"A Lack of Effective Communication" Adds $3 Billion to U.S. Census Cost

I have written several times about the upcoming 2010 U.S. census, to be conducted as required by the U.S. Constitution. In the April 05, 2006, newsletter, I wrote about the planned use of handheld computers to "help revolutionize how census information is collected." I also wrote, "For the 2010 Census, the Census Bureau plans to use automated systems to quickly capture information from interviews. That is designed to reduce the need for paper-based processing, improve efficiency and accuracy, and reduce costs."

Today the Census Bureau performed an about-face. Handheld computers are out, paper is in. And “reduced costs?” Well, hardly...

In 2006, the Census Bureau awarded a $595 million contract to Harris Corp. to develop more than 525,000 handheld computers that enumerators would use to collect data from Americans who did not send in their census forms. These handhelds would replace the millions of costly paper forms and maps that enumerators must carry when going door to door to visit Americans who did not mail their census forms. You can read more about that in my 2006 article at http://blog.eogn.com/eastmans_online_genealogy/2006/04/2010_census_to_.html.

Since awarding the contract, the project has experienced constant setbacks, including changing system requirements that led to increased costs and missed deadlines. Reports by the Government Accountability Office, the department's inspector general, and Mitre Corp. all issued warnings that the handhelds were at risk of not being ready by 2010 and may not work as planned.

Now Commerce Secretary Carlos Gutierrez gave testimony to the House Appropriations Committee on Commerce, Justice and Science. He said, “I am here today because the Field Data Collection Automation project has experienced significant schedule, performance and cost issues. A lack of effective communication with one of our key contractors has significantly contributed to the challenges.”

In his statement, Gutierrez called the situation with the handhelds “unacceptable.”

How do we resolve this quandary? The Commerce Department, parent organization of the Census Bureau, has created a backup plan that involves a lot of paper, hundreds of additional temporary employees, and a jump in expenditures of an additional $2.2 billion to perhaps $3 billion through fiscal year 2013. That will bring the total cost of the 2010 census to between $13.7 billion and $14.5 billion.

Not all of the extra expenditure is because of the issues with handheld computers, according to Gutierrez. He reports there are numerous other unexpected expenses affecting the bottom line.

So much for the Census Bureau's 2006 plan to "reduce the need for paper-based processing, improve efficiency and accuracy, and reduce costs." Oh well, it's only taxpayers' dollars.

You can read more at http://www.nextgov.com/nextgov/ng_20080403_9574.php.

Comments

[Ann Lamb]

Ah shucks. I guess they didn't notice that UPS and FedEx drivers have been using handheld computers to track packages for years. Same way as the government haven't noticed that the same system that handles money securely (credit cards)would work fine for verifying voter registrations.

[ellen b]

in the last census, no post office box was counted here, only mail boxes in front of houses,, there were some 2000 boxes in Salerno Post Office never counted...were the ones who,have mailboxes at MailBox stores counted? Is this accuracy?

[Linda Johnston]

I worked on the 2000 census, so perhaps, I can explain it to you. First of all, people DO NOT live in post office boxes. Yes, the census forms are sent to the mailing address, but there is a procedure that takes place, prior to the forms ever being mailed out. It is called "address listing". This is where the Census Bureau hires a number of workers, who go to specified territories and "map" buildings that exist. They must follow a specified route. If new streets or buildings are found, they are added to the "listing". If previously existing buildings are gone, they are, subsequently removed. This process will probably take place in early 2009.

Then, when the actual census enumerator goes out, in 2010, "selected counties" will be chosen. In each of those counties, every other building, or apartment is a "long form" candidate. If a resident indicates that they have sent in a short form, they are asked if they use a different mailing address, as opposed to the physical address, as of a certain date. If they do, a notation is made by the enumerator, to prevent duplication. If a long form is required of that household, the enumerator must then complete one. Often, one must be very diplomatic to get the additional questions answered. However, refusing to give that information is a violation of the law, so most people are very cooperative.

So, the answer to your question is, those people ARE counted, even though they don't live in the P.O. Boxes.

[Lee Hoffman]

I have no idea of the cost of the "laptops" carried by UPS, FedEx and others, but I do know that there are fully functioning laptops out there that cost $200. These are Linux based machines and primarily designed for use by children . But it sure would not cost a lot of money to convert this design to something for the enumerators to use. These little systems are great. While these are designed for kids, adults can enjoy their use also with only a little effort. I am not using mine for genealogy, but I think I could without too much trouble.

Lee

[Barnegat Blummis]

This is yet another example of the "business-oriented" Bush Administration not learning anything whatsoever from business experience. Bush's "The Bumbler", not "The Uniter" and not "The Decider." Bush/Cheney and their cohorts will certainly go down in history as the worst government the US has ever had for so many reasons including how the Census is being screwed up and how the so-called Patriot Act and other laws work to prevent and put obstacles in the way of legitimate genealogical inquiries. Even the FOIA requests are being blocked and done incompletely or inaccurately. What a mess!

[Gerald Eberwein]

I don't buy the Census Bureau's implication that the ineffective communications was the contractor's fault. My limited experience in dealing with the Census Bureau is they don't like to communicate outside of their own walls. I have, for official government reasons, tried to communicate with that agency several times only to be blown off (calls and messages not answered) or been given a huge referred to another person run around until they achieved their goal. I quit calling.

I was on the Census 2000 Committee for my County. We were organized by the Bureau and we had problems communicating with them. Things we were asked to submit were never included and accuracy updates to maps were ignored. It was incredible. The result is I have very little trust in the Census for accuracy.

In 1990 I observed census takers in a small city. They were standing on a street and looking through binoculars at houses on the hillside. The only way to the houses was to walk up several flights of stairs. I (as an on duty Police Officer) approached them and inquired what they were doing. They produced ID and said they were taking the census. I asked why they didn't go to the houses. They said because they looked abandoned and they weren't going to walk all the way up there to find that out.

They were there because that part of the city is served by PO Box only. Thus, they are supposed to visit every address.

Later on I testified in a hearing about the innacurate count for that part of the city. It seems there were more registered voters than people in the census. Through other records and actual visits, the accuracy of the voter registration was verified, so the city had a solid foundation. The Census Bureau was required to do a recount; they had missed over 2300 people.

On this issue, I also suspect there was a reliance on a technology concept that didn't materialize as well as anticipated. Often we (in this section of the nation) have things thrust upon us that can't happen due to the lack of technology to support the concept. One of these is the presence of broadband internet. Although 85% of the people my county have that capability in their homes, it is not readily available in public places. Plus, the 85% are centered in 15% of the land area. We are roughly the size of the southern 25% of New England. That means the rural areas are rather dry in that capability. Our neighbor to the East is equally as deficient. In fact their state finally got a second area code late last year. And, except for the interstates, cell phone communications has many black holes in New Mexico.

I have found the same thing exists in Texas and probably others equally as vast; when added up it is a huge chunk of America.

Big city folks simply have no clue regarding rural life. So they design things that won't work, and scratch their heads when they fail.

[Wayne Thalls]

Just how do you suppose historians will explain all this to future genealogical researchers anxiously awaiting the release of the census data in the year 2082? After all we are living in the era of technologial revolution, right? This is the computer age!

Ironically, I could argue that the computer age was launched when the Census Bureau contracted with a young engineer, Herman Holerith, to develop a mechanized system for handling the enormous amount of data included in the 1880 census records. The storage media was a system utilizing punched cards which could be sorted in many different ways. The Holerith enterprise went on to become International Business Machines (IBM). Those 2010 records will be released approximately 100 years after that very successful 1880 project.

Recent studies show that many governmental and business organizations are failing to make the organizational changes required to truly benefit from the technology investments they are making. The computer is only a tool--not a magic bullet.

I walked past a neighborhood dog grooming business the other day. A sign in the window says, "Check out my Website". Isn't the logic for having a web site to promote the business, not the other way around? Big organizations are doing the same sort of thing. Even some genealogists seem to think research success is assured with acquisition of the latest hardware and software.

[Fred Mussler]

Is there any way to volunteer for the long form in the next census? I'd like to leave as much information as possible about my household for my descendents who might have an interest in genealogy. I tried to do this in 2000 after the fact (having received the short form), but without success. The census workers I spoke with then seemed taken aback that anyone would actually prefer to complete the long form and had no solution for me, although one suggested that I write anything I wanted in the margins of the short form on the assumption that a photocopy of my form would probably survive for future generations. I thought that was unlikely (but I didn't blame Bush).

[Sue Wilson]

For 2000, I found the long form online and filled it out. I filed it with our personal papers so family members can see it later.

[Matt]

I made a graph of the costs of past census years on my blog post. It's very startling to see the rise in costs. http://exitcavehere.com/2008/04/april-fools-joke-from-the-us-census-bureau/

[Elizabeth Thomas]

There are quite a number of people who do not own a computer in their homes, and some are computer illiterate, nor do they want to have such an expense after hearing from others who have a computer and have needed to get them repaired. For those of us who enjoy using a computer for research, that others don't want to own a computer can be mind boggling to them. Yet, we must all understand that there are many people who have never typed in their lives simply because some are not as coordinated with their minds and hands.

In 2000, during the time of the census, we had all of our possessions in 2 storage units, and could not get to my computer and we were living in a rural area. The census taker came to the neighborhood & saw the next door neighbor, told us they would be back but never came back to interview us.

The 1990 census also missed me, since at that time I had moved because I had taken another job in another state. If the government thinks that they are getting a true reading of all the people who are in our country, then they are mistaken. This also reminds me the fact that I found that I could not find my father in the 1910, 1920, or 30 census, because he had been out of the country doing research work for Pierre S. Dupont and the Dupont Company. How do the states expect to get their supposed share of tax dollars for the number of people who really live in their state, and how do the states really think they will have a count of the illegal number of people who live in this country?

[Don Michel]

Unfortunately, this is a problem that runs rampant, not only in the government but with most all users who want to have a computer system designed. They simply don't know what they really want as they have not carefully considered the output they need. I have worked for two very large firms that did contracting with and for many government agencies, both state and federal. When you ask the sponsor what they want the output report to look like, they have no idea. If they do, the next question is "What columns do you want to be able to sort by?" always stops them dead. The census problem is a classic case of not thinking clearly about the output needed and how sorting is to be done. The system designer cannot make that decision as they don't have the experience to know how the data is to be used. Until someone educates the agency leaders about how to pick project sponsers and what they MUST define, we will waste billions of taxpayer dollars every year on such failures.

[Ann]

Article posted today on Wall Street Journal Editorial page: http://online.wsj.com/article/SB120959897575757749.html?mod=djemEditorialPage