printable banner

U.S. Department of State - Great Seal

U.S. Department of State

Diplomacy in Action

The 2010 Census: Why Do It, How Does It Work and How the Information Is Used

FPC Briefing
Daniel Weinberg
Assistant Director, Decennial Census Programs, U.S. Census Bureau
Foreign Press Center
Washington, DC
March 26, 2010


Date: 03/26/2010 Location: Washington D.C. Description: Daniel Weinberg, Assistant Director, American Community Survey and Decennial Census, U.S. Census Bureau, briefs about the 2010 Census at the Washington Foreign Press Center on March 26, 2010. - State Dept Image


Video

10:00 A.M. EST

MODERATOR: Welcome to the Foreign Press Center. Today we have with us Daniel Weinberg, who is Assistant Director for the Decennial Census program at the U.S. Census Bureau. And he’ll tell you everything about this 2010 Census.

MR. WEINBERG: Probably more than you want to know, also. (Laughter.) Well, I’ll start with the reason we do a census. This is in the U.S. Constitution. The Founding Fathers wrote this. The Constitution – let me give you a quote from the Constitution: The actual enumeration shall be made within three years after the meeting of the Congress of the United States and within every term of 10 years in such manner as they by – they shall by law thereat.

So why do they do that? Why did they have a census every 10 years? There’s another part. The representatives shall be apportioned among the several states according to their respective numbers, counting the whole number of persons in each state. So basically the key reason we do a census is to ensure that every community gets its fair share of political representation through congressional apportionment, in particular.

Basically, every state is entitled to one representative in our House of Representatives, but that accounts for only the first 50 representatives, and we have 435. The others are given to each state – apportioned is the word we use – on the basis of the population, and there’s a particular formula that’s been fought over many times over the past 200 years, but there is one in law that is used.

So the second reason is to draw the districts so that they’re of equal size, and that was a law that came – I’m sorry, that was a Supreme Court decision in 1964 that said that every district had to be the same size, so that all congressional districts are drawn so they have the same number of people based on the census, and that’s actually carried down to many other layers of the U.S. So the states will redraw their legislative districts for their assemblies and senate, their legislatures, so that they have districts of the same size. And then the local jurisdictions, whether, say a city, would redraw their districts so they’re the same size.

I live in Fairfax County, which has over a million people, and after every census they redraw their supervisory districts, is what they call them, so they’re all the same – have the same number of people.

Now, this census population data, as well as other census data, are also used to distribute federal funds. As a matter of fact, over $400 billion a year is distributed on the basis of census data and also other data that we collect, so that is a key use. And that gives each locality an incentive to make sure all this – all their residents are counted, because the more they have, the more money they will get. So that we get a lot of cooperation from the localities because of that incentive that’s built into the – in effect, built into the census.

Now, this is – because it’s in the Constitution, the Congress did make the census mandatory. Everybody in the United States, including yourselves, even if you’re from a foreign country if you live here, usually live or stay here, you are required to fill out the census form. We actually do make an exception for ambassadors who live in the foreign territory of the embassy. They – it’s optional for them because that’s actually not United States land. But everybody else who lives or stays here most of the time is required to fill out the census.

Now, we – because everybody’s required, and there is some personal information – it’s not a very complicated questionnaire. It’s only 10 questions. But we do have age, sex, race, date of birth, that sort of thing, name. We do go to very extreme lengths to maintain the confidentiality of these data. They’re only collected for statistical purposes. Every person who works with the Census Bureau who sees the data, including the enumerators who are out there, temporary employees, are sworn for life to protect the confidentiality of these data. And they could be subject to a fine of up to $250,000, U.S. dollars, if they violate the law. And we do protect these individual responses, put out basically just tables, just tabulations, and not even the Patriot Act can override these protections.

We do share the data with the National Archives and they keep them confidential as well. They have to be sworn Census agents, the people who see these data, and they can release them to the public after 72 years. And that is a big basis for a lot of genealogical research in the United States, tracing family trees and that sort of thing. So we do take that very seriously.

This is really the largest peacetime activity, domestic activity, the federal government undertakes, and it’s really not just now. It takes roughly 14 years from the time we start planning until the time we finish releasing the data. So for example, I’ve already starting working on our 2020 census. As a matter of fact, I started in 2008 working on our 2020 census, and we’re well along in planning that census.

The idea of the census is to count every person living in the United States on April 1, 2010, citizens and non-citizens. We estimate there will be about 309 million people counted in roughly 134 million housing units. It covers all 50 states, the District of Columbia. It covers Puerto Rico. It also covers the U.S. territories, which are the island areas of the Virgin Islands, the Northern Mariana Islands, American Samoa and Guam. They all have their censuses as well that we do at this time.

The key changes in the 2010 Census from before – we have what we’re calling a short form-only census this time. There are only 10 questions for the first person, seven questions for everybody else. We estimate the average household will take only about 10 minutes to fill out the questionnaire. It’s really the shortest census form since 1790, which I think had seven questions.

We used handheld computers equipped with GPS, global positioning system, for a nationwide address canvassing. About a year ago, in April through July of 2009, we sent about 140,000 people out to walk or drive every street in the U.S. to get a GPS coordinate for every housing unit, and we had already spent a good deal of money realigning all of our streets to GPS network. So the idea of this undertaking was to minimize what we call “geocoding” errors, to make sure the housing unit was in the right place, the right block, so that when people draw their lines there for redistricting, we have the people in the right place.

We also do a census in group quarters, not just housing units. Group quarters are places like prisons, hospitals, nursing homes, college dormitories, places where people live together but may not have a separate, like, kitchen. We have made a special effort this decade to identify the group quarters, and are about to start counting people in those group quarters.

We are mailing, for the first time, a bilingual questionnaire in – about 13 million households will get an English and Spanish questionnaire. We think that will improve response rates in the heavily Hispanic neighborhoods.

One other change we are making is for the first time we’re using a replacement questionnaire. I’ll go through the timing in a little bit, but we have found through research that sending replacement questionnaires to the people who haven’t responded, substantially increases the mail response, which then reduces the cost – overall cost of the census.

So we expect to – we’ve mailed questionnaires already. This was March 15th to the 17th. You should have gotten one. In our mail – what we call our mail out areas, about 20 million housing units got – households, got these questionnaires. We actually have to other operations. We have one operation in rural areas without good mailing addresses. In most of those areas, we will take a questionnaire and hang it on people’s doorknobs and ask them to mail it back, because we can’t mail to post office boxes and – or rural routes, but they can get mail picked up there. So we’ll hang it on their doorknobs, they’ll fill it out, put in their mailbox to be picked up and it will come back to us.

And there’s actually about one percent of the country we will do in the old-fashioned way. We will send somebody to somebody’s house, and knock on the door and ask them who lives there. We’ll – it’s called, you know, in-person enumeration.

We will have – I mentioned a total of about 134 million housing units. We have about 270,000 group quarters in the United States. It’s about 3.5 million square miles of land that’s covered, and every single bit of that land is covered somehow, nine million census blocks, which is our basic organizing geography. And to supervise the field work, we’ve actually opened 12 regional census centers, a separate office in Puerto Rico, and 494 local census offices. And at the peak, we’ll have about 1.4 million workers working on the census, not all of them enumerators. At peak we’ll be about 700,000 in May.

This is a challenging task because we don’t have the ability to delay the census. You know, if somebody is trying to launch – let’s take, example, the space shuttle. You’re trying to launch the space shuttle. You see there’s a leak. They say, “We got to postpone the launch. We will take it back in the hangar, fix the leak, and then we’ll launch it later.” We have a statutory deadline. We must report these numbers to the President by December 31st, the total population in every state. So we can’t delay things.

We obviously have language challenges in the United States. We produced the form itself in six different languages, but we have language assistance guides in 60 languages. So that’s a challenge, just to reach people who don’t speak English as their primary language.

Not everybody mails the form back. If they mail it back, it costs us under a dollar per person to get a form mailed back and processed. If we have to visit them, I think the estimate was roughly $57 to get each one. And so you can see there is a tremendous desire to get these forms mailed back, because it saves the Census Bureau a lot of effort, and it saves the country a lot of money. We estimate for every one percentage point increase in the mail response rate we save about $85 million.

We do have challenges, what we call coverage challenges. Not everybody wants to be counted. We have to go to special lengths to try and catch those people. We have something starting next Monday, actually. We have – on Monday night, we have shelter evaluation. We’ll go to shelters and try and enumerate the people without a home at those shelters. On Tuesday night next week we’ll go to soup kitchens and mobile food vans and try and get those people. And then Wednesday night next week we’ll actually – between midnight and 6:00 a.m., go to places where the homeless might congregate and try and get them counted as well. So we do go to long – to substantial lengths to try and count everybody.

After we deliver the apportionment counts to the President by the end of December, we then have another statutory deadline. We have to report the population by block to all the states by the end of March, so they can begin their redistricting operations. The numbers at the end of December determine how many seats they’ll have in the House of Representatives. Well, there will be some states that gain seats, some states that lose seats. And even the states that don’t get a change, the people move around, so the districts are going to be of different size. So they all have to redistrict. All the states have to redistrict to get districts of the same size. I guess the exception are states that are so small they only get one seat. They don’t have to redistrict. It’s the whole state anyway.

Then we will proceed to release the remaining data products starting roughly next June through the end of September of 2013. These are basic tabulations that people are interested in.

I mentioned the 1.4 million temporary employees. We will recruit approximately 3.6 million people to fill those 1.4 million jobs. And we are – last I checked we were ahead of our target. We were about 109 percent of our target as we move up by – our target is this 3.6, two five million applicants by April 25th. And we were – as of, you know, last week, we were at 109 percent of our target for that week. So we – that seems to be going well.

Now we’ve basically – the basic components of the census are – they really fall in four groups. One is to get the – what we call the frame right, and that’s to get the address list right. And we have a number of operations to do that, including the address canvassing I mentioned earlier, but other ones.

We then do a whole series of enumeration operations, some that focus on the housing units, but also some that focus on the group quarters. We have a special group quarters enumeration program, where people will go out to these nursing homes and the college dorms, as so forth, and collect that information.

And then we have support operations for that. For example, we have a telephone questionnaire assistance centers. We have four of those that – commercial call centers that – there’s a number on the form. If you have a question, you can call them and they will answer the question. So we have to have all those kind of support operations.

Obviously, there was a huge printing task to print out what we mailed out. We printed – and the enumerator forms, also, you know, 200 million forms in paper that had to be printed, plus all of the advance letters and the reminder cards and the training manuals and all that. It’s a huge printing activity. We had this – all of these 120 million mail out forms in this one warehouse in Chicago. You know, it’s putting all your eggs in one basket and watching that basket very carefully. And starting about a week before the forms were to go out in the mail, the U.S. Postal Service would start backing trucks up. We had, like, a thousand semis of mail forms going out from that Chicago warehouse all over the country. So it’s a lot of support, obviously, not just the people in the field.

And then we have, if you will, verification and – what we call verification in coverage operations. If there seems to be a discrepancy on the form, and I’ll give – we will call you. For example, if the first question says, “How many people live here?” And you might fill out three. And then you go on and ask questions about person one, and then you ask questions about person two. And they never filled out the answers for person three. They only filled it out – so we’ll call and find out that extra information. That’s an example of the follow-up operation. Every operation has quality control built in.

And then we have a follow-up operation we call census coverage measurement, which is, if you will, an independent census in a sample of blocks around the country that is then, using statistical techniques, used to assess how good the census is. And there will be reports that come out of that.

To complement the actual operations, we have a communications program. We have – goals are really three: to increase the mail response, as I said how much that saves; to improve the accuracy and reduce the differential undercount. We don’t want one group that is count – we don’t want to miss people differentially from one particular group versus another; and we want them to cooperate with the enumerators. If they don’t send it back, we want them to accept that people are going to come by.

We use paid media to do this. We’re spending roughly $140 million on advertisements. I don’t know if you’re basketball fans. If you see March Madness, you’ll see an ad from Frank, March-to-the-Mailbox. That will be on in the Elite Eight Saturday night and Sunday night. You’ll see that ad. We were at the Super Bowl. We also get – we’re sponsoring a driver in NASCAR, in the stock car racing. So he’s got the U.S. Census on his hood, on his rear bumper, that sort of thing. We hope he does well so he’s featured more, but can’t guarantee that.

We also have a partnership program, which we send – we’ve hired roughly 3,000 people across the country to work with others to help us do the census better. For example, I’ll just give you an example at the national level. We had somebody contact Best Buy, which is a consumer electronics store, and they have these – this huge wall of TV’s. Well, they’re going to be playing a Census ad. Instead of a Disney movie, they’ll be having a Census ad on. Target, a big retailer, they’ll have a Sunday – they put these Sunday inserts in the papers. They’ll devote one of the pages of their Sunday insert to the Census ad. And that is donated. That is – we don’t pay for that; they donate that. And we go down to local levels. We encourage these localities to form complete count committees where they get members from business, from the nonprofit sector, so forth, to work together to encourage counts in their communities. And they work throughout the Census. And their real target is to try to reach the hard to count and encourage them to participate.

We also have – they also provide locations for what we call questionnaire assistance centers. If you need help with your form, we’ve got 30,000 of those across the country, where you can go and get help.

This – one final note. I mention this is a short form, 10 questions, 10 minutes. What about all those other topics we used to collect information on? We had a long form also as part of the 2000 Census that had, I think, it was 59 questions. We made a design change in this past decade. We now do something called the American Community Survey, that we ask – it replaces the long form, basically. It’s for – it’s a very large household survey of about 250,000 households per month, which is almost three million over a year. We produce statistics every year from that survey to very low geographic levels. So we now basically do a monthly survey to replace the once every 10 years long form. And that seems to work very well. People are very cooperative, and we release data every year from that survey.

So that’s the overview. Think – I’ll be happy to take questions, try to give you a little more insight into some of the details if you’re interested. So feel free to ask questions.

You had your hand up first, so there’ll be plenty of time. Go ahead.

QUESTION: Okay. Nicholai Zimin, Itogi, Russia. It’s my first American Census, but I’ve never known that you start so early the preparation work.

MR. WEINBERG: Yeah.

QUESTION: So could you please give us the sense, what are you doing so early?

MR. WEINBERG: Okay.

QUESTION: On the planning.

MR. WEINBERG: Yes. Well, for example, we’re not using the internet at all for this census in terms of collecting data. We are using it in terms of publicity, but not for collecting data. We know that in 10 years, the internet will be an even bigger part of people’s lives than it is now, so we are planning tests of how to use the internet to collect data for the census.

Another example is automation. We attempted to automate most of the field operations for this census, and we weren’t successful. We did automate our – what I called our address canvassing. That was the updating of the address frame, using hand-held computers. But we weren’t successful in using that for non-response follow-up. And we know that we need to – since we weren’t successful, we know we need to continue to work in that area.

And the cost growth of the Census has been pretty substantial. And we want to look at as many methods as possible that will keep the costs down. For example, instead of sending out 140,000 people into 2019, and that cost – in 2009, that cost about $450 million, that one operation. Let’s see if we can spread that work over the decade, keep the address list continually updated, and only targeted areas would get that sort of last-minute – high-growth areas would get that last minute updating in 2019 if we keep the address list current. And in that one, we’ve proposed to the Congress – it’s in the President’s proposed fiscal 2011 budget that we be allowed to do that continual updating. And we’ll see if the Congress approves for our – when they consider our budget later this year.

So it’s that sort of thing we’re looking at. It’s clearly a fairly low level, but there are research questions that can contribute to this cost saving in 2020 that we need to start now.

Sure. Go ahead.

QUESTION: My name is Dagmar Benesova, World Business Press Online. I’m from Slovakia. My question is : You mentioned that this year there will be just 10 questions in the form and that it will be the shortest form since 1976, so –

MR. WEINBERG: 1790.

QUESTION: Oh, 1790. Thank you. So my question is the reason why there will be just 10 questions is because of the money, because of the economic crisis or because – less questions, less paper, less people to (inaudible), or because you want to address more people? It will be easier for them to answer the questions? And one more question is: You mentioned that this year, you are mailing the forms to the Americans, to the households, and in the previous years you used to visit the households. It is also the reason that you want to save money, or why?

MR. WEINBERG: All right. Two separate questions.

QUESTION: Yes, thank you.

MR. WEINBERG: We found in 2000 that when we sent out both the short form and a long form, the long form went to about one in six households. We got a much higher mail return rate from the people who got the short form than from the people who got the long form. So we figured it would be easier for people to send back just a short form, and it would be cheaper. We would save money in the long run of focusing on just the short form and letting the American Community Survey do the long form. So it was mainly for cost reasons that we went to a short form, because we now have this other mechanism for getting the long form information.

Now as far as the mail out/mail back strategy, we have used this for many decades. We actually started testing it in 1950 Census and continued testing it in ’60, and used it for a great part of the population in ’70, and it was really our main way to do the census starting in 1980. It’s now reached – we use the mail method for basically 99 percent of the population now, but we can see the writing on the wall that people, especially the younger age groups, would prefer to be able to fill this out on the internet or on a smart phone, or who knows what else will be around in 10 years? And we want to be ready for that. So we think that the – and we certainly would actually prefer to have the electronic data come directly into our system rather than having them fill out the paper. You have to send it to a huge data capture center, where it’s checked in, it’s scanned, any of the write-in entries is keyed, and then it comes to us.

So actually, the internet and other electronic means can be a big cost saver for us as well. But it depends on a number of things: the acceptance of not only the 19-year-old but the 90-year-old; it has to be easy to use; it should be in many languages; it should be – the important thing is we have to tie each response to a piece of ground, to where they live. So we can’t just say, “Come and fill out the census.” We have to send you an ID, so that we can tie your response to the piece of land that you live, because that’s what needed for this apportionment, for this drawing of the districts, so that everybody has the same number of people in the district. And that means we have to contact you somehow, so I think we’ll probably still mail you a letter, assuming the Postal Service is still around, but we want to move to the internet for many reasons. We think it’ll improve response and it will save money.

Okay.

QUESTION: One additional question. Do you have some estimate how much money will save using this new form?

MR. WEINBERG: The short form itself? Well --

QUESTION: And mailing (inaudible.)

MR. WEINBERG: We know because we had to stand up the American Communities Survey itself. That’s roughly $1.8 billion over the 10-year period, so that – there was a tradeoff. We said we’d get more data more often but – and we’ll make the census simpler. So I can’t really point to a dollar saving from just the short form.

QUESTION: Okay. Thank you.

MR. WEINBERG: Yeah. Go ahead.

QUESTION: Jane Morse, IIP, U.S. Department of State. I know that certain ethnic groups weren’t happy that there wasn’t a write-in, a line to be able to have some count on their representation in the United States. So how do you get figures for how many Russian Americans live in the United States or how many Slovakians, for example, live in Michigan?

MR. WEINBERG: Again, that’s the American Communities Survey. There is a question on ancestry on the American Communities Survey. So we have, again, Hispanic origin and race, but then that’s followed by an ancestry question where people can write in the – I mean, we have boxes for the most common but they can write in anything they want there and we’ll tabulate it.

QUESTION: And all of this gets folded into some U.S.-wide calculation?

MR. WEINBERG: We produce data for the whole country every year, as well as for much smaller levels of geography.

QUESTION: Then why do you need a 10-year census? You’re doing –

MR. WEINBERG: Well, the American Communities Survey is just a sample of the population.

QUESTION: Oh, I see.

MR. WEINBERG: And the only way to do congressional apportionment is through a complete census. So it’s a sample, and it has margins of error around every estimate. We could say that New York City has – I’ll pick a number, 250,000 people of Russian ancestry, but it’s somewhere between 220,000 and 280,000, because there’s a margin of error. I made that number up; don’t quote it. You can look it up on the website, and it will have the actual margins of error around that number.

The Census – you can’t say that this block has three or four, maybe five people. You can’t do that. You have to say it’s got exactly 15 people so they can draw the line. There is no margin of error associated with the Census. Now, it doesn’t say there’s no error, because we know we miss people from time to time. We count duplicates from time to time. We know there are errors, but the number is the best number we have, and because it comes from a complete count.

Yeah. I’ll let – you want to follow up? No, go ahead.

QUESTION: Yes. You talk about people who don’t want to be counted, and I assume lots of illegal immigrants don’t want to be counted. So is there – I mean, I don’t know if there’s any way to reassure them to answer this government form, but how do you make estimates there? Because I know it’s been widely publicized that we’ve got, like, 11 million illegals or something like that.

MR. WEINBERG: Well, okay. We don’t estimate the number of illegal immigrants in this country. What we do is try to count everybody. Now, you’re right, some illegal immigrants worry that any contract – contact with the government might result in some negative results for themselves, so we have to go to as great a lengths as we can to reassure them. Now, the way we do that is several ways: We target advertisements to them and focus on the privacy and confidentiality message. We hire enumerators from their community, so the people who come to their door are people they recognize. And we work with community leaders in their communities to get the word out. There might be a pastor in the local church who might emphasize how important it is to answer the Census. It might be somebody who runs the local health center who might be willing to put up signs in their health clinic about how important it is to answer the Census. We go to – we target extra funds to partnerships programs for those people who might worry about it.

When – we try to create forms in their language, which might help as well. And I’m – just for example, I’ll give you an interesting example. The Secretary of Commerce, which is Secretary Locke, is from Chinese ancestry. And we showed him the Chinese form and he said, “This is in simplified Chinese. Most of the Chinese in this country use traditional Chinese.” And we said, “Yes, that’s true, but most of the people of Chinese ancestry in the country already speak English. They don’t need this form. This form is for the recent Chinese immigrant who may have come over in a container and is on the – not – we want them to understand this form.” So we do – it’s not easy. I’m not trying to say it’s easy, but we sort of really make an effort to reach all of them.

QUESTION: One last question: Some anti-immigrant groups, mostly, argue that the opposite kind of result can happen in that lots of illegals report that they live here in the United States, and thereby certain districts in the United States would actually qualify for more representation in Congress simply because they have huge numbers of illegals. Is that political repercussion?

MR. WEINBERG: Well, the Constitution and all the interpretations we’ve gotten from the courts over the years says it’s the resident population, not citizen and non-citizen, it’s resident population. And if they live here, they’re residents, and that’s what should – those are the – everybody should be counted who lives here, not legal or illegal. The Congress, I suppose, could try to change the law if they really wanted to, and then that would have to be adjudicated through the courts, but the courts have upheld the current interpretation for centuries.

QUESTION: I’m Emmanuelle Richard with France 2 TV network, and my colleague is also a news producer. They tried to call to – so that he could follow an agent in the case when you have physical people going door to door, and we are told it’s not possible.

MR. WEINBERG: That’s correct.

QUESTION: Is it not possible because we’re foreign media and you only have that much resources and you let only national media, or –

MR. WEINBERG: Not at all. Don’t let any media – not any U.S. media at all.

QUESTION: Oh, okay.

MR. WEINBERG: Because it’s part of this protecting the privacy and confidentiality of the data. If you interview – if you had a picture of somebody filling out their form, basically, you are involved in the data collection process. And we are required by law to keep that private and confidential.

QUESTION: Yeah.

MR. WEINBERG: Now, you can, I’m sure, work with our information office to get a simulated interview to show what it would look like if it were real, but you can’t follow a real interviewer. They are instructed to – if the media approach, they have to stop work.

QUESTION: You can’t follow an agent? They can go and then you stop filming, because it’s private, after that?

MR. WEINBERG: Yeah, but why not use – if it’s just a picture of somebody knocking on a door you want, the Census Bureau will give you b-roll and somebody knocking –

QUESTION: (Inaudible) and then as the agents say, “Well, I’ve been turned today by this number people,” or, “It’s been difficult,” or, “I go around” – because we want to show the scale –

MR. WEINBERG: Yeah, yeah, yeah. Well, all I can suggest is work with our public information office and see what they – I mean we want publicity. Of course, we want publicity.

QUESTION: (Inaudible.)

MR. WEINBERG: Yeah.

QUESTION: You know, 300 million people. How do you count them? It’s –

MR. WEINBERG: Exactly, so –

QUESTION: Well, who do you recommend that – in your office who we –

MR. WEINBERG: Well, Stephen --

QUESTION: -- (inaudible) TV –

MR. WEINBERG: Stephen Buckner is the name I would contact.

QUESTION: Buckner?

MR. WEINBERG: Yeah.

QUESTION: Okay.

MODERATOR: New York, go ahead.

QUESTION: Hello. I’m Gabriel Plesea, for Romanian media. I have two questions. One, you said there is an exception, ambassador. Isn’t this – just the ambassador or the diplomatic mission staff? And also, you don’t have the international organizations. I’m referring to the United Nations. What about those? You do have national (inaudible) the counters. And the (inaudible) that the census refers to the residents. You do not go by citizenship or whatever. But what about those?

The other question is –

MR. WEINBERG: Let me do one. Let me answer that one. We’ll come – yeah. It’s – basically, it’s diplomatic, I suppose, because you – the people who live in the foreign soil within the United States, the Embassy compound. I don’t think it’s just the ambassadors. But we do want foreign nationals who live on U.S. soil to fill out the census. It’s – they live – if they live here most of the time. I mean some countries are different. I know in Ireland, it’s if you’re there that night, whether you – you know, you may be passing through. You’re there one night, they want to count you that one census night. But here, if you usually live or stay, we want you to count. If you’re a foreign tourist, you come in for a week, you see New York, you go to Washington for a week, we don’t – you’re not considered a usual resident. So go ahead with your other question.

QUESTION: Yeah. The second – you answered that already. It referred to nationalities, like, you know, like Romanian, Czech, Bulgarians, and what have you. You answered that. Who is – you know, you have this (inaudible). They have these numbered by religion, like Catholics, you know, Jewish, Hindus, or what have you. Is that also done by the American Community Survey? And also there are data about financial, you know, groups, U.S. population grouped by financial (inaudible) and so on and forth.

And the last one is – refers to U.S. born versus naturalized (inaudible). So you know you find those (inaudible) various locations. Is that also (inaudible) somewhere a function of the U.S. Census Bureau, or there are some other organizations that would do that?

MR. WEINBERG: Let me answer the religion question first. The Census Bureau is prohibited by law from asking questions about somebody’s religious affiliation. There’s actually a Congressional law. So we – the religion data that we publish in our Statistical Abstract are from other sources. The – you know, the Church World Council or wherever – you know, the individual churches might report their membership to some organizations and we might assemble it. But we don’t collect it.

The other information you asked about, the financial – the income of individuals or whether they were native-born or not, yes, those come from the American Community Survey. And they’re tabulated, and we have this – it’s called American Fact Finder on the census website where you can actually get these data at the national level, the state level, city level, and eventually, at the end of 2010, we’ll be publishing those statistics at the census track level which are – average about 4,000 people.

QUESTION: Thank you.

MR. WEINBERG: Sure.

MODERATOR: Any other questions, New York?

QUESTION: (Inaudible.)

MR. WEINBERG: Okay.

QUESTION: Hi. Thank you very much. I’m Hideo Miyawaki from Kyodo News Agency. You may already have explained to us, but (inaudible) concern. What was the earliest Congressional election to be conducted based on this 2010 census?

MR. WEINBERG: That would be the –

QUESTION: (Inaudible.)

MR. WEINBERG: That would be the 2012 election.

QUESTION: And my related question is whether the results of the census will affect the (inaudible) election in the future?

MR. WEINBERG: Definitely it would affect Presidential election in the future, also the 2012 election. The Electoral College, which is the number of votes that determine the President is based on a state level. So if – for example, let’s use California. California, I think, voted for President Obama rather than Senator McCain in the last election. If they got two more seats, they would get two more votes in the electoral college. So there’s a direct relationship between the number of representatives and the vote in the electoral college. So it would definitely affect the voting in 2012 as well. Sure.

QUESTION: Can you give us any sense about how the U.S. census compares to censuses done over – in other countries around the world?

MR. WEINBERG: That’s an unfair question. Of course, ours is the best. What do you want me to say? (Laughter.) No, I will – that’s a facetious remark, but we do have a lot of interchange with other countries. And there are similarities. There are differences. Specifics, Canada does a census – the Canadian census is every five years. They’ll be doing their next one in 2011. And they have had, for many years, an initiative to get broadband connections out as much as they can in the country. So they are focusing very heavily on the internet in 2011 to collect their data. And we – matter of fact, I mentioned how important that was to us. We’ll be going to work with them to learn from the Canadians about the internet.

In many countries, they use a long form only and not a short form. So they’ll – I think England and Wales, for example, they’ll use 100 percent long form. Not every country has a regular collection. They differ. I mean, you can’t – let’s take a country – unfortunate example of Haiti. They don’t have the wireless – they don’t have the internet connections at all. They don’t have mail right now. The mail system has broken down. You want to do a census in Haiti right now, you’d have to go door to door. So there are many differences. We try to assist other countries. We have – countries come here to observe us. Most of them are amazed by the amount of money we spend on advertising and try to learn from our experiences there. So in some ways we’re a leader, in some ways we may be lagging behind. I’ve given you, sort of – we’re a leader in the advertising, the partnership, the outreach efforts. We’re sort of lagging behind on the internet use.

QUESTION: Nickolay Zimin. (Inaudible) what is the total budget of the current census?

MR. WEINBERG: The 10-year – no, but we don’t think of it as a single year – I can give you those figures, but over the 10-year period of – well, it’s more than a 10-year period. I guess a 12-year period of 2001 through 2013, it’s 14 and a half billion dollars is what we estimate for all the parts of the census, including the American Communities Survey I mentioned. That’s about 1.8, $2 billion of that. We spent about a half a billion dollars on aligning all of our roads with a GPS system so that they’re now in the right place. That was about half a billion. And then the rest, about $12.3 billion over that 10-year cycle – sorry, 13-year cycle is 12.3 billion for the short form census itself.

QUESTION: But you cannot (inaudible) the particular – the budget for April 1st.

MR. WEINBERG: Oh, for this year alone, for 2010, we’ll be spending about 7 and a half billion dollars just this year on the census. So this is clearly – we spent about 85 percent of the total in just 9 and 10. That’s – it’s hiring people. It’s nice because we’re in a recession. We can give people, a lot of people, jobs, even if they’re just temporary jobs. But it’s really hiring one and a half – almost one and a half million people. That costs money. We also pay them for training. While they’re being trained, we pay them. We pay them for mileage, because we ask them to use their own cars. We pay them for that. And we obviously give them the materials; the training manuals, the forms, the pencils. It’s an amazing – we have a facility in Indiana, Jeffersonville, Indiana, our national processing center. And they have this amazing facility they built to assemble these kits that go to the enumerators. You know, it’s like an assembly line. This box rolls down, you put in two pencils, you put in a pencil sharpener, you put in a box of paperclips, you put in this and that, and at the end it’s packaged up and it’s put aside and then shipped out to all of these 494 local census offices who then ship them out to the trainings. We will have about 40,000 training sites that are open on April – what is it – 26th, I think. We’ll have 40,000 training sites around the country with training going on.

QUESTION: Dagmar Benesova. You mentioned also the jobs. It’s creating new jobs. Do you have some survey or estimate how many new jobs will be created this year for April 1st
MR. WEINBERG: There was a study by the Commerce Department, and I’m sure that Miriam that can get ahold of it – it’s on the website – that estimated that we might reduce unemployment by as much as half a percentage point during that period. But there was a very careful study done, and you can take a look at it. That’s just my recollection.

QUESTION: And just – thank you. And just for verification, as far as 13-year period, there will be 14.5 billion dollars, like, the budget? And this year it will be spending 7.5 billion dollars.

MR. WEINBERG: Approximately, yes. That’s right. Of that 14 and a half billion for the 13-year period, we’re spending 7 and a half billion this year alone.

QUESTION: Okay. Thank you very much.

QUESTION: Jane Morse. And that’s to count 309 million people.

MR. WEINBERG: Right. Well, it’s actually a little more than that. Because the 309 million is our estimate for just the U.S. and we have to add in Puerto Rico and all the island areas. And they’re in the budget, but they’re a relatively small proportion.

QUESTION: Emmanuelle Richard. And what is the estimated people who don’t have a mailing address, people who are nomads, who are – you know, who keep moving in trailers or who are hard to reach –

MR. WEINBERG: Well, we don’t – we have a separate – an operation called Enumeration at Transitory Locations. It’s a fancy name, but it’s people who live in trailer parks – I mean, campgrounds. It’s people who live in marinas. It’s people who live – who work in circuses and carnivals, that travel around the country. We have a separate operation for them that just started this week, actually, to go to those people. We’ve identified those places. We will go there and try to – and enumerate them in person. I don’t have an estimate for you. We’ll know that after. But –

QUESTION: It’s very small (inaudible) migrant workers who keep moving from –

MR. WEINBERG: No. They’re – well, no, no. They’re included, because we know – those are physical locations we will visit. It’s the people that move around. I know – I think it’s, like, 20,000 locations, transitory locations we’re visiting, something like that number. So it’s a substantial effort over a fairly short period of time.

Go ahead.

QUESTION: Dagmar Benesova. If I can ask maybe the last question. You mentioned that it lowered the unemployment by 0.5 percent.

MR. WEINBERG: I’d rather –

QUESTION: Would it be temporary just – so it’s for April unemployment rate or for the whole year?

MR. WEINBERG: No. No. It’s – don’t quote that number until you read the source, because I’m just – it was my recollection. But basically these people will start work the last week in April, and they’ll work for perhaps six to eight weeks, through May and June and then maybe some of them into July. So you won’t see it in the April unemployment rate that’s reported at the beginning of May. You should see some effect in the May unemployment rate that’s reported in the beginning of June. Because the way the unemployment rate survey is done is we ask people, “Were you working the week of in this case May 12th or April 12th?” And if we don’t hire them till the end of April, they won’t have been working April 12th. It will be not until May that they have the job. So will show up at the – in the May rate, is my expectation.

MODERATOR: Well, thank you very much for a very informative briefing.

MR. WEINBERG: Oh, you’re welcome.