Prepared Statement OF

Charles Louis Kincannon

Director

US Census Bureau

The 2010 Decennial Census Program

Before the Subcommittee on Federal Financial Management,

Government Information and International Security

Committee on Homeland Security and Governmental Affairs

U.S. Senate

6 June 2006

Good morning. On behalf of the U.S. Census Bureau, I want to thank Chairman Coburn and Senator Carper for the opportunity to provide the Senate with an update on our progress on the reengineered 2010 decennial program. The reengineered 2010 decennial program comprises three integrated components: the American Community Survey, which will provide timely, accurate data for states, towns, and even neighborhoods; the Master Address File (MAF) and TIGER Enhancement Program, which will serve our nation by updating the address list and modernizing the electronic maps by which we collect and disseminate census data; and most importantly the 2010 Census, a short-form only census, which is the Census Bureau’s core constitutional responsibility.

As you may imagine, the decennial program is the Census Bureau’s largest activity and its highest budget priority—in fact, it is the nation’s largest peacetime mobilization and is mandated by Article 1, Section 2 of the U.S. Constitution. However, the Census Bureau collects other censuses and surveys that contribute to the nation’s statistical system. These include the Economic Census that catalogues the nation’s economy by collecting business information, including number of employees, payroll, receipts, and product line revenues. The Economic Census is a detailed profile of the U.S. economy — from the national level to the local level and from one industry to another. The Economic Census is conducted every five years, for years ending in ‘2 and ‘7. The Economic Census provides information on over 23 million businesses and 96 percent of the nation’s economic activity, including data for over 1,000 different industries, including 8,000 manufactured products and 3,000 merchandise, commodity and service lines. These data inform economic and financial decisions in the private sector, as well as the federal, state, and local levels.

In addition to the decennial and Economic Censuses, the Census Bureau also collects other data, such as the Current Population Survey, the American Housing Survey, and the Service Annual Survey. These surveys provide information about a range of topics, from public finances to unemployment to housing conditions. Data from these censuses and surveys inform policy decisions at the federal level. All of these programs and the data they produce go to support a complex information infrastructure. This infrastructure supports informed decision making. Moreover, data such as these perform an important service to the American taxpayer because they provide accurate measurements of public needs and program effectiveness.

In fact, there is an increasing need for such data at the local level. The Economic Census and the decennial census provide such data. Yet, one of the historic limitations of the decennial census is that the information grows stale and increasingly less accurate as we move further from Census Day. The American Community Survey, one of the components of the reengineered 2010 decennial program, addresses this limitation. The American Community Survey is the most important evolution for the decennial census in over 60 years. The American Community Survey was fully implemented last year with the support of Congress and replaces the long form of the census—an important and crucial step in realizing a short-form only census. In the past we collected long-form data in the context of the decennial census. It was costly and it undermined our effort to conduct a basic enumeration. Now with the American Community Survey, we will collect the detailed data for socioeconomic characteristics over the course of the decade providing yearly, up-to-date information to federal users and our nation’s communities.

The American Community Survey collects information such as educational attainment, income levels, housing values, and other socio-economic and housing characteristics. Every question on the American Community Survey is mandated by law or fulfills federal requirements. With a three-million-household sample every year, approximately 250,000 households per month, the American Community Survey will deliver data to governments with populations of 65,000 or more beginning in August of this year. As the survey continues, we will publish long-form type data for places of 20,000 or more in 2008 and for all other areas, including census tracts, in 2010 and every year thereafter.

The American Community Survey is an important development providing timely data for states and local communities, replacing the old system that delivered data once a decade. These data are required to carry out an array of federal mandates, including the Voting Rights Act. The answers to population and housing questions support programs such as No Child Left Behind, Low Income Home Energy Assistance Program, and community block grants. These data support programs that are important to local communities such as Wilmington, Delaware and Tulsa, Oklahoma. The American Community Survey will provide timely, accurate information for every county, city, and neighborhood—the level where the most crucial decisions affecting American communities are made. American Community Survey data will help city and community leaders in Milford, Delaware and Wagoner, Oklahoma, and at the same time, allow the Census Bureau to focus its efforts in 2010 on the core, constitutional responsibility to conduct the enumeration.

The success of the 2010 reengineered census program effort will also depend on the MAF/TIGER Enhancement Program, an extensive, nation-wide operation to modernize and consolidate MAF/TIGER. MAF/TIGER is essential to the American Community Survey and the decennial census. The MAF is the address list that furnishes us with a list of households to mail questionnaires to or to contact as part of the other enumeration operations. TIGER—the street map for the census—is a digital geographic database that includes complete, consistent coverage of the United States and its territories. We use these tools to organize our work. The MAF/TIGER Enhancement Program is a multi-faceted effort, taking advantage of well-established technologies, such as Global Positioning System (GPS) capabilities, to improve the outdated, error-prone map system currently in place. We are working to align streets of the TIGER maps in order to exploit GPS capabilities and we are expanding our geographic partnerships with state, tribal, and local governments. To date, we have realigned the streets and roads for approximately 1,700 of the nation’s counties, with about 1,600 more to go in order to reach completion by April 2008.

This improvement program, along with other geographic activities, is important because ensuring the accuracy of the spatial location of the addresses is the only guarantee that political representation and resources can be distributed fairly, as they are allocated to geographic entities — states, cities, towns, census tracts, and blocks. The need for accuracy underscores the unique nature of the American census, and our constitutional and legal obligation to ensure the accuracy of the census. To collect high quality, timely, consistent data in a nation as diverse as ours is a challenge. I’m not only speaking in the sense of its difficulty, but also of the value and importance of this task. The challenge was issued in the Constitution of the United States, which states that an “actual enumeration shall be made within three years after the first meeting of the Congress of the United States, and within every subsequent term of ten years.” Starting in 1790 through today, we have honored the promise made by our Founders to ensure fair representation.

Our overriding goal for the 2010 Census is to improve the coverage and accuracy of the census; and we have developed a rigorous planning and testing program that includes many long-sought census improvements, such as bilingual questionnaires, a second questionnaire mailing, and targeted census coverage follow-up programs.

One significant improvement is automation and infrastructure. Part of our efforts have centered on two major systems, the 2010 Decennial Response Integration System (DRIS) and the Field Data Collection Automation (FDCA) system. Both of these are large information technology (IT) contracts, totaling together over one billion dollars. The purpose of the DRIS contract, which was awarded last year to Lockheed Martin Corporation, is to ensure accurate and protected collection and storage of Americans’ data whether by paper form, hand-held computer, or telephone. We are currently involved in Phase I of this program, which includes design and implementation of the system for the 2008 Census Dress Rehearsal.

For the 2010 Census, the Census Bureau also plans to increase the use of automation to directly capture information collected during personal interviews in non-response follow-up and eliminate the need for paper maps and address lists for the major field data collection operations. The FDCA contract was awarded this spring to the Harris Corporation. It provides automation resources to support field data collection operations, including an integrated IT infrastructure, as well as support for mobile computing devices and other aspects of the field activities.

The census testing program, as well as the Dress Rehearsal, is also central to our efforts. This year we are conducting the final Census Test in Travis County, Texas, and on the Cheyenne River Reservation in South Dakota. These sites provide an environment to further test and refine census operations and activities, such as the use of GPS-equipped hand-held computers and a replacement, second mailing of the questionnaires to nonresponding households. We will also focus on enumeration methods within an American Indian community, finding ways to improve coverage and testing improved self-response options. The testing program is valuable because it allows us to test operations separately in different environments to determine whether these operations can be used in census-like conditions and to prepare for the Census Dress Rehearsal.

We strive to make operations in the Dress Rehearsal closely resemble the actual census. We will conduct the 2008 Dress Rehearsal in two locations, San Joaquin County, California, and in Fayetteville and Eastern North Carolina, opening Local Census Offices in Stockton and Fayetteville. The Dress Rehearsal will feature the technology we plan to use in the decennial census, including the various data collection operations that are being developed through DRIS and FDCA. We will include a second mailing to encourage households to respond and potentially reduce the costly non-response follow-up workload. We will use a targeted mailing of Spanish/English bilingual questionnaires which were successfully tested in the 2005 National Census Test.

All of this underscores the importance of congressional support for all aspects of the 2010 Decennial Census Program, from the American Community Survey to the Dress Rehearsal. Thousands of individual operations and procedures must be successfully implemented before Census Day, less than four years from now, in order to ensure the success of the 2010 Census. The decennial census, as I mentioned before, is the largest peacetime mobilization undertaken by the government. It is our responsibility to count every community, every street, every household, and every person. It is, therefore, necessarily a complex and expensive task.

To fully understand the costs it is worthwhile to consider the scope of this task. To conduct a census of every household in the United States for Census 2000, the Census Bureau sent questionnaires to more than 117 million households and printed more than 1.5 billion pieces of paper. Approximately, 67 percent, or 80 million, of those households responded by mail. For the rest, we sent census takers, or enumerators, to collect the census information. We opened 520 Local Census Offices and hired more than 860 thousand temporary workers to conduct the census. For 2010, we are projecting there will be more than 310 million people living in America and that we will have to count more than 130 million households.

The President’s FY 2007 budget request for the Census Bureau is over $800 million dollars. This request includes approximately $184 million for salaries and expenses, as well as $182 million for other economic and demographic programs conducted by the Census Bureau. The majority of the budget request— $512 million—is for the decennial census program. This is an increase of $64 million from last year and includes $180 million for the American Community Survey; $74 million for MAF/TIGER; and $258 million for 2010 Census activities. Over the course of the decade, or the decennial life cycle, we expect the reengineered census will cost more than $11 billion. (An attached document, Estimated Life Cycle Costs for Reengineering the 2010 Decennial Census Program, provides further explanation of these costs.) This figure includes the cost of yearly data from the American Community Survey, the MAF/TIGER Enhancement Program, and the short-form only 2010 Census. It is also true that this figure is considerably higher than the cost of Census 2000. However, the rate of increase, about 50 percent (30 percent after inflation), is much less when compared to previous censuses.

We must also consider that our increasingly diverse population is more difficult to count. In addition, experience reveals that people have become more resistant to answering surveys and providing information to the government. As we plan and test new data collection methods, we try to make reasonable calculations about the impact they will have on public cooperation and the overall response rate, since the non-response follow-up operation is truly the cost-driver for the census. We have successfully tested and plan to implement bilingual questionnaires in selected communities; a second mailing to non-responding households; and automated field data collection.

We have also considered other data collection methods, including Internet data collection. Based on our research, as well as our own experience and knowledge of the experiences of other countries, we do not believe Internet data collection would significantly improve the overall response rate or reduce field data collection. The Census Bureau offers an electronic response option for the Economic Census and other economic surveys and we generally obtain high response rates. It is altogether different, however, when we consider household and population surveys and censuses. The 2003 and 2005 Census Tests offered an Internet response option, and in both cases, the response rates were low, and offering an internet response option did not increase the overall response rate. We have also consulted the statistical offices of Australia, Canada, and New Zealand. Each of these countries utilized the Internet in their most recent censuses. The Internet response rate ranged from 7 to 15 percent. Each of the statistical offices indicated that it was not possible to accurately anticipate the response rate, and that ultimately using the Internet did not affect the overall response rate. Anticipating the response rate has important operational considerations. Because they were unable to accurately anticipate the Internet response rate, the other countries were unable to reduce the paper data capture operations out of concern they would not have the capacity to fully process the census responses. This would be true for the Census Bureau as well. Moreover, while the Internet response option did not reduce the overall cost of data collection, and the cost for some specific activities, such as security and server capacity, increased.

We have seriously considered the lessons our colleagues have learned. We are also concerned that utilizing the Internet could jeopardize other planned improvements. At this point in the decade, efforts to develop an Internet response option would divert attention and resources from tested and planned improvements such as the second mailing—which we know can increase the overall response rate by several percentage points. It is also important to keep in mind that the 2010 Census utilizes only the short form. There are very few questions in this form, and most can be answered by checking a box.

A successful census is more than a technical achievement; it is the creation of a national resource that empowers decision making. The decennial census, including the American Community Survey, is a national resource—available to everyone. It is also important to remember that when we make data available, we have taken every step we can to protect the confidentiality of those data. The data we produce do not reveal individual identities. This is a legal requirement applicable to every household and business from which we collect data, including every person in the American Community Survey. At the Census Bureau, every person takes an oath not to disclose the data we collect. In fact, violators are subject to stiff fines and imprisonment. Every person is sworn for life. I took this oath over thirty years ago when I first came to the Census Bureau. It is an oath I honor and that the Census Bureau takes seriously because we believe this requirement is critically important in our society. People do question government motives and want to protect their personal privacy. Our most important relationship is with the respondent, because good data depends on their trust and their willingness to provide answers.

In fact, the trust of our respondents and your support are the most essential ingredients for the success of the decennial census; and I hope, Mr. Chairman, you will agree that it is a success worth supporting. I thank you for this opportunity to provide an update to the Senate and I would be happy to answer your questions.