2010 Census Stirs Debate In Washington More than 100,000 people started work for one organization this month: the U.S. Census Bureau. The 2010 census doesn't start until next year, but it's already generating strong debate in Washington. The results will affect many things, including congressional district borders.
NPR logo

2010 Census Stirs Debate In Washington

  • Download
  • <iframe src="https://www.npr.org/player/embed/103264526/103264522" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
2010 Census Stirs Debate In Washington

2010 Census Stirs Debate In Washington

  • Download
  • <iframe src="https://www.npr.org/player/embed/103264526/103264522" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


From NPR News, this is ALL THINGS CONSIDERED. I'm Jacki Lyden.

We'll begin tonight with a piece of news we haven't been hearing much of lately. More than 100,000 people started work for one outfit just this month.

Mr. FERNANDO ARMSTRONG (Regional Director, United States Census Bureau): We are doing it in waves. We have already deployed the first wave of listers. And then we are, right now, as we speak, training a second group that will start in the field on Monday.

LYDEN: Fernando Armstrong supervises a big chunk of those new workers for the Census Bureau. He started out as a canvasser in Puerto Rico. Four decades later, he's director of the five-state Philadelphia region.

The actual census doesn't start for a year, but the work and the debate here in Washington is already going strong. This time around, Fernando Armstrong has had no trouble filling all those temp jobs, 1.4 million of them by the time it's all over.

Many of those new employees spent this weekend trudging up and down the streets, verifying addresses - every address. Armstrong remembers the days when census takers unrolled enormous paper maps, like a ship's captain from centuries ago. They would take out a pencil and make…

Mr. ARMSTRONG: A little dot, a little mark on the map to identify where that address was located on that map and then, you know, thousands and thousands of these paper maps would be shipped to our processing office in Jeffersonville, Indiana.

LYDEN: And there, an army of workers recorded every little dot, a lot of room for error. These days, canvassers have GPS, a big change. And as mundane as the grunt work is, it matters.

Mr. ARMSTRONG: This gives us the advantage of collecting the GPS coordinates for every one of the addresses so that we can place the resident precisely in the right congressional or representative district that they need to be.

LYDEN: That's the key and why a simple count is generating such heat here in Washington. The 1990 census missed so many Americans that 10 years later the Clinton administration tried to employ statistical sampling, the kind of math pollsters use. That lit a political firestorm that's still burning.

Republicans argue that sampling violates the Constitution, which mandates a head count. Democrats said the census doesn't count every head and that Republicans were worried about minority groups gaining political power.

The congressional debate spilled into the Supreme Court, and ultimately, the Bush administration discarded sampling entirely. So, what's the bottom line here? How much federal money goes to your state? How many representatives your state gets in Congress, and how many people vote on your behalf in the Electoral College?

All that makes the census a high stakes game. And would you agree with that, Newt Gingrich?

Mr. NEWT GINGRICH (Former Speaker of the House): Absolutely. When you add in, it also affects how we design state legislative districts, how we design city council districts, county commission districts.

The census is one of the most important mechanisms we have for understanding ourselves, who we are, how many of us there are, and how to distribute power.

LYDEN: Newt Gingrich is, of course, the former speaker of the House for the GOP, and he joins us from his office in Washington.

Thanks for being with us.

Mr. GINGRICH: Oh, I'm glad to talk with you.

LYDEN: Now, there's been a fight over the census almost as long as there's been one. It was one of the two crucial issues Senator Judd Gregg cited when he withdrew as President Obama's nominee for Commerce secretary. You've been involved in fights over the census for more than a decade. Why does it remain such a touchstone issue for the Republican Party?

Mr. GINGRICH: Well, we want all Americans to participate and we want all Americans to be counted accurately, and we are very worried both that managerial incompetence will lead to stunning miscounts and that somebody will try to apply some kind of abstract, theoretical, statistical model, which will totally distort the system.

And the federal bureaucracies have been getting steadily less capable of doing anything. And so, there's a great fear you could have the Katrina effect applied to the U.S. census and suddenly have a U.S. census that is just fundamentally, profoundly wrong.

LYDEN: Well, let's talk about one that was. The 1990 census missed about 10 million Americans, and that is almost 2 percent of the population. The Clinton administration then introduced statistical sampling for the 2000 census, using a statistical tool to account for people who might have been missed by that traditional head count you're talking about.

Mr. GINGRICH: You know, I don't think there are abstract, theoretical models. If you didn't count somebody, how did you know who you didn't count?

LYDEN: But what you're talking about is research polling, and every single poll uses statistical sampling to give us data for all sorts of things.

Mr. GINGRICH: Polls are statistical samples. They are indicators. But we're not talking about an indicator. We're talking about the absolute number from which political power in America derives. So they're telling you that the ability to count people who weren't counted is sufficiently high that they can theoretically adjust the model based upon their interpretation of who should have been there.

LYDEN: Mm-hmm.

Mr. GINGRICH: I think that's frankly pretty whacked.

LYDEN: But you call for outsourcing the census to private companies. Why is that any better?

Mr. GINGRICH: Well, first of all, what I've said is that it's certainly better than paying ACORNs and other activist groups who have a political agenda. But second, there are institutions that go to every single neighborhood.

FedEx tracks every package, every day while it's moving. Now, that's the level of accuracy you'd like to have in the U.S. census.

LYDEN: Well, I'm not going to argue with your experience of FedEx versus mine, but I do want to bring this up. Democrats say the reason Republicans resist statistical sampling is that they just don't want to count people who aren't likely to vote Republican. What do you say to that?

Mr. GINGRICH: Well, first of all, you used the key word, count. I have no opposition to counting everyone. I want to count everyone. The Founding Fathers felt very deeply about a census. They require it in the Constitution.

The first census was taken in 1790. It was taken using assistance to the U.S. Marshals. They went around and counted everybody. It was the first really accurate census in the modern world. And they did it cleanly. And people generally thought they did it pretty honestly.

LYDEN: The operation we have in place right now was in place during the Bush White House. Do you think that President Obama has too much control? It's not really going to change, is it?

Mr. GINGRICH: Who knows? They're going to get down to a final opportunity to fundamentally rig the game, and they're going to try to use a statistical analysis to rig the game. But let me go back to what I said a minute ago…

LYDEN: But isn't that just conjecture on your part because the rules are actually clear here, are they not? And they really are Bush White House rules.

Mr. GINGRICH: Well, they're going to change those rules. Do you really believe they're going to implement precisely what the Bush White House left? You don't think they're going to go to statistical adjustment? The shifting comes at the end. The shifting comes after they count. The shifting comes when they announce suddenly, oh, there are actually 600,000 more people in Los Angeles than we thought.

Now, we haven't found them, but we're confident they're there. And so, we're now going to add 600,000 to Los Angeles based on theoretical people in our computer simulation, which has told us that they're there. And if they're not there, they should be there. And if they haven't gotten there yet, maybe they can arrive by tomorrow night from some foreign country.

LYDEN: Newt Gingrich, former House speaker. We want to thank you very much for your time today.

Mr. GINGRICH: Thank you.

LYDEN: Now, we turn to Robert Shapiro. He was an undersecretary of Commerce during the Clinton administration, and he supervised the 2000 census.

Mr. Shapiro, thanks for being with us.

Mr. ROBERT SHAPIRO (Former Undersecretary, Commerce Department): It's a pleasure to be here.

LYDEN: Now, you heard the Republican critique of President Obama's approach. How much can he shape the 2010 census?

Mr. SHAPIRO: He can't shape it very much. The census is the largest non-military operation the government puts on. The 2000 census involved over a million people. It involved $7 billion. It involved 500 operational centers.

LYDEN: Well, give us a little sense of how the statistical sampling works.

Mr. SHAPIRO: Right.

LYDEN: We understand it's not going to be done for 2010.

Mr. SHAPIRO: Well, one million households were chosen at random, that is with no particular bias. You go to those households. You, in effect, re-interview them directly, then you look at the differences between the populations that comes out of the sample and the populations that have come out of the total census. And where there are differences, there are statistical ways to adjust. And that's what you do. The sample is not in place of an enumeration. It's in addition to an enumeration.

LYDEN: If you were running things this time around as an economic advisor inside the White House, what would be the single most important reform for the census that you'd enact?

Mr. SHAPIRO: Well, I think the single most important reform would be the use of statistical sampling. The government, by law, must use census data in these formulas for the distribution of funds. The Bush administration quashed the use of the 2000 sample to correct the enumerated count. So they were perfectly happy with the count.

Let me say the political implications of sampling are vastly exaggerated. The notion here is that a count, which increased the numbers of minority people found, would somehow change congressional results. That's very naïve, politically.

LYDEN: Does that surprise you that as we approach this census coming up that this is still such a hot debate, still such a partisan debate?

Mr. SHAPIRO: Well, it doesn't surprise me because there's so much at stake. We live in a period in which we worship science or the results of science, but there's also a very strong reaction against science so that we have an extended national debate over matters, which - about which there is no scientific debate. It is the exact same thing with statistical sampling.

There is no science that supports the view that this would not improve the count, and yet the first President Bush rejected the use of sampling in 1990. The second President Bush rejected the use of sampling in 2000. And now, the second President Bush will have precluded the use of sampling in 2010. That's 30 years of ignorance being imposed on science and with significant political results.

LYDEN: Thank you so much. These days, Robert Shapiro is chairman and founder of Sonecon, a private finance consultancy.

(Soundbite of music)

LYDEN: He was an undersecretary of Commerce during the Clinton administration.

Thanks very much for your time.

Mr. SHAPIRO: My pleasure.

LYDEN: President Obama's choice to run the Census Bureau is still awaiting his Senate confirmation hearing. His name is Robert Groves. He worked on the 1990 census, and he's a University of Michigan professor who specializes in survey research.

Copyright © 2009 NPR. All rights reserved. Visit our website terms of use and permissions pages at www.npr.org for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.