Why Should You Care If A Statistical Agency is Being Reorganized? | Stats + Stories Episode 75 / by Stats Stories

Lisa LaVange is the 2018 President of the American Statistical Association and she is PhD, is Professor and Associate Chair of the Department of Biostatistics { add link to dept } in the Gillings School of Global Public Health { add link to Gillings SPH } at the University of North Carolina at Chapel Hill. She is also director of the department’s Collaborative Studies Coordinating Center (CSCC), overseeing faculty, staff, and students involved in large-scale clinical trials and epidemiological studies coordinated by the center.

Ronald L. (Ron) Wasserstein is the executive director of the American Statistical Association (ASA). Wasserstein assumed the ASA’s top staff leadership post in August 2007. Prior to joining the ASA, Wasserstein was a mathematics and statistics department faculty member and administrator at Washburn University in Topeka, Kan., from 1984–2007. During his last seven years at the school, he served as the university’s vice president for academic affairs.

+ Full Transcript

ROSEMARY PENNINGTON: Earlier this fall, the USDA announced plans to move two statistical agencies it relies on for data to offices outside Washington D.C. Scientists, economists and statisticians were among those protesting the move when it was first announced, suggesting the move would be detrimental to USDA science-based mission. Their concerns seems to have made no headway as the USDA still plans to move the agencies including the Economic Research Service outside D.C. The American Statistical Association recently issued a statement criticizing the decision with ASA president Lisa LaVange saying the move “undermines evidence based policymaking in the food, agriculture and rural sectors of our economy and society”. That's the focus of this episode of Stats and Stories, where we explore the statistics behind the stories and the stories behind the statistics. I'm Rosemary Pennington. Stats and Stories is a production of Miami University's departments of Statistics and Media Journalism and Film as well as the American Statistical Association. Joining me in the studio, our regular panelist John Bailer chair of Miami statistics department and Richard Campbell of media, journalism and film. Our guests today are ASA president Lisa LaVange and ASA Executive Director Ron Wasserstein. Lisa and Ron, thank you so much for being here.


LISA LAVANGE: Thank you for inviting us.

PENNINGTON: I'm just going to ask this question, I guess I'll direct this at Lisa. Why did the ASA feel compelled to issue this particular statement?

LAVANGE: The ASA feels very strongly that the statistical agencies within the federal system are key providers of accurate and objective data, data that are very much needed to set policy in our country. The Economic Research Service of the USDA is one such agency, and the idea to move this agency outside of Washington raised several alarms for us. One, the disruption itself and the possibility that staff, very skilled staff, and of course we focus on statisticians but all of the staff could choose not to move and we would have a period where we would be understaffed quite significantly at the Economic Research Service, but also because there was an internal realignment for the ERS to be placed under the group that they have sort of a regulatory authority for. And this brings up the independence of the ERS. You know, we comment on a lot of things at the American Statistical Association, but we focus on things that affect our membership very much. We are the largest organization of statisticians with almost 20,000 members and moves like this have a direct impact on our membership. We also feel very strongly about protecting the practices of statistics and we felt like this was touching both of those points for us.

PENNINGTON: So it's not just the move outside D.C., but it's also a move where the independence is going to be lost. So it's going to be a different space where it can't do the work that you think it needs to do?

LAVANGE: I think the issue isn't so much where the ERS is located, although there does tend to be a huge market of statisticians that are highly skilled and able to fill the jobs that the ERS needs, that is not to say agencies can’t exist in other parts of the country. Of course, there are plenty of wonderful statisticians and others in other parts of the country. The suddenness of the move and the announcement being made without adequate motivation so to speak, or justification for the move, we couldn't really understand why it was happening. But also the suddenness of it and the fact that the move could leave the agency very understaffed until things got settled down, but I also think, and Ron actually knows more of certain details than I do, that this was also accompanied with a budget cut, which just compounded our concern that we would end up with an understaffed agency in the middle of a move that could be fairly chaotic at least in the short term. And then coupled with that is the realignment organizationally, which would take away some of the independence of this agency and independence in our statistical agencies is terrifically important. You want the data to be handled very objectively and that was the other aspect. And Ron, you undoubtedly could explain this better than me.

RON WASSERSTEIN: Lisa, thanks for that great explanation and...John, I want to try an example out on you. Suppose that the governor of Ohio announced that Miami University was going to be relocated to an unspecified location, to solve problems that no one thinks that the university has. And while we are at it, the university will report to the lieutenant governor of Ohio instead of to the Department of Higher Education. Would that really improve an already excellent Ohio university? I don't think so. But I'd argue that it's the equivalent of what's going on here. Or, if highbrowed university examples don't work, how about this: let's say we're moving the Yankees from New York City to Cooperstown to attract more Hall of Fame players. Save a ton of money, being you know, upstate New York! It just doesn't make sense.

JOHN BAILER: Well, you know, Ron, let me follow up on that real quickly. And you know, the example that you give of a university moving, it's pretty clear what the mission is of a university and how it serves the community in which it operates. You know this is a harder story to tell in terms of a statistical agency, you know, why should the general public care about this story?

WASSERSTEIN: First of all, I think it's important to think about why the general public should care about the federal statistical system at all. After all, it's probably the most important government agency that no one knows about. It's really not an agency. It's a system of 13 agencies called principal statistical agencies and many other data collecting and reporting units in the government. But the policies and the decisions that this system informs, cover the entire public and private spectrum, from an 18 year old deciding on a career to a Fortune 500 company deciding where to place its new facility, all the way of course to the big one, the Census, determining representation in the U.S. House of Representatives. And John, these agencies were set up to provide us as Lisa mentioned, a trustworthy source of objective, relevant, accurate and timely information for where people have to make decisions, both inside and importantly, outside the government as well. These data help them understand where we are now, how we got here and where we need to go from there. And so your question about comparing this to a university and understanding its mission, I would say that ERS has a very clear mission, it's had it for a long time, it's a research institution. In fact, following on to the university comparison just a little bit farther, it's the 3rd highest ranking agricultural economics research institution in the world. All the major universities in the world ranked below ERS and the only two institutions that rank higher, the World Bank, and the International Food Policy Research Institute, are both located in D.C. So here's where the top organizations that study these problems are located and it makes good sense to have it be located in D.C.

RICHARD CAMPBELL: So following on your example there, if Miami were to be relocated, there would be a lot of news stories about this. So in this particular example I'm…I was puzzled and just doing some preliminary research, how few news agencies picked this up. I didn't see a Time story, didn't see a Wall Street Journal story, unless I missed it…I see the A.P. picked it up, but not until November and then I saw Politico wrote something about it. So part of this, you know, this is the most important agency that nobody knows enough about. What should journalism's role be here? What should, what could ASA do, for instance, to make this more of a sexy story, something that we should care about, if anything? This was my trouble. This does seem like a really good journalism story and I didn't see anybody pick it up.

WASSERSTEIN: Well, I'll make a stab at that and Lisa has some thoughts I'm sure too. But I would say the big challenge that we face is that the media, and the legislature and all of us are extremely distracted by everything else that's going on. So when you think about the totality of news items that are happening and things that can draw the attention of the press, this one is a little bit harder for people to get excited about, and there's also a subtle issue when it comes to the Congress itself. On the one hand, it's something that we feel like we can make a good case to members of Congress about but on the other hand there's a chance that their home area will be the place that’s selected for the new ERS Who wants to get in the way of that?! And I mean you know all those places didn't get Amazon.com, so maybe they'll get ERS

CAMPBELL: One of the stories that I picked up was from North Dakota where the Republican senator out there is already calling together a group to try to lure a part of this to North Dakota. So I see some things going on, on that front.

WASSERSTEIN: If I could pursue that just one step further, we argue that moving is a bad idea. But I suppose it could also be argued that moving is a good idea and our point is, why not have that debate? Why not have a process by which the idea is floated and feedback gets received? But that's not what happened. This was announced in August when Congress was away for summer recess. We are unaware of any consultation with stakeholders, with other agencies, with Congress, and the government doesn't have to work that way. There are plenty of examples of consultative processes that have involved the federal statistical agencies.

PENNINGTON: So you mention the Census and how important a Census is, for you know, helping decide how many representatives a state gets, right? So I think the reason people know about the Census is, there's a concrete understanding of what kind of data they collect. Economic Research Service, you know, it's important and fundamental but what exactly is it that they are collecting and doing? I think I saw it referred to a sort of information infrastructure for USDA What exactly does that translate to?

WASSERSTEIN: The ERS provides data and publications on a world of things related to agriculture and to our lives. Animal products, crops, the farm economy, farm management and farm practices, food and nutrition assistance, food choices and health, food markets and prices, food safety, something that's important to all of us, the rural economy and lots more. Those are things that people need to know but oftentimes we have this data and we actually don't give any thought to where it comes from.

BAILER: Well, you know ASA doesn't do this very often. You know, these kinds of statements. And so can you talk about kind of what are some of the triggers that would lead ASA to say wait, we have to weigh in on this?

WASSERSTEIN: I'd like to throw that one over to Lisa with just this one comment. The ASA was founded way back in 1839 as an organization to support the 1840 census. So there's a long history of the ASA. The entire history of the ASA is tied with the government in some ways. But Lisa has been on the board for a couple of years and has seen why the ASA takes up causes. Lisa, what are your thoughts?

LAVANGE: Thank you. So as I mentioned earlier we are particularly attuned to events that might negatively impact our membership, and our members are the statisticians working in these principal statistical agencies, the 13 agencies that Ron mentioned, including the Economic Research Service and the USDA. So any action on the part of the government or anybody else that we feel is to the detriment of our members, that is one of the tests that we put events to, to decide if we want to speak out against them. Another is just simply bad statistics, and we would speak out if we saw something happening, where information was getting manipulated or you know something untoward was happening with official statistics. The board actually set up a very rigorous…it's really a test where issues that are brought out, OK, can you speak about this issue, and we go through, Ron does this in advance of the board meetings and then the board decides, is this something we should speak out against, it matches, or, you know, it passes our test? If there's some endangerment to our members, there's something counter to good statistical practice, our mission is to promote and protect the practice of statistics in our profession and so when we see something like this you know if it's egregious enough we speak out and I think I mean when you mentioned that we are worried about the ERS being able to do its job which Ron already told you was ranked quite high in the world in terms of Ag Econ. We're worried about it losing its independence and then moved from being…from reporting up to another office inside the USDA and we're also worried that this was sort of sprung on the people pretty quickly and the reasons given for the move just don’t hold up under scrutiny, right? The reasons that were given included improving the USDA’s ability to attract and retain highly qualified staff. There's no attrition at the ERS that is any higher than the rest of the USDA and as Ron pointed out, you have a huge population of skilled statisticians in the Washington area because there are so many jobs in that area. To have to place these important USDA resources clustered with stakeholders was one of the issues and you could argue that that could be a good thing to move the economic research out of D.C. and closer to maybe the farmers that you know generates or is the source of a lot of the data that the ERS analyzes. So that's something that could be argued if a public argument was able to be had, and then it could benefit the American taxpayers and that would be you know somehow lowering the cost because it's cheaper to operate in North Dakota perhaps, for example, than it is in Washington D.C. but that doesn't take into account the cost of the move itself and in particular the cost of replacing the staff that we anticipate might not move with the agency and that I think that's what really gives us pause. In terms of all of the problems we could speak out about, I think this one caught our attention because we didn't see a public dialogue happening, we didn't understand the reason, the reasons that were given ostensibly just didn't seem to hold water so to speak, and we think this is a really good agency that we don't want to undermine. They do a valuable service providing economic data about agricultural systems that's a huge part of our country and so we felt like we had to say something about it.

PENNINGTON: You're listening to Stats and Stories and today we're talking about the American Statistical Association’s decision to issue a statement about USDA's announcement that it will move two statistical agencies including the Economic Research Service outside Washington D.C.

CAMPBELL: So Ron, listen, usually, I want to talk as a skeptical journalist about what's really going on here. So, part of it…and Ron mentioned it, the announcement in August when government likes to announce things that it wants to sort of fly under the radar, not have people notice, which happened, but how much of this is…I mean, if you relocate a lot of people are going to not want to relocate. So part of this is, how much of this is just savings that they're hoping to cut staff and this is one way to do it. We're just going to cut the number of statisticians we have, a lot of them won't want to move, so this is really part of what's going on. And how much is it in this age of sort of skepticism although that's really the wrong word here…the sort of anti-data, anti-science, that they don't, they're not really particularly interested in the data…the government that we have right now is not particularly interested in what statistics can reveal about sort of the state of climate change, the state of wage disparity, all the things that we should be concerned about today.

WASSERSTEIN: So that's a grand problem, isn't it? It's a science-wide problem. But the fact is that there's going to be other governments coming along, other administrations, there's going to be a need for this data. But it isn't just the federal government that needs this data. State governments need it. State Bureaus of agriculture and so on need the data, state legislatures need it, so that they can understand how the economy of their state is working, in order to predict, for example, what their revenue situation is likely to be further on. The question I have, and I think where there's a media story is, what you may have been hinting at, Richard, which is, what is it that we know when we have this data, that we'd rather not know? Would it be easier to make certain policies or to make some new laws, if we didn't have this data staring us in the face that says this program works or this policy helps ensure that more children are food secure? That's the question I have, is what is it about a high functioning data organization that's so problematic that it needs to be reorganized and busted up, when there's no clear problem that really needs to be solved here?

PENNINGTON: So I have a question. We've seen…we had Andreas Georgiou on the show earlier this year and there certainly seems to be a lot of stories in the news about statistical agencies, not just in the United States but in other places, being under attack or the work being questioned and I wonder given the statement that you’ve issued, even though I know you don’t issue it very often, how concerned is the ASA about what seems to be an increase in attacks on the validity of the work that statisticians do?

WASSERSTEIN: We're deeply concerned about that. We're concerned about it worldwide, and we're concerned about it in the U.S. Things that crop up in the U.S. are the sort of slow devaluing of these agencies as reflected in where they get placed within the organizational structure. You can sort of push down and downgrade an organization by just re-organizing it a little bit, putting it somewhere else, maybe making the appointee to head that organization not require Senate confirmation…that's another way of downgrading the status of an organization. And because we're concerned, we started a program with a whole bunch of other organizations. That's what called Count on Stats. It's a proactive campaign to distinguish federal statistics as absolutely essential to the functions of our economy, society and democracy, and to point out how much businesses count on stats, how communities count on stats and how we all count on stats to be credible and accurate.

BAILER: Are there other stat agencies or actions by stat agencies that ASA is monitoring and weighing in on, that are causing some concern?

WASSERSTEIN: There are, there are enough of them going on that I'm pretty sure I'm not going to remember all of them. But we have an interest in what's going on, with regards to, for example the Bureau of Transportation Statistics. There's been a proposed realignment of the Bureau of Transportation Statistics which is part of the Department of Transportation. BTS has faced budget constraints for years and now there's a proposal to move it out of the department into another department. That is a real downgrade to the organization, and once again would jeopardize its reputation as a policy neutral federal statistical agency.

PENNINGTON: Lisa, I know you're an educator, you're a professor. What can we do as educators to help people, students, you know these young adults who are going to go out and become consumers, have a better understanding of what it is these agencies do and how important they are? How can we communicate in a classroom, whether it's a journalism classroom or a stats classroom?

LAVANGE: That’s an excellent question. When I think about that, I'm in a department of biostatistics, which offers courses to our students on everything from basic probability and inference, to statistical genetics, to clinical trials and epidemiology and so forth. We do have a sample survey course that tries to educate the students about the quality and the value of large scale federal surveys on which policy is made. The National Agricultural statistics service, NASS is another agency adjacent to ERS that's currently inside the research education and economics group. They conduct a survey of farmers I believe, Ron, you can correct me if I'm wrong, Linda Young one of our very active ASA members currently directs service there. You already know about the census, there's surveys of health care expenditures or surveys of the health of the nation, there are surveys of economic indicators and so forth. These are all over the government, all of these federal statistical agencies, they collect data in scientific ways and then they inform policy with that data. I personally think courses on sample survey should be part of every statistical curriculum, they are in most graduate programs in statistics, and they get students interested in the function of the statistical agencies and then the students might go on after graduate school to be employed by them. I think, you know, all the students in the mathematical and statistical sciences these days are hearing and learning about data science, they're learning about big data, they're also hopefully learning about concepts like objectivity and reproducible research. These are all extremely important and they have become part of the vernacular of statisticians, but also now the curricula and they are being taught in more and more of our courses. These concepts are what's fundamentally at stake here. The people designing the studies to collect the data and the people analyzing the data need to be objective, so that the data, the results of that analysis are interpretable and useful for setting policy and that’s where this independence concept comes in. It's a lot easier to conduct reproducible research when you are objective and you are independent from the eventual users of the data. This realignment of the ERS going back to the original topic, is an attempt to make the ERS or a proposal from a research, education and economics area which is just a research and educational area of the USDA and to move it into the office of the chief economist where policy is set. And we’re arguing that that independence that is needed is going to go away. You are going to move the statistical agency that needs to be objective in collecting data under the user of the data that has policy setting agendas and that’s where the politics gets mixed in. So I think one of the one of the thrust of the ASA is to keep these federal statistical agencies independent and not affected by politics. The civil servants that work in these agencies don't change with the administration, they’re not elected to them. They’re all civil servants in many cases have spent their entire career or a good portion of their career being very dedicated in contributing to the work of the agencies. Their realignment into a policy arm of the USDA it just doesn't make sense to us. And that's a message that is very teachable to students, what it means to be objective, what are good practices for data collection, data analysis, data reporting, minimizing bias, improving precision and these are all the things that we teach our students in statistical classes. How do you do reproducible research, so that your research can be reproduced, verified by another statistician you know, in another part of the world? All of these things are very teachable and this move goes up against the things that we're trying to teach and I think that's you know one of the reasons I, anyway, get passionate about things like this.

PENNINGTON: Well that's all the time we have for this episode of Stats and Stories. Lisa, Ron, thank you so much for being here.

BAILER: Thanks guys.

WASSERSTEIN: Thanks for having us.

PENNINGTON: Stats and Stories is a partnership between Miami University’s departments of Statistics and Media Journalism and Film as well as the American Statistical Association. You can follow us on Twitter, Apple podcast or other places you can find podcasts. If you'd like to share your thoughts on the program send your e-mail to statsandstories@miamioh.edu or check out our fancy website at statsandstories.net and be sure to listen for future editions of Stats and Stories, where we discuss the statistics behind the stories and the stories behind the statistics.