[Senate Hearing 110-834]
[From the U.S. Government Publishing Office]



                                                        S. Hrg. 110-834
 
               REDUCING THE UNDERCOUNT IN THE 2010 CENSUS

=======================================================================


                                HEARING

                               before the

                FEDERAL FINANCIAL MANAGEMENT, GOVERNMENT
                   INFORMATION, FEDERAL SERVICES, AND
                  INTERNATIONAL SECURITY SUBCOMMITTEE

                                 of the

                              COMMITTEE ON
                         HOMELAND SECURITY AND
                          GOVERNMENTAL AFFAIRS
                          UNITED STATES SENATE


                       ONE HUNDRED TENTH CONGRESS

                             SECOND SESSION

                               __________

                           SEPTEMBER 23, 2008

                               __________

       Available via http://www.gpoaccess.gov/congress/index.html

       Printed for the use of the Committee on Homeland Security
                        and Governmental Affairs



                  U.S. GOVERNMENT PRINTING OFFICE
45-579                    WASHINGTON : 2009
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing 
Office  Internet: bookstore.gov Phone: toll free (866) 512-1800 
Fax: (202) 512-2250  Mail: Stop IDCC, Washington, DC 20402-0001




        COMMITTEE ON HOMELAND SECURITY AND GOVERNMENTAL AFFAIRS

               JOSEPH I. LIEBERMAN, Connecticut, Chairman
CARL LEVIN, Michigan                 SUSAN M. COLLINS, Maine
DANIEL K. AKAKA, Hawaii              TED STEVENS, Alaska
THOMAS R. CARPER, Delaware           GEORGE V. VOINOVICH, Ohio
MARK L. PRYOR, Arkansas              NORM COLEMAN, Minnesota
MARY L. LANDRIEU, Louisiana          TOM COBURN, Oklahoma
BARACK OBAMA, Illinois               PETE V. DOMENICI, New Mexico
CLAIRE McCASKILL, Missouri           JOHN WARNER, Virginia
JON TESTER, Montana                  JOHN E. SUNUNU, New Hampshire

                  Michael L. Alexander, Staff Director
     Brandon L. Milhorn, Minority Staff Director and Chief Counsel
                  Trina Driessnack Tyrer, Chief Clerk


FEDERAL FINANCIAL MANAGEMENT, GOVERNMENT INFORMATION, FEDERAL SERVICES, 
                AND INTERNATIONAL SECURITY SUBCOMMITTEE

                  THOMAS R. CARPER, Delaware, Chairman
CARL LEVIN, Michigan                 TOM COBURN, Oklahoma
DANIEL K. AKAKA, Hawaii              TED STEVENS, Alaska
BARACK OBAMA, Illinois               GEORGE V. VOINOVICH, Ohio
CLAIRE McCASKILL, Missouri           PETE V. DOMENICI, New Mexico
JON TESTER, Montana                  JOHN E. SUNUNU, New Hampshire

                    John Kilvington, Staff Director
                  Katy French, Minority Staff Director
                       Monisha Smith, Chief Clerk


                            C O N T E N T S

                                 ------                                
Opening statement:
                                                                   Page
    Senator Carper...............................................     1

                               WITNESSES
                      Tuesday, September 23, 2008

Hon. Steven H. Murdock, Director, U.S. Census Bureau.............     4
Robert Goldenkoff, Director, Strategic Issues, U.S. Government 
  Accountability Office..........................................     6
Hon. Kenneth Prewitt, Ph.D., Former Census Director, Columbia 
  University.....................................................    16
Roderick Harrison, Ph.D., Senior Fellow, Joint Center for 
  Political and Economic Studies.................................    18
Karen K. Narasaki, President and Executive Director, Asian 
  American Justice Center........................................    19
Joseph J. Salvo, Director, Population Division, New York City 
  Department of City Planning....................................    21
Arturo Vargas, Executive Director, National Association of Latino 
  Elected and Appointed Officials (NALEO) Educational Fund.......    23

                     Alphabetical List of Witnesses

Goldenkoff, Robert:
    Testimony....................................................     6
    Prepared statement...........................................    42
Harrison, Roderick, Ph.D.:
    Testimony....................................................    18
    Prepared statement...........................................    68
Murdock, Hon. Steven H.:
    Testimony....................................................     4
    Prepared statement...........................................    31
Narasaki, Karen K.:
    Testimony....................................................    19
    Prepared statement with an attachment........................    75
Prewitt, Hon. Kenneth, Ph.D.:
    Testimony....................................................    16
    Prepared statement...........................................    62
Salvo, Joseph J.:
    Testimony....................................................    21
    Prepared statement with attachments..........................   140
Vargas, Arturo:
    Testimony....................................................    23
    Prepared statement...........................................   151

                                APPENDIX

Questions and Responses for the Record from:
    Mr. Murdock..................................................   160
Charts submitted by Senator Carper...............................   168


               REDUCING THE UNDERCOUNT IN THE 2010 CENSUS

                              ----------                              


                      TUESDAY, SEPTEMBER 23, 2008

                                 U.S. Senate,      
        Subcommittee on Federal Financial Management,      
               Government Information, Federal Service,    
                              and International Security,  
                          of the Committee on Homeland Security    
                                        and Governmental Affairs,  
                                                    Washington, DC.
    The Subcommittee met, pursuant to notice, at 10:34 a.m., in 
room SD-342, Dirksen Senate Office Building, Hon. Thomas R. 
Carper, Chairman of the Subcommittee, presiding.
    Present: Senator Carper.

              OPENING STATEMENT OF SENATOR CARPER

    Senator Carper. The Subcommittee will come to order. I have 
just joined all of you from another hearing that is going on, a 
Banking Committee hearing. I serve on the Banking Committee, 
and we have before us there the Chairman of the Federal 
Reserve, the Secretary of the Treasury, and the Chairman of the 
Securities and Exchange Commission. I just spoke with Senator 
Coburn, who is on the floor of the Senate, and he going to join 
us in about half an hour or so. I told him that if the census 
operation continues to make progress, if we continue to make 
progress in addressing some of the woes that we learned about 
over the last year, we may detail both of you guys to go over 
and help Chairmen Bernanke, Cox, and Paulson with their duties.
    In any event, on that note, let's go ahead and get started. 
I have a statement, and then if we are joined by others, we 
will offer them the opportunity to give a statement, too.
    Today's hearing is the fourth, as you may recall, in a 
series of oversight hearings looking at the Census Bureau's 
preparation for the 2010 census. One of our responsibilities in 
the Congress is to conduct oversight. When things are going 
well, I think it is a good idea to do oversight and acknowledge 
that. When things are not going well, we have an obligation to 
find out why and to make sure that we put a spotlight on 
whatever is going badly in the hopes that we will find out ways 
to make it go better. And I think with respect to the census, I 
think we have rounded a turn, and we are heading in the right 
direction now. But today's hearing will evaluate the Census 
Bureau's plans to ensure the accuracy of the 2010 census.
    On April 1, 2010, the Census Bureau will conduct its 23rd 
Decennial Census of our Nation's population. The decennial 
census is a constitutionally mandated activity that is designed 
to produce a baseline of information on the number of U.S. 
residents and their characteristics. Census results are used to 
apportion seats in the House of Representatives, where I once 
served. I represent a State, Delaware, where we only had one 
Representative. In fact, there is a total of seven States that 
have only one. And for us, the apportionment of seats in the 
U.S. House of Representatives is not a big deal. For other 
States, like California, where I think they have 53, and other 
States who have between 1 and 53, it can be a very big deal 
indeed.
    But our census results are used to apportion seats in the 
House to redraw Congressional and State legislative boundaries 
and to allocate billions of dollars in Federal assistance to 
State and local governments.
    Census data also provide information on population growth 
patterns and demographic information that are used by both the 
private sector and by Federal, State, and local officials.
    With such substantial reliance on census data, accuracy is 
critical. Unfortunately, every census in the Nation's history 
has failed to count all of our residents, resulting in an 
undercount of the general population.
    Looking back at the 2000 census, it was unprecedented in 
terms of its budget. More money was spent on it than any other 
previous census. As a result of the hard work of the Bureau, 
though, it was able to reduce both the number of Americans who 
went uncounted.
    Despite the Census Bureau's success, undercounting still 
remained an issue for many communities throughout this country. 
In 2000, the official census count was 281.4 million, and the 
adjusted estimate was just over 284 million. The Bureau 
reported a net error of just under a half percent. And that 
sounds pretty good. The truth, though, is that there were large 
errors in 2000 that I do not believe we can afford to repeat 
this time around.
    The Bureau's own Accuracy and Coverage Evaluation Survey 
revealed that 6.4 million people were missed and 3.1 million 
people were counted twice. In other words, the 2000 census 
produced a net undercount of some 3.3 million people.
    The undercount would be less problematic if it were evenly 
distributed among all Americans. However, studies show that 
undercounting tends to have a disproportionate impact on racial 
and ethnic minorities, on children, and on immigrants. In 2000, 
Asian Americans were missed nearly twice as often as whites, 
African Americans nearly three times as often as whites, and 
Hispanics were missed four times as often as whites.
    Although today's hearing is focused on the undercount, I do 
not think that we should minimize the overcount issues that 
exist. In past censuses, the Bureau reduced its net undercount 
by letting the people counted twice substitute for those who 
were missed. That may work for statistical purposes, but it is 
problematic for a number of reasons. The people counted twice 
are not like the people who were missed.
    I will say that again. The people who were counted twice 
are not like the people who were missed. They are not the same 
race. They do not have the same income. And they do not live in 
the same places. The importance of getting this thing right is 
striking when we think about the countless ways in which we 
depend on census data.
    For starters, the undercount affects the distribution of 
Federal funds that are allocated on the basis of population. A 
study performed by PricewaterhouseCoopers found that the 
undercount in 2000 would cost States $4.1 billion. This loss of 
Federal funding taxes the resources of States and local 
governments and compromises the level of services provided to 
residents--the people that all of us work for.
    Underenumeration in the census also has serious political 
implications. In political representation that is based on 
population, undercounted people get less credit for their 
population than they are due. This skews the make-up of the 
House and results in some communities being underrepresented, 
while others get more of a voice than they are due.
    Reaching out to those who were historically hard to count 
is even more important when we consider that for every 1 
percent of the population that does not respond to the census, 
we are going to have to spend about $75 million, I am told, to 
go door to door to get everyone counted, or just about everyone 
counted. As a result, it is vitally important that we do the 
necessary hard work now so that we can get an accurate, cost-
effective count in 2010 that will serve us well into the next 
decade.
    So as the Census Bureau begins its final preparation for 
2010, we need to make sure that you are reaching out as 
aggressively as we can to historically undercounted groups. 
This Subcommittee looks forward to hearing from our witnesses 
today and gaining your perspectives as we work together to 
ensure that this happens.
    We have two panels today. On our first panel of witnesses 
is Steve Murdock who serves as Director of the Census Bureau. 
Mr. Murdock officially became the Director of the Census Bureau 
on January 4, 2008. I do not know about you, but it seems a lot 
longer ago than that to me. It probably does to you, too. And 
you have the responsibility of overseeing the planning and 
implementation of the operation for the 2010 census. Prior to 
becoming Director, Mr. Murdock was a demographer for the State 
of Texas, and he played a leadership role in his State's 
coordination activities in the 1980, the 1990, and the 2000 
decennial censuses.
    Robert Goldenkoff is the Director of Strategic Issues at 
the Government Accountability Office, where he is responsible 
for reviewing the 2010 census and governmentwide human capital 
reforms. Mr. Goldenkoff has also performed research on issues 
involving transportation security, human trafficking, and 
Federal statistical programs. He received his Bachelor's of 
Arts in Political Science and Master's of Public Administration 
degrees from the George Washington University.
    Gentlemen, we are delighted that you are here. We will ask 
that you keep your comments to about 5 minutes, and if you run 
a bit over, that is OK. If you run a lot over, that is not OK. 
The full testimony of both of you will be entered into the 
record, and I would just invite you to proceed as you see fit. 
Thanks so much for joining us.
    Mr. Murdock, would you like to go first?

 TESTIMONY OF HON. STEVEN H. MURDOCK,\1\ DIRECTOR, U.S. CENSUS 
                             BUREAU

    Mr. Murdock. Mr. Chairman, Ranking Member Coburn, Members 
of the Subcommittee, thank you for the opportunity to come 
before you again to discuss our ongoing efforts to address 
issues related to the undercount in the 2010 census.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Murdock appears in the Appendix 
on page 31.
---------------------------------------------------------------------------
    Although in discussing our efforts to meet this goal, we 
rightfully stress our outreach and promotional efforts, 
including the advertising campaign and the Partnership Program, 
today I want to stress that the Census Bureau's commitment to 
improving coverage and addressing the undercount encompasses 
wide-ranging activities conducted before decennial field-based 
operations begin, during field operations, and after the 
conclusion of field operations.
    The census is based on addresses from which we identify 
households in which we count individuals. Our goal is to count 
everyone once, only once, and in the right place. This begins 
by taking the address list, referred to as the ``Master Address 
File,'' from the past census and updating it. For 2010, updates 
from the U.S. Postal Service have been obtained twice annually, 
and we improve that list by a process that allows local 
governments to review our address list for their areas in a 
process we call the Local Update of Census Addresses (LUCA).
    In 2010, for the first time, this review included both 
addresses for individual households and for group quarters, 
which are housing arrangements such as college dormitories, 
nursing homes, military barracks, jails, prisons, and other 
facilities.
    We also work with local communities to form Complete Count 
Committees, that operate during the census to ensure that 
residents of their communities respond to the census. Through 
our partnership and other programs described below, we 
establish with cooperating organizations Be Counted Centers 
that allow persons who believe that they have been missed to 
fill out and submit a census form and Questionnaire Assistance 
Centers that will help people to complete the census through 
assistance in multiple languages. The census questionnaire has 
also been designed with the goal of reducing undercount and 
overcount. Questions have been added to the 2010 form that can 
check for consistency in completing the census form and then 
improve our ability to identify both overcounts and 
undercounts. The questionnaire is available in five languages 
in addition to English, and for the first time, we will mail 
bilingual--Spanish-English--questionnaires to about 13 million 
households. Language guides are available for more than 50 
additional languages.
    Prior to beginning census operations, we make concerted 
efforts to hire local residents for census jobs because we know 
that people are more likely to respond to the census when local 
people are asking for their assistance and participation.
    Our first extensive field operation in the decennial census 
is the formal address listing process which begins in 2009. It 
involves census employees in a process of verifying all 
existing addresses on the Master Address File, adding new 
addresses identified, knocking on every door to determine, 
where possible, the number of households living at that 
address, and obtaining GPS coordinates for each household which 
expedites finding such households for future census operations.
    In March 2010, the massive mailing of forms to households 
begins, and for the first time, we will employ a second 
mailing, which has been found in many surveys to increase 
response rates. We will also employ strategies that differ by 
type of area so that in rural and other areas that do not have 
the city style addresses that are most convenient for mail 
delivery, we can obtain a good count as well.
    When the non-response follow-up process used to obtain data 
from households that did not respond to the census by mail 
begins, we complete numerous operations that result in making a 
minimum of six attempts to obtain data from households. During 
this period, we also conduct enumeration of service-based 
programs, such as shelters, soup kitchens, and mobile food 
vans. Regional offices develop and implement hard-to-enumerate 
programs specifically designed to reach specific hard-to-count 
groups, and coverage follow-up operations are conducted that 
involve telephone and other contacts with households where 
inconsistencies were found in their questionnaires.
    Finally, after the major decennial census operations are 
nearly complete, we will initiate and complete a census 
coverage measurement process through which we measure the 
extent to which we have over- or under-counted various 
population groups.
    Beginning in 2008 and running through the census operations 
described above is the 2010 communication program that 
integrates a mix of mass media advertising, targeted media 
outreach to specific populations, as well as national and local 
partnerships, grass-roots marketing, the Census in Schools 
program, and special events.
    Members of the Subcommittee, be assured that everything we 
do in the 2010 census is aimed at improving accuracy and 
coverage with an eye toward reducing undercounts and counting 
everyone. Whether the challenges are in remote Alaska, in 
densely populated urban areas like New York or Chicago, or in 
the colonias in South Texas, the Census Bureau will marshal the 
efforts necessary to include their residents in the 2010 
census. To us it does not matter how hard it is to reach 
everyone. It matters that we reach everyone. I am happy to take 
questions.
    Senator Carper. Say that last sentence again.
    Mr. Murdock. OK. To us it does not matter how hard it is to 
reach someone. It matters that we reach everyone.
    Senator Carper. That is pretty good. Who writes your stuff? 
[Laughter.]
    Did you write that yourself?
    Mr. Murdock. Not that one, no. The gentleman is here that 
wrote that, though.
    Senator Carper. Will that gentleman raise his hand? All 
right. You may have a second career. Thank you for that 
testimony.
    Mr. Goldenkoff.

TESTIMONY OF ROBERT GOLDENKOFF,\1\ DIRECTOR, STRATEGIC ISSUES, 
             U.S. GOVERNMENT ACCOUNTABILITY OFFICE

    Mr. Goldenkoff. That is a tough act to follow, but, Mr. 
Chairman, thank you for the opportunity to be here today to 
discuss reducing the undercount in the 2010 census. As you 
know, an accurate enumeration is a daunting task as the 
Nation's population is growing larger, more diverse, and 
increasingly reluctant to participate in the census.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Goldenkoff appears in the 
Appendix on page 42.
---------------------------------------------------------------------------
    An undercount occurs when the census misses an individual. 
An overcount occurs when an individual is counted in error.
    What makes these errors problematic, as you have already 
noted, is a differential impact on various sub-groups. 
Minorities, renters, and children, for example, are more likely 
to be missed by the census, while more affluent groups, such as 
people with vacation homes, are more likely to be double-
counted. As census data are used to apportion seats in Congress 
and a number of other important purposes, improving coverage 
and reducing the differential undercount is critical.
    As requested, my remarks today will focus, first, on key 
activities the Census Bureau plans to use to reduce the 
differential undercount in 2010; and, second, the various 
challenges and opportunities that might help or hinder these 
efforts. Importantly, in my remarks this morning I want to 
stress the following: Although the Bureau has developed a range 
of activities aimed at reducing the differential undercount, 
these activities are generally in the planning or early 
implementation stages, and a variety of uncertainties and 
challenges lie ahead.
    Reducing the undercount begins with a complete and accurate 
address list. The Bureau develops its address list over the 
course of the decade using a series of operations that include 
partnerships with the Postal Service as well as with State, 
local, and tribal governments. One such operations is address 
canvassing where thousands of temporary Bureau employees known 
as ``listers'' verify the addresses of all housing units by 
going door to door across the country. To help find hidden 
housing units, such as converted basements where hard-to-count 
groups might reside, the Bureau trains listers to ask if there 
is more than one residence at a particular address or to look 
for clues such as an outbuilding or two doorbells that could 
indicate additional living quarters.
    Another effort aimed at reducing the undercount is the 
Bureau's Integrated Communications Campaign that consists of 
paid advertising, earned media and public relations, Census in 
the Schools, which is a program aimed at reaching parents 
through their school-aged children, and partnerships with key 
national and grass-roots organizations that have strong ties to 
their communities.
    The Bureau also operates a range of special enumeration 
programs that target hard-to-count populations. They include 
the Be Counted Program, Questionnaire Assistance Centers, and 
Service-Based Enumeration, which aims at including the homeless 
and other individuals without conventional housing. Other 
activities, such as offering in-language questionnaires, can 
help improve coverage among people with limited English 
proficiency.
    While each of these activities can help the Bureau improve 
the differential undercount, they also face open questions 
that, if not resolved, could reduce the effectiveness of the 
Bureau's efforts. For example, with respect to address 
canvassing, the Bureau plans to provide listers with GPS-
equipped handheld computers to verify and correct addresses. 
However, the companies have experienced shortcomings such as 
freeze-ups and data transmission issues, and the reliability 
has been problematic. The Bureau plans to conduct a limited 
field test of the handhelds this December. However, if the 
performance issues persist, the Bureau will have little time to 
make any refinements as address canvassing is scheduled to 
start early in 2009.
    Another challenge is that several operations aimed at hard-
to-count groups, such as the Be Counted Program, Service-Based 
Enumeration, and Group Quarters Enumeration, were not tested 
during the dress rehearsal of the 2010 census, which was held 
earlier this year. Consequently, the Bureau missed an important 
opportunity to see how they might perform in concert with other 
activities planned for 2010 as well as identify the need for 
any improvements that might enhance their effectiveness.
    In summary, if the various activities aimed at the hard to 
count are implemented as planned, they will help the Census 
Bureau achieve its goal of improving coverage. At the same 
time, a number of uncertainties and challenges lie ahead, and 
the success of the Bureau's efforts to reduce the undercount 
will depend in large part on the extent to which the activities 
are adequately tested, start and finish on schedule, get 
implemented in the proper sequence, and receive appropriate 
staffing and funding.
    It will also be important for the Bureau to develop 
effective monitoring programs to ensure the various operations 
are on track and enable the Bureau to quickly respond to any 
contingencies that might arise. In the months ahead, it will be 
important for the Census Bureau and Congress to focus on these 
issues as well as be alert to newly emerging challenges. And, 
as always, we look forward to assisting the Subcommittee in 
this regard.
    Mr. Chairman, this completes my prepared statement, and I 
would be pleased to respond to any questions that you might 
have.
    Senator Carper. You used exactly 4 minutes and 58 seconds. 
That is pretty good. You could have slowed down there right at 
the end and just nailed it right at 5 minutes. [Laughter.]
    The first series of questions I am going to ask actually 
flow from the two charts that are to my right,\1\ to your left. 
But before I do that, I will just mention something. Another 
subcommittee I chair deals with clean air and nuclear safety. 
And I am forever encouraging the nuclear industry to focus on 
safety, to adopt at every one of our 104 nuclear power plants a 
culture of safety. One of our core values in our Senate office 
is ``If it is not perfect, make it better.''
---------------------------------------------------------------------------
    \1\ The charts referred to appears in the Appendix on page 168.
---------------------------------------------------------------------------
    These first three columns--this goes back to the census 
from 1940, 1950, 1960, 1970, 1980, 1990, and 2000, and the 
darker line, actually the blue line, blue column, reflects the 
undercount for black Americans, African Americans, and the 
total undercount is reflected here I guess in the gray. You see 
the total undercount from 1940, it looks like about 5 million 
people. Compared to 2000, the total undercount was essentially 
null and void, which is a great improvement. Unfortunately, 
there is still a significant undercount among African Americans 
of about--I think it is about 3 million people. So if it is not 
perfect, make it better.
    I would just ask, Mr. Goldenkoff, you just ran through a 
series of things, steps that the Census Bureau has taken to 
make it better; if not make it perfect, at least to make it 
better. Are you encouraged that we are going to see--for the 
most part the improvement, if you look at these numbers here 
for undercount of African Americans from 1940, 1950, 1960, 
bounced up a little bit in 1970, dropped way down in 1980, 
bounced up a little bit in 1990, and then dropped down again in 
2000.
    Are you encouraged that the Census will, if not make it 
perfect, make it better in 2010?
    Mr. Goldenkoff. We are certainly encouraged from what we 
have seen so far in terms of the level of effort that the 
Bureau has put forth. The good news is that they are applying a 
lot of the lessons learned from the 2000 census. A lot of what 
they are doing is data driven. They are taking information, 
demographic information, to target their resources, which is 
extremely important now that the budget is constrained. They 
also recognize the importance of working with local members of 
the community. So for all those reasons, we are certainly 
encouraged by what we are seeing.
    What we do not know is how well these various operations 
will work once they go live, and what we have seen, for 
example, in 2000 that a lot of the plans on paper, they look 
really good on paper, but things happen. And so one of the 
things that concerns us is that some of these activities were 
not tested during the dress rehearsal. And while it is true 
that the Bureau has performed a lot of these activities before, 
some of them encountered glitches in 2000. Also, the dress 
rehearsal provides an opportunity to make improvements as 
another data point there. Every census is different, so even if 
the Bureau has performed these operations before, they have to 
be--we do not know how they are going to perform in 2010 
because the environment is different. And there are a bunch of 
new operations that, again, while good--and we certainly 
commend the Bureau and give them a lot of credit for things 
like the bilingual questionnaires--the census has a lot of 
moving parts. It is a big, complex machine, and without testing 
these different operations, it is unclear whether they are 
going to work in concert with one another.
    Senator Carper. All right. Thank you.
    Dr. Murdock, why don't you just briefly respond to some of 
the comments that Mr. Goldenkoff has just made?
    Mr. Murdock. Well, we certainly appreciate the comments 
regarding the additional efforts that we are making because 
there are an extensive number of those. And as he has pointed 
out, the census is a very complex process. We have attempted in 
every way that we could to discern how well operations will 
work. We did have to curtail some things in terms of the dress 
rehearsal because of rescoping efforts, because of some funding 
issues last year. But we are making every effort possible to 
assess how these various activities will do, and we are using 
our databases from previous censuses to look at where we are 
going.
    One of the things we have from that is a very good idea in 
most of these operations from looking at the past how a change 
in those will likely affect the response we get because many of 
these are spin-offs of programs and ideas that we have been 
following for some time.
    So, yes, we did not in all cases have the full dress 
rehearsal kind of event that we would have liked, but these 
activities are ones we feel confident that we are going to be 
able to perform well within all of the constraints and all of 
the issues that are involved in a decennial census and all of 
the imponderables, the economic, social, and other 
circumstances that may exist at the exact time when the census 
occurs.
    Senator Carper. All right. Thank you.
    Dr. Murdock, can you tell us, if you will, the Bureau's 
specific plans for reducing both the net and differential 
undercounts now that you will not have the handheld computing 
devices to conduct non-response follow-up?
    Mr. Murdock. Well, in large part, we believe that the 
program that we have instituted will get us an adequate 
undercount; that is, we would, of course, like to have no 
undercount, but we do have a wide variety of things that we are 
doing that are new. And let me just for a minute mention those, 
because I think what tends to happen is to look at just a few 
operations. But let me just give you an idea of some of the 
things we are doing this time that we did not do last time.
    We have been using more frequent updates of USPS 
information to update our mailing lists. We are using a GPS 
during the address canvassing that will help us locate 
households and go back to households who do not participate. We 
have had a much more consolidated and much more activistic LUCA 
program. That is the Local Update of Census Addresses. And we 
have had very good participation in that. We estimate that 
about 85 percent of all the addresses in the country are 
involved in LUCA governmental entities that have agreed to 
cooperate and look at our address lists, see if they are right, 
make corrections, tell us where we are wrong, help us adjust 
those.
    We have an address canvassing process this census that 
involves knocking on every door. The last census we knocked on 
every third door in terms of verifying where we were at. We 
have a coordinated Communications and Partnership Program. Last 
time, we had a program--two different programs that sometimes 
meshed and sometimes did not mesh, meaning that things 
sometimes were not available on time or were available in large 
quantities after they were needed. The new questionnaire, with 
the questions that we have put on there specifically to help us 
identify undercount and overcount, to look for consistencies or 
inconsistencies in the responses. And very important, we 
believe, is the multiple language, not only the bilingual 
questionnaire that we make available, but we are making 
questionnaires readily available in five languages. And we are 
making questionnaire assistance guides available in over 50 
languages.
    So as we look at the diversity of the U.S. population and 
the challenge that creates for work in terms of counting the 
hard to count, we believe we have initiated a large number of 
activities this census that are better than last census that 
should help us address these needs.
    That is not to say that this is not a tremendous challenge, 
because the increasing diversity of the United States--I should 
say it this way: The diversity of the United States has 
increased substantially from 2000. And as a result, the 
challenges for us in terms of making sure we count all of the 
individual population groups is extensive.
    Senator Carper. That is a pretty impressive list that you 
just ran through for us.
    Let me stick with, if I can, the handheld computers for 
maybe one more minute. I am going to ask you to be brief in 
responding to this, and this will be for both of you. And, Mr. 
Goldenkoff, why don't you go first? But just how confident are 
you in the functionality of the handheld computers for the 
upcoming address canvassing operations?
    Mr. Goldenkoff. Well, I think it remains to be seen. The 
key event is what happens in the test in December. So far, as I 
said, they have had a bunch of technological problems that have 
reduced their reliability, reduced the productivity. So what 
happens in December, that I think will be key to what we can 
foresee for when the address canvassing actually starts in the 
spring. So we will be there. We are planning to observe the 
field tests.
    Senator Carper. All right. Good.
    Mr. Goldenkoff. But right now it is a big unknown.
    Senator Carper. All right. Thank you.
    Dr. Murdock, same question. How confident are you?
    Mr. Murdock. Well, as you know, when we began the process 
of rescoping early this year, we had found a number of 
difficulties. We have instituted a much more complete and 
comprehensive management program and a testing program so that 
we have, for example, during September and October of this 
year, we are doing some product integration testing. During 
October and November, we will be doing validation system 
testing. And we have scheduled for December an operational 
field test, and that will be a case where we will take the 
handhelds into the field to be used exactly as they will be 
used in the 2010 decennial census.
    When you look at census operations, one of the things that 
we recognize is that the census remains a high-risk activity. 
It always is, and there is no way to reduce all of that risk. 
Nothing is going to change this. But I believe the program of 
oversight that we have instituted, the program of testing that 
we have instituted, which is much more rigorous than what 
existed previously, we are confident that those sets of 
procedures put in place will lead us to a good result related 
to the 2010 census.
    Senator Carper. All right. OK. Thanks to both of you for 
your responses there.
    I am going to telegraph a pitch. I am going to ask you just 
before we conclude this portion of the hearing, so that the 
folks who are with you can be thinking about this, and maybe if 
you need some information, they can be pulling it together. But 
I am going to ask you right before we excuse this panel, Dr. 
Murdock, I am going to ask you a question in conjunction with 
the continuing resolution that we will adopt here in probably 
the next several weeks, and maybe even several days. To what 
extent are the needs of the Census Bureau reflected or likely 
to be reflected? What are your needs going to be? And so just 
be thinking about that as it pertains to the continuing 
resolution. Don't answer the question now, but just know that 
before I excuse you, I am going to ask you to give us some 
thoughts on that. And if you are not comfortable in answering 
it right now on the record, then we will just ask you to 
respond in writing.
    OK. In its final report to Congress, the U.S. Census 
Monitoring Board in 2000 made several recommendations that were 
offered in an effort to improve future censuses. And one of 
those recommendations was that gross error rather than net 
error should be used as the primary basis for evaluating the 
accuracy of the census.
    For example, the net undercount in the 2000 census, I think 
it was about 3.3 million, but the number of people missed is, I 
think, 6.4 million, while the number of people counted twice 
was around 3 million. At first glance, it may seem that the 
people counted twice cancel out those that were missed. And 
while that may work for statistical purposes, it just does not 
work out when we are trying to understand the characteristics 
of the population that we are counting.
    The people who were missed, as I said earlier in my opening 
statement, are unlike those who were counted twice. They live 
in different places. They may speak different languages. They 
come from different socioeconomic backgrounds. And I think both 
of you have mentioned it, the idea that folks who are fortunate 
to own vacation homes may get counted twice. Families that are 
fortunate enough to be able to send their children to school 
away from home, in some cases those students get counted twice. 
This can be problematic for State and local governments that 
rely on Federal funds that are allocated on the basis of 
population estimates.
    Has the Bureau decided to implement the board's 
recommendations and use gross error as its standard measurement 
of census error?
    Mr. Murdock. We have not really made a decision in terms of 
how that will specifically be reported, but historically, 
although it is the net that tends to get the attention, the 
Census Bureau's reports have indicated both, so that we have 
looked at both in our ongoing operations because we are very 
aware that there are groups that we have overcounted, and that 
has been reported, and there are groups that we have 
substantially undercounted. So we look at both of these as we 
look at our census and evaluate how well we have done.
    Senator Carper. Mr. Goldenkoff, do you have any comment on 
that?
    Mr. Goldenkoff. We would agree with the Census Monitoring 
Board. Since the early 1990s, we have been recommending that 
the Bureau calculate gross error as that is a better measure of 
the actual amount of error in the census. Just because you 
count somebody twice, that does not compensate for missing one 
person because of the demographic differences between the two 
groups.
    Senator Carper. I do not underestimate how difficult it is 
to count everybody. And I appreciate all the efforts that are 
ongoing, that are undertaken and ongoing to do a better job. 
But just in terms of basic equity, the folks that we are 
counting or overcounting are really people that are more 
privileged than those that we are undercounting. And that 
violates my sense of what is fair and equitable, and I am sure 
that it does that of almost everybody, maybe everybody in this 
room. So I would ask that we just keep focusing on that.
    Mr. Goldenkoff, GAO has done several studies on how 
inaccurate census data can affect Federal funding to States and 
to local governments. What was the estimated impact of the 
projected census 2000 undercount on the allocation of Federal 
funds to State and local governments? What was the estimated 
impact of the 2000 census on the allocation of Federal funds to 
State and local governments? That is the first question. The 
second one is: Which States were expected to receive the 
biggest dollar losses as a result of incorrect population 
estimates?
    Mr. Goldenkoff. Sure. What we did was we did not look at 
all grants. What we did was we looked at social services block 
grants because that grant program relies exclusively on 
population to allocate funds, and so it would be the most 
sensitive to any changes in population. And what we found, 
there would have been a total of--I think it was a $4.1 million 
shift. Some States, a group of States--and I forget the exact 
proportion--would have gained $4.1 or $4.2 million, and another 
set of States would have lost $4.2 million. So it was pretty 
much--it was a wash from that perspective, and I think the 
District of Columbia would have lost the most. And I have some 
information in my formal statement. We have a graph there that 
shows more of the details.\1\ So it would have had a modest 
impact.
---------------------------------------------------------------------------
    \1\ The chart referred to appears in the Appendix on page 60.
---------------------------------------------------------------------------
    Senator Carper. All right. Thank you.
    Let me sort of pivot here and go in a little bit different 
direction. Maybe I will ask two more questions, and then I am 
going to excuse this panel and bring on our next panel.
    I think we all agree that having a complete and accurate 
address list is the cornerstone of a successful census. Nearly 
half of the undercount arises from missed housing units and 
households. As a result, we need to know every address in the 
country--a daunting task. We need to get a questionnaire to 
each address. That is not easy either. We need to have them 
return it, and then we need to be able to capture that 
information provided accurately.
    Again, for Dr. Murdock, just to start here, how does the 
Bureau plan to develop a complete mailing list so that a mail 
survey can reach each household? You have spoken to this a 
little bit, but I just want to ask you to go back and pick it 
up again.
    Mr. Murdock. OK. Certainly. There are several activities 
involved in this. One is that we start with the address list 
from the last census, but twice a year since then, we have 
obtained from the U.S. Postal Service their delivery sequence 
file, which gives us an update of addresses. And in the last 
couple of years, as a result of a process called the Local 
Update of Census Addresses (LUCA)--we have shared our 
information on addresses with local communities--cities, towns, 
etc., and with States, who have reviewed those data and are 
telling us of addresses that they have that we do not have on 
our list, providing information on new subdivisions that may 
have come into place. So with a combination of our ongoing 
efforts and theirs, we are building this list.
    We also have an ongoing effort on census address update 
which involves all of our other field activities that are non-
decennial, and as we locate addresses that are not on our 
Master Address File, those are added as well.
    So we are using the best, if you will, palette that we can 
start out with, with the U.S. Postal Service data. And then we 
are adding to it what we think is the best of all in many ways, 
and that is locals' information about their own communities and 
their own cities about where addresses are and where the new 
developments have begun. And we will be including that, 
continuing that throughout this decade. And, in fact, we will, 
even in 2010, be getting some materials on new additions from 
local communities.
    So it is really a very good partnership, if you will, 
between us and other Federal agencies and local governments in 
assisting us, and their assistance has been instrumental to 
having a good address list. In addition, as we indicate when we 
go out to do address listing, which is this massive effort that 
we will begin next year, our people, our listing people, will 
find addresses. They will note those addresses that will go 
onto our list. And even during the non-response follow-up, 
which is the part that occurs after the 2010 response to the 
mail questionnaire has occurred, we will be identifying 
addresses and ensuring that households at those addresses get 
census forms to complete.
    So we have an ongoing set of field operations combined 
with, we think, a quite good base from the Postal Service 
combined with assistance from local jurisdictions.
    Senator Carper. Well, I am encouraged to hear about the 
cooperation of the local jurisdictions, the State and local 
governments. But when you think about it, they do have a dog in 
this fight, and they do have some skin in the game. They have 
an obligation to help, but they also have a special interest in 
helping. And I am glad to hear that they are meeting that 
obligation, or at least a bunch of them are.
    A question really for both of you, and then I am going to 
go back to that question I telegraphed earlier. But what 
additional measures, Dr. Murdock, is the Bureau taking to 
ensure hard-to-count populations in communities affected by a 
bunch of hurricanes--Katrina, Rita, Ike, and the list goes on--
are counted given that many residents are displaced from their 
homes in various States of rebuilding? And your comments--I am 
just going to ask you to comment on that.
    Mr. Murdock. Well, these incidences we really have to take 
on a case-by-case basis. But what we have found, for example, 
in Katrina-impacted areas is that we are going to use somewhat 
different procedures than we used there. We cannot simply rely 
on Postal Service addresses. So we will be using a combination 
of Update/Leave, and that is where we go in and find a housing 
unit, leave a questionnaire to ensure that they get it. And in 
other cases, in other places, we will use an Update and 
Enumerate, which means that we will actually locate the housing 
unit, come back and do an enumeration of that. So it will be a 
much more intense kind of effort and a recognition that 
standard addresses and standard address lists will not work 
necessarily in such areas. So we are giving them very special 
attention.
    For example, the Dallas office that covers both the 
Hurricane Katrina- and Hurricane Rita-impacted areas and now 
the Hurricane Ike-impacted area are working with local 
jurisdictions to discern how best to ensure that all persons 
are counted.
    Obviously, given the recency of Ike, we are developing 
those procedures right now for those areas, and we will do 
those over the next few months. But we recognize that these 
raise unique challenges, and it means that we are going to have 
to put more people in the areas doing door-to-door kinds of 
work than we would in an area that had not experienced such 
devastating circumstances.
    Senator Carper. Mr. Goldenkoff, you may have a comment on 
that. You may not. If you do----
    Mr. Goldenkoff. I think that what the Bureau is doing is 
very encouraging. They know what needs to be done. They have 
done it before. They recognize that the housing stock is not 
stable. And I think most importantly is that under the Bureau's 
protocols, they always err on the side of inclusion. So even if 
something does not look inhabitable, even if it just looks like 
a concrete slab, it will still be included in the address list 
because a mobile home or a house can be built on that slab come 
census day. So I think that we can be encouraged by the efforts 
the Bureau is putting forth.
    Senator Carper. All right. Thank you.
    As I said earlier, Dr. Murdock, as you know, we are working 
on a continuing resolution that is likely to take us some time 
into the first part of next year, maybe as far as into March. 
We have been told that the Census Bureau will be taken care of, 
but you may want to take this opportunity to tell me just what 
that means to be taken care of and what are your needs likely 
to be, an interim funding measure that will carry us, we will 
say, through March? What do you need to get through until then? 
Do you think you will get what you need in terms of 
recommendations from the Administration? I know you are working 
with the Commerce Secretary and with OMB.
    And, last--and you may want to answer this for the record, 
but what happens if you do not get what you believe you need?
    Mr. Murdock. Well, let me answer it, first of all, that we 
have received good support from the Administration in 
recognizing that the Census Bureau will need an anomaly in 
order to go forward, and that is because if we were to have to 
continue at our budget, which we have for this year, which is 
about a third of what we need for next year, it would have 
devastating effects on the 2010 census. We would, for example, 
not be able to open all of our local census offices, the first 
150 that are critical for the address canvassing operation. If 
it were delayed long enough, it could impact the very address 
canvassing process itself that begins early next year.
    So it is very critical that we not operate under a 
circumstance of continuing resolution.
    Senator Carper. All right. OK. I think that does it for 
this panel for now. Thank you for an encouraging report. If it 
is not perfect, make it better. Keep working hard.
    Mr. Murdock. We will.
    Senator Carper. And maybe someday we will have a hearing 
and say you got it perfect. I hope to still be around. I hope 
you will be, too. Thank you both.
    Mr. Goldenkoff. Thank you very much.
    Senator Carper. I will just invite the second panel to go 
ahead and take your seats, if you would, please, and I will 
begin introductions.
    I am going to start with the introduction of the Hon. 
Kenneth Prewitt. Mr. Prewitt served as the Director of the 
Census Bureau from 1998 to 2001. As Census Director, Dr. 
Prewitt managed decennial operations in the 2000 census. Dr. 
Prewitt is now the Vice President and Carnegie Professor of 
Public Affairs at Columbia University. Is that true?
    Mr. Prewitt. Of course.
    Senator Carper. Is it Dr. Prewitt?
    Mr. Prewitt. Yes.
    Senator Carper. Congratulations. And I understand you have 
served on many professional advisory committees, and you are 
currently most active on the Committee on National Statistics 
of the National Research Council. Welcome.
    Mr. Prewitt. Thank you.
    Senator Carper. Good to see you.
    Next, Roderick Harrison is a Senior Fellow at the Joint 
Center for Political and Economic Studies, a research and 
public policy institution that focuses exclusively on issues of 
particular concern to African Americans and other people of 
color. Mr. Harrison was the founding Director of DataBank, an 
online clearinghouse of data on African Americans and other 
ethnic populations. Previously, I am told, Mr. Harrison served 
as Chief of the Census Bureau's Racial Statistics Branch. When 
did you serve in that role?
    Mr. Harrison. From 1990 to 1997.
    Senator Carper. OK. And in that role you helped to expand 
the content and number of the Bureau's publications and 
releases on racial and ethnic populations. Mr. Harrison, we are 
glad you are here. Thank you.
    And Karen Narasaki, the President and Executive Director of 
the Asian American Justice Center, one of the Nation's leading 
voices advocating for the rights and interests of Asian 
Americans. I understand you serve in a number of leadership 
positions in the civil rights and immigrant rights community, 
and that you are Vice Chair of the Leadership Council on Civil 
Rights.
    Ms. Narasaki. Correct.
    Senator Carper. In addition, I am told that Ms. Narasaki is 
the Chair of the Rights Working Group, a coalition of human 
rights, civil rights, civil liberties, and immigrant rights 
groups, and working to address a variety of issues important to 
our immigrant community in this country. Welcome. We are happy 
you are able to join us.
    And Joseph Salvo, Director of the Population Division of 
New York City's Department of City Planning. The Population 
Division serves as the city's in-house demographic consultant 
and provides expertise in the development of population 
estimates, projections for infrastructure and capital planning. 
The division is also working closely, I am told, with the 
Census Bureau on the tactical preparation for the 2010 census 
and evaluation of the new American Community Survey. Mr. Salvo 
currently serves on the Census Advisory Committee of 
Professional Associations.
    And, finally--last, but not least--Mr. Arturo Vargas is the 
Executive Director of the National Association of Latino 
Elected and Appointed Officials, a national membership 
organization of Latino policymakers and supporters. Prior to 
joining NALEO, Arturo was Vice President for the Community 
Education in Public Policy at the Mexican American Legal 
Defense and Educational Fund, where he supervised and directed 
the organization's community education and leadership 
development programs. Mr. Vargas is a nationally recognized 
expert in Latino demographic trends, electoral participation, 
voting rights, the census, and redistricting.
    We are delighted that each of you made time to be here 
today. I just want to thank a number of you who have been 
working for years to make sure that we do, as best as we can, 
count everyone in this country, count them accurately. And so 
thank you for those past efforts and thank you for your 
willingness to be with us here today.
    Mr. Salvo, I am tempted to put you at the front of the line 
so you can be our ``opening Salvo.'' [Laughter.]
    But I am going to withhold from that temptation, and we are 
going to turn to Dr. Prewitt and ask him to lead us off. Mr. 
Prewitt.

  TESTIMONY OF HON. KENNETH PREWITT, PH.D.,\1\ FORMER CENSUS 
                 DIRECTOR, COLUMBIA UNIVERSITY

    Mr. Prewitt. Well, of course, what that means is that Mr. 
Salvo can correct me when it gets down to his turn.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Prewitt appears in the Appendix 
on page 62.
---------------------------------------------------------------------------
    Just a quick prefatory comment, if I might, Mr. Chairman, 
since you raised the issue of the CR in talking to the last 
panel. There are three dimensions to the CR that really matter. 
One, is it adequate? Second, is it timely,? Is it going to be--
because if it is adequate but late, it is the same consequences 
for the census. And I hope it has some flexibility. Certainly 
in 2000, if we did not have some budget flexibility, we could 
not have responded to unexpected circumstances which happened. 
So as the Congress deliberates about the CR, I would hope all 
three of those dimensions are attended to.
    Senator Carper. Good. That is good advice, and I did not 
say this but I am going to say it now. Your comments do not 
count against your 5 minutes that we have asked you to use, so 
we will start the clock over again. But I would ask everybody 
to try to stick to the 5 minutes if you could. That would be a 
big help.
    Thank you.
    Mr. Prewitt. OK. Thank you very much.
    The word ``accuracy'' has been around in census 
conversation since the beginning. In 1790, Thomas Jefferson 
gave the first count to George Washington, and the President 
was unhappy with that count because he said it was too low. 
This was not an idle concern on Washington's part. He felt like 
the European powers would think that we were weak, that our 
independence was fragile. So he actually instructed Jefferson 
to report higher numbers, and Jefferson did some complicated 
arithmetic and actually sent out the diplomatic corps, and he 
actually adjusted the census in 1790 upward to try to 
compensate for the undercount.
    Senator Carper. So that is how we ended up with 72 million 
people that year? [Laughter.]
    Mr. Prewitt. And then as late as the most recent census, 
the 2000 Census, it was Commerce Secretary Evans who described 
it as ``the most accurate census this Nation has ever 
conducted.''
    Well, based on your opening remarks, and I think my own 
judgment of the Census 2000, I think that is--I appreciate what 
Commerce Secretary Evans said, but I think it is an inaccurate 
statement itself about accuracy because of your focus on the 
gross error. Anytime you have a census which overcounts and 
undercounts in the magnitude that we do it, it is not fair to 
call it ``the most accurate census ever conducted.''
    I want to try to focus on--you have talked about the 
undercount. You understand it. But I think it would be very 
helpful if the full Congress understood that you can have a 
very good census down to the last 1 percent, and that is when 
the undercount kicks in. But you can also have a very good 
census all the way from top to bottom, but if the address file 
leaves out those households which are disproportionately 
undercounted, you can go in at the top level and get an 
undercount or you can wait until the last 1 percent and get an 
undercount. So the problem with the census is you have got to 
get each one of those steps, from the address file all the way 
down to counting that last 1 percent, if you want to try to do 
something about the differential undercount.
    Now, of course, in that last 1 percent, you can also get 
your overcount because that is when you are counting, double-
counting the college kids and so forth and so on.
    So the important thing, it seems to me, is to understand 
the distributional accuracy and numerical accuracy are two 
dimensions of the accuracy conversation. And numerical accuracy 
is how close do we get to the true count, but distribution 
accuracy is the proportionality, of course. And anytime things 
are being allocated on a share basis, as congressional seats, 
of course, and as Federal funding, then distributional accuracy 
is the name of the game. Because in some fundamental sense, you 
could undercount; but if you undercount evenly across all 
geographic areas and across all demographic groups, then you 
would not be as unfair as you are in trying to get a better 
census, a more numerically accurate census, but one which 
builds in distributional inaccuracies. And it is extremely 
difficult for the Census Bureau to manage the tension between 
trying to reach everybody and yet worrying about getting 
distributional inaccuracies in that. So I just want to focus 
the Subcommittee's attention, if I can, on constantly as you do 
your oversight, your question has to be about distributional 
accuracy, not just accuracy.
    I would like to conclude by saying that the Census Bureau 
takes much pride in finding its mistakes as it takes in not 
trying to make the mistake in the first place. It is a 
scientific organization. It is in its DNA to be self-
correcting, self-improving. And the most important report card 
that it produces is its Coverage Evaluation report card. It 
wants to know how it did. I certainly strongly concur with your 
judgment about the gross error. I do think we ought to be 
basing this on gross error, not net error. But I would just 
again urge the Subcommittee to pay attention to how well the 
Coverage Evaluation Survey is funded, is designed, is timely, 
and so forth. It is the only report card. There is nobody else 
to tell the Census how poorly it did other than the Census 
Bureau itself and, therefore, to tell the American people. So 
focus on distributional accuracy, and the net error matters but 
so does the gross error, and worry about the Coverage 
Evaluation as a measure.
    Finally, I would like to say that we actually did do well 
in 2000 on many dimensions. Oddly, we gave back almost half a 
billion dollars to the Federal Government. We came in under 
budget.
    Senator Carper. Did you ever get a thank-you note for that?
    Mr. Prewitt. Well, can I take a few more seconds? Or I will 
not stop on time.
    Senator Carper. All right.
    Mr. Prewitt. What was really interesting is no one wanted 
to admit it because no one wanted to `fess up that somebody had 
not sort of gotten this down to the right last penny and so 
forth. It is what gave us the flexibility. I would much rather 
have given money back and had flexibility to cover some of the 
problems we ran into. But that is obviously not the way the 
government normally thinks.
    Thank you, sir.
    Senator Carper. Thank you so much. Mr. Harrison.

TESTIMONY OF RODERICK HARRISON, PH.D.,\1\ SENIOR FELLOW, JOINT 
           CENTER FOR POLITICAL AND ECONOMIC STUDIES

    Mr. Harrison. Yes, I am going to begin by quoting Dr. 
Prewitt's comment before the 2000 census. The 2000 census was 
the first that the Bureau planned to actually use all this 
technical apparatus that it has for estimating the undercount 
after the census is conducted, to do a one-count census where 
it would adjust--it would use the surveys that it uses to 
estimate the undercount to adjust the numbers to be more 
statistically accurate. And the Supreme Court ruled in 1999 
that this would not be constitutional for purposes of meeting 
constitutional requirements for apportioning the Congress. And 
the Bureau had to change, within a year, years of planning for 
the one-count census.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Harrison appears in the Appendix 
on page 68.
---------------------------------------------------------------------------
    Dr. Prewitt said at the time that using traditional 
counting methods, the Bureau must run harder to stay in place. 
It will run harder. It hopes to stay in place. I think you did 
much better than staying in place. And, actually, I think 
during the questioning one thing that you might ask is about 
the very specific steps--again, I think the testimony earlier 
today that this is a very complex sequence of events that build 
upon each other, of operations that build upon each other, and 
it requires success at any stage. If you start failing with the 
address list, with the canvassing, the probability of 
undercounts growing will accumulate and your likelihood of 
overcoming them in later stages.
    So I think you want to pay particular attention to the test 
in December, to the results of the handhelds, and, if the test 
suggests that everything is not resolved, to alternatives for 
making sure that the address list is as complete as possible.
    That said, a few points that have come up. Yes, the 
undercount/overcount, the net undercount is a poor measure. The 
thing that is important to note is that the undercount and 
overcounts for the small geographies for which they are 
calculated in the estimates of the undercount, or for 
geographies small enough that it does not affect apportionment, 
so what it does affect is all the other uses of the census. So 
if you are switching off individuals, you still are counting 
the same number of people, roughly, for purposes of 
apportionment and redistricting. It is when you get to the 
characteristics of those people, what their needs are, the 
planning that is built upon that, that this matters.
    The block grants mentioned, the difference is about a 
quarter of a percentage point in funds, and the amounts 
involved--this is heretical to say, but I think we must realize 
that to get greater accuracy to correct those errors might not 
be possible and might not be cost-effective. You are talking 
about something like $67,000 in the district's budget. You 
could allocate this by other means much more effectively than 
getting a more accurate count.
    So I do think, however--I think we should not overestimate 
the degree to which--we should not underestimate the 
consequences, but we should not overestimate it either. Some of 
these issues would be more directly and effectively dealt with 
as social policy issues than as census count issues.
    That being said, clearly one thing that contributes to the 
undercount, and these hard-to-reach populations particularly, 
is distrust of the government, feeling that the government is 
not necessarily fair. And the census, it is absolutely 
essential, I think, that people participate in the census and 
leave it feeling that the count has been fair, that they have 
counted equally with everyone else. And that is something that 
I do not think you can put a price tag on. I think the entire 
integrity of the Nation, its commitments from the very 
beginning of the census, does depend on that, and that is 
priceless.
    Senator Carper. Thank you. You were exactly 5 minutes.
    Ms. Narasaki, I do not know how you top that, but good 
luck.

  TESTIMONY OF KAREN K. NARASAKI,\1\ PRESIDENT AND EXECUTIVE 
            DIRECTOR, ASIAN AMERICAN JUSTICE CENTER

    Ms. Narasaki. Thank you for inviting the Asian American 
Justice Center to testify on an issue that is one of the top 
priorities for the civil rights community over the next 2 
years. And I have longer testimony that I am asking for it to 
be submitted for the record.
---------------------------------------------------------------------------
    \1\ The prepared statement of Ms. Narasaki with an attachment 
appears in the Appendix on page 75.
---------------------------------------------------------------------------
    Senator Carper. Your entire testimony will be part of the 
record.
    Ms. Narasaki. In 2000, we helped lead a census outreach 
campaign with our partners--the Asian Law Caucus the Asian 
Pacific American Legal Center, and Asian American Institute. 
And over the next 2 years, we will be working with the 
Leadership Conference on Civil Rights Education Fund, NALEO, 
the NASP, and the National Congress of American Indians to 
really help reduce that differential undercount.
    Many civil rights laws, as you know, rely on the census 
data for their enforcement and implementation, and one of the 
keys for the decennial is the enforcement of the Voting Rights 
Act. This is what makes it even more critical, as you point 
out, that the right people are counted in the right places.
    For these reasons, we think it is particularly important to 
make sure that there is going to be adequate resources and 
strategies to try to reach the growing minority and other 
historically undercounted populations. I am going to focus on 
language barriers, fears, and concerns about privacy. In 
addition, as we saw, as you noted in Hurricane Katrina and now 
Hurricane Ike, many immigrants and minorities live in non-
traditional households and are more likely to be displaced. So 
we are very concerned about what the Bureau will be doing to 
address that.
    In 2006, almost 55 million persons spoke a language other 
than English at home. Almost half have a difficulty speaking 
English well, and this lack is one of the biggest barriers for 
immigrant households.
    In addition, many immigrants have fled countries with 
corrupt or oppressive governments. Moreover, the hostile 
rhetoric of some elected officials in the heated immigration 
debate we hear and the institution of some hostile policies and 
widely publicized raids on the local level of homes as well as 
of businesses is going to discourage immigrant participation.
    Finally, concern about privacy overall, confidentiality, 
and misuse of data is held by many Americans, but particularly, 
as Mr. Harrison notes, minority communities. This concern has 
increased because of the controversial post-September 11, 2001 
policies and practices, and the Bureau's 2004 provision to the 
Department of Homeland Security of Arab American data at the 
zip code level reinforced this concern.
    The communications plan, the partnership plan, and the 
language assistance plan, therefore, are really critical to 
overcoming these challenges and reducing the differential 
undercount. Also, these policies, though, have to be put in 
place early to enable the Bureau to hire people locally to the 
communities and with the needed language skills.
    The Bureau has wisely sought to integrate the partnership, 
outreach, and paid marketing campaign in one communications 
strategy. However, the plan is not yet final, and it is not 
clear to us yet whether there will be sufficient resources 
targeting each minority and hard-to-count community.
    In 2000, the Partnership Program is credited for having 
reduced the differential undercount. It has empowered locally 
known and trusted messengers to be able to speak knowledgeably 
and persuasively. And because of this lack of trust, it is very 
important to have locally known people be able to tell their 
communities why it is important and safe to participate.
    The partners also helped the Bureau to hire a diverse 
talent pool with the needed skill and local knowledge of their 
communities. And we believe to maximize the effectiveness of 
this program the Bureau needs to mandate that regional offices 
share and consistently use best practices, and certainly more 
resources are needed for this program.
    Chairman Carper, we very much appreciate your commitment to 
ensuring that we achieve an accurate count, and we are counting 
on your leadership in ensuring that there will be sufficient 
funding for the census, particularly in this critical year that 
is coming up.
    We also commend the Bureau's work to date to ensure that 
Spanish speakers are adequately assisted, but we are concerned 
about the other language minorities that need to be counted, 
particularly the smaller Asian American ethnic communities who 
have the highest levels of linguistically isolated households.
    Finally, we think the Bureau needs to begin to work with 
the necessary agencies to set citizenship and retiree exemption 
hiring policies in place as soon as possible. Without the 
citizenship exemption, the Bureau may not have a sufficient 
pool of bilingual community partner specialists and enumerators 
who can overcome the language barriers and the fears of census 
in these immigrant communities.
    Finally, while this hearing is focused on the preparation 
of 2010, I would be remiss if I failed to urge the Subcommittee 
to review the implementation of the American Community Survey, 
which is, as you know, replacing this year the traditional 
decennial long form. We have concerns about the accuracy of 
data for small and particularly migrant populations.
    Thank you.
    Senator Carper. Thank you very much, Ms. Narasaki. Mr. 
Salvo.

TESTIMONY OF JOSEPH J. SALVO,\1\ DIRECTOR, POPULATION DIVISION, 
           NEW YORK CITY DEPARTMENT OF CITY PLANNING

    Mr. Salvo. Thank you, Chairman Carper. It is a pleasure to 
be here. On behalf of Mayor Michael Bloomberg, I want to thank 
you for the opportunity to speak about issues affecting 
undercount in the 2010 census.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Salvo with attachments appears in 
the Appendix on page 140.
---------------------------------------------------------------------------
    Regardless of your political leanings or assessments about 
past performance of the Census Bureau, I think we can all agree 
that the Bureau is struggling right now with the daunting task 
of trying to engineer a successful 2010 census within a very 
tight timeline. At this point, those of us who are on the 
outside looking in are very concerned. We are concerned because 
we would not like to see the census reduced to making a census 
happen as opposed to creating a high-quality enumeration, of 
course, that counts all Americans.
    We have heard the Census Bureau's pledge that the 2010 
census will fully enumerate the population of the Nation. I am 
here today to ask the Subcommittee to hold the Census Bureau to 
its pledge in two areas of concern to those of us who are avid 
users of the data.
    The first relates to the fact that the Census Bureau needs 
to provide us with their plan regarding how it intends to mail 
questionnaires to millions of housing units that lack apartment 
information--information that links a questionnaire to 
occupants of a specific housing unit. Remember, when 
questionnaires are delivered, there are no names on those 
questionnaires. This goes to the heart of what you just heard 
about how the address list is the foundation for the census. 
From a local community standpoint, which is where I stand, 
March 2009 is as important as April 2010.
    This problem affects many communities in the Nation because 
there are problems with addresses in many communities--
addresses that, frankly, may not cut it for census delivery 
purposes. So I would like to refer you to a series of photos 
that I have in the back of my written testimony. Chairman 
Carper, do you have my----
    Senator Carper. I do.
    Mr. Salvo. At the end my written testimony, there are four 
pictures of two buildings that contain housing units without 
apartment numbers.\1\
---------------------------------------------------------------------------
    \1\ The pictures referred to appears in the Appendix on page 147.
---------------------------------------------------------------------------
    Senator Carper. Yes, got them.
    Mr. Salvo. OK. The first two pictures show a building with 
three apartments, each with its own separate entrance. Since 
there are no apartment numbers, the Census Bureau needs to have 
a procedure to create labels when the building is examined in 
the 2009 address canvass so that each questionnaire can be 
attached to each individual apartment. These need to be labels 
that the Postal Service can use to deliver questionnaires, for 
example, Apartment 1R, for one right; Apartment 2, upper, 
lower, basement.
    In the second example, we see two pictures of a larger 
building with five doorbells and one mailbox, which is common 
in places where people sort or retrieve their own mail by name 
of occupant. Putting all five census questionnaires into a 
single mailbox with apartment designators can cause confusion 
when trying to attach a questionnaire to each apartment, 
especially when non-response follow-up is required. The Census 
Bureau can avoid this problem by issuing specific instructions 
to field workers in the address canvass operation on how to 
label each apartment using descriptors.
    To its credit, the Census Bureau has undertaken extensive 
research in the post-2000 period to explore the best ways of 
dealing with apartments like these that are not labeled in 
preparation for the address canvass. Indeed, I have had 
conversations with the New York Regional Director Lester 
Farthing, regular discussions about the frustration that this 
can produce in address canvassing.
    A resolution is at hand. The Census Bureau has done 
research, but they have yet to formally adopt a procedure to 
handle this problem. Right now the danger is that with all the 
pressure on the Bureau to keep its time frame and with the 
block canvass of the Nation fast approaching, this innovative 
work may fall by the wayside; therefore, having a serious 
ripple effect on the enumeration itself.
    A recent GAO report points out that the cost of the address 
canvass can dramatically increase if the Census Bureau cannot 
accurately anticipate the number of housing units per hour that 
can be examined. The problems that exist in small, multi-unit 
buildings can complicate the address canvass and greatly 
increase costs in many areas if the Census Bureau fails to 
implement a strategy that deals with these problematic 
addresses.
    Therefore, I would like to request that the Subcommittee 
ask the Census Bureau if they plan on implementing a procedure 
in the block canvass to label problematic housing units in 
these buildings and, if so, when the details of this plan will 
be released.
    Can I run over a little?
    Senator Carper. I am going to ask you to go ahead and wrap 
it up, please.
    Mr. Salvo. OK. The other point I wanted to make is that the 
Census Bureau needs to make good on its promise to form 
meaningful partnerships with local governments. It is hard to 
overstate the importance of proactive community involvement in 
the census. The only way to produce what would be considered a 
good enumeration, as we actually have heard, is that local 
residents need to produce messages for the people in their own 
neighborhoods. And the Census Bureau needs to reach out to the 
locals in a way that is, frankly, going to be more difficult 
than in the past because of all the issues involved in privacy 
and everything that has happened since 2000.
    So as a second point, I would like to ask the Subcommittee 
to require that the Census Bureau provide in detailed terms 
their plans for the outreach and to consult with organizations 
that deal with the local communities so that they can report to 
the Subcommittee exactly what those plans are and how they are 
manifesting themselves at a local level.
    Thank you very much.
    Senator Carper. Those are two reasonable requests. I am 
going to ask at the end of the hearing if you will spend a 
little time with the folks on our Subcommittee who work with me 
on the Census Bureau and just discuss those requests further.
    Mr. Salvo. A pleasure. Thank you very much.
    Senator Carper. We will see what we can do. Thank you for 
bringing them to our attention, and thanks for the pictures, 
too. Pictures are worth a thousand words, and these are good 
ones. They help make your point well.
    Mr. Vargas, you are our last witness today. Thank you. 
Please proceed.

  TESTIMONY OF ARTURO VARGAS,\1\ EXECUTIVE DIRECTOR, NATIONAL 
 ASSOCIATION OF LATINO ELECTED AND APPOINTED OFFICIALS (NALEO) 
                        EDUCATIONAL FUND

    Mr. Vargas. Thank you, Mr. Chairman. The NALEO Educational 
Fund is a nonprofit, nonpartisan organization that facilitates 
the full Latino participation in the American political 
process, and we are the leading census policy development and 
public education organization in the Latino community. Since 
2000, we have been working with the Census Bureau as a member 
of the Census Advisory Committee on planning for Census 2010, 
as well as the American Community Survey. Next year, we will 
transition our award-winning campaign ``Ya es hora''--which 
means ``It is time''--to educate Latinos about the importance 
of being counted in the 2010 census. We will devote the full 
weight of our capacity and influence to help make the 2010 
census a success because we know that the Latino population, 
the Nation's second largest population and fastest growing, is 
at greater risk than ever before of being undercounted in the 
next census. An undercount of the Latino community will mean a 
failed census, and to avoid such a disaster, we offer the 
following recommendations:
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Vargas appears in the Appendix on 
page 151.
---------------------------------------------------------------------------
    First, Congress must fully fund the Census Bureau. The 
groundwork done in the final year before the census will 
determine its success. Any delay in these activities, as we 
have heard this morning, will undermine the count. The Bureau 
faces a daunting challenge with respect to its decision to 
abandon the use of the handhelds. The Bureau must now undertake 
intensive and more costly preparations to switch back to the 
traditional enumeration method, such as hiring more enumerators 
than originally planned. We urge Congress to fund the 
administration's revised fiscal year 2009 request of 
approximately $3.1 billion. We are concerned that if full 
funding is not provided, other critical programs will be 
shortchanged, such as partnership programs or the paid media 
outreach.
    Two, the Census Bureau must implement a communications and 
outreach plan that uses culturally appropriate outreach 
materials and takes into account the special challenges in 
reaching Latino sub-groups. Earlier this month, the NALEO 
Educational Fund helped the Census Bureau convene several other 
Latino national organizations to meet with the communications 
vendor, GlobalHue Latino, to help coordinate these efforts. 
These firms need to continue to meet with us to create a wide 
range of promotional materials in language appropriate for 
Latino audiences, and utilize print, broadcast, and digital 
media in their efforts.
    Three, special strategies and preparations will be required 
to enumerate the Nation's immigrant population. Our Nation's 
current debate about immigration as well as actions by State 
and local governments have created a climate which has 
exacerbated immigrants' distrust of Government, including the 
Census Bureau. It will be formidable challenge to convince 
immigrants, legal permanent residents, and U.S. citizens who 
live in households where family members have varying 
immigration status to answer the census and that the census is, 
in fact, safe and confidential.
    Four, the Census Bureau must ensure that its Census 2010 
workforce reflects the diversity of the Nation's population, 
from its highest managerial position to its field enumerators. 
The Bureau must strengthen its existing efforts to increase the 
number of Latinos at the Census Bureau where they are currently 
the most underrepresented population group among the workforce. 
We also urge the Bureau to implement a waiver in its hiring 
practices to allow the hiring of work-authorized non-citizens 
to ensure that there are sufficient enumerators with non-
English-language speaking abilities. This will be especially 
critical in areas that have experienced population increases of 
the non-English-speaking population.
    Five, the Census Bureau must quickly adjust its plans based 
on the outcomes of the 2008 dress rehearsal. As we heard 
earlier today from the GAO, they are offering several 
recommendations. We are concerned that many aspects of the 2010 
plan were not included in the dress rehearsal, which raises so 
many uncertainties.
    Six, Congress must reject any proposals which would prevent 
the full enumeration of every person living in the United 
States on census day. There have been a number of legislative 
and policy efforts to exclude the undocumented from the census 
enumeration. These proposals are contrary to one of the 
fundamental precepts of our Constitution. We urge the 
Administration and all Members of Congress to reject these 
flawed proposals and to go about the business of promoting a 
full enumeration of all persons living in the United States, as 
required by the Constitution.
    And, seven, the U.S. Senate must support a seamless and 
expeditious transition in the leadership of the Census Bureau 
once a new Administration takes place in January. It is 
critical that there be no disruption in census operations with 
the advent of a new administration. Although various components 
of these operations will be overseen and carried out by career 
employees who will stay on regardless of changes at the White 
House, the head of the Census Bureau plays a key role by 
inspiring confidence and trust in the Bureau. Any delay in the 
Bureau's leadership transition will impair the Agency's ability 
to keep its 2010 operations completely on track.
    We are devoted to working closely with the Subcommittee, 
with the Administration, and with the Census Bureau in ensuring 
a full count come Census 2010.
    Senator Carper. Mr. Vargas, thank you. This is a really 
good panel, and each of you have given valuable testimony. We 
thank you for that.
    Sitting here listening to you--and some of you know each 
other and have worked together before. Mr. Vargas, do you know 
this cast of characters? Have you worked with these people 
before?
    Mr. Vargas. Yes, I do, sir.
    Senator Carper. All of them?
    Mr. Vargas. I think this is the first time I have met Mr. 
Salvo and Mr. Harrison, but I have worked with Mr. Prewitt, 
during the 2000 census, when he was the Director there. And Ms. 
Narasaki and I have a long history of working together on a 
variety of civil rights issues.
    Senator Carper. All right. Mr. Salvo, do you know these 
other folks?
    Mr. Salvo. Well, I certainly know Ken Prewitt and Roderick 
Harrison from his days at the Census Bureau.
    Mr. Vargas. Have we met?
    Mr. Salvo. You know, I am not sure. Mr. Vargas and Ms. 
Narasaki I have not met.
    Senator Carper. All right. Ms. Narasaki, recognize these 
guys?
    Ms. Narasaki. Yes, and it is a great team and cast. The 
Census Bureau--and the people who work on the censuses--is 
actually a fairly small family in terms of who works on it all 
10 years and not just when the actually decennial is happening. 
So I am pleased to be up here with my colleagues.
    Senator Carper. Mr. Harrison, how about you?
    Mr. Harrison. Yes, I know everybody by name. I am terrible 
by sight. And we know half the people in the audience, too.
    Senator Carper. Good. All right. I saw a bunch of them 
nodding their heads yes or no.
    Dr. Prewitt, let me come back to you, if I could. My 
colleague Dr. Coburn has been detained over in the Senate 
chamber, unfortunately, but one of the issues that he has been 
very much interested in, as have I, has been our ability to use 
technology to enable us to do a better job in conducting our 
census in 2010 as compared to 10 years earlier.
    One of our questions we have asked of the Census Bureau and 
Dr. Murdock and the Secretary of Commerce over and over again 
is why are we not able to make some greater use of the 
Internet. And the question, rather than--let me just add, going 
back 8 years when we were doing the 2010 census, did you have 
an expectation, kind of looking ahead from that point up to 
now, that we would be able to use the Internet more broadly 
than we are planning to do?
    Mr. Prewitt. Yes. I will preface it by saying that the 
Bureau has been a technological innovator for a very long time. 
The huge technological innovation in 2000 was data capture, 
intelligent character recognition, enormously sophisticated, 
and we did extremely well with our contractors on that.
    So I certainly think the mood going into preparation for 
2010 on the basis of that success was that it would be a 
technologically more based, if you would, census especially 
because it is the short form only. It is much more complicated 
if you are carrying the long form. But only with the short 
form--I myself in 2000 answered by Internet. It was an option 
presented but not publicized. But since I knew it was there, I 
recognized it, I did it immediately, and it must have taken me 
10 seconds. I got just the short form.
    So, yes, I think the expectation was we would be using the 
Internet by now, and I have not followed the pros and cons, 
although I have followed the handheld issue. I would like just 
quickly to say on the handheld, using the handheld in address 
canvassing is itself a technological breakthrough innovation. 
So, in some respects, we will look back on 2010, if it is 
successful in the address listing thing, as another 
technological step forward. It is too bad it cannot be used in 
non-response follow-up, but that is what has happened.
    With respect to the Internet, I personally am disappointed. 
On the other hand, I am certainly not second-guessing the 
decision of the current management at this stage of the game. 
At this stage of the game, the strongest thing I can urge is do 
not try to put in any new procedures. We already have 
procedures that are going to go out on the field untested, as 
you heard from GAO, and we simply cannot burden it with any 
untested procedures.
    Senator Carper. All right. Thank you.
    I have one question that I am going to ask all of our 
witnesses to respond to, if you would. And then I am going to 
promise you some follow-up questions and ask for you to respond 
to those in writing. Much of the Bureau's success in 2000 was 
attributed to its partnerships with community-based 
organizations. Let me just ask you to describe the value of the 
Partnership Program, the value of that Partnership Program or 
those Partnership Programs, and to assess its overall 
effectiveness in ensuring full participation of hard-to-count 
groups.
    Mr. Vargas, would you like to lead us off?
    Mr. Vargas. Thank you, Chairman, for that question. The 
Partnership Program is absolutely critical to the success of 
the census because, in essence, what happens is that the Bureau 
counts on local organizations, local leaders to become the 
messengers for the census that the census is safe and 
confidential. In many ways, we put ourselves on the line 
relying on the Census Bureau's faith that they will, in fact, 
keep this safe and confidential. So not only does there need to 
be a partnership at the Bureau, but a level of trust and 
working relationship. And in order for that to happen, it needs 
to be developed over time. We cannot have the Census Bureau 
show up in January 2010 saying, ``OK, it is time for the 
census.'' We need to develop that trust beginning today, which 
is why funding for the Partnership Program is so critical so 
that we develop those relationships over time, so that come 
April 2010 we can stand up and, with a certain amount of 
certainty, tell our people, our constituents, ``Fill our your 
census form. It is safe and confidential.''
    Senator Carper. Mr. Salvo.
    Mr. Salvo. I am in a position where almost on a daily basis 
people come to me for their numbers. And these are the people 
that know that numbers matter, and these are the people that 
need to communicate the message within their own communities 
that the census message or the census process actually could be 
very invigorating and exciting. I know it sounds strange when 
you talk about numbers, but the fact is that once people 
understand the linkage between their own destiny, living within 
their own communities, and the numbers that are produced in the 
decennial census--and also, I might add, in the American 
Community Survey--once they get it, then response rises. And 
the only way to get it is to have it delivered by one of their 
compatriots right in their own community.
    All of the national media attention that the Census Bureau 
wants to generate, all those TV ads at a national level will 
not work unless that connection is made.
    Senator Carper. Good point. Thanks. Ms. Narasaki.
    Ms. Narasaki. Yes, I would like to start out by commending 
the Census Bureau who, over the last 10 years, has done a lot 
in terms of doing focus groups of many different new or 
emerging communities to really understand what the barriers are 
and working much more closely with the advisory committee in 
trying to work on what the outreach and advertising campaign is 
going to be.
    Last time around, for example, the Bureau came out with a 
very catchy slogan: ``It's our future. Don't leave it blank.'' 
But it turned out not to translate very well into many Asian 
languages.
    So it is that partnership at the national level and at the 
local level that will help make sure that the money that the 
Bureau is investing in the advertising is actually going to 
effectively reach our communities.
    But as Mr. Salvo notes, some of the strongest persuaders 
are going to be not the government, and not the paid 
advertising. It is the earned media who is going to turn to 
people in the community, the faith-based leaders, the small 
business owners, the teachers and others--they are the people 
who are going to be most effective in reaching out to those in 
our community who are afraid and do not understand what the 
census is about.
    Senator Carper. Thank you, ma'am. Mr. Harrison.
    Mr. Harrison. I think the major points have been well made. 
The partnership is perhaps the only way that many of these 
hard-to-reach populations get the message of the importance of 
answering the census, how it pays off. And, yes, some of the 
people who know the concerns of the communities, what issues, 
the education, if we have undercounts of children and do not 
plan our schools, etc., for the population that has grown, 
where it is. So it is that, the importance of answering the 
census.
    Second, particularly now in the post-PATRIOT Act 
environment and in many immigrant communities, the reassurance 
that, in fact, answering the census will not lead to any kind 
of difficulties with government agencies at the national or 
city--Federal, city, local, that is absolutely essential in 
many communities.
    And then, third, I am not sure that this has been 
mentioned, but there are--answering the census is not simple, 
particularly if you do not speak English. There are complex 
issues of who belongs in the household, who does not. Partners 
need to learn--I think the Bureau is going to have to rely more 
in 2010 than even in 2000 on partners to be able to convey some 
of the complexities of residence rules to households that might 
not know whether they count this child as part of the household 
or not. And these are often, again, more complex households 
than the standard nuclear two-children/two-adult family.
    Senator Carper. I am reminded of in our State, when I was 
governor, we launched an effort to recruit 10,000 mentors to 
work with kids who just needed an extra person in their life. 
The young man I have mentored for the last 10 years does not 
live with his nuclear family, either his mom or dad, but he 
bounces back and forth between a great aunt and a grandparent.
    Mr. Harrison. Very likely to be undercounted.
    Senator Carper. Yes. He is just a prime example of the kind 
of person that can slip through the cracks.
    Dr. Prewitt, you have the last word. I am going to ask you 
to just use it briefly.
    Mr. Prewitt. Certainly, I think the odds of accurately or 
reasonably accurately counting the new immigrant population, 
especially the undocumented, is zilch without a good 
Partnership Program. The Bureau itself cannot do that 
population group at all. And other population groups are 
equally important, but not like--we are going to have a huge 
undercount in that population in 2010.
    On the other hand, what I would quickly say about the 
Partnership Program is that from the point of view of the 
Census Bureau Director, when you really are exhausted in 
Washington, being beaten up by the Subcommittee and having a 
GAO----
    Senator Carper. Not by this Subcommittee.
    Mr. Prewitt. No. Let me do a detour here. Oddly, the Senate 
forgot us after I was confirmed. I did not have Senate 
hearings. I had some 23 House hearings, but the Senate just let 
the census go.
    Senator Carper. Wow.
    Mr. Prewitt. Yes, it is interesting. But the whole IG, GAO, 
the whole apparatus, it was an unusual census environment 
politically. You cannot imagine how nice it was to go out to 
the partnership meetings where people were excited about the 
census, cared about the census, making noise about the census, 
understood the census, and so forth.
    The thing I would most hope to happen in 2010 with respect 
to partnership is a different kind of partnership with the U.S. 
Congress, which was a bifurcated--individuals very good on both 
sides of the aisle, but as an entity it was not very good.
    Just imagine the following: April 1st, live feed, 535 
Members of Congress sitting at their desks filling out the 
census form as a statement to the country. Or just imagine 
April 1st, the President and his Cabinet and the leadership of 
the Congress on the steps of the Thomas Jefferson Memorial 
filling out their census forms. Some statement like that by the 
Congress and the Executive Branch would be enormously important 
for the census. And I hope something like that can happen.
    Senator Carper. Thank you.
    When I look at this panel--well, let me back up a little. 
When I put together my staff in the Senate, when I put together 
my administration as governor, and in teams that I led before 
that, I always tried to put together a team that was reflective 
of my State in terms of its age, its gender, its ethnic 
background. And I look at this panel, this is a pretty diverse 
panel, and I think one that has provided excellent testimony 
for us today. And I thank you very much for that. And for those 
of you who have been working in these vineyards for a while, to 
try to make sure that we do come as close as we can to counting 
everybody, I thank you on behalf of all those people who did 
not know you were working for them. But you are doing the 
Lord's work there.
    We are going to leave the hearing record open for about 2 
more weeks, and that will enable some of my colleagues who have 
statements that they wanted to give or will give, and also to 
ask a couple of questions. And I would just ask if you do 
receive some post-hearing questions--you will probably get one 
or two from me as well. But I would ask that as you receive 
those, you try to come back with your responses in a timely 
way.
    I want to thank the Members of my Subcommittee staff, some 
of whom are here today, and Dr. Coburn's staff, for helping put 
together, I think, a very good hearing. With that having been 
said, I have two other hearings that are underway right now, 
and I am going to head back to them and see if we cannot figure 
out how to get our Nation's economy and our banking system and 
our financial and credit systems functioning again.
    Mr. Prewitt. We do not want to hold you up.
    Senator Carper. But I feel better about the work that we 
are doing to address the undercount in the census.
    Thank you all. This hearing is adjourned.
    [Whereupon, at 12:10 p.m., the Subcommittee was adjourned.]


                            A P P E N D I X

                              ----------                              

[GRAPHIC] [TIFF OMITTED] 45579.001

[GRAPHIC] [TIFF OMITTED] 45579.002

[GRAPHIC] [TIFF OMITTED] 45579.003

[GRAPHIC] [TIFF OMITTED] 45579.004

[GRAPHIC] [TIFF OMITTED] 45579.005

[GRAPHIC] [TIFF OMITTED] 45579.006

[GRAPHIC] [TIFF OMITTED] 45579.007

[GRAPHIC] [TIFF OMITTED] 45579.008

[GRAPHIC] [TIFF OMITTED] 45579.009

[GRAPHIC] [TIFF OMITTED] 45579.010

[GRAPHIC] [TIFF OMITTED] 45579.011

[GRAPHIC] [TIFF OMITTED] 45579.012

[GRAPHIC] [TIFF OMITTED] 45579.013

[GRAPHIC] [TIFF OMITTED] 45579.014

[GRAPHIC] [TIFF OMITTED] 45579.015

[GRAPHIC] [TIFF OMITTED] 45579.016

[GRAPHIC] [TIFF OMITTED] 45579.017

[GRAPHIC] [TIFF OMITTED] 45579.018

[GRAPHIC] [TIFF OMITTED] 45579.019

[GRAPHIC] [TIFF OMITTED] 45579.020

[GRAPHIC] [TIFF OMITTED] 45579.021

[GRAPHIC] [TIFF OMITTED] 45579.022

[GRAPHIC] [TIFF OMITTED] 45579.023

[GRAPHIC] [TIFF OMITTED] 45579.024

[GRAPHIC] [TIFF OMITTED] 45579.025

[GRAPHIC] [TIFF OMITTED] 45579.026

[GRAPHIC] [TIFF OMITTED] 45579.027

[GRAPHIC] [TIFF OMITTED] 45579.028

[GRAPHIC] [TIFF OMITTED] 45579.029

[GRAPHIC] [TIFF OMITTED] 45579.030

[GRAPHIC] [TIFF OMITTED] 45579.031

[GRAPHIC] [TIFF OMITTED] 45579.032

[GRAPHIC] [TIFF OMITTED] 45579.033

[GRAPHIC] [TIFF OMITTED] 45579.034

[GRAPHIC] [TIFF OMITTED] 45579.035

[GRAPHIC] [TIFF OMITTED] 45579.036

[GRAPHIC] [TIFF OMITTED] 45579.037

[GRAPHIC] [TIFF OMITTED] 45579.038

[GRAPHIC] [TIFF OMITTED] 45579.039

[GRAPHIC] [TIFF OMITTED] 45579.040

[GRAPHIC] [TIFF OMITTED] 45579.041

[GRAPHIC] [TIFF OMITTED] 45579.042

[GRAPHIC] [TIFF OMITTED] 45579.043

[GRAPHIC] [TIFF OMITTED] 45579.044

[GRAPHIC] [TIFF OMITTED] 45579.045

[GRAPHIC] [TIFF OMITTED] 45579.046

[GRAPHIC] [TIFF OMITTED] 45579.047

[GRAPHIC] [TIFF OMITTED] 45579.048

[GRAPHIC] [TIFF OMITTED] 45579.049

[GRAPHIC] [TIFF OMITTED] 45579.050

[GRAPHIC] [TIFF OMITTED] 45579.051

[GRAPHIC] [TIFF OMITTED] 45579.052

[GRAPHIC] [TIFF OMITTED] 45579.053

[GRAPHIC] [TIFF OMITTED] 45579.054

[GRAPHIC] [TIFF OMITTED] 45579.055

[GRAPHIC] [TIFF OMITTED] 45579.056

[GRAPHIC] [TIFF OMITTED] 45579.057

[GRAPHIC] [TIFF OMITTED] 45579.058

[GRAPHIC] [TIFF OMITTED] 45579.059

[GRAPHIC] [TIFF OMITTED] 45579.060

[GRAPHIC] [TIFF OMITTED] 45579.061

[GRAPHIC] [TIFF OMITTED] 45579.062

[GRAPHIC] [TIFF OMITTED] 45579.063

[GRAPHIC] [TIFF OMITTED] 45579.064

[GRAPHIC] [TIFF OMITTED] 45579.065

[GRAPHIC] [TIFF OMITTED] 45579.066

[GRAPHIC] [TIFF OMITTED] 45579.067

[GRAPHIC] [TIFF OMITTED] 45579.068

[GRAPHIC] [TIFF OMITTED] 45579.069

[GRAPHIC] [TIFF OMITTED] 45579.070

[GRAPHIC] [TIFF OMITTED] 45579.071

[GRAPHIC] [TIFF OMITTED] 45579.072

[GRAPHIC] [TIFF OMITTED] 45579.073

[GRAPHIC] [TIFF OMITTED] 45579.074

[GRAPHIC] [TIFF OMITTED] 45579.075

[GRAPHIC] [TIFF OMITTED] 45579.076

[GRAPHIC] [TIFF OMITTED] 45579.077

[GRAPHIC] [TIFF OMITTED] 45579.078

[GRAPHIC] [TIFF OMITTED] 45579.079

[GRAPHIC] [TIFF OMITTED] 45579.080

[GRAPHIC] [TIFF OMITTED] 45579.081

[GRAPHIC] [TIFF OMITTED] 45579.082

[GRAPHIC] [TIFF OMITTED] 45579.083

[GRAPHIC] [TIFF OMITTED] 45579.084

[GRAPHIC] [TIFF OMITTED] 45579.085

[GRAPHIC] [TIFF OMITTED] 45579.086

[GRAPHIC] [TIFF OMITTED] 45579.087

[GRAPHIC] [TIFF OMITTED] 45579.088

[GRAPHIC] [TIFF OMITTED] 45579.089

[GRAPHIC] [TIFF OMITTED] 45579.090

[GRAPHIC] [TIFF OMITTED] 45579.091

[GRAPHIC] [TIFF OMITTED] 45579.092

[GRAPHIC] [TIFF OMITTED] 45579.093

[GRAPHIC] [TIFF OMITTED] 45579.094

[GRAPHIC] [TIFF OMITTED] 45579.095

[GRAPHIC] [TIFF OMITTED] 45579.096

[GRAPHIC] [TIFF OMITTED] 45579.097

[GRAPHIC] [TIFF OMITTED] 45579.098

[GRAPHIC] [TIFF OMITTED] 45579.099

[GRAPHIC] [TIFF OMITTED] 45579.100

[GRAPHIC] [TIFF OMITTED] 45579.101

[GRAPHIC] [TIFF OMITTED] 45579.102

[GRAPHIC] [TIFF OMITTED] 45579.103

[GRAPHIC] [TIFF OMITTED] 45579.104

[GRAPHIC] [TIFF OMITTED] 45579.105

[GRAPHIC] [TIFF OMITTED] 45579.106

[GRAPHIC] [TIFF OMITTED] 45579.107

[GRAPHIC] [TIFF OMITTED] 45579.108

[GRAPHIC] [TIFF OMITTED] 45579.109

[GRAPHIC] [TIFF OMITTED] 45579.110

[GRAPHIC] [TIFF OMITTED] 45579.111

[GRAPHIC] [TIFF OMITTED] 45579.112

[GRAPHIC] [TIFF OMITTED] 45579.113

[GRAPHIC] [TIFF OMITTED] 45579.114

[GRAPHIC] [TIFF OMITTED] 45579.115

[GRAPHIC] [TIFF OMITTED] 45579.116

[GRAPHIC] [TIFF OMITTED] 45579.117

[GRAPHIC] [TIFF OMITTED] 45579.118

[GRAPHIC] [TIFF OMITTED] 45579.119

[GRAPHIC] [TIFF OMITTED] 45579.120

[GRAPHIC] [TIFF OMITTED] 45579.121

[GRAPHIC] [TIFF OMITTED] 45579.122

[GRAPHIC] [TIFF OMITTED] 45579.123

[GRAPHIC] [TIFF OMITTED] 45579.124

[GRAPHIC] [TIFF OMITTED] 45579.125

[GRAPHIC] [TIFF OMITTED] 45579.126

[GRAPHIC] [TIFF OMITTED] 45579.127

[GRAPHIC] [TIFF OMITTED] 45579.128

[GRAPHIC] [TIFF OMITTED] 45579.129

[GRAPHIC] [TIFF OMITTED] 45579.130

[GRAPHIC] [TIFF OMITTED] 45579.131

[GRAPHIC] [TIFF OMITTED] 45579.132

[GRAPHIC] [TIFF OMITTED] 45579.133

[GRAPHIC] [TIFF OMITTED] 45579.134

[GRAPHIC] [TIFF OMITTED] 45579.135

[GRAPHIC] [TIFF OMITTED] 45579.136

[GRAPHIC] [TIFF OMITTED] 45579.137

[GRAPHIC] [TIFF OMITTED] 45579.138

[GRAPHIC] [TIFF OMITTED] 45579.139