<DOC> [109th Congress House Hearings] [From the U.S. Government Printing Office via GPO Access] [DOCID: f:27016.wais] APPORTIONMENT IN THE BALANCE: A LOOK INTO THE PROGRESS OF THE 2010 DECENNIAL CENSUS ======================================================================= HEARING before the SUBCOMMITTEE ON FEDERALISM AND THE CENSUS of the COMMITTEE ON GOVERNMENT REFORM HOUSE OF REPRESENTATIVES ONE HUNDRED NINTH CONGRESS SECOND SESSION __________ MARCH 1, 2006 __________ Serial No. 109-131 __________ Printed for the use of the Committee on Government Reform Available via the World Wide Web: http://www.gpoaccess.gov/congress/ index.html http://www.house.gov/reform ______ U.S. GOVERNMENT PRINTING OFFICE 27-016 WASHINGTON : 2006 _____________________________________________________________________________ For Sale by the Superintendent of Documents, U.S. Government Printing Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800; (202) 512ÿ091800 Fax: (202) 512ÿ092250 Mail: Stop SSOP, Washington, DC 20402ÿ090001 COMMITTEE ON GOVERNMENT REFORM TOM DAVIS, Virginia, Chairman CHRISTOPHER SHAYS, Connecticut HENRY A. WAXMAN, California DAN BURTON, Indiana TOM LANTOS, California ILEANA ROS-LEHTINEN, Florida MAJOR R. OWENS, New York JOHN M. McHUGH, New York EDOLPHUS TOWNS, New York JOHN L. MICA, Florida PAUL E. KANJORSKI, Pennsylvania GIL GUTKNECHT, Minnesota CAROLYN B. MALONEY, New York MARK E. SOUDER, Indiana ELIJAH E. CUMMINGS, Maryland STEVEN C. LaTOURETTE, Ohio DENNIS J. KUCINICH, Ohio TODD RUSSELL PLATTS, Pennsylvania DANNY K. DAVIS, Illinois CHRIS CANNON, Utah WM. LACY CLAY, Missouri JOHN J. DUNCAN, Jr., Tennessee DIANE E. WATSON, California CANDICE S. MILLER, Michigan STEPHEN F. LYNCH, Massachusetts MICHAEL R. TURNER, Ohio CHRIS VAN HOLLEN, Maryland DARRELL E. ISSA, California LINDA T. SANCHEZ, California JON C. PORTER, Nevada C.A. DUTCH RUPPERSBERGER, Maryland KENNY MARCHANT, Texas BRIAN HIGGINS, New York LYNN A. WESTMORELAND, Georgia ELEANOR HOLMES NORTON, District of PATRICK T. McHENRY, North Carolina Columbia CHARLES W. DENT, Pennsylvania ------ VIRGINIA FOXX, North Carolina BERNARD SANDERS, Vermont JEAN SCHMIDT, Ohio (Independent) ------ ------ Melissa Wojciak, Staff Director David Marin, Deputy Staff Director/Communications Director Rob Borden, Parliamentarian Teresa Austin, Chief Clerk Phil Barnett, Minority Chief of Staff/Chief Counsel Subcommittee on Federalism and the Census MICHAEL R. TURNER, Ohio, Chairman CHARLES W. DENT, Pennsylvania WM. LACY CLAY, Missouri CHRISTOPHER SHAYS, Connecticut PAUL E. KANJORSKI, Pennsylvania VIRGINIA FOXX, North Carolina CAROLYN B. MALONEY, New York ------ ------ Ex Officio TOM DAVIS, Virginia HENRY A. WAXMAN, California John Cuaderes, Staff Director Ursula Wojciechowski, Professional Staff Member Juliana French, Clerk Mark Stephenson, Minority Professional Staff Member C O N T E N T S ---------- Page Hearing held on March 1, 2006.................................... 1 Statement of: Kincannon, Charles Louis, Director, U.S. Census Bureau; Brenda S. Farrell, Acting Director, Strategic Issues, U.S. Government Accountability Office; and David A. Powner, Director, Information Technology Management Issues, U.S. Government Accountability Office........................... 8 Farrell, Brenda S........................................ 14 Kincannon, Charles Louis................................. 8 Powner, David A.......................................... 37 Rector, Ralph, Ph.D., senior research fellow and project manager, Center for Data Analysis, the Heritage Foundation; Andrew Reamer, Ph.D., deputy director, Urban Markets Initiative, the Brookings Institution; and Margo Anderson, professor, history and urban studies, University of Wisconsin-Milwaukee........................................ 126 Anderson, Margo.......................................... 155 Reamer, Andrew, Ph.D..................................... 135 Rector, Ralph, Ph.D...................................... 126 Letters, statements, etc., submitted for the record by: Anderson, Margo, professor, history and urban studies, University of Wisconsin-Milwaukee, prepared statement of... 157 Clay, Hon. Wm. Lacy, a Representative in Congress from the State of Missouri, prepared statement of................... 6 Farrell, Brenda S., Acting Director, Strategic Issues, U.S. Government Accountability Office, prepared statement of.... 16 Kincannon, Charles Louis, Director, U.S. Census Bureau, prepared statement of...................................... 10 Maloney, Hon. Carolyn B., a Representative in Congress from the State of New York, prepared statement of............... 115 Powner, David A., Director, Information Technology Management Issues, U.S. Government Accountability Office, prepared statement of............................................... 39 Reamer, Andrew, Ph.D., deputy director, Urban Markets Initiative, the Brookings Institution, prepared statement of......................................................... 137 Rector, Ralph, Ph.D., senior research fellow and project manager, Center for Data Analysis, the Heritage Foundation, prepared statement of...................................... 128 Turner, Hon. Michael R., a Representative in Congress from the State of Ohio, prepared statement of................... 3 APPORTIONMENT IN THE BALANCE: A LOOK INTO THE PROGRESS OF THE 2010 DECENNIAL CENSUS ---------- WEDNESDAY, MARCH 1, 2006 House of Representatives, Subcommittee on Federalism and the Census, Committee on Government Reform, Washington, DC. The subcommittee met, pursuant to notice, at 10:06 a.m., in room 2154, Rayburn House Office Building, Hon. Michael R. Turner (chairman of the subcommittee) presiding. Present: Representatives Turner, Foxx, Clay and Maloney. Staff present: John Cuaderes, staff director; Ursula Wojciechowski, professional staff member; Juliana French; clerk; Adam Bordes and Mark Stephenson, minority professional staff members; and Jean Gosa, minority assistant clerk. Mr. Turner. A quorum being present, this hearing of the Subcommittee on federalism and the Census will come to order. Welcome to the subcommittee's oversight hearing entitled, ``Apportionment in the Balance: A Look into the Progress of the 2010 Decennial Census.'' Today we will consider the status of the Census Bureau's preparations for the 2010 decennial census. This hearing is a followup to our April 19, 2005, hearing last year entitled, ``Halfway to the 2010 Census: The Countdown and Components to a Successful Decennial Census.'' Since then the Census Bureau has achieved and is nearing completion of several key milestones. The Bureau has successfully carried out the American Community Survey for 1 full year. Additionally, the MAF/TIGER Enhancement Program is nearing what we all hope will be a successful completion. As the Bureau continues its preparation for a short form only census, it is undertaking two major contracts: the Field Data Collection Automation program and the Decennial Response Integration System. These two technology contracts have a combined value of over $1 billion. These major contracts signal the first real ``hi-tech'' census, and the subcommittee will examine how the successful implementation of these contracts is critical to the 2010 decennial census. Furthermore, the subcommittee will explore several other issues such as the Local Update of Census Addresses [LUCA], program and the intergovernmental partnerships required to facilitate the program. There are a number of important issues that can impact the successful implementation of the census, including personnel and infrastructure matters, hiring and training temporary workers, and establishing temporary field offices. Testing for the 2010 decennial census is already underway. The Bureau is testing policy and technology concepts in Travis County, TX, and the Cheyenne River Reservation in South Dakota. Canvassing in Texas was to be completed in 6 weeks, and the subcommittee understands that this goal was not met. In today's hearing, we will examine this issue, as well as the issue of using handheld technology in the testing environment. It is our understanding that the handhelds failed to perform adequately and the activity was concluded without finishing the address file that is needed for the next test phase. These issues must be resolved before the 2008 dress rehearsal. I am eager to hear what the Census Bureau is doing to address the problems of their tests and other issues related to the 2010 decennial census. On our first panel, we welcome remarks from the Honorable Charles Louis Kincannon, director of the Census Bureau. Then we will hear from Ms. Brenda Farrell, Acting Director of Strategic Issues, and Mr. David Powner, Director of Information Technology Management Issues, both from the Government Accountability Office, regarding their assessment of the Bureau's planning for the decennial census. On our second panel, we will hear from Dr. Ralph Rector from the Heritage Foundation, Dr. Andrew Reamer from the Brookings Institution, and last, we will hear from Dr. Margo Anderson, professor of history and urban studies at the University of Wisconsin-Milwaukee. [The prepared statement of Hon. Michael R. Turner follows:] [GRAPHIC] [TIFF OMITTED] T7016.001 [GRAPHIC] [TIFF OMITTED] T7016.002 Mr. Turner. With that, my colleagues on the subcommittee and I welcome you and look forward to your testimony, and I now yield to the gentleman from Missouri, the distinguished Member Mr. Clay, for any opening remarks that he may have. Mr. Clay. Good morning, and thank you, Mr. Chairman, for calling today's hearing to review the Census Bureau's efforts for carrying out the 2010 decennial census. I welcome our witnesses, especially Director Kincannon of the Census Bureau. I would like to begin by expressing my unwavering support for the American Community Survey and its goals of delivering more timely and effective data to the Bureau. As the ranking member of this subcommittee, I am pleased to have been a part of the efforts to bring the ACS into reality. The Census Bureau is now collecting data in every county in the United States, and this effort will make the 2010 census less complex and more efficient. Information from the decennial census provides an important benchmark for the formulas governing many of our domestic programs, thus helping us serve the needs of our citizens. The Bureau continues to face pressing challenges, however, as preparations for the 2010 decennial census begin in earnest. First, it remains unclear if their acquisitions for new IT infrastructure will be tested and ready for the 2008 rehearsal of 2010 census. It will be difficult to ensure an accurate census if we cannot rely on the new technologies being implemented to aid in agency efforts. Furthermore, it remains unclear to me if past problems concerning the undercounting of minority populations or the accuracy of the Master Address File have been remedied. With only 4 years left until field work begins, plans and mitigation strategies must be established to ensure the most accurate and reliable census possible. Once again, I look forward to learning more today about the plans for the 2010 census and how we in Congress can be of assistance to the Bureau. This concludes my statement, Mr. Chairman, and I yield back. [The prepared statement of Hon. Wm. Lacy Clay follows:] [GRAPHIC] [TIFF OMITTED] T7016.003 [GRAPHIC] [TIFF OMITTED] T7016.004 Mr. Turner. Thank you, Mr. Clay. I appreciate your interest and your professionalism on the committee and your work with the issues of the census. We will now start with the witnesses. Each witness has kindly prepared written testimony, which will be included in the record of this hearing. Witnesses will notice there is a timer with a light on at the witness table. The green light indicates that you should begin your prepared remarks, and the red light indicates that time has expired. It is the policy of this committee that all witnesses be sworn in before they testify. If you would please rise and raise your right hands. [Witnesses sworn.] Mr. Turner. Please let the record show that all the witnesses have responded in the affirmative, and we will begin with Mr. Kincannon. STATEMENTS OF CHARLES LOUIS KINCANNON, DIRECTOR, U.S. CENSUS BUREAU; BRENDA S. FARRELL, ACTING DIRECTOR, STRATEGIC ISSUES, U.S. GOVERNMENT ACCOUNTABILITY OFFICE; AND DAVID A. POWNER, DIRECTOR, INFORMATION TECHNOLOGY MANAGEMENT ISSUES, U.S. GOVERNMENT ACCOUNTABILITY OFFICE STATEMENT OF CHARLES LOUIS KINCANNON Mr. Kincannon. Thank you, Mr. Chairman. Good morning, and on behalf of the Census Bureau, I want to thank you and the ranking minority member and the whole committee for inviting me to testify. Today we are 4 years and 1 month from census day. The success of the decennial census is the Census Bureau's largest and most important priority. It represents 60 percent of the President's 2007 budget request for the Census Bureau. The budget for the decennial census program covers the American Community Survey [ACS]; the MAF/TIGER enhancement; and the 2010 census activities themselves. Because of strong congressional support, the American Community Survey is on track and moving toward its goals. We will release the first annual estimates from the full ACS this August for approximately 8,000 communities with populations of 65,000 or more and, for the first time, data for every congressional district in the country. In 2008, we will release data for communities of 20,000 or more, and in 2010, data for every census tract--2 years before equivalent data would be available from a traditional census long form. The ACS is crucial to the overall success of the decennial census because it replaces the long form and allows us to focus our attention on a complete count of the American population. Another critical component of the success of the 2010 census is the address list and map. The Census Bureau is conducting an extensive nationwide operation to modernize and consolidate MAF/TIGER. We are using GPS to align the streets of the TIGER maps and working with communities to ensure that we do not miss a new neighborhood. To date, we have realigned the streets and roads for about 1,700 of the Nation's counties, with about 1,600 more to go in order to reach completion by April 2008. We will complete this task on time. We are also working to improve our most significant partnership opportunity--the Local Update of Census Addresses program [LUCA]. In 2007, we will invite the Nation's 39,000 municipalities to help update the address list for their communities for use in the 2010 census. Knowing that communities differ, we will offer different options for governments to participate, ranging from a full review of the address file to a simple review of housing unit counts. We are working to strengthen the infrastructure of the 2010 census through technology. As the chairman said, this will be the first truly high-tech census. Our efforts have centered on two major systems: the 2010 Decennial Census Response Integration System [DRIS], as we call it, and the Field Data Collection Automation system [FDCA]. These large information technology contracts together total over $1 billion. The purpose of the DRIS contract, which was awarded last year to Lockheed Martin Corp., is to ensure accurate and protected collection and storage of Americans' data whether by paper form or handheld computer. The FDCA contract, which will be awarded in the next month or so, provides automation resources to support field data collection operations. As we move forward, it is essential to remain on schedule. This year, we will conduct a final test census in Travis County, TX, and the Cheyenne River Reservation in South Dakota. These tests are important to our ability to conduct a successful dress rehearsal and ultimately a successful census. We will conduct the 2008 dress rehearsal in two locations: San Joaquin County, CA, and in nine counties surrounding Fayetteville, NC. In fact, some aspects of the dress rehearsal are already underway, including LUCA. The dress rehearsal will use the technology we plan to use in the decennial census, and this is quite important. No last- minute experiments. We will include a targeted second mailing of questionnaires to encourage households to respond and reduce costly non-response followup. We will also send a targeting mailing of Spanish-English bilingual questionnaires in selected neighborhoods. It is important to note that many of the 2010 census operations and procedures, and especially decisions, those involving technology, need to be in place before the dress rehearsal. The President's budget recognizes that we cannot postpone improvements or tests without introducing risk to the census. All of this underscores the importance of congressional support for all aspects of the 2010 decennial census program from the ACS to the dress rehearsal. Thousands of individual operations and procedures must be successfully implemented before census day in order to ensure the success of the census. Mr. Chairman, I thank you for this opportunity to provide an update to the Congress, and I look forward to answering your questions. Thank you. [The prepared statement of Mr. Kincannon follows:] [GRAPHIC] [TIFF OMITTED] T7016.005 [GRAPHIC] [TIFF OMITTED] T7016.006 [GRAPHIC] [TIFF OMITTED] T7016.007 [GRAPHIC] [TIFF OMITTED] T7016.008 Mr. Turner. Thank you. Ms. Farrell. STATEMENT OF BRENDA FARRELL Ms. Farrell. Mr. Chairman, Mr. Clay, thank you for the opportunity to be here today to discuss the Census Bureau's preparations for the 2010 census. Full and comprehensive planning is crucial to the success of any large, long-term project, especially with the costs, complexity, and high stakes of the decennial census. The 2010 census projected life-cycle costs span 13 years and total over $11 billion, and its recruitment goals are similar to the 2000 census--2\1/2\ million applicants could be recruited to carry out census operations. Given the escalating costs of the census in an era of serious national fiscal challenges, oversight will be particularly important. As shown in the figures on the screen and on page 6 of the testimony, the projected average cost is $72 per housing unit for 2010, and it is nearly 5\1/2\ times greater than the $13 it cost to count each household in 1970 in constant fiscal year 2000 dollars. My remarks today are based on findings from our prior report and preliminary results from ongoing work that we plan to issue in the near future. First, I will describe the overall progress that the Bureau is making toward preparing the 2010 census. Second, I will note some issues that pose a risk to a successful census. Most importantly, the Bureau is further along in planning the 2010 census compared to a similar point in time during the 2000 census cycle. Early in this decade, the Bureau developed a promising design to achieve its principal goals for the 2010 census. The fundamental design of the census has the potential to control costs and improve coverage and accuracy. Also noteworthy is the Bureau's greater willingness to outsource key census-taking operations that would be difficult for it to carry out on its own. It will be important for the Bureau to focus on its acquisition activities to help ensure the 2010 contractors fulfill the Bureau's expectations. While the Bureau should be commended for the progress that it has made, it will be important for the Bureau to resolve issues that pose a risk to a successful census. For example, the Bureau plans to use handheld mobile computing devices to help develop the census address list and collect data from millions of households that do not respond to the initial census questionnaires. These handheld devices are an important step forward because they are designed to replace many of the paper questionnaires and maps that were used in past censuses and are a key element of the Field Data Collection Automation program, one of the acquisition contracts that my colleague Mr. Powner will discuss. The Bureau has never before used these devices in a decennial. In tests held to date, census workers found the devices easy to use for such things as using the electronic maps to find their assignment areas. On the other hand, the reliability of the devices proved troublesome as the devices experienced transmission problems and memory overload. The Bureau has taken steps to address these issues, and next month the devices will be evaluated again, and we will be onsite to assess the extent to which the Bureau has addressed these reliability issues. Further, I would like to note that Hurricanes Katrina and Rita highlight the importance of contingency planning and examining whether the Bureau's existing operations are adequate for capturing the demographic and physical changes that have occurred along the Gulf Coast. We have had a preliminary discussion with the Bureau on this topic and will continue to assess the Bureau's contingency planning as part of our oversight of the 2010 census. In conclusion, while the ramp-up to 2010 is making progress, past experience has shown that Congress has every reason to remain vigilant. As we have done throughout the past several decades, we look forward to supporting the subcommittee in its decisionmaking and oversight efforts. Mr. Chairman, this concludes my statement. I will be happy to take questions from you or Mr. Clay at your convenience. [The prepared statement of Ms. Farrell follows:] [GRAPHIC] [TIFF OMITTED] T7016.009 [GRAPHIC] [TIFF OMITTED] T7016.010 [GRAPHIC] [TIFF OMITTED] T7016.011 [GRAPHIC] [TIFF OMITTED] T7016.012 [GRAPHIC] [TIFF OMITTED] T7016.013 [GRAPHIC] [TIFF OMITTED] T7016.014 [GRAPHIC] [TIFF OMITTED] T7016.015 [GRAPHIC] [TIFF OMITTED] T7016.016 [GRAPHIC] [TIFF OMITTED] T7016.017 [GRAPHIC] [TIFF OMITTED] T7016.018 [GRAPHIC] [TIFF OMITTED] T7016.019 [GRAPHIC] [TIFF OMITTED] T7016.020 [GRAPHIC] [TIFF OMITTED] T7016.021 [GRAPHIC] [TIFF OMITTED] T7016.022 [GRAPHIC] [TIFF OMITTED] T7016.023 [GRAPHIC] [TIFF OMITTED] T7016.024 [GRAPHIC] [TIFF OMITTED] T7016.025 [GRAPHIC] [TIFF OMITTED] T7016.026 [GRAPHIC] [TIFF OMITTED] T7016.027 [GRAPHIC] [TIFF OMITTED] T7016.028 [GRAPHIC] [TIFF OMITTED] T7016.029 Mr. Turner. Thank you. Mr. Powner. STATEMENT OF DAVID POWNER Mr. Powner. Chairman Turner, Ranking Member Clay, we appreciate the opportunity to testify on key acquisitions supporting the 2010 decennial census. The use of automation will be critical to the success of the upcoming census. Nearly a quarter of the 2010 respondents are expected to use the Internet. Key technologies will be used to standardize responses, and field-based enumerators plan to use nearly half a billion mobile computing devices. However, acquiring technologies can present enormous challenges and risks if not managed effectively. These technology acquisition risks have been highlighted in numerous oversight hearings by Chairman Davis at the full committee, and your early attention to and leadership over the decennial acquisitions, Mr. Chairman, will hopefully ensure greater Bureau and contractor accountability. This morning, as requested, I will summarize the importance and status of two key acquisitions that are critical to the 2010 decennial and key management activities that the Census Bureau is establishing that are crucial to delivering this technology on time, at cost, and with the promised functionality. The Census Bureau has initiated efforts to acquire the Response Integration System and the Field Data Collection Automation program. The integration system is intended to receive and standardize census data from the various response modes, including census forms, telephone agents, and the Internet. It is also intended to standardize data collected from mobile computing devices, which are key to capturing non- response followup. The mobile devices are part of the data collection program, which is also expected to provide office automation for regional and local census offices, as well as the telecommunications infrastructure. The integration contract was awarded to Lockheed Martin and its seven subcontractors in October, and the Field Data Collection contract was to be awarded at the end of this month. However, the Director's testimony this morning indicates the contract will occur in late spring. Both projects' life-cycle costs are expected to total over $1 billion. Both acquisitions involve ambitious schedules to deliver the needed functionality to support the planned 2008 dress rehearsal and are absolutely essential to achieving the goals of the decennial, including increased coverage, accuracy, and timeliness of the data. Key management activities and processes are needed to effectively manage these acquisitions. Last June, we reported to you, Mr. Chairman, that the Census Bureau's institutional information technology management capacity still had room for improvement. Given these weaknesses and the importance of the integration system and the Field Data Collection program, you asked for a detailed review of these acquisitions to assist in your oversight of the decennial. While both projects have initiated steps to establish key project management activities, neither has the full set of capabilities needed to effectively manage these acquisitions. Incomplete management activities include those for requirements management, risk management, and contract monitoring. These increase the risk that these projects will encounter problems in meeting cost and schedule expectations. Given the immovable deadline for performing the 2010 decennial census, if unexpected problems occur, the Bureau will be faced with two options to address these problems: one, throwing more money at them; or, two, accepting systems with reduced functionality. To address these program management shortfalls, my written statement includes a number of specific recommendations that focus on further defining exactly what is expected from these acquisitions, establishing robust risk mitigation programs that include early escalation and quick resolution of risks, and establishing clear metrics to oversee contractor performance. The decennial management team appears to be dedicated to bolstering its management capabilities and have told us that they plan to complete these important activities as soon as possible. I would like to stress that these endorsed management activities ultimately are about placing the Government in charge of defining what it wants, being on top of risks, and having metrics to measure contractor performance. Relying on contractors for technical solutions is fine. Relying on them for requirements and performance metrics is not. Mr. Chairman, the Response Integration and the Field Data Collection program are crucial to the success of the decennial. Although we commend efforts to date to establish these key contracts, additional management attention is needed to effectively oversee these acquisitions. Establishing the recommended management activities are critical to ensuring that the Bureau is in the driver's seat as these acquisitions process forward. This concludes my statement, Mr. Chairman. Thank you for your leadership and oversight of the decennial census. [The prepared statement of Mr. Powner follows:] [GRAPHIC] [TIFF OMITTED] T7016.030 [GRAPHIC] [TIFF OMITTED] T7016.031 [GRAPHIC] [TIFF OMITTED] T7016.032 [GRAPHIC] [TIFF OMITTED] T7016.033 [GRAPHIC] [TIFF OMITTED] T7016.034 [GRAPHIC] [TIFF OMITTED] T7016.035 [GRAPHIC] [TIFF OMITTED] T7016.036 [GRAPHIC] [TIFF OMITTED] T7016.037 [GRAPHIC] [TIFF OMITTED] T7016.038 [GRAPHIC] [TIFF OMITTED] T7016.039 [GRAPHIC] [TIFF OMITTED] T7016.040 [GRAPHIC] [TIFF OMITTED] T7016.041 [GRAPHIC] [TIFF OMITTED] T7016.042 [GRAPHIC] [TIFF OMITTED] T7016.043 [GRAPHIC] [TIFF OMITTED] T7016.044 [GRAPHIC] [TIFF OMITTED] T7016.045 [GRAPHIC] [TIFF OMITTED] T7016.046 [GRAPHIC] [TIFF OMITTED] T7016.047 [GRAPHIC] [TIFF OMITTED] T7016.048 [GRAPHIC] [TIFF OMITTED] T7016.049 [GRAPHIC] [TIFF OMITTED] T7016.050 [GRAPHIC] [TIFF OMITTED] T7016.051 [GRAPHIC] [TIFF OMITTED] T7016.052 [GRAPHIC] [TIFF OMITTED] T7016.053 [GRAPHIC] [TIFF OMITTED] T7016.054 [GRAPHIC] [TIFF OMITTED] T7016.055 [GRAPHIC] [TIFF OMITTED] T7016.056 [GRAPHIC] [TIFF OMITTED] T7016.057 [GRAPHIC] [TIFF OMITTED] T7016.058 [GRAPHIC] [TIFF OMITTED] T7016.059 [GRAPHIC] [TIFF OMITTED] T7016.060 [GRAPHIC] [TIFF OMITTED] T7016.061 [GRAPHIC] [TIFF OMITTED] T7016.062 [GRAPHIC] [TIFF OMITTED] T7016.063 [GRAPHIC] [TIFF OMITTED] T7016.064 [GRAPHIC] [TIFF OMITTED] T7016.065 [GRAPHIC] [TIFF OMITTED] T7016.066 [GRAPHIC] [TIFF OMITTED] T7016.067 [GRAPHIC] [TIFF OMITTED] T7016.068 [GRAPHIC] [TIFF OMITTED] T7016.069 [GRAPHIC] [TIFF OMITTED] T7016.070 [GRAPHIC] [TIFF OMITTED] T7016.071 [GRAPHIC] [TIFF OMITTED] T7016.072 [GRAPHIC] [TIFF OMITTED] T7016.073 [GRAPHIC] [TIFF OMITTED] T7016.074 [GRAPHIC] [TIFF OMITTED] T7016.075 [GRAPHIC] [TIFF OMITTED] T7016.076 [GRAPHIC] [TIFF OMITTED] T7016.077 [GRAPHIC] [TIFF OMITTED] T7016.078 [GRAPHIC] [TIFF OMITTED] T7016.079 [GRAPHIC] [TIFF OMITTED] T7016.080 [GRAPHIC] [TIFF OMITTED] T7016.081 [GRAPHIC] [TIFF OMITTED] T7016.082 [GRAPHIC] [TIFF OMITTED] T7016.083 [GRAPHIC] [TIFF OMITTED] T7016.084 [GRAPHIC] [TIFF OMITTED] T7016.085 [GRAPHIC] [TIFF OMITTED] T7016.086 [GRAPHIC] [TIFF OMITTED] T7016.087 [GRAPHIC] [TIFF OMITTED] T7016.088 [GRAPHIC] [TIFF OMITTED] T7016.089 [GRAPHIC] [TIFF OMITTED] T7016.090 [GRAPHIC] [TIFF OMITTED] T7016.091 [GRAPHIC] [TIFF OMITTED] T7016.092 [GRAPHIC] [TIFF OMITTED] T7016.093 [GRAPHIC] [TIFF OMITTED] T7016.094 [GRAPHIC] [TIFF OMITTED] T7016.095 [GRAPHIC] [TIFF OMITTED] T7016.096 [GRAPHIC] [TIFF OMITTED] T7016.097 [GRAPHIC] [TIFF OMITTED] T7016.098 [GRAPHIC] [TIFF OMITTED] T7016.099 [GRAPHIC] [TIFF OMITTED] T7016.100 [GRAPHIC] [TIFF OMITTED] T7016.101 [GRAPHIC] [TIFF OMITTED] T7016.102 [GRAPHIC] [TIFF OMITTED] T7016.103 [GRAPHIC] [TIFF OMITTED] T7016.104 Mr. Turner. Thank you. I want to recognize that we have been joined by Carolyn Maloney from New York. Thank you for being here today. Mrs. Maloney. Nice to see you. I would ask permission to place my opening comments in the record? Mr. Turner. Without objection. [The prepared statement of Hon. Carolyn B. Maloney follows:] [GRAPHIC] [TIFF OMITTED] T7016.105 [GRAPHIC] [TIFF OMITTED] T7016.106 Mr. Turner. Recognizing that we have an 11 o'clock address by the Prime Minister of the Republic of Italy, we are going to try to conclude our questions to panel one within that time period, which you have been very precise within the timeframe of your oral remarks, and I appreciate that, giving us what should be ample time to ask questions. I would like to start with Director Kincannon. Does the Bureau have any plans to adjust the 2010 census count? This includes working with any outside nongovernment entities to plan for adjustment. Mr. Kincannon. No, Mr. Chairman, we do not have plans to address the census results. We spent more than 3 years working on what we thought was a well-designed system to provide improvements to the quality of the count. But at the end of that more than 3-year period of our best experts working on it, our conclusion was that it was not possible with the technology and means at our disposal to adjust the census for the main products of the census which required block-level statistics and place statistics. We simply cannot do that. So we are not planning on doing that, sir. We do plan to measure coverage for purposes of continuing to assess and improve the techniques we use in the census, however. Mr. Turner. Director Kincannon, I believe that in your written testimony you did not talk about the power of the Internet and its assistance it can provide in the 2010 census. Many response-driven organizations have gone the way of the Internet as a way of collecting their data. Certainly other panel members have referenced the Internet. Will you share with us how the census will be harnessing the Internet to allow people to respond to their census questionnaire? Mr. Kincannon. Well, I am aware that the Congress has much optimism about the Internet as a way of reducing the cost of many Government activities, and sometimes that has worked. Our experience with household surveys and with census tests has been that it is not something that increases response or improves the quality of the data that we get. We tested this in the 2003 national census test. There was no difference in the response rate in the panel that had no option to respond by the Internet from the one that had the option to respond by the Internet. About 10 percent of the responses in the Internet test panel came from the Internet, but it did not increase response at all. Furthermore, the concerns about the complexity of dealing with the Internet make me very cautious in how much we would depend on that. The well-documented effort that phishing and spamming on the IRS Web site are a caution to me. I nearly fell victim to that myself, and I can see how a person who tries to respond to the census might well be duped into providing information that would lead to identity theft on the Internet. And if it does not increase our efficiency or effectiveness, increase our response rates, or measurably reduce our costs, then the cost of protecting against that is probably not warranted. We continue to explore and we contemplate having an option for Internet response, but I do not see it as a main component of what we will do. Mr. Turner. Ms. Farrell or Mr. Powner, would you like to respond to that, on the prospects of the Internet use? Mr. Powner. If I could, I think it is great that we acknowledge the security considerations with using the Internet for this next census, although some of the Bureau's internal documentation claim that up to a quarter of the respondents could, in fact, use the Internet to respond to this upcoming census. I think what is key is if you look in particular at one of the key acquisitions, which is the response system, which the contract was already let in October. You have a contractor that is working toward integrating Internet, phone, and paper forms, and it is very important that, although I see where the Director is coming from in terms of it may be lower than that quarter of the respondents, it is very important that we be prepared from a systems perspective to respond to a higher number of Internet responses and that the systems have the capability and the contractors are well prepared to integrate those Internet responses. Mr. Turner. There have been some concerns about the handhelds and if they should fail in 2008 or 2010. Mr. Powner, are you comfortable, or Ms. Farrell, with the Bureau's contingency planning with respect to the handhelds if they should not perform? Ms. Farrell. The handheld computing devices are a key part of the design for the 2010 census, and to date, for the 2006 and 2008 dress rehearsal, we are not aware of any backup plans in the event that there are widespread problems with these devices. If they fail, it will cause serious operational challenges for the Bureau to back these up with paper questionnaires or whatever else will be necessary for the verification for address canvassing. Mr. Powner. Mr. Chairman, if I can expand on Ms. Farrell's response, if you look at the FDCA contract, which is to be awarded some time this spring--and handheld is a part of that contract--it is very important to address the problems that have been identified to date that the Census Bureau specifies performance requirements. We have had serious performance issues with the handhelds to date, so understanding what the availability of those handhelds should be, what the response time is, what our peak loads are, that clearly needs to be specified in these contracts so that we could hold contractors accountable for those specific requirements. That ties to one of our management activities where you look at the requirements management area. Mr. Turner. Director Kincannon, would you like to respond? Mr. Kincannon. Yes, I would be happy to. The handhelds that we have tested in the field to date have been ones developed at the Census Bureau, and they have confirmed our opinion of a couple of years ago that we were not able to develop a device that would meet all the requirements for the dress rehearsal or the census ultimately. We simply do not have those technological capabilities. We advanced in the process of procuring those services, and I am happy to say that all of those companies who have sought to submit a bid have submitted devices, prototype devices, that exceed what we were able to do and that appear that they can fully meet the functional requirements, including security, ease of use, and communications, both wireless and landline. So we are confident that they will be able to meet the requirements that we have set out. Mr. Turner. So if I could rephrase the question, my understanding of your answer, are you saying that you trust them enough that you don't believe that there is a contingency plan that is necessary or your contingency planning has not been completed as you are looking to trying to nail down the greatest efficiency of these units? Mr. Kincannon. Well, it is the latter, Mr. Chairman. We do have contingency planning, and we are concerned about that. But we are now at this stage of things much less concerned than we were a year ago about widespread failures. We will know that more when the contract is awarded, I believe as GAO indicated, they will be closely involved in understanding the capabilities offered and in monitoring the testing of those capabilities. Mr. Turner. Mr. Clay. Mr. Clay. Thank you, Mr. Chairman. Director Kincannon, can you tell me what will be done to test the Local Update of Census Addresses in the 2006 census test? Also, can you update us with an outline of Bureau plans on what it will do to help local governments prepare for the address correction program through the Local Update of Census Addresses program? Mr. Kincannon. Yes, Congressman. We did use the LUCA process in preparation for the 2006 test, and we evaluated any problems that we identified in that. We did not consider it a test per se because we have had a well-working LUCA process. We have already begun rolling out the steps for conducting LUCA for the dress rehearsal, and we will continue to fine-tune that. But it has proved a very useful process in 2000 and in the test censuses for ensuring that we do not miss neighborhoods and that we have the best local information incorporated in what we plan to do. Mr. Clay. Can you tell me about how you reach out to local governments and assist them or work with them? Mr. Kincannon. On LUCA? Mr. Clay. Yes. Mr. Kincannon. Yes, sir. Well, we offer an array of options. There are different degrees and sophistication of local governments according to their scale and other factors, and we offer a variety of options that they can choose within the LUCA depending on what they think is their best way of checking those addresses. We give them information in advance of LUCA about what materials will be provided and how they might use them, and, you know, so we work with them in that way. Mr. Clay. There have been many individuals forced to leave the Gulf Coast after the hurricanes last fall. What efforts are underway to account for those who have left areas affected and include them in the 2010 census? Will there be a measure to determine if these individuals have left permanently or only on a temporary basis? Have you all addressed that at Census and figured how you are going to count these individuals? Mr. Kincannon. Yes, Mr. Clay, we had procedures that have been used in the past for persons displaced in hurricanes in the census year. Hurricane Floyd hit North Carolina pretty hard in 1999. A lot of people were displaced. For census 2000 we had procedures that we made sure we contacted both former residences or sites that may have been temporarily or permanently abandoned and people living in various shelters. We counted them where they said they intended to be. That was identified their usual place of residence. We are farther away, and let us hope that with the best human efforts and God's aid that the people displaced have settled in a permanent way either in their former homes or in new homes by 2010. But we do have procedures; they are arduous, and they cost extra money. But we have procedures that work with persons who are displaced. Mr. Clay. I am concerned about provisions in the President's budget that would lead to the elimination of the Survey of Income and Program Participation [SIPP], as you call it. SIPP is the only large-scale survey explicitly designed to analyze the impact of a wide variety of Government programs on the well-being of some of America's poorest families. What justification can you offer us for the elimination of the SIPP program? Mr. Kincannon. Well, first let me give the context. We operate, as always, in a setting of constrained resources. The Congress does not give us all the money that we ask for. Sometimes the policies that we follow in the executive branch mean that we are accorded lower priorities than some other things. That is not surprising. We know that. In the formulation of the budget for 2007, we did not have room within our allowance for all the things that we had done and wanted to do. Then we followed what is our practice and what is mandated by Congress over the last several years, which is to look at all our programs, do the things that are of highest priority, deal differently with things that are of lowest priority or troubled by quality and so on. When we assessed this within the Census Bureau, we realized that the SIPP program is rather mature; more than 20 years it has been in place. It has been useful. It has some chronic problems that we either have not been able to solve or have not been resourced sufficiently to solve, and that has troubled us. So what we are looking for is in a world that has changed over 20 years, with new methods, much more successful means of using administrative records from programs intended to assist those in poverty or with other kinds of difficulties, and with the successful appearance of the American Community Survey providing yet another source of data on the condition of families, that we want to put together a new program, a re- engineered program that will continue to meet the needs of Federal agencies for longitudinal studies of income dynamics. This includes the HHS Assistant Secretary for Planning and Evaluation, the Food and Nutrition Service in Agriculture, the Administration on Children and Families, the Social Security Administration, and other agencies that have as their responsibility meeting the needs of people who may be disadvantaged or at risk. So we want to re-engineer what we are doing, take recognition and build on the new sources of data that we have, and find a way to continue to meet those needs within the resource constraints we have. Mr. Clay. Just out of curiosity, what were some of the difficulties in gathering the data? Mr. Kincannon. The SIPP is a very complex survey deigned to produce both cross-sectional and longitudinal data, and the need for longitudinal data means that you have to continue interviewing the same household over a period of a couple of years, actually. Mr. Clay. Four years. Mr. Kincannon. You are better informed than I am. Your briefing notes are better than mine. Mr. Clay. I am a speed reader. [Laughter.] Mr. Kincannon. Well, I am being too long-winded if I am giving you a chance to read all that. It is very difficult, we have found, and increasingly difficult to keep up the response of families, follow them when they move, and so on over that 4-year period of time, so that there is severe attrition. And we need to find some way differently to address that. Mr. Clay. And you think you will come up with a more efficient manner or condense the way you take the survey? Mr. Kincannon. Well, we think that we can come up with a better model of using tools now available to us, both from survey results in the ACS and CPS and our gained experience in modeling and improved access and capability with regard to administrative records, and providing something that will help these agencies meet their responsibilities. Mr. Clay. I thank you for your response. I appreciate it. Mr. Kincannon. Thank you. Mr. Turner. Ms. Foxx. Ms. Foxx. I would like to followup on the question about what do you do about people who have been displaced by something like Katrina. One thing I am concerned about is it seems to me that if people have been away from the home for 5 years, what does that do to distort the numbers? But I would like to know something about the cost of that. You said it is much more expensive. And who is making the decision on the cost/benefit analysis of that program versus another program where we might be able to gather better data? I am very interested in that. Mr. Kincannon. Well, I can talk a little bit more along those lines and would be very happy to followup with a special briefing with more details, both about what we did in Hurricane Floyd and about what we have done subsequently with Katrina victims. We have, for example, in our current surveys, we continued in the areas affected by Katrina and Rita and Wilma to keep up good response rates, so we know how to find people when they have moved and we have ways that are very successful in doing that. In the American Community Survey, we have added some instructions--we did not change the questionnaire--to make sure that people who were evacuated and staying in other people's homes do get identified and surveyed so that we can then tell from the questions already on the survey something about their condition and in some limited circumstances about where they were before. We plan to use those data to produce information on the characteristics of affected areas for the 8 months of 2005 before Katrina struck and for the 4 months after. This should be of some use both to Federal and local officials in assessing the condition. We worked with the Bureau of Labor Statistics to identify in a similar way in the Current Population Survey people who had been evacuated, and the Labor Department was then able to assess differential rates of unemployment for people who were settled differently. The CPS is limited in sample size, and so it does not give for that small a universe of people information below the national level. The American Community Survey will be able to give more information at smaller levels. Both of them are limited in that the Current Population Survey does not address people who live in group quarters. They are not handled in the labor force survey traditionally. The American Community Survey began covering group quarters only when funding was provided for fiscal year 2006, so it will not be able to tell us much about people who were in shelters and such like, but if they are living in trailers, FEMA trailers and so on, yes, we will get information about them. Ms. Foxx. Thank you. Mr. Turner. Mrs. Maloney. Mrs. Maloney. I want to thank the chairman and ranking member for calling this, and I would like to direct my questions to Mr. Kincannon. But I would first like to ask Ms. Brenda Farrell--and it is a followup of correspondence I have had with the GAO--to get back to me in writing. Does the GAO have the technical expertise to evaluate the Census Bureau's techniques for measuring the accuracy of the 2010 census? If not, when do you expect to have that expertise? This is a followup of letters that I have sent to GAO requesting this analysis. They say they are not capable of doing it. If you could get back to me on what exactly--what hurdles they need to go through so that they can have the technical expertise to evaluate the accuracy of the 2010 census. I would get it to you in writing, but I really want to get to Mr. Kincannon because we have been called to a meeting right now, a very important one. We have a foreign head of state who will be addressing a joint session of Congress. I would like a yes or no answer. Without SIPP, we will not be able to adequately study many policy issues such as the long-term effects of welfare reform or the effects of recent budget cuts and program changes. Is the Census Bureau concerned about helping Congress on both sides of the aisle evaluate public policy? Yes or no. Mr. Kincannon. Yes. Mrs. Maloney. OK. Then going back to the question that was raised by Ranking Member Clay about the decision to really stop the SIPP program because of budget constraints, there is no guarantee that in the future the budget constraints will not be worse. And many Members of Congress and really the scientific community--the research community is very concerned about the elimination of the SIPP program and recognizes that the Census will not be able to replace such a unique and important survey with one costing less. The SIPP took over 7 years to develop, as did the American Community Survey, as you pointed out, and Congress as well as private foundations, research institutes, have invested millions in understanding and processing the data. My question is: Is there any other place that researchers can get comparable information on program participation and income on a sub-annual basis? And I again would like this answer in writing because the research community is telling me and the scientific community is telling me that there is not comparable information. And I feel that this is very important. We need to know what is happening in the country. We need to know what is happening with our populations in certain areas, and the SIPP provided valuable information. I know you mentioned American Community Survey and a lot of other surveys when you responded to Mr. Clay, but the research community is telling my office and me personally that this will not give the same information. And so I would like it in writing, the answer to this question, because I think this is so serious that we should really look at it in depth. I would like to note that Ranking Member Clay and I, along with Members on both sides of the aisle, have sent a letter to the President--we are sending one to have this money reinstated to the budget, because we believe this research is very important. Would you like to elaborate? Is there one that gives you the exact comparable information on program participation and income on a sub-annual basis? Mr. Kincannon. The dimension that is missing in existing other surveys than SIPP--that is, the CPS and the ACS--is the longitudinal dimension. We have to find a way to craft that element using data from those surveys, but probably with follow-on surveys or independent surveys, and using administrative records from the programs affected. We cannot describe now in detail exactly how that will be done. We will work on that with the Federal agencies that have quite important needs, with the Congress, which has important needs and we understand that. We are the servant of the Congress in this regard. My ``yes'' was not an idle yes. And we will work with the research community. Whether we can replicate every topical nuance of the SIPP is another question, but we can find a way to substitute for the longitudinal element with a new longitudinal element, and that is quite important to do. I would also like to say---- Mrs. Maloney. Before we abolish it, I would like to see in place what it is you are going to put out there, because the longitudinal is very, very important to understand where we are, where we are going, where we have been in the past. And I just put that out there for the scientific community. I would like to followup that the Census Bureau has released at least two memos discussing why they are discontinuing the SIPP. In both cases, the memo states that the reason for eliminating the SIPP have to do with the lateness of the data and the problems with attrition and nonresponse. This memo does not acknowledge that the SIPP's nonresponse rate is the same as the Current Population Survey, yet there is no talk about getting rid of that, which is also conducted by the Census Bureau for the Bureau of Labor Statistics, and that compared to the two other national longitudinal surveys, attrition is lower than the panel study of income dynamics and about the same as the National Longitudinal Study of Youth. Furthermore, data for 2004 was released prior to data from the March Current Population Survey, so it appears Census has worked hard to get the data out quickly. If these reasons for eliminating the SIPP are invalid, is the reason SIPP is being cut purely due to budget constraints rather than research needs or substantive issues with the data? Mr. Kincannon. No, it is not solely for budget constraints. It does stem from longstanding concerns about the robustness of SIPP. The Census Bureau field staff is extremely capable. They get a higher response rate than almost anybody else working to collect data, and I am proud of that. But, still, the attrition in SIPP is a serious problem. Attrition is a characteristic problem of longitudinal surveys, but it has, in our view, become more serious here. Our choice in a constrained resource environment was to cut everything in a kind of an unmanagerial, mindless sense, reducing everything equally, or to apply priorities, as the Congress instructed us to do. That is why after 2 or 3 years of asking Congress for funds to cover 20 percent of the economic activity in this country between economic censuses we finally stopped asking for that. We have not stopped asking for money for longitudinal data on income dynamics. We expect to ask for a program of a substantial level to continue doing that, and we just have to accommodate the realities of weaknesses in SIPP and what we will have in resources. Certainly, in terms of priorities, we rank the censuses above others. The population and housing census and all the components, including the ACS, the economic censuses--these are fundamental. The economic censuses are the only time, twice a decade, when we measure almost all the economic activity in the country. The other years, we are making policy, the ES calculating GDP and so forth, missing 20 percent of the economic activity in the country, and it is the part in the service sector where jobs are being created much more than in other sectors of the economy. We have to make our best set of priorities within the constraints placed on us by the Congress, among others. Mrs. Maloney. But you stated that attrition and---- Mr. Turner. Mrs. Maloney, I hate to interrupt, but the House and Senate will convene shortly in a joint meeting to receive the Prime Minister of the Republic of Italy. Because House rules do not allow committees to meet during a joint meeting of Congress and out of respect for the Prime Minister, the subcommittee will be recessing, subject to the call of the Chair. If you are unable---- Mrs. Maloney. Mr. Chairman. Mr. Turner. I was just going to say, if you were unable---- Mrs. Maloney. Point of personal privilege? May I make a request? Mr. Turner. I was just going to suggest---- Mrs. Maloney. OK. Mr. Turner. I would just say if you are not able to--and we are going to let this panel go, if it is OK with you. Perhaps if you could take the next 2 minutes and ask your questions for the record to which they could respond. Mrs. Maloney. OK. Thank you very much, and I will get them in writing. Again, last year, we asked for specific information on the plans for measuring the accuracy of the 2010 census. After a long delay, I got a document that did not provide very much information. Please provide the committee, respectfully, with specific milestones and deadlines for decisions on how you will measure the accuracy, when the operational procedures for that measurement will take place, and when you will report to Congress on the accuracy of the 2010 census. Also, the Census Bureau is including a question on ancestry on the 2010 census. Consequently, the Census Bureau will be in a position to provide agencies like the Department of Homeland Security with counts on the number of Greek Americans, Indian Americans, Arab Americans, Irish Americans on a block-by-block basis. What is the Census Bureau's policy on providing this kind of block-level information to law enforcement agencies? Again, I thank you for your very difficult job. We rely on the statistics that you give us. I am particularly disturbed by the pattern of the gap between the haves and the have-nots in our country. It is growing in a way that I believe people on both sides of the aisle are tremendously concerned. It is not good for the wealthiest people in our country. It is not good for the poorest people. And that SIPP program was the document that really gave the information of what was exactly happening with this gap, and I think that is important for policymakers because we certainly want our country to prosper, all of our citizens to prosper, and I think it is important to track it. So I just want to underscore that I really do not want to see it eliminated unless you have in place something that really takes account for that data. And if you are going to eliminate it, I feel that there would be members on both sides of the aisle that would do a budget amendment that would restore specifically what was needed for that data, and if you could get us the specific costs, we could work on it. Thank you very much. Mr. Turner. Thank you, Mrs. Maloney. The subcommittee will now recess subject to the call of the Chair. The subcommittee will reconvene immediately after the joint meeting of Congress, and we will adjourn this panel and commence with panel two when we return. Thank you. [Recess.] Ms. Foxx [presiding]. Thank you all for coming back after the brief recess we had. It is the policy of this committee that all witnesses be sworn in before they testify, so I ask that the second panel of witnesses please rise and raise your right hands. [Witnesses sworn.] Ms. Foxx. Thank you. Let the record show that all witnesses responded in the affirmative. Dr. Rector, we will begin with you. STATEMENTS OF RALPH RECTOR, PH.D., SENIOR RESEARCH FELLOW AND PROJECT MANAGER, CENTER FOR DATA ANALYSIS, THE HERITAGE FOUNDATION; ANDREW REAMER, PH.D., DEPUTY DIRECTOR, URBAN MARKETS INITIATIVE, THE BROOKINGS INSTITUTION; AND MARGO ANDERSON, PH.D., PROFESSOR, HISTORY AND URBAN STUDIES, UNIVERSITY OF WISCONSIN-MILWAUKEE STATEMENT OF RALPH RECTOR Mr. Rector. Mr. Chairman and members of the subcommittee, thank you for inviting me to testify today. I am a senior research fellow and project manager in the Heritage Foundation's Center for Data Analysis. I participate in professional organizations that deal with Federal statistical issues. However, the testimony presented today reflects my own views, not necessarily those of the Heritage Foundation or any other organization. Research within Heritage's Data Center focuses primarily on policy debates at the national level. As a result, I will limit my remarks to examples showing why census data are useful to researchers analyzing Federal policies. However, I want to begin by discussing the overall importance of producing an accurate and complete decennial census and for continuing the American Community Survey. The constitutional Framers intended the decennial census to play a key role in ensuring the representative nature of the Federal Government. The Census Bureau relies on the MAF and TIGER programs to produce an accurate and complete census. As explained more fully in my written testimony, the need to coordinate MAF and TIGER programs raise serious issues. However, implementing the ACS can help in updating and verifying the systems that are used to collect decennial census data. These benefits will only be realized if the ACS is adequately funded, and they will also only be realized if the Census Bureau works closely with governmental entities and other groups at the State and local level, and this is the reason why I think the program LUCA is so important. In the remaining portion of my testimony, I summarize several reasons why census data are so useful to nongovernmental analysts who are studying national issues, and I illustrate these examples with research conducted at the Heritage Foundation. Nevertheless, I believe that these examples are typical of ways that many researchers, from a variety of political perspectives, use census data. To begin, census data help localize national issues to regions that are meaningful for decisionmakers and ordinary citizens. Traditionally, census long form data have been the primary--if not the only--source of information for demographic and socioeconomic information for regions that interest policymakers and the public such as congressional districts and Zip codes. Social Security critics, for example, highlighted the program's general low rate of return. CDA economists used congressional district data produced by the census in combination with data from other sources to estimate Social Security's rate of return for retirees in each State and in each congressional district. Census data are also important in evaluating the effectiveness of Federal grants. CBO has indicated that researchers should control for the independent effects when analyzing the outcome of Federal initiatives. For this reason, CDA analysts often use census data in their statistical evaluations of Federal programs. Census data have also been used to analyze proposals that would change Federal policies. For example, to examine the potential for a Social Security reform plan intended to produce wealth for low and moderate-wage earners, CDA analysts wanted to create a representative demographic profile, a set of those, from a data base that was large enough to permit a very detailed set of classifications. Fortunately, the first national-level American Community Survey, micro-file, was available, and using this micro-file, CDA analysts developed up-to-date profiles that would otherwise not have been available. Finally, the ongoing ACS will benefit smaller but more detailed special-purpose surveys administered by the Census Bureau, such as the current Population Survey, the Consumer Expenditure Survey, American Housing Survey, and the Survey of Income and Program Participation [SIPP]. Analysts and research organizations and universities routinely use these other census surveys to study issues such as welfare, education, and taxes. These smaller household surveys are adjusted to be consistent with data from the Census Bureau's population estimates program. And the ACS provides a valuable source of information for updating these population estimates. In conclusion, census data are the backbone of a constitutionally mandated apportionment process, and census data are also vital to information that analysts and policymakers at all levels uses. Thank you. [The prepared statement of Mr. Rector follows:] [GRAPHIC] [TIFF OMITTED] T7016.107 [GRAPHIC] [TIFF OMITTED] T7016.108 [GRAPHIC] [TIFF OMITTED] T7016.109 [GRAPHIC] [TIFF OMITTED] T7016.110 [GRAPHIC] [TIFF OMITTED] T7016.111 [GRAPHIC] [TIFF OMITTED] T7016.112 [GRAPHIC] [TIFF OMITTED] T7016.113 Ms. Foxx. Thank you. Dr. Reamer. STATEMENT OF ANDREW REAMER Mr. Reamer. Madam Chairman and members of the subcommittee, good morning. I am Andrew Reamer, deputy director of the Urban Markets Initiative at the Brookings Institution, and I appreciate the opportunity to appear before the subcommittee to discuss the elements necessary to the design and implementation of a successful 2010 census. UMI's mission is to stimulate greater public and private investment in urban communities through improving the availability of data for small areas, and in this regard, the single most important data set is the decennial census. The vitality of America's businesses and economy relies significantly on a successful census. As the title of this hearing suggests, the decennial census has an even more essential public purpose: enabling apportionment and redistricting. The decennial census is the platform on which we build our democracy. Seats in Congress, in State legislatures, and in city councils are allocated on the basis of the census, and the census is the fundamental mechanism for re-creating our democracy every decade. In my opinion, achieving a true and precise 2010 census depends upon four elements. First, we need a complete and accurate Master Address File. Simply put, we cannot count people if we do not know where they live. Second, we require minimal coverage error, reducing omissions and double-counting. Third, we need a fully, consistently funded American Community Survey. Taking the long form out of the decennial census will do much to improve coverage. And, fourth, we need to automate field data collection through the use of handheld computers. I will review each of these elements in some detail. Regarding the Master Address File, the completeness and accuracy of the MAF was affected by three issues in 2000: difficulty in capturing fast-growing areas; many group quarters had geocoding and categorization errors; and numerous housing units in small, multi-unit urban buildings were missed. The good news is that the Census Bureau has in place the elements to address these issues, and there are five important elements to recognize. The first is the Community Address Updating System [CAUS], which uses American Community Survey field staff on an ongoing basis to update addresses. Our understanding is that CAUS has been successful. Second, the Bureau has provided a thoughtful, detailed plan to address issues regarding the accuracy of group quarters enumeration, categorization, and geocoding. Third, Congress passed a law in the 1990's to enable the creation of LUCA, the Local Update of Census Addresses Program, and LUCA provides a framework within which local governments can give addresses to the Census Bureau and still improve accuracy of the MAF; however, experience indicates that local government participation in 2000 was not nearly what it might have been. There are several barriers to local participation, including a lack of staff resources, capacity, and training. The smaller the community, the greater the barriers. And it is clear that LUCA can be a much more effective program for 2010, and achieving this potential is going to require some active, cooperative relationship between the Census Bureau and local governments, and getting LUCA underway by late 2007 is really a tight window here for getting LUCA up and going--late enough to capture addresses and soon enough to incorporate them into the census. The fourth element for the MAF is the Update/Enumerate program to capture units in small buildings in urban areas, small multi-unit buildings. And, last, the Census Bureau should look at working with State governments as a resource for updating the MAF to use detailed administrative records the State governments have available to update address lists. In combination, these five elements can bring about a more accurate MAF, and I would suggest that this committee, for its own edification, ask the Bureau to report on its approach for preparing the MAF for 2010. With this full understanding, my hope is that Congress can provide the resources to make that happen. With regard to coverage improvement, in 2000 there were more duplicates and omissions than was optimal, and the Bureau has embarked on a series of efforts to correct these problems, and we at Brookings support these, including testing alternative short forms and approaches for flagging households that have coverage problems. The American Community Survey, the value of that is that it removes the long form from the decennial census and in doing so allows the Census Bureau to focus entirely on doing an accurate population count. So full funding of the ACS in and of itself will help a more accurate decennial census. Last is the realm of technologies. It is time to apply 21st century methods of data collection to the decennial census, and the use of handheld computers should lower the cost of data collection quite significantly. So, in conclusion, an accurate census is vital to our democracy, and I think these four elements will help make that happen. And I will be happy to answer any questions that you might have. [The prepared statement of Mr. Reamer follows:] [GRAPHIC] [TIFF OMITTED] T7016.114 [GRAPHIC] [TIFF OMITTED] T7016.115 [GRAPHIC] [TIFF OMITTED] T7016.116 [GRAPHIC] [TIFF OMITTED] T7016.117 [GRAPHIC] [TIFF OMITTED] T7016.118 [GRAPHIC] [TIFF OMITTED] T7016.119 [GRAPHIC] [TIFF OMITTED] T7016.120 [GRAPHIC] [TIFF OMITTED] T7016.121 [GRAPHIC] [TIFF OMITTED] T7016.122 [GRAPHIC] [TIFF OMITTED] T7016.123 [GRAPHIC] [TIFF OMITTED] T7016.124 [GRAPHIC] [TIFF OMITTED] T7016.125 [GRAPHIC] [TIFF OMITTED] T7016.126 [GRAPHIC] [TIFF OMITTED] T7016.127 [GRAPHIC] [TIFF OMITTED] T7016.128 [GRAPHIC] [TIFF OMITTED] T7016.129 [GRAPHIC] [TIFF OMITTED] T7016.130 [GRAPHIC] [TIFF OMITTED] T7016.131 Ms. Foxx. Thank you. Dr. Anderson. STATEMENT OF MARGO ANDERSON Ms. Anderson. Thank you for inviting me to testify. I agree with many of the comments that my colleagues have made, and so I will elaborate on some new ones. It is quite clear from the busy agenda that the Census Bureau never really stops taking a census; rather, its work is cyclical, and as they move from one plan to the next, they look back and forward. Relatedly, the world in which a census is taken also changes. Most notably the population grows, but often we are also quite surprised because not only does the population grow, it shifts in differential ways that means that the apportionment and redistricting mechanisms and allocation mechanisms lead to policy changes as a result of it. In this context, the Census Bureau faces something of a catch-22. Until the count is complete, the true dimensions of change are not clear. And yet the catch-22 the Bureau faces is that it must anticipate that change as it builds the plan, and that is the situation that we are in now. I think we need to keep that dilemma in mind as we do discuss the plans for 2010 and, in particular, keep the goal in mind of an accurate, efficient, and useful census. And as a result, I agree very much with the comments that my colleagues have made about the need to make sure that LUCA stays on track, that the Master Address File has development technology, and so forth. I want to make a few more comments about what I call risks and surprises that I see on the horizon and I think others do as well. The two big risks that are quite new and outside the realm of the Bureau: one is, of course, the budgetary environment, which we have heard much about today; and the second is that we may in 2010 be taking a census while the country is at war for the first time. In other words, the United States has never taken a census when the homeland was under threat, and we have little experience as a Nation anticipating if the war might affect the 2010 count. We need, as my colleagues have indicated here, to maintain funding in the development of ACS, and the risks to the 2010 count if in some sense anything goes wrong with the ACS off stage, if you will, are substantial, which is another reason to keep the funding moving. The long-term issue here is for the goal--even though we are talking very much about operational issues at this point in the planning process, is still accuracy and a fair count. And the historical record suggests that the Bureau does well when it has its planning processes under control and when stakeholders, be they Members of Congress, State and local government, advocates for particular demographic groups, feel that the process is a good one and under control. If they do not, those stakeholders have alternative mechanisms to affect the count, including filing lawsuits, changing through legislation the plan, or challenges to participation. So the building of trust in the operational plan is very, very important, and it needs to be integrated into what we are seeing right now. Some surprises that we already saw in 2000: The duplicate enumeration issue is a very new one for the census, and I would suggest that even now at the operational stages that we be looking at that one in particular and say, OK, how are we going to solve the problems of duplicates as we move toward 2010? Group quarters is another one, which you have already heard about. Are we going to change the short term, in particular, the measurement of race and ethnicity, because the long form is no longer there. There is discussion of moving the ancestry question to the short form. What impact will that have? Will there be pressure to add information about citizenship status or alien status of the population to the short form census? In general, we need to sort of keep our eyes on the prize and continue to focus on evaluation of accuracy and fairness. That is always an open question. We can always do better. And I expect that we will be talking about that a great deal in the next 4 years. Thank you, and I will take questions. [The prepared statement of Ms. Anderson follows:] [GRAPHIC] [TIFF OMITTED] T7016.132 [GRAPHIC] [TIFF OMITTED] T7016.133 [GRAPHIC] [TIFF OMITTED] T7016.134 [GRAPHIC] [TIFF OMITTED] T7016.135 [GRAPHIC] [TIFF OMITTED] T7016.136 [GRAPHIC] [TIFF OMITTED] T7016.137 [GRAPHIC] [TIFF OMITTED] T7016.138 [GRAPHIC] [TIFF OMITTED] T7016.139 [GRAPHIC] [TIFF OMITTED] T7016.140 [GRAPHIC] [TIFF OMITTED] T7016.141 Ms. Foxx. Thank you very much. Dr. Anderson, the first question is for you. What is your opinion of the Bureau's efforts for the decennial census thus far? And you mentioned the issue of trust in your testimony. Do you think the Bureau is building trust for its operational plans with more testing prior to this decennial as compared to the last? Ms. Anderson. I think that they are--you know, as we are sitting here today, they are involved in--I mean, this hearing itself is involved in that process. I would like to see, again, a bit more, mostly because I am very sensitive to the fact that even, as I say, issues off stage, as you heard in the first session about SIPP, have a way of oozing back, if you will, into discussions of the decennial. So I would like to see a more systematic approach to those issues and how it would move forward. I think the planning process for the 1990's at this point is not as good a guide for what we are doing in this decade simply because the planning process of the 1990's was fraught by changes in direction as the political makeup of Congress and the Presidency shifted over that decade. The planning started with a Republican President and a Democratic Congress, moved to a Democratic President and a Republican Congress, a Republican President and a Democratic Congress and so forth. Right now we have what looks like a stable planning environment of moving forward. If that continues, it adds grounding to the plan. Personally, I would like to see a little more discussion of the evaluation and adjustment issues, but that is, again, certainly a political decision that can be made by the administration and Congress together. Ms. Foxx. Thank you. This question is for all of you, but I will start with Dr. Reamer and I will work backward that way. If the Bureau plans to start LUCA in June 2007, does that give local governments adequate time to confirm, correct, and add addresses before the decennial? Mr. Reamer. I am not an expert in the process of the step- by-step process. My understanding from other people who are is that it certainly would be sufficient. But I think a lot has to happen between now and June 2007. What happened last time, my understanding is the National Academy hosted a panel on LUCA that the local governments were not adequately prepared for the LUCA process. A lot of them are small and resource-poor. So I think a lot of planning has to happen between now and mid/late 2007 so that the local governments are aware of what their responsibilities are, what their opportunities are, and how they might work with governments at a higher level, a town working with a county, a county working with a regional planning council, so that the LUCA process--the burdens can be shared locally and participation could rise as a result. So I would like to see a lot of planning between now and 2007. Ms. Foxx. Dr. Rector. Mr. Rector. Well, like Dr. Reamer, I do not have immediate experience with the LUCA program, but I have heard concerns raised. I think that the 2007 date, what I can tell, is sufficient, but I would want to emphasize that it is important for the Census Bureau to do an effective job of communicating. Some of the stories that I have heard indicate that it has not always been clear what information the Census Bureau is actually requesting and how, in fact, the local communities are supposed to supply it to the Census Bureau. And so I think the communication is as important as timing. Now, with regard to timing, I do think that the 2007 data is important for these local communities, particularly those that need additional resources to actually take advantage of the LUCA program. They have to build that into their budget. And so it is not just a planning of them assigning resources, but actually making sure that there are sufficient funds available to cover the expenses required. And so I think that as much lead time as possible is helpful, but from what I can tell, the 2007 date is sufficient. Ms. Foxx. Thank you. Dr. Anderson, do you want to add? Ms. Anderson. A similar kind of response, which is that the communication of the program and one thing that can be done right now is find out how aware State and local governments are that this is coming. It is understandable that they were ill prepared before 2000 because the law was only passed in, I think, 1994. So there should be a reservoir of experience that really needs to be built on, but it is, again, that kind of integration. This is a very hard task to do this national-to-local kind of communication and support. So anything you can do now to help it along, you know, will be wonderful as we move toward 2000. The places in the country that did take best advantage of LUCA were ones that knew it early and were sensitive to it. Ms. Foxx. Thank you. One more question. Do you all agree that the short form census will provide a more accurate census than a combined short form and long form decennial census? I will start with you, Dr. Rector. Mr. Rector. I think it certainly can, and given the plans that the Census Bureau has put in place, I think that it probably will, and the main reason for that is that they will be able to devote their resources on the short form. And so I think that, certainly given the programs that they have in place, should produce an accurate, complete census. Mr. Reamer. I agree and will add a couple things, I think. The ACS contributes in a number of ways to a more accurate decennial census. One is, as Dr. Rector says, that by taking the long form activity out of the short form process, you can have staff at the census focus on counting people and not worrying about other things. But, second, I think importantly is that the ACS itself has enabled the Census Bureau to put in the field a professional permanent staff so that for 2010 it can rely less on temporary workers. It will still have to rely on temporary workers, but it will have a professional staff in the field which will allow it to have a more accurate census. And then the third aspect is back to the MAF. One component of the American Community Survey is this CAUS program I mentioned, the Community Address Update program. So ACS staff throughout the decade are updating the MAF through this program, and that also will lead to a more accurate decennial census. So it is for those three reasons, I think, that we will have a more accurate census. Ms. Anderson. I think if the address listing and development work proceeds well, yes, you are going to do better. Again, as my historian says, there used to be only six questions on the census, and in some ways that is what we are going back to. I would also urge the Bureau to talk about that and to talk about what it is doing, because for most of the American population, of course, this is going to be a big surprise. I mean, they do not know about the ACS and, frankly, are not very interested. So, again, I also think that there is a programmatic and communications issue here that needs to be addressed. Ms. Foxx. Thank you. I am sorry. I do have one more question, and it is for Dr. Reamer. The Community Address Updating System is part of the ACS and, as you said, aims on an ongoing basis to use ACS staff to update address information. As of this date, 1,475 county TIGER maps have been updated to improve the Master Address File, and the contractor plans to deliver the remaining 1,758 county maps by 2008. Would you consider these two updating programs promising? And if you do, would you still recommend that Congress fund the Census 2000 Experimental Update/Enumerate program for this upcoming decennial? Mr. Reamer. Yes, I do think the two efforts you mentioned will be very helpful, and I think the Update/Enumerate program is separate from those two and is, therefore, important in its own right. It was found in the 2000 census that, particularly in urban areas, small multi-unit buildings where there are not city-style addresses, where there is a single mail drop-off, it was difficult to enumerate because Census was not quite sure how many units were actually in the building. And what Update/ Enumerate does is actually targets neighborhoods in which those types of buildings are prominent and then sends people in the field to actually go to the building and go inside and count the doors. So I think that there is a need over and above CAUS and the TIGER updating for Update/Enumerate, and the 2000 experience seemed to be quite positive in that there were--in the target areas they added 14 percent--the number of addresses went up by 14 percent because of people in the field. They also deleted 6 percent of addresses, but the result was there were more units to count. So I would support funding for that? Ms. Foxx. Thank you. Do any of you have any brief closing remarks that you would like to make? [No response.] Ms. Foxx. Well, before we adjourn, I want to thank our distinguished panels of witnesses for their participation today. I appreciate your willingness to share your knowledge, experiences, and thoughts with us. I would also like to thank my colleagues for their participation today. Clearly, there is a lot involved in planning the 2010 census, and I am pleased to see that the Bureau is making every effort to ensure that the decennial census is the most successful yet. However, we are not out of the woods yet. Clearly, obstacles remain, but I am confident that by working together we can ensure that the 2010 census is the best ever. Again, I want to thank our witnesses for their time today. In the event that there may be additional questions we did not have time for today, the record shall remain open for 2 weeks for submitting questions and answers. Thank you all. We stand adjourned. [Whereupon, at 12:18 p.m., the subcommittee was adjourned.] [Additional information submitted for the hearing record follows:] [GRAPHIC] [TIFF OMITTED] T7016.142 [GRAPHIC] [TIFF OMITTED] T7016.143 [GRAPHIC] [TIFF OMITTED] T7016.144 [GRAPHIC] [TIFF OMITTED] T7016.145 [GRAPHIC] [TIFF OMITTED] T7016.146 [GRAPHIC] [TIFF OMITTED] T7016.147 [GRAPHIC] [TIFF OMITTED] T7016.148 [GRAPHIC] [TIFF OMITTED] T7016.149 [GRAPHIC] [TIFF OMITTED] T7016.150 [GRAPHIC] [TIFF OMITTED] T7016.151 [GRAPHIC] [TIFF OMITTED] T7016.152 <all>