open data and evidence informed policymaking. And we we get up in the morning and we start thinking about data and and our work, but that takes us into circles that involve folks at the White House, members and staff at Congress. as well as individuals in the business community and the statistical enterprise and the research community. So we do our best to collaborate across a lot of different people and trying to understand how we can better use government data. That's great. I mean, you've seen, i mean, working, I mean, thinking back to Ryan's work, like he was one of the first, I think at least on the Hill of really thinking publicly about better federal data.
00:04:07
Speaker
So you've sort of seen this big, change and evolution along with, you know, changes in social media and iPhones and all this
Advancements in Data Utilization
00:04:15
Speaker
stuff. Like, I'm not really sure what my question is here, but I guess looking back over the last, like, what would that be like 10, 15 years? Like, do you have a big takeaway from that, from some, from seeing that this, you know, sort of recent history with dramatic change?
00:04:29
Speaker
I think there's a couple of key takeaways, but I would take this back even further to the founding of our country, which includes a decennial census. I mean, data is quite literally written into the Constitution of the United States. So founding fathers and those who laid out many of the principles of the country have been thinking about this question.
00:04:48
Speaker
We're about this question all along. In the last 15 to 20 years, I would say the US government has really revolutionized its capabilities to better use data. And that's the culmination of lots of things that are happening across a really big ecosystem, particularly as technology has evolved. That doesn't mean, and don't get me wrong here, but government is definitely not in the place that it should be to do all this work. We have so much more to do in terms of advancing new technologies, policies and procedures, privacy protections. But almost a decade ago now, so and back in 2016, Paul Ryan and Patty Murray's Evidence Commission launched. And for those that aren't that familiar with how that came about, it was basically a frustration about a knockdown, dragout budget fight when they were both the respective budget chairs in the the House and the Senate.
00:05:37
Speaker
And they were trying to figure out better strategies to get the information that they actually wanted as decision makers. And so they made an agreement that once the budget fight was done, they were going to figure out how to better ah ah change the ecosystem for the better.
00:05:52
Speaker
yeah And I'd say the great success of the Evidence Commission, which existed for just over a year, is that it culminated in real and meaningful legislation that fundamentally changed the way that government has the capacity and thinks about pushing out everything from open data, but also better using the statistical information. So this is linking together really confidential or private information so we can get general insights or aggregate insights that are beneficial to society. So if you were to compare where we were 20 years ago to where we are today, it's really night and day. I think some of the the tea leaves were there, but both with technology and the great leadership from both politicians, but also individuals that are career civil servants, folks who have provided advice through advisory committees and other mechanisms. There is just a lot that has changed for the better that has led us to have much better knowledge than we had 20 years ago in a lot of sectors from financial services to agriculture to um you all really, you name it. It's kind of across the board.
Challenges for Statistical Agencies
00:06:54
Speaker
Yeah. um You mentioned the last ah sort of earlier, you know, the last 12 months or so have been a tough year for statistical agencies, right? there's been mass layoffs, retirements, forced retirements, removing data sets, stopping surveys, attacks on civil servants, you know, like Erica McIntarfer from the BLS. Folks who listen to this show know I've focus a lot on that. um And I should also say to to broaden it a little bit is also not just from the administration, but also from policymakers and other stakeholders and other thought leaders really attacking you know a lot of these statistical and data agencies like CBO comes under attack ah a lot of the time, um which they always have, but now seems like a different tenor. um And so I'm curious, maybe you can
00:07:37
Speaker
tell us a little bit about what the the Data Foundation does like every day, especially over the last 12 months. Like, like has has your work changed dramatically or yeah, I'll stop there. So the work changed dramatically. what does it What does a day look like when when folks come in the building in the day?
00:07:57
Speaker
Well, first of all, I don't think there's a normal day in the work that we do. Particularly over the last year, there were some twists and turns and yeah pace was was definitely fast. yeah um but Let me maybe give an example. So you mentioned the Congressional Budget Office. And CBO is a fantastic agency housed within the legislative branch charged with essentially estimating the cost of legislation and providing projections for the Congress. And as a former person who worked at the Office Management Budget, I have love-hate relationship with CBO because I spend a lot of time arguing with them about why my number was right and their number was wrong and then trying to convince them. But the the truth is they have a fantastic staff, they have fantastic leadership, and they approach all of their work with an incredible sense of rigor and that bolsters their credibility. A couple of years ago, Congress provided a new statutory authority to the CBO, which is essentially a mechanism that lets them reach into executive branch agencies, so administrative agencies, to use their data to improve budget scores.
00:09:01
Speaker
So on its face, that may either sound like a good thing or a huge privacy risk, depending on how you how you approach the question. And I think from our perspective, it's actually both things can be true at the same time. It can be hugely beneficial, but must be implemented and managed correctly.
00:09:17
Speaker
So when this bill was moving through the Congress, it was overwhelmingly supported by Republicans and Democrats, unanimous out of the respective budget committees. And that's a sign that Congress wanted better information actively. This is just in the last couple of years.
00:09:31
Speaker
Well,
CBO's Data Breach and Privacy
00:09:32
Speaker
guess what? In the last year, in 2025, CBO had a data breach. And that's a major intersection with why we also have to talk about strong privacy protections at the same time.
00:09:42
Speaker
At the same time, CBO was fully transparent about the fact that that happened. They are incredibly good at telling all Americans and the Congress about how the data are being used. And when you take all of these things together and a whole bunch of other data governance concepts, this is exactly what we do. We are here to help agencies, but also the partners on the outside of government, better understand what a healthy, strong, robust data governance ecosystem really looks like. And if you don't track what all those words mean, that's okay. The point is, we want data to be used responsibly, ethically, legally, and importantly, used.
00:10:19
Speaker
So if we're collecting all this information and we're just letting it sit in ah in a mainframe somewhere, or even worse in some cases, literally on paper files collecting dust, that's a problem. we're not We're not capitalizing or catalyzing the value of that information. So every single day we're now engaged in helping to build the chief data officer function, which is relatively new. It's a creation out of the Evidence Act that President Trump signed back in 2019. It establishes new data leaders that sit alongside the statistical agency leaders in every single agency of government.
00:10:55
Speaker
So not even every agency of government has a statistical agency. There's 13 principal statistical agencies and about 100 other units that are statistical. but every agency is collecting data, which is why having a chief data officer, regardless of the kinds of data that are collected, is really important so that we have standards and quality. We can sync it to scientific integrity and ensure that the information that's being used for decision-making is also relevant for decision-maker. But then anything that's being published satisfies Information Quality Act. It's fit for purpose. It's fit for use. And the American people can then have, hopefully, increased trust that this is good data.
Roles in Data Ecosystem
00:11:34
Speaker
right
00:11:34
Speaker
So we work with all of these actors and more in terms of trying to promote that. Over the last year, though, as you acknowledged, and I think I've alluded to a couple of times here, there were pretty big changes in capacity in statistical agencies and really every agency of government, if you look across the board. And that's one of the challenges that that I think folks have had to navigate. And it's not just about the data officers. There's also career statisticians and data managers and IT folks, evaluation officials, and many others that are supporting this knowledge ecosystem that were, in some cases, they were fired. In some cases, they were transferred. some cases, people voluntarily left. But that's knowledge and infrastructure that we actually need in government. And we we want our government to have
00:12:18
Speaker
to be able to produce and maintain all this information responsibly. So we're actively advocating and trying to support that that infrastructure so that it can can exist not just today, not just last year, but continue in the future, and then also help prioritize it.
00:12:34
Speaker
So one of the unique things that happened the last 12 months was, ah well, there were a lot of unique things in the last 12 months. That's right.
00:12:45
Speaker
The one that comes to mind because it's been in the news the last few days is there's now evidence that ah Doge, Elon Musk's Doge team went into Social Security Administration.
00:12:55
Speaker
They did get their hands on administrative operations. private information, people's social security numbers and other information. And that information appears to have been shared with outside firms. How do you think about that when you're, you know, I mean, you're sort of, um you know, working within this world of like government data, and then there's sort of like this outside actor coming in, you know, it's it's just a very strange place to be. So how do you work around that? Do you do do you talk to people about that? Or is that like,
00:13:25
Speaker
This is like moving beyond what we're sort of set up to do. Yeah, that is exactly the kind of issue that we we work on, both in a policy framework, but also trying to make sense and particularly over the last year, ah identify the truth in what is happening. And the example that you just mentioned, ah for folks that may not be aware, is also the subject of a fair amount of litigation. There were temporary restraining orders that were issued related to the Department of Government Efficiency or to the DOGE.
00:13:55
Speaker
that ah effectively said hold off on doing certain things until you have met other requirements that are in law or can validate that those requirements are met. That's what a lot of temporary restraining orders will do when it comes to data processes. yeah The new information seems to suggest fairly clearly that there were violations of our expectations and also potentially laws. Now that's for the courts to ultimately decide yeah and how to interpret that. And I think we we try to walk a very fine line here in calling balls and strikes when we see them. On this particular issue, we were very actively working with both parties in the Congress and also having interactions and conversations with folks in the White House, as well as Social Security Administration,
00:14:41
Speaker
and in advance of things like this happening. And one of the areas that arem very proud to say that we leaned in a lot last year was being able to promote what robust privacy protections look like, but also clarify what the law actually requires. We have a fantastic law called the Privacy Act of 1974. And while it is 50 years old, the core principles the processes that are there, they might need to be updated in many ways, don't get me wrong, but the core core processes are still fairly robust. It's also how we have other processes in place in government that ensure people are trained and they know whether they're supposed to have access, and then we manage that access appropriately.
00:15:19
Speaker
um I like to talk about ah a framework, particularly in the statistical agencies called the five safes. And I shouldn't have said this out loud because now I'm going to at least skip one of them if I do the list off the top my head. But it's it's a way of thinking about a heuristic for having safe data and safe people and safe spaces and places and basically laying out a policy framework where we recognized you can't just lock all of the data in a lockbox and hope that nobody ever touches it because we're actively talking about people using the data.
00:15:49
Speaker
SSA, the Social Security Administration, uses data every single day to make thousands, hundreds of thousands of decisions. And we need them to do that. We want them to do that. Some of those are fairly insignificant, just as processes go. yeah They're looking at like the wait times on a call, the individual information that they're collecting, not particularly sensitive. The final metric matters a lot for policy. And we want to know how how long are people waiting to get the services that they expect? but they also have really sensitive data at the Social Security Administration about disability insurance claims. Some the most sensitive information that American people have, particularly vulnerable populations, and that is exactly the kind of information that we must ensure is protected and the people who are approved for access are actually the ones accessing it and that we're not inadvertently or intentionally violating laws for political purposes or otherwise.
00:16:39
Speaker
So this is squarely within the realm of issues that we track, we monitor, we try to find policy mechanisms to support. Sometimes that involves quite literally picking up the phone to the oversight committees to say, hey, just want to give you a heads up, this thing's happening. but What we're really clear about doing is not trying to throw a political football.
00:16:57
Speaker
That's not how we operate. And in particular, for this SSA example, SSA is important to every American that's paying into the insurance program. yeah And it is an an absolutely critical part of government infrastructure for data to ensure that that is protected and the privacy protections are robust.
Data-Driven Administration
00:17:16
Speaker
And I think everyone that we talked to on the Hill, ah they got it. And so there there were a lot of people paying attention, which is exactly what we want.
00:17:25
Speaker
Candidly, we always needed that too. So in in some ways, while I think people will look at the last year and say, oh my god, I can't believe x happened. In many other ways, I think it gives us new insight to say people should really care about data, always.
00:17:41
Speaker
And these are the things we're going to continue to build out to make the system better. These are the things that we can do to protect what's already there. And also, these are all the things that weren't working before, so let's fix those too.
00:17:52
Speaker
Right. So I want to come to the fixing part, but I also want to ask you about looking back the foundation. And you've written a bunch of pieces about some of the concerns that you've had with the that the administration has done um with data. But you've also written that this administration is the most data driven in American history. and i'm And I'm curious how you sort of square the two when you have an administration that's taking data down and maybe giving some private data away, but also being sort of data driven.
00:18:20
Speaker
Yeah, so I realized that many, and including those who reached out to me after I've said this, find that statement to be particularly controversial. I consider it a statement of fact.
00:18:32
Speaker
And when you look at just the proliferation of data across the executive branch, so strike the word administration for a second, The executive branch of the United States government is collecting more data than it's ever collected.
00:18:44
Speaker
That's from citizens, non-citizens, businesses, also ingesting information that's being scraped from the web. There is information that the government purchases.
00:18:55
Speaker
When you take that entire body of knowledge and recognize it's going to continue to go that way. So will I be able to say the same statement in a couple of years about the next president? Well, gosh, I hope so. But the point is, there's a lot of debate in particular over the last year about individual decisions that were made about data elements, so particular fields or columns in a data set. There were concerns about, you know should this thing be included or that thing be be included?
00:19:21
Speaker
Race and ethnicity and sexual orientation being the ones that have gotten a lot of of ah focus. And i would say the decision to ah collect or not collect data is actually a policy decision.
00:19:34
Speaker
It's a reflection of what our society values. How we use that information can also be a reflection of societal values. This a democratic republic. We need values to always be at the heart of our discussions. What we don't want, and this is really core in things like the Committee on National Statistics principles and practices for federal statistical agencies, is that there should not be manipulation of this information. We should be taking strides to ensure the data that are collected are also being presented back as accurately, as timely, and as responsibly as possible. And that requires relevance. And it one of the core things in a part of the Evidence Act that became a regulation last year in the public trust regulation.
00:20:18
Speaker
So there there are certainly places where some government data assets have been removed. Over the last year, we know that. There are data elements that have been ah adjusted based on their presentation. But let me give some more context, which is to say, this is not the first administration to do that. This is not the first time the executive branch has done that.
00:20:38
Speaker
But there are a lot more people paying attention to how we're using data right now because we're using so much data in our society every day. There are more data assets today data.gov than there were a year ago.
00:20:50
Speaker
There are 60,000 new data assets on data.gov as of the beginning of January, for just January of 2026. There is a lot happening when we talk about this broader ecosystem and all of the actors and all the people who are ultimately producing and hopefully also using this data. And it's not just about one person. should never be about one person driving and what we expect them, her, him to use the information for. We're huge advocates of open data and more specifically, openness of information, ah which means that some things will never be able to be made fully open. And I personally want a lot of my financial information to be private to me for lots of reasons. um And I think there are good cases to to be made for that. And I think it's also important for folks to have that expectation that that private information can be used responsibly. So this is the other core thing that I think has come up a lot is what it means to use data responsibly. We're having a very vibrant discussion in our society right now about a lot of very important policy matters.
00:21:56
Speaker
just going to say the word immigration without going into any details. And how we use data to support or not support immigration policy is, it that's actually a policy question to address and talk about. And when certain actions are being undertaken within the confines of the current structure of law, there may be people that are very concerned about that happening.
00:22:18
Speaker
However, if it complies with current law, then that should be a policy conversation, not criticizing the data itself. or the systems that are being used for the data, or most importantly, the civil servants who are excellent, who are executing the work that they're being told to execute.
00:22:34
Speaker
So this is a long way say it's a very complicated ecosystem, but there's a lot happening, which is why I can confidently say The administration is really leaning in on the use of data, great executive order on breaking down barriers at silos to data sharing, which covers a lot of the frustrations that we've had in statistical agencies, but more broadly across government agencies for approving permits and pesticide licenses and drug applications. There are so many reasons that we need to bring this entire infrastructure fast forward another 50 years because the the ecosystem is pretty dated.
00:23:09
Speaker
Yeah, I mean, i think I think one of the issues people might key in on is I think that phrase data driven, that phrase itself, I think sort of has implies a group or an individual, whatever it is, whatever it's ah you know whatever the noun is that the person or that organization is using data in some way, ah I would probably put in parentheses responsibly and objectively to further some goal. And I think this administration
00:23:40
Speaker
I think just having more data doesn't necessarily make the administration more data-driven. They have more data. And as you mentioned, in a year from now, they'll have even more data. So I don't know. It's tricky. But you make a really good point about the data sharing piece.
00:23:55
Speaker
And I think it was the Evidence Act where they're going to combine Census Bureau, Bureau of Economic Analysis, and... Maybe the BLS, maybe those were the three. Yeah. Okay.
00:24:08
Speaker
So I wanted to ask you a broader question though. So we sort of have that as like one possible sort of combination of agencies. In your mind, how do you envision what like a statistical or a government data system 2.0 would look like?
Future of Government Data Systems
00:24:24
Speaker
Is it like a statistics Canada type model where there's one agency that's responsible for all the data? You you mentioned earlier that like it's important to have every agency, no matter what, have a CDO. like If you were elected, if you were CDO or if you were president, really, I think is where you have to go. Absolutely not. what like
00:24:43
Speaker
Where should we be in five or 10 years? like What is your vision for for that? as you said, an updated model. Yeah, well, first of all, I hope to not be elected. i will definitely not be president. But if you give me the magic wand to make all these things happen, i think there are some core components of things that we we would expect and like to see. Having said that, the US system is often viewed as the model for the rest of the world. yeah Right after we released the the final report from the Evidence Commission,
00:25:10
Speaker
We had really great conversations with the statistical agencies and and Europe and and parts of Africa, where some of their their agencies were working on building out new systems. The US has long participated in UN statistics.
00:25:22
Speaker
And whats what's interesting is that even though the model in the United States is federated, meaning there's a bunch of different agencies that are out there compiling all this knowledge, it's not necessarily as clear that a single agency solves for all the problems.
00:25:36
Speaker
It's one of the things that we considered way back when in the Evidence Commission, and there also been just dozens of other committees and commissions and task forces over the years that have looked at this question.
00:25:47
Speaker
And the reality is it's probably going to be a continually changing context. And this is my opportunity to drop terms like artificial intelligence. yeah Things that we didn't really, we've known about AI for 50 plus years, but how to actually deploy it at scale is something that's relatively new for most of our data ah friends and family. Right? so um the The ecosystem that we're really talking about, it it builds on what exists. It does not imagine that we're just destroying and starting over. And I just say that maybe as my headline and and yeah enforcing all of these things that we really do need what's there to exist today. You mentioned this idea of better connecting Bureau of Economic Analysis, Labor Statistics, and Census for economic statistics. And this is an idea that's been floated for years. it was actually embedded in concept first in the Confidential Information Protection and Statistical Efficiency Act of 2002, SIPC.
00:26:43
Speaker
And it was one of the ideas from then chief statistician, Catherine Wallman, who was really committed to finding ways to better link and share information across the economic statistical agencies. So we just have better economic data. So unemployment could be better and gross domestic product would be higher quality. And guess what? Fast forward, we didn't fix the authorities the way that we thought we did. needed to adjust other underlying authorities and the individual agencies. And the provision, while helpful, didn't actually solve the problem.
00:27:12
Speaker
So that was reauthorized by Congress back in 2018. And President Trump signed that in 2019. We have this proposal that's out there. That is one idea for improvement. There are lots of groups like the America Statistical Association just put out a great report looking at the health of the modern federal statistical system with some ideas and suggestions. Here are some core things that that we're tracking. um The infrastructure has to be sustainable and efficient. What I mean by that is we've long suffered to find the real financing mechanisms that work. There are certain places where we expect users to pay for data because there's a direct, actionable ah value that they can add to that.
00:27:55
Speaker
Patents being a good example of that. But we also maintain an open patent database that's run at relatively low cost and can spur billions of dollars of innovation across industry. It's a really refined example. But what are the other places where we might need to be more creative with those financing mechanisms? We just had the longest government shutdown in history, 43 days, and we know that our economic indicators for the US suffered as a consequence of that. There were data collections that didn't happen. If this information is so vital to the success of the United States and our business community, we should look at whether those need to either be exempt from shutdowns or alternatively funded through
00:28:35
Speaker
the mandatory side of the budget, the same way that we treat entitlements. And where are those places in the policy ecosystem where we might prioritize that? It's certainly not everything. Don't get me wrong. i don't want anybody to say I believe in yeah know overspending on data because I'm also a huge advocate of data minimization, which is part of what we mean for the efficiency. We collect a lot of data that's really low value.
00:28:57
Speaker
And particularly data that doesn't have appropriate data standards applied where it's common sense. A good example of this is actually back in the Paycheck Protection Program, if you remember that. yeah One of the great critiques was we had all this administrative information, and we were able to see who was getting all the the loans and the grants.
00:29:15
Speaker
But basically, we didn't apply the fundamental data standards to ensure people coded states right. I think there were seven different ways that people spelled Philadelphia. These are things that we should be able to get right if we want to have good data. Now, if you map on top of that things like artificial intelligence, the AI is not going to solve all our problems if the underlying data are bad.
00:29:36
Speaker
And so it really, just to double down on this point, that sustainability, but also improving the efficiency is really key. We're also talking a lot right now about what permanence of this infrastructure looks like. And again, this harkens a little bit back to the shutdown, but actually also back to the conversations we've had over the last two decades about making the ecosystem better.
00:29:56
Speaker
So Congress in the Chips and Science Act a couple of years ago gave the National Science Foundation authority to launch something called the National Secured Data Service. Maybe scaling the and NSDS is going to be a component of that.
00:30:07
Speaker
The NAR, the National AI Research Resource, is another component of of how we can better use administrative data. When you take these pieces and put them together, they're increasingly innovative capabilities that we have in government. And so this is a lot of process conversation I'm giving you. But if you extrapolate that and say, OK, but how does that make weather data better?
00:30:26
Speaker
So the weatherman can tell me to actually have the umbrella today or that I can leave it at home. And so much of what the weather forecasts and the models are using is just government data. And that's ah that's a case where we have a really good system in place today. It's not perfect. Don't get me wrong. However, it can still be better and it can integrate more data assets.
00:30:47
Speaker
And then the last thing that I'll say is transparency, transparency, transparency. Especially over the last year, clued on a lot of challenges that folks articulated because they didn't know what was happening with data.
00:31:02
Speaker
And part of that, I think, was a function that the administration was just going so fast that they weren't communicating. Maybe they had incentives to not communicate. But long before this administration, we had really clunky mechanisms to tell people, the government's going to collect this.
00:31:17
Speaker
ah Is this valuable? How should we improve this data collection? For the the data matches and the linkages and systems of record, all of this material shows up what's called the Federal Register.
00:31:28
Speaker
And my parents don't read the Federal Register. They shouldn't read the Federal Register. They don't care about the Federal Register, but they do care about how government's using their data. So how do we create a place where they can go find out how the government is actually using their data? Well, I do this for a living, and I actually cannot tell you all the places that government has my data, even though I'm supposed to know under the Privacy Act.
00:31:49
Speaker
That's a problem, and we should fix that. So better transparency cuts in lots of ways and ultimately, i think, makes government more responsible. Hopefully that also lets people trust more of how governments really using in the data.
State and Local Government Roles
00:32:02
Speaker
Yeah. I mean, we could spend a whole hour talking about trust in data as well. i do want to ask, though, so far we have focused on the federal level. I'm curious what you think state and local governments what their role is and what their role is in the long run, especially over what's happened in the last year. I mean, are states gonna be able to rely on the on the federal data system? Should they be doing more of their own? We've seen more and more sort of consortia being built in different sort of topic areas. i mean, is that that's sort of fragmenting the system a little bit? Is that a good thing, a bad thing? Like where do you see the state and the local actors playing in this broader discussion?
00:32:45
Speaker
Yeah, so this is a really key part. And i think it it shouldn't be just the conversation around 2025 that makes us raise our hands and say, oh, yeah, state local governments, municipalities, counties are really important for all this stuff to work right. And all the other government jurisdictions that I can't remember what what they are off top of my head.
00:33:02
Speaker
um Tribes also being a major major input for for the federal government. um There are existing systems that while the federal government says this is the federal system, it's actually state and local data that effectively support it, in some cases fund it.
00:33:19
Speaker
yeah Good examples of that include the vital records system. So we use vital records through reporting that happens to the Social Security Administration, which we were talking about earlier, as well as to the National Center for Health Statistics at HHS, Health and Human Services. And it's how we do epidemiological intelligence. But we also use that same information when people are, ah when they pass away and we get a death certificate, for example, to be able to do fraud detection and adjust for improper payments. So you have one source of local information that just has a phenomenal host of benefits to government.
00:33:54
Speaker
But the private sector uses the same data. ah we We also use that information for developing new vaccines. We use that same information for banks who are trying to verify accounts. So there there are very clear cases for things like that example.
00:34:10
Speaker
At the same time, we have really robust sources that exist today that could be better, like the National Directory of New Hires. We created it in child welfare, sorry, child reforms back in the mid-90s. And the system is what we use to find non-custodial parents. We call them deadbeat dads and not always dads. um But we we know because every time I get a new job, I file a form, it gets it gets submitted there. My quarterly wages are there. There's unemployment information information that's there. There's also really sensitive information that states provide about kids.
00:34:42
Speaker
So there's lots of information that is aggregated up to the federal level because the states are providing it already. I'm using these two examples because, one, they're really expensive to operate.
00:34:53
Speaker
Two, they're really good partnerships between the federal government and state and local entities. Some of it was legislation, some of it was goodwill, but most of the data standards that we have for vital records, for example, that tell everyone to report you know this characteristic in and this particular way, that's because of the partnership that happened with an organization called NAFSCIS, the state and local entities, and the National Center for Health Statistics. It wasn't an accident. It required deliberate planning in order for that to succeed. So as as local governments and these consortia are emerging, i think it's really important to also keep in mind that's a great and fabulous starting point.
00:35:31
Speaker
And we definitely need regional collaborations in lots of places. I'm from Kansas City, right? Like Kansas and Missouri need to share some information for those of us that just used to bounce back and forth between the states.
00:35:42
Speaker
But there's there's much larger regions that have that same vested interest for economic purposes, for social services, and for a number of other reasons. So ensuring that we find efficient ways to do it, and it's not just saying like, good luck, everybody, best wishes. But in in part, the more that we can consolidate at least the policy mechanisms, we can also promote strong privacy mechanisms, help promote data minimization. And these are are those core principles that are back in the Privacy Act that I mentioned. Yeah. Do you worry a little bit that there'll be groups of states that come together to share data and there will be other groups of states that will come together and share data and they won't share? So ah Missouri is probably not a good example, but if we did, um I'm trying to think Washington state, the Washington and Oregon and Idaho and Montana. this so So deep blue, deep red states that don't share data. Are you worried about that? Do you see a path where...
00:36:40
Speaker
yeah and we sort of have our data bifurcated in that way kind of way. i am super worried about this. i call this the hyper politicization of data. It's not new, but it's becoming more emergent, at least in the contemporaneous environment. yeah We're having this very active conversation right now about what SNAP data, so for food stamps, should be reported to the federal government. And ah this administration, the Trump administration, has requested an increasing volume of information from states. Some states are complying. Some states are not. I think we can fill in the gaps of who is and who isn't just kind of loosely. And the the worry is what happens on the flip side of this administration if it switches to another political party? Do do the states then invert? Right. ah And I think that is one of the greatest risks to a sustainable and robust ecosystem that can continue to actually address the needs of the American people because we can't we can't build federated national scale models of things
00:37:38
Speaker
when those who are providing the inputs are constantly being turned on and off. so on So on SNAP, I'll just tell you, when the Evidence Commission was issuing its recommendations, it was unanimous. Every Republican and every Democrat supported there being a national database of SNAP information.
00:37:51
Speaker
But that was also built on really strong privacy protections, knowledge, and what we call radical transparency. And those are pieces that this administration and any future administration should continue to lean in on, particularly for vulnerable populations. And then we're still going to have lots of debate about how the information is used. Are we increasing access to Snap? Or are we using this to enforce Snap?
00:38:14
Speaker
And right both things can actually be true at the same time. Yeah. um OK, so um I want to swing back ah before we finish up to talk about some of the more public programs or products that that the Data Foundation is
Public Programs for Data Engagement
00:38:28
Speaker
doing. and when I say public, I mean open to the public to participate in. um So I think there are two things that that we should probably talk about. First is the Evidence Act hub that Data Foundation has. And then there's also People's Data 100, which i I think are two separate things, but related, big overlap. But um maybe you could talk a bit about both of those and how people can get involved.
00:38:50
Speaker
Yeah, so um both initiatives that we launched in the last year as part of our ongoing work to promote the health of the national ecosystem, but also talk more about the value of what's available to the American people. So we just crossed in mid-January the seven year anniversary of the Foundations for Evidence-Based Policymaking Act or Evidence Act. And one of the things that we observed over the last seven years is that as agencies were really leaning in and implementing and still are implementing and publishing documents and guidance information, plans around how they're going to build more open data assets or evaluation plans, the information was increasingly dispersed.
00:39:31
Speaker
So if, John, you know the agency you care about, you can Google their website and maybe magically find the right thing. However, most people don't know what the Evidence Act is, and that's OK. And probably far fewer people would know how to find all the resources in one place. So our main goal as a starting point was to help consolidate some of the information that was already out there about the Evidence Act, how agencies were doing, and importantly, for the questions that actually aren't just about one agency. that we are able to better synthesize the information across the entirety of the federal government.
00:40:03
Speaker
So good example, chronic absenteeism is not just an issue of the Department of Education. It's a topic that we know has huge implications for social services, health and well-being. the list goes on and on. And so we we need a lot of people to come to the table to tackle problems around chronic absenteeism in schools.
00:40:21
Speaker
So how does this ecosystem around open data and our learning agendas, these strategic plans for what we're going to build knowledge on in the future, as well as the evaluations actually tied to that?
00:40:32
Speaker
so we're hoping this Evidence Act Hub, it's available at evidenceact.org, is a great resource for our community to begin delving more into the details about what has been generated out of the Evidence Act in recent years, but also help us better understand how to look across all of these resources and documents. the Current administration, they're still producing evaluation plans and learning agendas and open data plans. And we are one organization with a perspective on how to interpret the information, but none of this really matters if people aren't using it.
00:41:02
Speaker
So researchers need to be able to actually find the learning agendas to say, well what questions does agency really care about? And I will say, sometimes we have to look past the rhetoric that might be in the document to identify the core questions that folks are interested in. is true in any political administration. All of these documents can go through political and policy reviews because they're tied to strategic plans, which are political narratives.
00:41:24
Speaker
So look past that and look at the questions that are relevant that might give us some insights about things that either you're working on or interested in or where you can help agencies do better.
00:41:35
Speaker
That's great. That's great. I mean, it's it's something that comes up in conversations I have a lot, but i but I'm with you. I don't think people a lot of people know the... For people, even for people who are interested in it and work in this space, so I don't think they know all about the exact pieces of it and how it will impact their ability to use data, put it simply.
00:41:58
Speaker
And it's complex. there's There's really no expectation from me or many others that any person out there would actually know how to find all this stuff or what to do with it. So that's one thing that we're hoping to also be able to take as the next step. We've been talking about some AI machine learning integrations, improving the machine readability, but also just overall the usability. So this is functional in the long term.
00:42:20
Speaker
Yeah. um The other project you have that maybe excites me even more is the People's Data 100. So can you explain what that is and then maybe give me like your favorite story that's come out of that so far?
People's Data 100 Initiative
00:42:33
Speaker
Yeah, i um I love this initiative. I wish we would have done it 10 years ago, yeah but we did just launch it in 2025 amongst these emerging conversations about what data sets are getting funded at what levels and how do we prioritize if there's limited capacity and resources. So we launched this initiative to help articulate better the value of the government data that we've been talking about for this whole time. And one of the observations that we made was that there's no PR agency for government data. So bunch of this stuff happens behind the scenes. It's boring. It's not observed unless something goes wrong. and that was a lot of the emergent discussion over the last year is that there were conversations or potentially threats that something might change.
00:43:18
Speaker
And so what we're hoping to do with the People's Data 100 is to fill in some of that gap. We're not going to claim that we'll cover it all. But we are identifying 100 of what we consider the most valuable data sets across sectors for the American people and businesses that are produced by the government.
00:43:37
Speaker
So this is available right now at peoplesdata100.org. It's a great resource for being able to also learn more just about the initiative broadly. We're getting hundreds of submissions and from, in some cases, government agency staff that really want us to prioritize their particular data asset. And in some cases, they're right. It is one of the most important things, in my opinion. and In other cases, people that are experiencing data in their own unique way out and and in Main Streets.
00:44:06
Speaker
And we've gotten a pretty wide variety of of submissions. um I think one of the one of the challenges that we'll have is narrowing to 100. So as we think about this as like the Time 100 for data, there are vastly more data sets that are valuable to American society and even internationally. And I think what we want to be very honest about is just because it doesn't make this a list doesn't mean it's not valuable. But we're hoping to also tell the story better. So once something's on this list, we can tell people about the things that are in the Federal Register.
00:44:37
Speaker
the information collection requests or the system of record notices that there's no reason you would understand what laws those die to or or need to if you're a normal person, but it matters for how government uses the data. So we can help connect the dots for the American people a little bit better and then articulate those value propositions and value statements.
00:44:54
Speaker
So I'd say, you know, there's hundreds of stories, literally, for things that we could talk about. One of the most obvious that's that's out there is probably the decennial census, which I referenced to the top of this call. It's written into the Constitution.
00:45:08
Speaker
um i won't pre-state it's definitely on the list, but I expect it's probably on the list. Right. Do you have a favorite one? You know, i I was actually talking about one of the data systems that i I really value a lot from my past work called the National Directory of New Hires.
00:45:26
Speaker
okay And one of the reasons that I think it's really valuable. and So this is run by the Department of Health and Human Services Office of Child Support Enforcement. And it's the system that enables us to not just have those partnerships with state and local governments like we were talking about, but really help kids. This is all about helping improve the lives and the outcomes for children who are vulnerable. And we do that by identifying the parents and ensuring they're paying their fair share of support for the children.
00:45:54
Speaker
And in some cases, also helping to manage child welfare issues. And these are like the really sensitive and important issues the government addresses. And having a value statement on that alone is so significant. But particularly we stack that right up there with things like global positioning satellites, GPS, that is a very wickedly expensive system for the government to operate. And yet, it then in turn generates billions, maybe trillions of dollars in the global economy. So we we have, and those are very different examples that I'm throwing out here intentionally, but ah they're very important for all the work that we do in our society.
00:46:31
Speaker
Yeah. Nick, this is great. So we've got a couple places folks can find you. So evidenceacthub.com and peoplesdata100.org. If people want to get in touch with you or with Data Foundation more broadly, um any place in addition to the, it's obviously the website, but anywhere else they should try to connect with you or others on the team.
00:46:52
Speaker
Well, you can always reach out to us at info at data foundation.org with questions, comments, concerns. We're on all social media. And our team is very active out in the professional speaking circuits and conferences. And would really hope that if you run across any of them at major conferences, say hi and let's have conversation.
00:47:11
Speaker
Awesome. Nick, thanks so much for coming on the show. It was really great chatting with you. And I think anyone who's working with data and is wishing you luck to improve how these data are collected and published and used. So thanks so much for coming on the show. Appreciate that. Thank you.
00:47:27
Speaker
Thanks for tuning in, everybody. Hope you enjoyed that episode. I hope you will check out the Data Foundation website and some of the efforts that they have underway, including their Evidence Act Hub.
00:47:37
Speaker
If you have a few minutes while you are getting ready for your next podcast, if you could click a button. a link somewhere here or there on your app, on your website. Give this show a rating or a review. It helps me make the show better with higher quality editing, a little more marketing, and of course, allows me to get more guests onto the show. And I should also say, If you have suggestions for guests, people you would like to hear from in the data or data visualization space, please let me know. I'm always looking for good guests to bring on the show so that I and you can learn more about how to be good data communicators. So until next time, this has been the policy of this podcast.
00:48:19
Speaker
Thanks so much for listening.