Trending:
The community surrounding the Census Bureau is concerned about the disbanding of the bureau's external advisory groups.
Commerce Secretary Howard Lutnick has disbanded three Census Bureau external advisory groups. Gone are the Census Scientific Advisory Committee, the National Advisory Committee on Race, Ethnic, and Other Populations and the 2030 Census Advisory Committee. The community surrounding the Census Bureau is concerned. Mary Jo Mitchell, the director of government affairs at the Population Association of America, joined the Federal Drive with Tom Temin to discuss what this all means.
Interview transcript:
Tom Temin: And you were a member of the 2030 Census Advisory Committee. Tell us what these committees did, the types of people that are on them. Start there.
Mary Jo Mitchell: Yeah, thanks so much for raising awareness about these committees. There were three committees at the Census Bureau. The Census Scientific Advisory Committee, which long-standing committee established in 1994, had a unique mission to provide scientific and technical advice to the Census Bureau on issues such as cyber infrastructure, survey methodology. The National Advisory Committee on Racial, Ethnic, and Other Populations had a broader mandate to really advise the bureau across all of their surveys and included not just scientific experts, but also members of the public. But the committee I was on, the most recent committee that the Bureau established, the 2030 Census Advisory Committee, was established with a more narrow mandate to really focus attention on issues surrounding the construct of the 2030 census. So we were providing feedback to the bureau at their behest on issues such as online responses for group quarters, the 2026 field test, which are going to be conducted next year, and virtual field operations. They were really rethinking ways in which they might structure the offices that are open nationwide when they conduct the decennial census.
Tom Temin: Right. And what you say then implies something that, having been around Washington as you have for several censuses now, we are close in an operational sense to 2030, aren’t we?
Mary Jo Mitchell: Very close, five years out and the ramp up both in terms of preparations and funding really began in earnest a couple of years ago. But we’ve got a big test ahead of us next year and the success of 2030 hinges really on the outcome of that test. So we were doing important work we felt and so did the bureau to help them prepare for that.
Tom Temin: And Howard Lutnick, the Commerce Secretary, said, ‘Well, the committee’s roles had been fulfilled.’ He didn’t say they’re not needed anymore. That seemed kind of strange language. What do you make of that?
Mary Jo Mitchell: Yeah, I mean, I had to have a giggle. The 2030 census, unless I woke up five years from now, hasn’t happened yet. And I think, too, with respect to the other committees that were eliminated, the work is ongoing. It doesn’t end. We were volunteers. We had been vetted thoroughly. We we’re doing important work and serving a unique role. And so I really feel that the bureau has been undermined by this decision because they were relying on us and on our expertise and time to advise them on a wide range of issues.
Tom Temin: And the bureau does face some of the issues that every polling and surveying organization is facing. People don’t respond to mail so much anymore. They don’t even respond to online because nobody trusts anything. You get to click online anymore and it’s increasingly difficult and expensive and almost impossible really for enumerators to walk around knocking on every door. There’s an image, maybe, that worked in the 1940s or something. And so what are the big issues facing the census as the ultimate statistical organization for the government?
Mary Jo Mitchell: Well, there are several. First of all, they’re not just are conducting the largest civilian peacetime operation, the decennial census, they’ve got other mandates within their mission. I mean, they are in the field all the time conducting the American Community Survey, which in 2005 replaced the decennial long form. They are doing other surveys, such as the Current Population Survey, which is really the driver of our understanding of unemployment, employment in this country. They’re doing the National Health Interview Survey. They’re doing this in collaboration with other federal agencies. So their work is very complex and not just necessarily focused exclusively on the decennial census.
Tom Temin: Right. And many sectors rely heavily on the reliability and regularity of that data across many domains, medicine, economics, consumer purchasing, all these things, fair to say?
Mary Jo Mitchell: Oh yes, definitely. The ACS combined with the decennial census is what informs over $1.5 trillion in annual federal spending and is used by businesses to determine where they’re going to build factories, how to understand characteristics of workforces, child care providers, medical providers as you noted, scientists who I represent. These are foundational data across all industries and data user communities.
Tom Temin: We’re speaking with Mary Jo Mitchell, she’s director of government affairs at the Population Association of America, and you have a project there called the Census Project. Tell us about that and what the aims there are.
Mary Jo Mitchell: The Census Project is a separate organization that I co-lead. It’s a coalition of 800-plus national, state and local organizations representing census stakeholders and data users from the public, private, and nonprofit and scientific sectors. We are supported by a consortium of foundations that decided years ago that it was important to have a platform that could not only provide information to census stakeholders about Census Bureau operations and activities, but also could come together under a big umbrella to make decisions about what we think the Census Bureau needs by way of funding. So we develop an annual funding recommendation that we communicate to Congress and to census stakeholders in hopes that they will embrace it and communicate it as well.
Tom Temin: All right. And the issues connected to the Census Bureau, the lack of maybe external objective, we hope, advice to them, the falling survey results, the difficulty of getting data in the world the way it is, points to the larger issue of the statistical infrastructure of the United States government generally. And I know you’ve thought about that, and what’s your assessment of where things are headed?
Mary Jo Mitchell: We’re very concerned. We’ve been concerned for years as have colleagues of mine representing other scientific research and statistical organizations, because we’ve seen, as you noted, Tom, declines in budgets for these agencies, declines and response rates, stagnant sample sizes, workforce challenges that these agencies have been facing with retirements, the inability to sort of innovate and expand surveys and their core missions. So we’ve taken on a new urgency even just since the end of January where we saw, for example, data being pulled from federal sites. We saw whole websites going down. That’s something I never thought in my career I would ever see. These are public goods. These data were paid for by the taxpayers. We expect, just as we do when we turn on the tap to get water, we expect we go to a federal agency’s website to be able to access data, and we saw that occur. Many of these data products and data have been restored as a result of legal challenges, but we’re still seeing that some data variables are missing and data documentation that’s necessary to make sense of these data are missing. So this is undermining trust in the federal statistical system. And when you undermine trust, you’ve got an existential threat then that you’ve got to contend with. That’s on our minds, as well as worrying about the fate, quite frankly, of some statistical agencies. I mean, essentially, the National Center for Education Statistics has been shut down. And I represent scientists and I’ve been hearing from scientists whose graduate students are unable to get simple administrative requests approved to access data in order to do their dissertations right now. These are problems I never thought we would be talking about vis-à-vis threats to the Federal Statistical Agency, and yet we are. And yet we haven’t even seen the president’s fiscal year 2026 budget released yet, and I don’t know what that will include.
Tom Temin: Yes. Washington I think is holding its collective breath, or I guess outside of Washington, they say the swamp is holding it’s collective breath. I guess it depends on your point of view. With respect to the statistics and the gathering and I think of places like the Bureau of Labor Statistics, which has been an essential source for business planning and analysis for decades, just rates of return of surveys are going down. New methodologies are always being developed to get that data. What’s your feeling? I guess we’re getting a little off topic, but it came to my mind. Does artificial intelligence have any role in the future of data gathering? I’m worried that it could produce nonsense that looks like good data.
Mary Jo Mitchell: You’re absolutely right. And the population scientists I represent, demographers, economists, sociologists, these are folks who you need to bring to the table in order to make sense of AI. AI is certainly an important tool, but you still need the experts who can help, I think, construct the best ways to use AI so that you aren’t getting nonsense data, as you say.
Tom Temin: So people are thinking about this idea. It comes up and gets swatted away every year of extrapolation in connection with the decennial count because I think you have to count every nose. Is that getting to be kind of an anachronism of the 21st century to count every nose? Is there thought that extrapolation could be used in a equitable manner?
Mary Jo Mitchell: I haven’t heard it in the context of AI, certainly there’s a lot of attention paid to the potential use of administrative records for being able to count nonresponsive households. There are only so many times you can go back to a household to try to knock on the door and follow up with them. And Congress had its way, they tried to do last year. They would further constrain the Census Bureau’s ability to do nonresponse follow-up by limiting them to two contacts for surveys. So that would be problematic. But I hadn’t heard about AI going the way of actually producing our census. I have heard about interest this administration may have in revisiting using or relying on the United States Postal Service to do that.
Tom Temin: Well, that’s a whole other topic. I wish we had time.
Copyright © 2025 Federal News Network. All rights reserved. This website is not intended for users located within the European Economic Area.
Tom Temin is host of the Federal Drive and has been providing insight on federal technology and management issues for more than 30 years.
Follow @tteminWFED
