[Senate Hearing 111-726]
[From the U.S. Government Publishing Office]



                                                        S. Hrg. 111-726

                   COUNTDOWN TO CENSUS DAY: PROGRESS
                     REPORT ON THE CENSUS BUREAU'S
                    PREPAREDNESS FOR THE ENUMERATION

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


                                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


                                 of the

                     ONE HUNDRED ELEVENTH CONGRESS

                             SECOND SESSION

                               __________

                           FEBRUARY 23, 2010

                               __________

       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
  56-842 PDF               WASHINGTON : 2010
___________________________________________________________________________
For sale by the Superintendent of Documents, U.S. Government Printing Office, 
http://bookstore.gpo.gov. For more information, contact the GPO Customer 
Contact Center, U.S. Government Printing Office. Phone 202-512-1800, or 
866-512-1800 (toll-free). E-mail, [email protected].  






        COMMITTEE ON HOMELAND SECURITY AND GOVERNMENTAL AFFAIRS

               JOSEPH I. LIEBERMAN, Connecticut, Chairman
CARL LEVIN, Michigan                 SUSAN M. COLLINS, Maine
DANIEL K. AKAKA, Hawaii              TOM COBURN, Oklahoma
THOMAS R. CARPER, Delaware           JOHN McCAIN, Arizona
MARK L. PRYOR, Arkansas              GEORGE V. VOINOVICH, Ohio
MARY L. LANDRIEU, Louisiana          JOHN ENSIGN, Nevada
CLAIRE McCASKILL, Missouri           LINDSEY GRAHAM, South Carolina
JON TESTER, Montana                  ROBERT F. BENNETT, Utah
ROLAND W. BURRIS, Illinois
PAUL G. KIRK, JR., Massachusetts

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

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

                  THOMAS R. CARPER, Delaware, Chairman
CARL LEVIN, Michigan                 JOHN McCAIN, Arizona
DANIEL K. AKAKA, Hawaii              TOM COBURN, Oklahoma
MARK L. PRYOR, Arkansas              GEORGE V. VOINOVICH, Ohio
CLAIRE McCASKILL, Missouri           JOHN ENSIGN, Nevada
ROLAND W. BURRIS, Illinois

                    John Kilvington, Staff Director
               Velvet Johnson, Professional Staff Member
    Bryan Parker, Staff Director and General Counsel to the Minority
                   Deirdre G. Armstrong, Chief Clerk













                            C O N T E N T S

                                 ------                                
Opening statements:
                                                                   Page
    Senator Carper...............................................     1
    Senator Burris...............................................    16
Prepared statements:
    Senator Carper...............................................    33
    Senator McCain...............................................    35

                               WITNESSES
                       Tuesday, February 23, 2010

Robert M. Groves, Director, U.S. Census Bureau, U.S. Department 
  of Commerce....................................................     4
Todd J. Zinser, Inspector General, U.S. Department of Commerce...     9
Robert Goldenkoff, Director, Strategic Issues, U.S. Government 
  Accountability Office..........................................    11

                     Alphabetical List of Witnesses

Goldenkoff, Robert:
    Testimony....................................................    11
    Prepared statement...........................................    71
Groves, Robert M.:
    Testimony....................................................     4
    Prepared statement...........................................    36
Zinser, Todd J.:
    Testimony....................................................     9
    Prepared statement...........................................    53

                                APPENDIX

Questions and Responses for the Record from:
    Mr. Groves...................................................    98
    Mr. Zinser...................................................   104
    Mr. Goldenkoff...............................................   109

 
                   COUNTDOWN TO CENSUS DAY: PROGRESS
                     REPORT ON THE CENSUS BUREAU'S
                    PREPAREDNESS FOR THE ENUMERATION

                              ----------                              


                       TUESDAY, FEBRUARY 23, 2010

                                 U.S. Senate,      
        Subcommittee on Federal Financial Management,      
              Government Information, Federal Services,    
                               and International Security  
                      of the Committee on Homeland Security
                                        and Governmental Affairs,  
                                                    Washington, DC.
    The Subcommittee met, pursuant to notice, at 2:32 p.m., in 
room SD-342, Dirksen Senate Office Building, Hon. Thomas R. 
Carper, Chairman of the Subcommittee, presiding.
    Present: Senators Carper and Burris.

              OPENING STATEMENT OF SENATOR CARPER

    Senator Carper. Good afternoon, everyone. Welcome, one and 
all, especially to our witnesses and to those who are seated in 
our audience.
    Today is the continuation of our oversight efforts relating 
to the 2010 Decennial Census, one of the few things that is 
actually spelled out in the Constitution that we are required 
to do, and we have been doing it for a long time.
    I have a button that Dr. Groves gave me just a minute ago 
and it says, ``United States Census 2010.'' The ``United 
States'' is in very small letters, but ``Census 2010'' is 
larger. It is going to try to count everybody. It is simple and 
it is important. Simple, easy, and important. In terms of 
messaging, that is a very good way to message.
    But as many of you know, the road to the 2010 Census has 
been anything but easy, and this Subcommittee has held a number 
of hearings on many of the operational and organizational 
challenges that are threatening the success of this particular 
census. My guess is if we went back over time, we would find 
that none of them have been all that easy, and it is even more 
so given how many people live in our country today, how many 
different languages they speak, and how many different kinds of 
living arrangements that we have, and kids in one State and 
parents and families in other States and a lot of people who 
aren't related living in the same group housing facility.
    While we are far from done, I think we can all take pride 
in the excellent work of Dr. Groves, his predecessor, Dr. 
Murdock, and the career professionals at the Census Bureau who 
have worked very hard over the past several months to get the 
census back on track.
    This year's census will be by far the most expensive in our 
Nation's history, even taking inflation into account. So far, 
the cost of the 2010 Census has been estimated to be about 
$14.7 billion. That reflects an increase of a little over $3 
billion in just the last 2 years. And although there are only 
37 days remaining until Census Day--is that when April first 
is?--Census Day, April 1, the 2010 Census has at long last 
begun in certain parts of the country.
    The population tally officially began in late January in 
remote parts of Alaska and the Bureau is now revving up for 
full-scale operations. I don't know if Dr. Groves will say this 
in his testimony, but the amount of free publicity that the 
2010 Census received by virtue of kind of the trek, if you 
will, that was taken through parts of Alaska by small airplane, 
by dog sled, or however folks got there to start counting 
people, a lot of coverage of that, free media, and very smart. 
I said, why are we starting in Alaska, and that really kind of 
answered my question because it is a great way to get the 
message out that we are starting and that every vote--not just 
every vote counts, but we want to make sure that every person 
counts.
    Overall, things seem to be going according to plan. 
Recruiting is--and that is not to say everything is perfect, 
and we will hear about some of the things that aren't. In fact, 
we will probably focus more on the things that aren't perfect 
than the things that are going well. But I should point out 
that a number of things are going according to plan.
    Recruiting is on track. There is a silver lining in every 
cloud. The cloud is our economy, the high unemployment. The 
silver lining is there is a lot of talent out there. People are 
anxious to work on the census who 10 years ago may not have had 
any interest in doing that, but today, they are signing up and 
we have got some very good talent coming to work on this.
    But census questionnaires have been printed and are 
scheduled to be mailed out the middle of next month. Local 
census offices are opening and operational, and the advertising 
campaign is moving smoothly into its active phase.
    However, given the sheer magnitude of such an undertaking 
as the Decennial Census, problems are to be expected. 
Investigations performed by GAO and the Commerce Department's 
Inspector General have raised concern that the Bureau is behind 
on testing and the full development of some of its key 
information technology systems. In December, the Bureau 
conducted two operational tests of the computer networks 
supporting decennial operations which revealed critical defects 
and IT performance problems. More recently, a quarterly report 
issued last week by the Commerce Inspector General noted that 
the Bureau wasted millions of dollars on workers who were hired 
and trained last year for temporary positions by the Census 
Bureau but never worked for the agency and others who 
overbilled for travel expenses.
    In addition to the operational issues that I previously 
mentioned, undercounting remains a serious challenge for many 
communities throughout our country. In 2000, about 6.5 million 
people were missed, many of whom were minorities and children, 
and reaching out to those who are historically hard to count is 
even more important when you consider that for every one 
percent of the population that does response to the census, we 
are going to have to spend about $85 million extra, I am told, 
to go door-to-door and get everyone signed up and counted. It 
is vitally important, then, 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 in the next decade.
    Sometimes people say to me, what can I do to help better 
ensure that kids coming out of our schools can graduate and 
read and write and do math and use technology? How can I, as 
one person, help? And I say, you can mentor. People say to me, 
what can I do in order to reduce our dependence on foreign oil 
and to do something that is good for the environment and our 
climate and I say, well, you can recycle. And people say to me, 
what can I do to reduce the budget deficit? I am just one 
person. What can I do? Well, everybody can fill out their 
census forms and turn them in, and to the extent that we do 
that, for every million people who are counted, that is $85 
million less we have to spend to go out and count them.
    With that said, I look forward to the expert testimony that 
our distinguished panel of witnesses will provide today. It is 
my hope that today's proceedings will provide us with a clear 
assessment of the complications facing the Census Bureau, how 
Congress can best partner with the Bureau as it works toward 
achieving its goal of an accurate and cost-effective census in 
2010.
    And I want to welcome, again, our panelists, none of whom 
are strangers here. I am going to take just a moment and 
introduce each of them.
    Dr. Robert Groves was nominated by President Barack Obama 
to be Director of the Census Bureau in April of last year. He 
was confirmed by the Senate in July of last year, an easy lift, 
as I recall. Dr. Groves is an expert in survey methodology and 
has spent decades working to strengthen the Federal Statistical 
System, improve its staffing through training programs, and 
keep it committed to the highest scientific principles of 
accuracy and efficiency. Having once served as Associate 
Director of the Census Bureau as a child--well, maybe not--Dr. 
Groves knows how the agency operates and what it needs to be 
successful and to successfully implement the Decennial Census 
and other related programs.
    Todd Zinser serves as our Inspector General for the U.S. 
Department of Commerce. As Inspector General there, Mr. Zinser 
leads a team of auditors, investigators, attorneys, and 
administrative staff responsible for detecting and preventing 
waste, fraud, and abuse in the vast array of business, 
scientific, economic, and environmental programs that are 
administered by the Department of Commerce and its 13 bureaus. 
Mr. Zinser holds a Bachelor's degree in political science from 
Northern Kentucky University and a Master's degree in political 
science from Miami University. Is that Miami University in 
Oxford, Ohio?
    Mr. Zinser. Yes, sir.
    Senator Carper. Yes. As a Buckeye, I am always happy to 
welcome another Buckeye, and especially having a guy here from 
the University of Michigan at our table.
    Robert Goldenkoff is the Director of Strategic Issues at 
the Government Accountability Office, where he is responsible 
for reviewing the 2010 Census and government-wide human capital 
reforms. Mr. Goldenkoff has also performed research on issues 
involving transportation security, human trafficking, and 
Federal statistical programs. He received his Bachelor of Arts 
degree in political science and Masters of Public 
Administration degree from the George Washington University.
    Normally, we would swear you guys in and give you that 
oath, but since you have such honest faces, we will forego that 
this time and just go right to your testimonies. We indicate 
that we would like you to keep your testimonies to 5 minutes.
    But welcome. Dr. Groves, why don't you go first?

TESTIMONY OF ROBERT M. GROVES,\1\ DIRECTOR, U.S. CENSUS BUREAU, 
                  U.S. DEPARTMENT OF COMMERCE

    Mr. Groves. I am delighted to be here. It is great to see 
you again, Senator. I think the most important thing that I 
should start with is to reiterate what you said. The 2010 
Census has indeed begun. We began in a little village in Alaska 
called Noorvik, 30 miles north of the Arctic Circle. We have to 
do Alaska first because many of the native villagers will 
disperse for hunting and fishing activities during the spring 
thaw.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Groves appears in the Appendix on 
page 36.
---------------------------------------------------------------------------
    Second, our advertisements are on the air. These are 
necessary to notify the American public that the census is 
coming and that we urge them to participate.
    There have been a lot of things that have happened since we 
were last reporting to you and your Subcommittee. We have 
opened all the local census offices. They are up and running 
and staffed. As you mentioned, recruiting for field operations 
is really a wonderful story this decade for the Decennial 
Census.
    What you didn't note is that we have just last week gone 
over the 200,000 mark in terms of partnership organizations 
that are working with us on a volunteer basis to get the word 
out among their constituents----
    Senator Carper. How does that compare with previous years? 
That seems like a lot of organizations to be part of the 
partnership.
    Mr. Groves. It was almost half this amount last decade.
    Senator Carper. So basically, you doubled. That is good.
    Mr. Groves. It is really amazing, and they are all over the 
country, small neighborhoods, big multinationals, and all sorts 
of things.
    We have nearly 10,000 Complete Count Committees. These are 
local committees often appointed by local officials that help 
get the word out in their city and locale.
    We are involved in two operations already. One is Group 
Quarters Advance Visit. This is a new thing we are doing this 
decade to help improve the counts in large institutions like 
prisons, assisted living facilities, and so on.
    And then just last week, 10 million letters went out to 
houses throughout the country for the more rural parts of our 
measurement.
    So a lot has happened. There are challenges that remain. I 
want to focus on those, with your permission. And before I talk 
about technical aspects, I want to talk about two things that 
have received some attention. One, hiring procedures, the 
background checks that we do on our staff, and then, two, the 
2010 Census media campaign.
    Since we last met with you, we have put in place additional 
procedures that we believe will assure both Congress and the 
public that we will be taking every measure possible to protect 
the American public during the phase where our census takers 
will visit individual households. Let me go through those a 
bit.
    Two steps are the same as what we did in the 2000 Census 
during the application process. Each applicant is required to 
accurately disclose information about any conviction, 
imprisonment, probation, or parole in the last 10 years. 
Failure to disclose this information will disqualify an 
individual.
    We then submit to the FBI database name, date of birth, 
Social Security number, and gender to do a record check in the 
FBI data set. This was done in 2000, as well. But in 2010, we 
are doing an extra couple of steps. For the 2010 Census, we 
will conduct a separate fingerprint check against the FBI 
database. We did this in a large operation before I got to the 
Bureau called Address Canvassing in the summer of 2009.
    We learned in that first use of fingerprinting that some 
folks did not generate readable fingerprints. They tended to be 
older employees. They tended to be women. We have beefed up 
training. We worked with the FBI and OPM and we are beefing up 
the training of the fingerprinters and we are also using some 
FBI-recommended lotion to help the ridges of fingers stand out 
for older people to get readable prints. If we still have 
unreadable prints after this initial effort, we have just made 
the decision to reprint using electronic equipment that will be 
stationed in each of the 500 local offices. This is a decision 
we made over the last few weeks.
    Senator Carper. Would you just explain what you just said 
there?
    Mr. Groves. Yes. One thing that is clear is that a set of 
electronic equipment on which you can take fingerprints gets a 
better read rate from those who have very worn-down ridges on 
their fingers.
    Senator Carper. Those older women you were telling us 
about.
    Mr. Groves. I didn't say that. I believe they were older 
people and they also tended to be women----
    Senator Carper. Probably some older men, too?
    Mr. Groves. Yes. Actually, people who have worked in manual 
labor tend to suffer from this problem. We will take the first 
set--we take two sets of fingerprints using cards, normal ink. 
If we can't read those, then we will ask the staff member to 
come in and take electronic fingerprints. We think we can get 
the unreadable rate down to about 10 to 12 percent given that, 
and that is an improvement over our experience in Address 
Canvassing.
    Senator Carper. And for the 10 or 12 percent that we still 
don't get a good read on their fingerprints, what do we do with 
those?
    Mr. Groves. The news on those is that based on our 
applicant pool, about 16 percent of the applicants when we 
submit their name and other identification generate an FBI 
record. There is some sort of history connected to their name 
and Social Security number there. When we do the added 
fingerprints----
    Senator Carper. And so they fall out in many cases?
    Mr. Groves. We look to see exactly, and I will say in a 
minute exactly how we handle those cases.
    When we then fingerprint the people who pass the name 
check--those are the only ones we fingerprint--there is a 
little over a 99 percent chance that they will pass the 
fingerprints, too. We pick up about 0.5 percent on top of that 
16 percent that have a criminal history. So even though the 
fingerprint check does pick up many, the biggest bang for the 
buck is the name check itself.
    For the 2010 Census, we have sharpened the criteria for 
disqualifying applicants with prior criminal histories. We will 
now automatically disqualify any applicant whose screening 
indicates prior convictions or a pending charge for certain 
categories of crimes, such as murder, sex offenses, robbery, 
voter fraud, and other crimes that suggest a threat to public 
safety or to the integrity of the census data.
    In addition, those who have been convicted or have charges 
pending involving crimes of dishonesty, burglary, theft, and 
vandalism are disqualified from employment, except when the 
person conclusively demonstrates that he or she doesn't present 
a threat. We will also use the e-Verify process to confirm 
employment eligibility.
    Let me note that the safety of the American public and of 
our staff is of paramount concern to me during this process. I 
fully support these unprecedented improvements in the screening 
of applicants.
    Let me turn to the Integrated Communications Campaign. We 
are buying paid media because the 2000 Census taught us that 
strategy succeeded in reversing a multi-decade-long decline in 
response rates. It worked.
    In 2007, the Bureau contracted with a professional 
advertising firm that retained 12 subcontractors to research 
and design this program. We are advertising in 28 different 
languages and across eight major audiences. Based on experience 
and the research that preceded this effort, we are focusing our 
advertising on so-called hard to enumerate or hard to count 
populations.
    When we enter into negotiations for media buys for national 
and local outlets, they follow industry practices of seeking 
added value from the media outlets. Examples of this term 
``added value'' for the 2010 Census campaign include additional 
broadcast spots provided for free, celebrity endorsements, or 
mentions of the census in programming or through public service 
announcements.
    In all, there were 2,100 requests for proposals that were 
issued for this media campaign, with over 61,000 media outlets 
responding, and each outlet engaged was asked to provide some 
added value. As of late January, the census team has negotiated 
almost $30 million in added value from media outlets. This 
represents leveraging the taxpayer money on media by about 23 
percent. We expect that will get up to about 25 percent. So in 
addition to paying for this, we are getting 25 percent extra 
value from these added value negotiations.
    In a perfect world, I would note that where every resident 
was completely aware of the constitutional underpinnings of the 
census, there might not be a need to spend taxpayer money on 
advertising. I get that fact. Unfortunately, we don't live in 
such a world. The last census proved with little doubt among 
the profession that the value of paid media, it demonstrated 
that value because of this reversal of the decline in response 
rates. So we spend advertising money in an attempt to save 
salary costs on Non-Response Follow-Up activities.
    Now, let me move to more technical topics. My full 
testimony reviews a lot of internal challenges. I want to focus 
on two or three that I am most concerned about. One has to do 
with a variety of IT systems. We have been conducting load 
tests of key components of the software that we will rely on to 
manage field operations. We ran a load test on December 3. It 
was partially successful. We weren't able to test one component 
of the system. We found defects in other parts that we were 
testing. This involved a national network, about 8,000 people 
in 400 offices banging on the system under a scripted set of 
protocols.
    One glitch in that test prevented us from testing the 
payroll system. We discovered there is a network problem that 
we have now fixed and we are probably going to add hardware to 
that payroll system as a solution. Solutions were developed to 
address each of the glitches we found in the December test, and 
then we did another one on December 15.
    The results of that test gave our technical folks and the 
independent assessment group that I appointed in August or so 
the belief that the basic infrastructure had the capability of 
handling the peak network traffic, but problems remain that I 
will review right now.
    One has to do with a piece of software called the Paper-
Based Operational Control System. What is this thing? This is a 
set of software that allows us to do the Non-Response Follow-Up 
activity, most importantly. After we receive all the 
questionnaires from mail returns, we will send out census 
takers. That system allows us to make assignments, keep track 
of the progress, and so on. This was a late add to the 
development when the handheld machines were chosen not to be 
used in 2008. It has been on the High-Risk List since that 
moment. It remains on the High-Risk List.
    Let me tell you where we are on this. It is going to be 
released in three phases. Release one has been released. It is 
in production, supporting the activities of Remote Alaska 
Enumeration Group Quarters Advance Visit that is going on that 
I mentioned and a couple of other things. The performance of 
the system at this point is not taking the load that we would 
like.
    Senator Carper. Say that again. What do you mean by that?
    Mr. Groves. The current performance of the system in the 
offices in the production of these activities looks to be such 
that it will not accept the load that we are going to need to 
give it in a month or so. So we are working on the performance 
capabilities of the system right now. It is satisfactory for 
what we are doing right now. We are executing the programs that 
we need to. But if we needed to ramp up today to the level we 
are going to have to ramp up in May, it wouldn't support that, 
is our estimation. So that is a key focus of everyone working 
on it, and it will probably be partially a software solution 
and partially a hardware solution.
    The second iteration, the second release of the software 
was released Friday, on schedule. We are now getting the first 
glimmers of what happens when it is in production. Part of that 
will not be released--the Non-Response Follow-Up support won't 
be released until about March 22, about a month before we need 
it. So we are still working on that.
    Then there will be a third iteration that will be released 
later for operations that are needed further down the line.
    This remains a high-risk development, as I anticipated at 
our last hearing. I reported to you that this was one of my key 
concerns. It remains so, primarily because of the compressed 
time that is available to develop the systems and the hard 
deadlines we face. We can't move any of our deadlines.
    So to aggressively mitigate and manage the risks, we have a 
steering committee that represents all the stakeholder 
divisions. A key component of the decision process that we are 
executing right now is a set of trade-offs. What are the core 
functions of the system that we need for the operations in May 
through July, to make sure those go? What are non-core 
functions that we can have work-arounds? That is the process we 
are using right now to manage this, and we want to make sure 
the right folks are at the table to contribute to those 
decisions and make sure they are wise ones.
    This is really a daily management oversight task. We have 
appointed a new group based on the advice of this external 
assessment team that we have brought in that is helping watch 
the process to make sure deliverables day by day stay on 
schedule. If there are any impediments to getting something 
done, we get rid of those and fight through the bureaucracy.
    Finally, I want to talk about cost estimation in the Non-
Response Follow-Up phase. I promised you and the Subcommittee 
that I would do a scrub of this and we have done that. We 
finished our work. We focused on the Non-Response Follow-Up 
phase. It was budgeted at roughly $2.7 billion, the May through 
July operations. I wanted to make sure that in addition to what 
was a top-down way of estimating the cost, that we did one that 
was bottom-up. We did that bringing in expertise from field 
operations.
    We learned from them, from their viewpoint on the process, 
what were the sources of uncertainty. We had to estimate 
through that process the fall in response rates over the 
decade. Our surveys are losing about 5 percentage points over 
the decade of cooperation rates. The effect of the short form 
versus the short and the long form, the effect of the 
replacement questionnaire, the effect of the bilingual form. We 
also needed to re-estimate the vacancy rate, given the 
foreclosure issues in the country.
    We did all of that, and then what we did next was to 
simulate about a thousand different scenarios, different 
combinations of things that might happen. When we finished that 
exercise, it was a comforting result, I can tell you. I feel 
much more comfortable that we are budgeted at a level that 
allows us to successfully complete the operations.
    One indicator of that is that only 9 percent of these 
thousand different scenarios, a perverse set of combination of 
events--very low return rates, bad productivity rates, higher 
vacancy rates--would produce--would be necessary for us to go 
into our contingency funds on Non-Response Follow-Up. So I feel 
better about that.
    Let me conclude by noting that the pace is picking up. 
There are hundreds of important tasks that will be completed 
across all components of the Decennial Census program. Folks in 
Suitland and throughout the regions are working very hard, 
night and day, to make sure this is a good census. There is a 
lot to do and the pace is quickening.
    I stated at the beginning of my testimony something that I 
believed when I entered this office and I still believe. The 
biggest risk to the 2010 Census is the uncertainty posed by the 
American public's response to the questionnaire, and we need 
your help and your continued help. We thank you for what you 
have done already to encourage everybody living in the United 
States to participate in this census to make it successful.
    Thank you, Mr. Chairman.
    Senator Carper. Thanks very much for that testimony.
    My staff and I met yesterday with the IG from the Commerce 
Department, Todd Zinser, and it was a very helpful 
conversation. Thank you for doing that yesterday.
    And Dr. Groves, I just want you to know, a guy who has some 
roots in Ohio actually said some very nice things about someone 
who has some roots in Michigan. You should be comforted by 
that. Not ready to rest on any laurels, but he was very 
complimentary.
    Mr. Zinser, please proceed.

    TESTIMONY OF TODD J. ZINSER,\1\ INSPECTOR GENERAL, U.S. 
                     DEPARTMENT OF COMMERCE

    Mr. Zinser. Thank you, Mr. Chairman. Thank you for inviting 
us to testify today on the Census Bureau's progress and 
preparation for this year's decennial count.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Zinser appears in the Appendix on 
page 53.
---------------------------------------------------------------------------
    Last week, we released our most recent quarterly report to 
Congress on the status of the 2010 Decennial Census. While our 
report and our testimony today discuss serious challenges 
currently faced by the Census Bureau, we are mindful and 
appreciative of the extraordinary efforts being made by a very 
dedicated staff at the Census Bureau to achieve a successful 
outcome. I would like to summarize our quarterly report in my 
testimony in four points.
    First, the development of the Paper-Based Operational 
Control System remains a significant risk area. The Paper-Based 
Operational Control System is a critical computer system for 
managing numerous decennial operations, including the Non-
Response Follow-Up operation. That operation is estimated to 
cost well over $2 billion and is projected to require 
approximately 600,000 census takers to visit almost 50 million 
households during a 10-week period in May, June, and July.
    We found that the development and testing of the system 
continues to suffer from setbacks that will reduce 
functionality and require the development of work-arounds to 
manage the Non-Response Follow-Up operation. The testing of the 
system is continuing to reveal critical defects. Schedule 
delays have also hindered the development of training manuals 
and technical support guides.
    Our second point in the quarterly report concerns the 
Decennial Application Load Test carried out by the Census 
Bureau for each of the various computer systems that will 
support the Decennial. The tests attempted to simulate the 
expected demand on its computer systems. They showed that the 
networks and devices were able to successfully handle peak 
loads, but there were performance and functionality problems 
with two of the more critical systems, the Paper-Based 
Operational Control System, which I have mentioned, and the 
Decennial Applicant and Personnel Payroll System.
    Our third point concerns budgeting and cost containment. We 
found that while Census reports staying within its budget 
during the most recent quarter we reviewed, spending among the 
local census offices remain a concern. We examined the 25 
percent cost overrun experienced by the Bureau for its address 
canvassing operation last year and found wide disparities in 
wages and mileage reimbursement in some of the local census 
offices. For example, when we examined mileage costs incurred 
by local offices for the quality control operation for address 
canvassing, we found that costs among the offices ranged from 
less than 1 percent of their budgets to 878 percent of their 
budgets.
    The 25 percent cost overrun for address canvassing 
indicates a problem with the original budget estimate, a 
problem managing the containment of costs, or perhaps some of 
both. In contrast, the Bureau reports that it spent only 59 
percent of its budget for group quarters validation. 
Inaccuracies of this magnitude in estimated budgets, combined 
with wide spending variances among local census offices, 
indicates significant weaknesses in the Bureau's budget 
estimation and cost containment capabilities.
    Which leads to our fourth and final point concerning the 
estimated budget for Non-Response Follow-Up. Census has 
projected a revised cost estimate of $2.33 billion for the Non-
Response Follow-Up operation, which is $410 million less than 
the prior estimate. However, this is partially offset by an 
estimated 40 percent increase of $137 million for the Vacant 
Delete Check operation, which is now projected to cost $482 
million.
    In addition, the final costs of the Non-Response Follow-Up 
operation remain largely dependent on the mail response rate, 
which is a significant uncertainty. We would also add the 
unknown impact on operation costs of the Paper-Based Control 
System with reduced functionality and performance.
    In brief, although much of the Bureau's plan is on track, 
the efficiency and accuracy of the Non-Response Follow-Up 
operation are at some risk because of the development problems 
with the Paper-Based Operational Control System and final 
Decennial costs remain uncertain.
    That concludes my summary, Mr. Chairman. I would be happy 
to answer any questions you might have.
    Senator Carper. Good. I have some, and again, thanks so 
much for the work that you and your colleagues have done in the 
IG's office. We look forward to asking you some questions. 
Thank you.
    Mr. Goldenkoff, welcome back. Thank you.

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

    Mr. Goldenkoff. Mr. Chairman, I want to thank you for the 
opportunity to be here today to provide an update on the Census 
Bureau's readiness for the 2010 Census.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Goldenkoff appears in the 
Appendix on page 71.
---------------------------------------------------------------------------
    As you know, just over 5 weeks remain until Census Day. The 
decade-long build-up to the Nation's largest peacetime 
mobilization has come to a close and the complex and costly 
business of data collection is now underway.
    The road to Census Day has been a rocky one, fraught with 
operational setbacks and cost overruns. The hurdles the Bureau 
has experienced to date, including weaknesses in the Bureau's 
IT systems and uncertainty over the ultimate cost of the 
census, which is now estimated at around $15 billion, led us to 
designate the 2010 Census a High-Risk Area in March 2008. As 
requested, I will update the Subcommittee on the state of the 
census, paying particular attention to, first, the rollout of 
key IT systems; second, the steps the Bureau has taken to 
revise its cost estimates; and third, the extent to which 
critical enumeration activities, particularly those aimed at 
hard-to-count populations, are on track.
    Senator Carper. Let me interrupt just for a second. Would 
you just go back a couple of sentences? You indicated a cost 
of, I think----
    Mr. Goldenkoff. I said around $15 billion.
    Senator Carper. OK.
    Mr. Goldenkoff. It is about $14.7 billion.
    Senator Carper. Thanks so much.
    Mr. Goldenkoff. The Bureau's readiness for a successful 
head count is mixed. Mr. Chairman, it is deeply troubling that 
with Census Day right around the corner, key IT systems, 
notably the Workflow Management System, this Paper-Based 
Operational Control System (PBOCS) that we have been talking 
about, which is essential for the Bureau's field operations, 
and DAPPS, the Personnel and Payroll Processing System that 
will be used to pay more than a million temporary workers, 
continue to be plagued by performance problems.
    When the Bureau held a critical load test this past 
December, the Workflow Management System experienced log-in 
problems, slow performance, and communication issues, while the 
Automated Personnel and Payroll System offered from server 
problems and sluggish performance. The Bureau is going to great 
lengths to address these issues, taking such steps as 
performing additional tests and purchasing new hardware, but 
the Bureau needs to move promptly. Just weeks remain before the 
systems need to support peak operations.
    In addition, the Bureau revised its cost estimate from $2.7 
billion to $2.3 billion for Non-Response Follow-Up, the largest 
and most costly field operation, where census workers go door-
to-door to follow up in person with non-responding households. 
However, the Bureau's cost analyses are not complete. According 
to the Bureau, it continues to reexamine the costs of two other 
Non-Response Follow-Up related operations, so at this point, 
estimates of the ultimate cost of Non-Response Follow-Up in the 
Decennial Census are still uncertain.
    Other functions, however, are faring better. Key 
enumeration procedures are generally on track, past problems 
are being addressed, and some activities aimed at improving the 
participation of hard-to-count groups are more robust compared 
to similar efforts during the 2000 Census. For example, the 
Bureau plans to fingerprint employees to better screen its 
enormous temporary workforce. However, as you know, in earlier 
operations, a number of fingerprint cards were unreadable, 
which prevented the FBI from conducting a complete background 
check. In response to this issue, among other actions, the 
Bureau plans to improve training procedures on how to take 
fingerprints.
    Efforts to boost response rates are also more aggressive 
compared to the 2000 Census. For example, the Bureau has 
increased staffing for its Partnership Program, which is an 
effort where the Bureau engages government and community 
leaders to gain their support for the census.
    Likewise, the Bureau's plans to mail a second replacement 
questionnaire to census tracts that had low or moderate 
response rates in the 2000 Census should help enhance 
participation in 2010, as will plans to hand-deliver an 
estimated 1.2 million census forms in areas along the Gulf 
Coast that were devastated by recent hurricanes.
    Moving forward, it will be important for the Bureau to 
quickly identify the problems affecting the key IT systems and 
test the solutions. Further, given the complexity of the census 
and the likelihood that other glitches might arise, it will be 
important for the Bureau to stay on schedule, monitor their 
operations with appropriate performance metrics, and have plans 
and personnel in place to quickly address operational issues.
    Now, these operational considerations aside, I want to 
stress that the Census Bureau cannot secure a successful 
enumeration on its own. The public must also fulfill its civic 
responsibility to mail back their census questionnaires in a 
timely fashion. As we have already discussed, according to the 
Census Bureau, each percentage point increase in the mail 
response rate saves taxpayers around $85 million and yields 
more accurate data compared to information collected by 
enumerators during Non-Response Follow-Up.
    The bottom line, Mr. Chairman, is that in a few weeks, a 
key determinant of the success of the 2010 Census will be both 
literally and figuratively in the hands of the Nation's 
residents.
    This concludes my prepared remarks and I will be happy to 
answer questions that you might have.
    Senator Carper. Great. Thanks so much.
    Let me just go back and try to get a handle on a couple of 
points. One is the question in my own mind of how many 
temporary employees are we talking about hiring, like at the 
peak? I don't know if it is April, May, or June. At the peak, 
how many additional temporary employees will we have on the 
payroll?
    I have heard 600,000. Mr. Goldenkoff just said over a 
million.
    Mr. Groves. Well, I think the reason it gets confusing is 
that for this entire fiscal year, it will be about 1.2 million 
jobs that we will have filled, and they go in and out. So right 
now, we are doing an operation that will close out pretty soon.
    Senator Carper. Some of the people that might be hired, you 
brought on board, say, in February, might not necessarily be 
still there in March, April, or May?
    Mr. Groves. Right, and some of them----
    Senator Carper. And vice-versa?
    Mr. Groves [continuing]. Will be rehired for a new job that 
they will continue with the Bureau into July. The peak will be 
the May 1 through July 10 time period, where we will have 
between 600,000 and 700,000 people, and they will be the census 
takers who will visit individual households.
    Senator Carper. OK. Thanks for that clarification.
    I am going to ask Mr. Zinser and Mr. Goldenkoff just to 
tell us what you think may be the--I bring this up in your 
shoes and in our shoes here on this side of the dais--what the 
most significant improvements that you have noted in the 
performance of the Census Bureau and their operation as we 
approach this count. What are you most encouraged about, each 
of you? Maybe give us two or three examples. And then give us 
two or three examples of what you believe we should continue to 
be concerned about, as you are.
    So, first of all, the good news, and then the things we 
ought to continue to be concerned about and keep our eye on. 
That is the question.
    Mr. Zinser. Thank you, Mr. Chairman. I think one thing that 
we have consistently commended the Bureau for is taking on a 
more sophisticated or rigorous risk management approach to the 
Decennial. They have got a team that has been analyzing risks, 
coming up with mitigation efforts, and developing contingency 
plans, and in our report, we comment on that and we have looked 
at the four plans that they have completed and think that they 
have done a good job there.
    I think that the other thing that probably carries over 
from one census to another that is a strength of the Bureau is 
their workforce. They have a highly dedicated workforce. You go 
out and meet the workforce in the local offices, they are very 
focused. They know their territories and they are committed to 
making this a successful census.
    On the downside, I think the IT problems that we have all 
talked about, and Mr. Goldenkoff alluded to in his statement 
are far and beyond the most troubling aspects of this 
Decennial, going back to the decisions that had to be made on 
the hand-held computer and the problems that continue to plague 
the Bureau.
    Senator Carper. Give us one more.
    Mr. Zinser. One more negative?
    Senator Carper. Example of things to be troubled about or 
concerned about as we go forward.
    Mr. Zinser. Well, I think the cost controls. I think that, 
on the one hand, I think that cost controls are always an 
issue. I think they are especially so in these times, but I 
would like to see more done from the senior levels of the 
Census Bureau to emphasize cost controls within their local 
offices.
    Senator Carper. OK. Good. Thanks. Mr. Goldenkoff, first, 
the good news, and then----
    Mr. Goldenkoff. I would agree with everything that Mr. 
Zinser has said and I would add to that. Very broadly, I think 
culturally, we would like to commend the Bureau for being much 
more open and much more transparent than it has been. GAO has 
had a relationship with the Census Bureau for a number of 
years. Certainly, I have been involved with the Census Bureau 
since 2000 and we have been very impressed with the outreach 
with Dr. Groves and some of his immediate predecessors over the 
last few years, and certainly since 2008, the Bureau has been 
much more open to outsiders, and outside advice. Dr. Groves and 
I, we get together on a regular basis to discuss not only 
issues with the 2010 Census, but also planning for the 2020 
Census.
    I think one of the things that is so important is the first 
step in dealing with a problem is recognizing that you have 
one, and I think that the Census Bureau is much more open to 
that, particularly in dealing with outside audit organizations. 
So that is definitely a plus I want to get out there, just the 
cultural change.
    I would also say the ability to develop work-arounds very 
quickly when they have identified problems. The Census Bureau 
has a lot of expertise on board that they can quickly get to 
the heart of a problem and develop a solution quickly, and so 
we commend them for that.
    Senator Carper. OK. And the things that--some of the 
aspects of the census that keep you up at night or should keep 
us up at night?
    Mr. Goldenkoff. IT issues would be No. 1, particularly with 
PBOCS. What we have seen is that they are facing people 
issues----
    Senator Carper. Paper-Based Operational Control System?
    Mr. Goldenkoff. That is correct.
    Senator Carper. I just want to thank you, while I think of 
it, when I read the testimony of, I think it is Dr. Groves, 
there are a lot of acronyms in the testimony. I just want to 
thank you so much for not mentioning those. [Laughter.]
    There was one sentence, or two sentences with five 
acronyms. I am just so grateful that you didn't use that 
sentence.
    Mr. Groves. Should I say, OK?
    Senator Carper. You may, as long as you spell it out. 
[Laughter.]
    Mr. Goldenkoff. So, as I said, the operational control 
system, that is the nerve center of the census, the brains of 
the census, and there are some people issues there, staffing 
issues, hardware issues, software issues, scheduling issues, 
and so that is definitely something that is keeping us up at 
night, and I know it is certainly keeping the Census Bureau up 
at night, as well.
    Senator Carper. OK. Back to you, Dr. Groves. You have heard 
the good and the not-so-good. You don't have to respond to the 
good. You can, if you want. But focus on the concerns 
especially with the IT problems, the concerns that were raised 
with respect to the Paper-Based Operational Control System. You 
have talked about it already, but just come back and give us 
some reason to feel more assured, and also talk with us about--
the point was made on focusing on cost controls. I think the 
comment was, like to see some more top-down concern or focus on 
cost control.
    Mr. Groves. On the IT side, I think there are--actually, we 
discussed this in the last hearing, I believe, how IT 
activities within a large government agency can be handled 
efficiently and how you can stay on the cutting edge and do 
developments that are needed in a way that are spending the 
taxpayers' money wisely and get in production in a timely 
fashion. Some of these are really big issues that deserve a 
hearing of its own, probably, at one point.
    On this particular software development, I think it is 
important to separate pieces out and look at them separately. 
There is a system for the payroll and personnel processing that 
I think is logically viewed as a separate issue. It is running 
on its own network, a separate network from others. It was 
taken really lock, stock, and barrel from our contractor in 
2008 when the decision was made not to continue with the 
handhelds and placed in Suitland. So it is a rather unique 
computer network.
    Part of the problems in the load test that we found was 
there were basically glitches in that network so that when 
people were accessing and sending requests to that network, the 
network was actually idle. It wasn't even getting those 
requests. That was fixed with some approaches to the operating 
system. When that is up and running, its performance is still 
not satisfactory. So on that component of the problem, we will 
be installing some new hardware starting March 1, and the hope 
and belief of the technical experts, both working on the team 
at the Census Bureau and brought in from the outside, is that 
particular problem has a good shot of being solved with that 
operating system fix and the added hardware.
    The Paper-Based Operational Control System is a separate 
issue, I think. That is new software being developed using this 
philosophy of agile programming that allows you to write 
separate little modules in one week cycle time. You do a little 
widget. You insert it in the overall whole. You test it. It is 
compatible with everything else that was in there before.
    We are entering a phase now that often happens, in my 
experience in software development. You have the initial 
release and production. Whenever you put something in 
production, users find things that they don't like, things that 
don't work that everybody thought worked, but with a 
combination of commands don't work, and those begin to--those 
are communicated to the software development group and so they 
have a list of fixes that is adding up as users find these 
things. At the same time, that same team is developing new 
functions for the next release. So there is a competition for 
that precious resource of programming skills, and the 
management procedures we put in place is to deal with that.
    Now, the one good thing that both of my colleagues 
mentioned about the staff of the regions is a component of the 
solution at this point in time, given where we are, and that is 
decisions to remove a function that someone wanted originally 
from this software and to have a work-around for it, some 
manual operation.
    It is comforting to me, as I travel around the country 
talking to our regional folks, that they are quite confident 
they can handle this. In fact, on many of the functions we are 
talking about removing from the software to get time to fix the 
existing functions, they said, well, we did that in 2000 
manually. We did it in 1990 manually. We know how to do that. 
It would have been nice to have this computer assistance, but 
we can live without it. So that is the good side of that.
    The wisdom that is required in this process is to make 
those tradeoff decisions in a way that costs and efficiencies 
aren't hurt, but we give the programming staff time enough to 
do the functions that are really core. And as I said, this is a 
high-risk enterprise. What I can promise all of us is that it 
has the full attention of management up and down the line. But 
this is a high-risk enterprise.
    Senator Carper. OK. Thank you. Thanks for raising those 
points and for your response to them.
    We have been joined by Senator Burris from Illinois. It is 
great to see you and thanks so much for joining us and for your 
attendance at many of our hearings. Senator Burris, you are 
recognized for a statement, if you like, and then questions.

              OPENING STATEMENT OF SENATOR BURRIS

    Senator Burris. Thank you, Mr. Chairman, and to the 
distinguished panel, I must say that having met with Director 
Groves recently on the various issues, I hope that things are 
moving as we had discussed, Director Groves, during our private 
meeting, especially in the area of the minority, underserved, 
or undercounted communities. I mean, that is a major concern of 
mine across the Nation.
    Just for the record, I want it to be known that I was able 
to hook up the Chicago operation with Stan Moore and we had a 
major event there to kick off the necessity for having 
everybody counted. It was a very successful event. We had over 
2,500 individuals who can contact or reach out to various 
sections, primarily in the African-American community in 
Chicago. And, of course, these are influential people that have 
access not only throughout the State, but throughout the 
Nation, in trying to make sure that some of our underserved and 
undercounted communities would be counted.
    They really have not, to my estimation, Mr. Groves, seen 
the value of what that census count would do for that 
community. Not only does it take care of the congressional 
redistricting issues and all that, but they don't understand 
how it does relates to the redistribution of funds for various 
governmental programs. So we have been trying to get that 
message out and I am just hoping that will take place.
    What I am also hearing, though, is a lot of backlash on the 
commercials that have been run and the expenditures. What are 
you all hearing about the amount of money that you are putting 
out in terms of trying to educate the voters--I mean, educate 
the citizenry that they should mail back that form when they 
get it?
    Mr. Groves. Well, we hear both praise and criticism, as you 
might imagine, Senator.
    Senator Burris. Especially on the Super Bowl commercial?
    Mr. Groves. The Super Bowl commercial did appear to be 
noticed by several people in the country, both positively and 
negatively. I think the way we think about this is, first of 
all, my personal opinion on this is I wish we didn't have to 
spend a dime on advertising. I wish that every resident of this 
country knew that the census was planned for this April and 
they couldn't wait to get their questionnaire in the mail.
    We have been doing surveys of the American public over the 
past few months, and other people are doing them, too. There 
are some shocking results there. So there was a Pew Center 
survey that was done just a few weeks ago. It showed among 
people 18 to 29 years old, many of whom are established in 
their first household by themselves, they are out of the 
parents' household----
    Senator Burris. Thank God. [Laughter.]
    Mr. Groves [continuing]. The 31 percent of them didn't know 
what this word ``census'' meant. And then the interviewer was 
instructed to say, well, a census is a count of everyone in the 
country. It is done every 10 years to reapportion the House of 
Representatives. Now that I have told you what it is, have you 
ever heard of that? And that 31 percent goes down only to 17 
percent.
    So we have a massive burden. You and I and everyone in this 
room know that the census is coming. We have known it for 
years. We know what it does and what it is about, but there are 
segments of our population that don't know anything about the 
census. And so we have to get the word out to them somehow, and 
this campaign that we are mounting has a lot of partner 
organizations that are trusted voices in communities all around 
the country. We have 200,000 of them now--it is a wonderful 
accomplishment, I think, on the part of the country--to get the 
word out.
    But in the 2000 Census, we learned that if we did paid 
media, that worked. In the 1990 Census, we relied on Public 
Service Announcements. What happened was we made a lot of them, 
but they aired at three o'clock in the morning. Not very many 
people saw them. And so for the first time in 2000, we reversed 
a long-term trend of declining participation because of paid 
advertising.
    The way I think about this is this communications campaign 
is going to spend about a dollar per person in the country. If 
that dollar per person encourages them to return the 
questionnaire, we save for each person about $25 to measure 
them by sending somebody out. So that is the trade-off 
decision. If we can spend a little money to save a lot of 
money, this makes a lot of sense. And in the 2000 Census, that 
advertising campaign more than paid for itself, many-fold over.
    The Super Bowl ad cost us 2.1 cents per viewer. That was 
the biggest audience in the history of this country. A hundred-
and-thirteen million people saw that ad. Now, we can debate 
whether that was a good ad or not----
    Senator Burris. Or whether they really related to it in 
terms of the census----
    Mr. Groves. That could be true, but in terms of----
    Senator Burris [continuing]. Rather than a part of 
something about football.
    Mr. Groves [continuing]. Cost efficiency of getting the 
message out, it is hard to beat that. If we took out an ad in 
the Washington Post, it is about five cents per person.
    Senator Burris. Yes.
    Mr. Groves. So I don't know a lot about advertising. I 
don't know what makes a good ad or a not-good ad. I do know 
that purchasing on the Super Bowl is a pretty cost efficient 
way to get the word out.
    That is a long-winded answer. I am sorry.
    Senator Burris. No, I appreciate that, because that did 
generate a few more questions. I see that I am a little bit 
over my time here, but I was just concerned about the overall 
cost in terms of the budgeting. You did have some of it in your 
remarks in terms of per--so you based that on about 310 or 320 
million in the population is what we are going? And then what 
would it cost to do the difficult part of the door-to-door? Has 
that been budgeted and projected as to how much that is going 
to cost us?
    Mr. Groves. Yes. It is hard to get the variable costs on 
this, but the number that seems to be most defensible is for 
every one percentage point of households that do not return the 
questionnaire, and we will have to go out and visit those, we 
will spend about $85 million.
    Senator Burris. Yes.
    Mr. Groves. So we are spending about $300 million----
    Senator Burris. In total.
    Mr. Groves. So if we can get four percentage points out of 
this through the advertising, it pays for itself.
    Senator Burris. What about the various activities after the 
census is over and there has to be an undercount, or an 
assessment of the undercount. Do you all get involved in that? 
Sometimes there is litigation involved. What type of plans do 
you have to try to head some of that off?
    Mr. Groves. Well, as we all know, this decade, our design 
is guided by a Supreme Court ruling before the 2000 Census that 
said that the reapportionment will be done based on the counts 
that we are getting. So all of our counts right now are focused 
on getting the best counts we can. Every fiber of our being is 
focused on getting the word out and encouraging people to 
return the questionnaire, and then we will do non-response 
follow-up.
    We, indeed, will do an evaluation of how well we did. There 
will be a large sample survey that will see whether--to 
estimate what proportion were missed, what proportions were 
double-counted. The results of that won't be ready until 2012 
because it is a very complicated statistical matter, but we 
will have that. So the country will know through that and 
through other ways of knowing how well we did.
    Senator Burris. Thank you, Mr. Chairman. I will wait for a 
second round.
    Senator Carper. Thank you, Senator. And again, thank you 
for being a faithful participant in these hearings and for 
adding your voice and presence.
    Senator Burris. I am very much interested in this count, 
Mr. Chairman.
    Senator Carper. Yes. All of us should be.
    Senator Burris. Absolutely. This is crucial, especially to 
the underserved communities.
    Senator Carper. Sure, it is.
    In the testimony that we received a bit earlier, Senator 
Burris, we heard that participation of partnership 
organizations in this census as compared to 10 years ago, we 
are witnessing about almost a doubling of the partnership 
organizations that are active and involved this time as 
compared to a decade ago, which speaks very well.
    There was some discussion of the paid advertising that is 
being used, and I was pleased to hear that the paid advertising 
is focused on the folks in our country who are the least likely 
to respond, the hardest to count. So that makes sense. You have 
some money to spend for advertising, less--rather than spending 
the money on the people we think are almost sure to be counted 
and to stand up and be counted, the idea of saying we are going 
to invest our money where it is going to be real hard to drag 
people out, it is sort of like getting people out to vote. We 
had get out the vote. This is like a get out the count kind of 
deal.
    And the other positive here was just in terms of talent 
pool. We have some really smart, able people to work in the 
census, and we have had those before, but I don't think ever, 
maybe at least in recent years, to the extent that we have now. 
We have got some great talent. My hope is, with that kind of 
talent on board and people who are used to working hard and 
being productive, that we might actually surprise ourselves in 
their ability to get things done.
    I want to ask a question, but I think you wanted to say 
something.
    Senator Burris. Yes. I just wondered, on the advertising 
side, which I probably should have raised with Mr. Groves, in 
the advertising, have they really zeroed in on minority radio 
stations or minority TV stations with the advertising to make 
sure that there is a proportional commitment to these media 
outlets and get it--believe it or not, I know it is a little 
expensive, but some of the local newspaper ads, and you can 
probably get word out to your district offices to look for 
those weekly newspapers that you can put some advertising in 
that we might reach these communities that we are speaking of.
    Senator Carper. OK. Good.
    I want to ask a question of Dr. Groves, if I could, just 
about out of this 600,000 to a million people that we have on 
board at the census, new people, temporary employees--we will 
say it is 600,000 people--how many of those folks will actually 
be enumerators and actually out there counting on a daily 
basis?
    Mr. Groves. That figure that I gave you, it is probably 
680,000 or so, that is the enumerator count. That is the census 
taker count.
    Senator Carper. If I were to look for the months, say, 
April, May, June, would you say that on an average for those 3 
months, we have 600,000 people counting as enumerators?
    Mr. Groves. As enumerators, right. Yes. They will be 
trained the last week of April. They will start work May 1, 
generally.
    Senator Carper. I tried to do some sort of like back-of-
the-envelope math just to see how many folks they would need to 
count. Let us say they were counting--enumerating, if you 
will--20 days a month. And let us say they counted each of 
those 600,000 people, counted five people each day. Five people 
a day, 20 days in a month, that is 100 people. And we will say 
we have 600,000 people counting and each of them count in a 
month 100 people. When I multiply 100 times 600,000, I come up 
with 60 million people a month that the enumerators were able 
to be counting. If they are doing that over 3 months, that 
would be 20 days a month of counting five people a day, that 
would be about 180 million people over 3 months at 60 million a 
month.
    Now, five people a day doesn't sound like much for an 
enumerator. A hundred a month, working 20 days a month, doesn't 
sound like a whole lot, either. A hundred-and-eighty million 
people sounds like--that is more than half the people we are 
going to count. But yet we know that about two-thirds of the 
people are going to respond anyway. Maybe, we will say, about a 
third will be non-responders, so those are the folks we have to 
go out and count, which would be about 100 million people.
    Why do we need 600,000 people a month over 3 months to 
count roughly 100 million people?
    Mr. Groves. Yes. The figures from 2000 are about 40 million 
households, and they are, on average, like, say, 2.3 people per 
household, so that is very close to your 100 million people.
    This is tough work, first of all. It is work that is mainly 
nights and weekends work because you have to call on houses 
when people are at home. The houses you are calling on are 
houses that, for one reason or another, chose not to return a 
questionnaire. And then they are also calling on a lot of 
vacant houses. The vacancy rate this decade is different than 
it was in 2000.
    When we call on a vacant house, we don't know it is vacant. 
All we know is we sent a questionnaire to this address and 
nothing came back. So we have to make really quite sure that 
when we call on a house, it is not just that people are--and no 
one answers the door--they are at work or somewhere else, but 
truly, no one lives in that house. That takes multiple visits.
    It is also true on the occupied houses that it takes 
multiple visits. We actually allow as many as six different 
visits to a housing unit.
    I can tell you that from all the things we know about 
measuring the American public, the houses that are most 
difficult to contact and get cooperation from are those that 
are susceptible to the differential undercount we have seen 
historically in the country. So you could, indeed, do Non-
Response Follow-Up a lot more cheaply, but what will happen is 
you won't measure those houses that are so difficult to 
contact.
    And so even though it is fantastic to imagine that it takes 
this much effort and that you can't just rip out five a day----
    Senator Carper. It is five people. It is not five houses, 
but five people.
    Mr. Groves. That is right. It is, indeed, the fact--most of 
these people, by the way, are working about 17 to 19 hours a 
week, so it is not a full-time job because there aren't that 
many hours that are really peak hours to call on houses.
    Senator Carper. I am going to ask Mr. Zinser and Mr. 
Goldenkoff, any comments on my questioning and the response we 
heard from Dr. Groves?
    Mr. Zinser. Well, Senator, I think one of the points that--
or one of the things you were putting your finger on was how 
efficient we can make the operation and how efficiently we can 
use the 600,000 employees that we hire, and I think that is the 
key point and I think that is why, to go back to what we have 
talked about here, why this operation control system is so 
important, because to get maximum efficiency out of that 
workforce, we need that kind of management information.
    Senator Carper. OK. Mr. Goldenkoff.
    Mr. Goldenkoff. Following up on non-responding housing 
units is an extremely tough job. During the 2000 Census, we 
went around with enumerators all over the country. We could see 
for ourselves how difficult it is. Right off the bat, you are 
dealing with non-responders, so those are people who already 
rejected the Census Bureau once. They haven't returned their 
mail questionnaire. And so it sometimes requires multiple 
visits.
    And on top of that, when you add some of the difficulties 
of going into certain neighborhoods, because of gated 
communities, other aspects, it makes it difficult to get into a 
neighborhood. Sometimes the Census Bureau has to have folks to 
help them get into an area, to help facilitate access to an 
area. The country is a big place. In more rural areas, you 
spend a lot of time in your car driving from one house to 
another. It is not just going door-to-door, but the next 
housing unit you go to could be miles away.
    And so for all those reasons, the amount of time goes up 
and the efficiency tends to go down. But I will say this. 
Everybody we have observed was extremely diligent and extremely 
conscientious about their task.
    Senator Carper. Well, that is encouraging to hear.
    I don't know, Senator Burris, if you have the kind of 
experience in Illinois that we do in Delaware. Delaware is a 
small State. We tend to be, at least in the northern part of 
the State, more densely populated. The southern part is a half-
mile, a mile between the houses or farms.
    But we do a lot of door-to-door get out the vote efforts in 
our State around election time, or just campaigning prior to 
elections to identify who is at home. You have a voter 
registration list. It doesn't always marry up to who is in the 
home. People come and go. Homes are foreclosed on. Folks grow 
up. Kids grow up and leave. They no longer live with their 
parents. There are all kinds of changes.
    And one of our challenges has always been to maintain the 
voter files so that on election day, who lives in a particular 
house and if they are registered and the idea is to try to get 
out the folks who are more likely to vote for you or your side.
    So we have a little bit of experience with--this is not 
enumerating, but in a way enumerating, but for not just to 
count people but actually find out who does live there, who is 
registered, and who is more likely to vote. And we know it is 
not easy. But I would say just the idea of doing five a day 
doesn't sound like a heavy lift. I am encouraged by what Mr. 
Goldenkoff has said about actually seeing what is involved in 
this work. But the focus on cost control, which has been raised 
as an issue here, obviously, it is something that Dr. Groves is 
mindful of and we just need to continue to be mindful of that.
    Senator Burris. Mr. Chairman, in the urban communities or 
the suburban communities, door-to-door sometimes works. But in 
the Chicago community or the major markets, people don't open 
their doors, even when you try to do door-to-door canvassing 
for political situations.
    I remember I had a little issue, and I live in sort of a 
middle-class neighborhood in Chicago. Some people had turned 
the street sign the wrong way and they made it a one way for 
one block, and we had three one-ways going west. And so I had 
to then try to go around and turn it back around, and I walked 
four or five blocks trying to get my neighbors with my 
recognition, and they all knew me, and to even get them to come 
to the door, to even open the door--and, of course, some were 
home and they open the door and saw it is me, then they would 
respond.
    But I found that very difficult in the metropolitan areas 
for those persons to open up their doors to anybody. They could 
be at home and the doorbells would ring or they would knock on 
them and they will not open the doors for anybody. Some of that 
is security purposes. Some of it is just, I don't want to be 
bothered. But in the old days, we used to do that. Precinct 
captains in Chicago would knock on your door. You could knock 
on doors, pass out literature, and people would open the door. 
That was in my early days in politics. But I have seen during 
my days that to have changed substantially.
    Let me raise a couple other points, if I may.
    Senator Carper. Before you do, I remember being invited to 
Illinois to campaign door-to-door with a candidate for office 
and----
    Senator Burris. Was this in Chicago?
    Senator Carper. We were going door-to-door and finding a 
hard time to get people to actually come to the door and open 
it. A couple of times, you knew somebody was home and on 
occasion people would say, ``Who is it?'' and I was finding it 
very difficult to get people to open their door. I would say, 
like, my name. Nothing happened. And then I would say, Roland 
Burris, and it just opened doors. It was just pretty amazing 
how it worked. [Laughter.]
    Senator Burris. Yes. That is how it----
    Senator Carper. Imagine their surprise when they opened the 
door. [Laughter.]
    Senator Burris. Mr. Chairman, I just had another couple of 
points here because I heard about the tough door-to-door, and 
we also now have a lot of foreclosed homes. And you won't be 
able to locate some of these people because you don't know 
where they have gone. Hopefully, you can find them. They have 
moved into rental units or somewhere. I mean, are they living 
with a parent or a friend or something? So that has to also be 
taken into consideration.
    Second, I just know that we in Illinois in our General 
Assembly have introduced legislation that states that the 
census would readjust the count for incarcerated persons back 
to their residence, not in the areas where the facility is. And 
we have several prisons in Illinois that were built in 
downstate, and 70 to 80 percent of those residents in those 
prisons are from Chicago. Those residents will count those 
prisoners as being a part of their community for census 
purposes and they will then try to get the revenue based on 
those calculations. I hope that we will look at that 
legislation--we are talking about in Illinois--and say that 
those persons are to be counted where they are going to go back 
to the community and put a drain on that community rather than 
being incarcerated and more than likely are not going to 
settle, for instance, in my home town of Centralia. My home 
town of Centralia has about 14,000 people in it. Well, in the 
prison, I forget what the prison population is now, but if they 
go and count those 800 or 900 prisoners, then that is going to 
make us have 15,000 people in Centralia and 90 percent of those 
are from Chicago. So Centralia will benefit. Chicago will not 
get the benefit of those people coming back.
    So, Mr. Groves, can you do this adjustment without 
legislation and they can make some type of adjustment in the 
count?
    Mr. Groves. This concern that you raised about where to 
count prisoners, we have heard from others. And we also realize 
that one of--a wonderful quality control process, we have to 
have outside people review counts, could use similar data. So 
for the first time, we will make available publicly, early in 
the process, in time to handle for redistricting purposes at 
the State level, counts of what we call group quarters. This 
would include prisons, but also dormitories and assisted living 
facilities, things like that, down to the block level.
    It is not exactly what you were saying. It isn't attempting 
to identify what the home place of a prisoner is. But for the 
first time, States will have available to them during the 
redistricting process the ability to identify prisoner counts 
down to block levels and use that however they wish----
    Senator Burris. That data is very readily available through 
State sources, of where that last residence was for that 
person. I don't see why that cannot be made an adjustment. My 
hometown has 14,000 people and there are 900 in the prison. Of 
those 900, 800 of them are from Chicago. Then it is not a 
14,900 population that is in Centralia. It is a 14,000 
population with probably another 50 of them coming from Decatur 
or Rock Island or Rockford or even from out of State.
    So I hope that there is some--even in this instance, 
because this is a complaint that I have heard about, the urban 
markets and how our smaller communities are taking advantage of 
these situations and the urban communities will need the money 
when these people return back to, for various allocations, and 
they are being shortchanged.
    So I am hoping that we can look at the adjustment, and I 
don't think this will take very much of--even if the census is 
taken, it shouldn't be very costly. You just adjust the 
numbers. Once the address is determined within that census 
tract area, you have 50 people in Centralia from that tract 
area, then you add that to the Chicago population, or the Rock 
Island population, or the Kankakee population, or the Joliet 
population. OK? And I am speaking for primarily the prisoners.
    Mr. Groves. Yes.
    Senator Burris. Of course, assisted living, the people are 
in the nursing homes and they are pretty close to home in those 
assisted living facilities.
    Mr. Groves. It might be good to describe how we do counts 
within prisons, because it is a special operation. We have 
something called an Individual Census Report. It is the 
preferred method of measuring within censuses because each 
person incarcerated would fill out their own form.
    In some prisoners, for security reasons, both security of 
our folks and others, the prison management says, we will take 
data off of our administrative records, the prisoner records 
themselves, and we do that when we are required to by the 
prison officials.
    Our findings are--and I don't know the Illinois record 
system, but I do know nationally, our findings are that across 
the States, the records miss large portions of the attributes 
we are trying to measure on the individual census form and so 
we greatly prefer to have people fill them out.
    So one of the problems that we face nationally with the 
issue that you raise is that States vary in the record systems 
of prisons, and what might be easy for one State to put in 
place to identify what the home of the prisoner is is very 
difficult for another State. That is our problem at the 
national level.
    We do one thing pretty well at the Census Bureau, I think, 
and that is after each census, we reevaluate what we call 
residence rules. Where do we count different people? One of the 
topics of great interest going forward for the 2020 Census is 
to use records systems more efficiently because that would 
reduce the burden on the American public and the cost of the 
data collection.
    So I believe that we must evaluate this placement, where do 
we place prisoners, along with how do administrative records 
consider different households. What does the address on the 
records system really mean going forward? And so I think we 
will.
    But unfortunately, we are not measuring right now--for your 
purposes, Senator, we couldn't do the adjustment that you are 
asking for because we are not measuring the home of an 
individual prisoner the way we are going about----
    Senator Burris. Yes, but when that prisoner is counted, Mr. 
Groves, it is added to that community.
    Mr. Groves. We follow with prisons the same rule that we 
follow for the vast majority of others, and that is we count 
you in your usual residence. Now, what do we mean by that? This 
is actually in the Census Act of 1790 that gave us this rule, 
and that means where people usually eat and sleep----
    Senator Burris. Sure, and for the prisoner, you are eating 
and sleeping in prison.
    Mr. Groves. That is why we count them where----
    Senator Burris. But that is no benefit--the community is 
benefiting by that in various ways, and one way is totally 
unfair to the community. We have 40,000 prisoners in Illinois. 
Sixty percent of them are from the Chicago community. You are 
talking about 24,000 individuals that are undercounted in 
Chicago and overcounted in those downstate communities. That is 
a number, an impact on that urban area, and we have already got 
urban problems as it is.
    Mr. Groves. And what I am saying, with this new tabulation 
for redistricting for the first time this decade, every State 
can decide how they are going to treat their prisoner 
population, whether they are going to keep them in the rural 
areas where they----
    Senator Burris. So that is a state-to-state decision?
    Mr. Groves. According to how we do redistricting, it is----
    Senator Burris. No, but then can that data be turned over 
to you and it be transferred to the particular urban area where 
these people are really housed? Will you make that count 
adjustment?
    Mr. Groves. If the State had those data certified in a way 
that the redistricting process found acceptable to that State, 
then they could do it themselves. We are not involved in 
redistricting, as it turns out. We provide the information that 
allows people at the State level to----
    Senator Burris. Now, I am not talking about redistricting. 
I am talking about the count of that head. I am sure these 
numbers might be used in redistricting, but they are also used 
for distribution of HUD funds and other various dollars and 
that goes into the count of what Centralia has. In order for 
Centralia to get the grant, it is based on what you turn in for 
the population of Centralia, as I understand it.
    Now, if I am wrong in that area, please correct me. But I 
think that you all ought to take a very hard look at how even 
nationwide, with over some two million individuals 
incarcerated, how they are being counted. And I think this is 
an issue that the census ought to take up in consideration and 
not just leave it to the States because you are getting a false 
report on the community from which this person is from because 
Centralia in no way is being--there is no use of the city 
streets. There is no use of the--because the State is paying 
for the water that Centralia provides it. So they are 
benefiting, and now they benefit unjustly because of the prison 
population.
    And I think that is something that you ought to look at 
specifically, not dealing with whether it is apportionment, not 
dealing with whether or not it is dealing with the 
redistricting. This is an area that we ought to look at for an 
accurate census count. Where does that prisoner actually have a 
home and where is that person going back to, and it will be a 
drain on that community as opposed to leaving Centralia and 
coming back to Chicago. I mean, that ought to be looked at, 
Director. I would suggest you do that.
    Mr. Groves. Well, I can promise you, going forward, we 
will. But I also must, in all honestly, note that from our data 
themselves, that won't be possible for the 2010 Census to----
    Senator Burris. GAO, have you all done any analysis--I 
mean, any comment in that regard?
    Mr. Goldenkoff. We have done some limited analysis, and we 
don't have a position on where prisoners should be counted. It 
is Congress' prerogative to work that out in consultation with 
the Census Bureau and the States.
    What I do want to point out, though, that there are some 
operational issues and feasibility issues that need to be kept 
in mind. One is that administrative records, as Dr. Groves has 
already mentioned, there is a lot of variation from State to 
State in the quality or completeness of administrative records. 
So that has to be considered.
    Another issue that needs to be taken into account is that 
there is, at least at present, no definition of a person's home 
on record. I mean, it is one thing to say, well, we should put 
them back where they came from. That is where they should be 
included in the census count. The issue there is, if someone 
has been in prison for a number of years, well, how do we know? 
Is that person going to go back to that location? Does that 
location, if they say, oh, their last previous residence before 
being incarcerated was 123 Main Street, does 123 Main Street 
even exist anymore? Was it a rental property that now somebody 
else is living in that same unit? So it would create--somebody 
would have to go out and verify that address still exists.
    Senator Burris. Well, certainly there is going to be some 
work in keeping records on it, but that person comes out of 
that jurisdiction, he is not going back to that jurisdiction, 
because the census is only done every 10 years. So there is 
going to be a big turnover in the prison population in 10 
years, and some of those people are going to be released during 
that 10-year period.
    Mr. Goldenkoff. Well, some will be, some won't be. Some are 
in Federal prison and won't even be going back to that State 
where they are housed. They are going back to some other State.
    Senator Burris. Well, I was thinking of this going to the 
40,000 that are in Illinois State prisons, as your example.
    Mr. Goldenkoff. OK.
    Senator Burris. You all could make a study by using 
Illinois because there are quite a few small communities where 
we have built in the last 20 years, they built 20 new prisons 
in Illinois.
    Mr. Goldenkoff. Right.
    Senator Burris. That is a prison a year. And they have been 
populated. Sixty to 70 percent of the urban areas, and then 
most of those people are five or--and for the lifers, that is 
another issue you have to deal with. Some of them might be 
getting life in those prisons, and then that would be 
difficult. But most of those prisoners are time certains and 80 
to 90 percent of them are going to be released. We just had 
1,100 of them released because of budget cuts, and where did 
they all head back to? They all headed back to Chicago. And it 
is a major drain on the urban community when that happens and 
the local community is taking advantage of that and that person 
is not even there anymore.
    Mr. Goldenkoff. Right.
    Senator Carper. I am going to ask you to go ahead and 
respond.
    Mr. Goldenkoff. It is something that requires further 
study. As I said, GAO doesn't have a position on it, but what 
we do have a position on is just what needs to be considered 
would be the operational and feasibility issues and it is 
something that will require further study, because as Dr. 
Groves said, it is really too late in this census cycle to get 
down to that level of detail.
    Senator Burris. I think that adjustments could be made, 
even in this cycle.
    Senator Carper. OK. Thank you, sir.
    I have a couple of short questions, and then if Senator 
Burris has anything else that he would like to add, he is 
welcome to, and then we are going to call it a day.
    One of the things I will be asking at the very end is 
anything that any of you want to add yourselves that comes to 
mind as a result of this conversation. Be thinking about that, 
but you will have that opportunity.
    Dr. Groves, we talked a little bit about this already, but 
could you give us some idea of how well recruitment and hiring 
are going, specifically with respect to bilingual communities? 
We all know not everybody in this country speaks English. Some 
speak it not very well. But there are 100 languages that I 
suppose are spoken in this country, easily. But I think you 
focus on maybe a dozen or so languages in terms of the ability 
to enumerate and to count folks.
    But how are we doing in terms of hiring folks that are able 
to go and be effective in bilingual communities? Is the Census 
Bureau having difficulty attracting enough eligible and 
qualified bilingual enumerators to work in these communities, 
and if so, are there any languages for which recruitment is 
lacking? If you are having certain problems trying to fill the 
slots that are needed for folks that can speak, maybe, less 
commonly spoken languages, what is the Census Bureau doing to 
address those difficulties?
    Mr. Groves. That is a great set of questions. I think the 
first thing to note is that we attempt to hire people from the 
neighborhoods in which they will be given assignments. So we 
want people who know the neighborhoods they will work in. This 
places a geographical constraint on the recruitment process.
    Nationally, we are in fine shape on counts. We are 
progressing better than we feared. When you go down to local 
areas, there are problems. It does appear to be the case for 
some groups, bilingual folks are the target that we want in the 
neighborhood. We are having some problems in rural areas 
greater than in urban areas through the recruitment process. I 
think that has been true in prior decades, as well.
    We are not, I don't think, in trouble on this. When I talk 
to the regional directors, they are concerned about bilingual 
recruiting, but remain optimistic that we will hit our goals.
    So the bilingual skills and the rural recruiting are the 
toughest areas for us right now, but we are actually--this is 
not what keeps me awake at night.
    Senator Carper. OK. Any steps that you all have taken at 
the Census Bureau to address either the bilingual or rural 
recruitment----
    Mr. Groves. We are advertising in in-language newspapers, 
some of the weekly newspapers that Senator Burris talked about, 
and we are advertising on in-language radio in areas. So there 
are a lot of census ads when we are after bilingual folks that 
use the media of that language as a way to do it.
    Senator Carper. I have seen some press--this is a change in 
the focus here a little bit, to ``hard to count.'' But we have 
seen some press reports that some Hispanic advocacy groups have 
launched a grassroots campaign calling for a boycott of the 
census unless immigration laws are changed. In my view, it is 
not, in the big picture, not an enlightened position to take, 
but nonetheless. But what strategies does the Bureau have in 
place to combat fears on the part of some immigrant communities 
that participating in the census will be harmful to them?
    Mr. Groves. We spend a lot of our time talking about this 
and working on this issue. It is an issue that arises in every 
census where the decade that preceded it had waves of 
immigration, people coming to this country from other countries 
with different cultures and different relationships to their 
central government are special educational targets for us.
    I think there are several things that we have learned over 
the decades that are appropriate in meeting this challenge. One 
is although I can go throughout the country giving speeches 
about the safety that is involved in participating in the 
census, the fact that you can't be harmed, while that is 
necessary, it is not sufficient and reaching out to these 
partner organizations is really key.
    We are so gratified by the organization of partner groups 
in a lot of in-language subculture groups, especially new 
immigrant groups. I was in Minneapolis just a few days ago 
where they have a large Somali population. You wouldn't imagine 
it. You wouldn't automatically associate that with Minneapolis. 
This was a long set of discussions. They have the concerns 
about whether those who might be undocumented need to fear that 
enforcement agencies would get this data, and we have a 
wonderful thing to say in this country about that, that we have 
very strong laws that protect that information from getting in 
the hands of any enforcement agency at the local, State, and 
national level.
    Repeating that was good, but when the Somali leaders of 
Minneapolis said that, it had different meaning than when I 
said it, and so the fact that we are using in-language ethnic 
media and using the leadership of those communities helps.
    We have hired partner specialists. The stimulus money 
allowed us to quintuple the number of partners specialists we 
have. We took them from the communities that were hard to 
count. They have their own network and ties, and that has been 
a wise decision in retrospect, I think, for us.
    So it is a challenge that never goes away. It is a question 
that arises all the time. Will I be harmed by participating in 
the census? We have a wonderful answer to give in the country, 
but all of us need to give that answer in a unified way.
    Senator Carper. Thank you. Staying on this point again for 
hard to count, in the past, I am told that the Census Bureau 
has worked with the Immigration and Naturalization Service to 
refrain from conducting raids during response follow-up. The 
thinking, I presume, is that if one arm of our government is 
actively arresting people believed to be here illegally, they 
and their friends and families might be less interested in 
helping another.
    What is the status of the Bureau' outreach to immigration 
officials on these issues, if you will?
    Mr. Groves. I believe the Secretary spoke to this, and I 
also did. We can't, as one Federal agency, ask another Federal 
agency to stand down on their mission, nor will that happen. 
Secretary Napolitano, in response to a query from one of the 
Catholic bishops, noted that the intent of her agency at this 
time was to not focus on individuals but to focus on other ways 
to fulfill their mission and that statement addressed some of 
the issues of concerns of these groups.
    Senator Carper. OK. Mr. Zinser or Mr. Goldenkoff, do you 
have anything you want to add on hard to count? You don't have 
to, but if you have anything that you want to add. Good.
    The last issue I want to raise, cost overruns. Dr. Groves, 
last week, the Commerce Department IG issued a report noting 
that the Census Bureau paid millions of dollars to temporary 
employees who never performed any field work and others who 
overbilled for travel. Let me just ask, how would you like to 
respond to the IG's findings? And as you begin to gear up for 
the upcoming field operations involving significantly more 
people, as we know, what types of internal controls do you plan 
to put in place or have you put in place to avoid these types 
of cost overruns in the future?
    Mr. Groves. Actually, the IG report focused on an operation 
that was done in the summer of 2009 and the results of their 
investigation on that. We testified with regard to that overrun 
in September, October--I have lost track of time. We learned 
from that overrun various things. One is that we are dealing in 
2009-2010 with a different labor market. We over-recruited, 
clearly, anticipating the labor market of 2000 and the 
attrition that was built into that kind of labor market. So we 
learned that lesson and we have adjusted our hiring and 
recruiting going forward.
    Second, I learned as a new director that the cost 
estimation process that led to the staffing decisions could be 
improved through updating components of the cost model, and 
that is what I testified to with regard to Non-Response Follow-
Up. Although it has been reported that we have a $2.3 billion 
cost estimate, we actually have a thousand cost estimates on 
Non-Response Follow-Up because there is no one cost estimate. 
We haven't yet seen what is going to happen that will produce 
the actual costs. As we see those events fall into line, we are 
going to narrow our range of cost estimates, and we will be 
completely transparent on this. I will tell you this as soon as 
we know it. So cost estimation is another thing that we are 
doing.
    I also want to point out that the address canvassing was an 
operation that was unique in the experience of the Census 
Bureau. Those handhelds had to work. There was great concern. 
There are probably testimonies in front of this Subcommittee 
that the handhelds weren't going to work at all. So our field 
folks were really quite concerned that they had enough staff to 
do the work. That is a natural concern of those involved in 
production processes that will never go away.
    I think our job is to make sure we are as cost efficient as 
possible, and I care deeply about this. It is notable that the 
two big field operations we have had since then have been on 
time and on budget or under budget.
    Senator Carper. That is encouraging.
    Mr. Zinser, Mr. Goldenkoff, anything you would like to add 
on cost overruns?
    Mr. Zinser. Thank you, Senator. I think in our report, we 
did try to put it in the right context, that these kinds of 
inefficiencies occur, and the message was with an operation 
coming up that is four to five times greater than what you just 
went through, attention to these areas is very important to 
cost controls.
    Senator Carper. OK. Thank you.
    Senator Burris, are you all in, as we say in the game, all 
in?
    I promised Dr. Groves, Mr. Zinser, and Mr. Goldenkoff, if 
you wanted to take a minute and just add a closing comment, 
this might be a good time to do that. If you don't want to, 
then I will wrap it up. Go ahead. Mr. Goldenkoff, anything you 
wanted to add or take away?
    Mr. Goldenkoff. No. I think that we are all in general 
agreement here on what the issues are, so in terms of anything 
new or dramatic that no one has ever realized before.
    What I would like to stress, though, is that the census, as 
large as it is, as complex as it is, it is an inherently 
fragile operation. It doesn't take much to derail it. So that 
is why, moving forward, there is really not a whole lot that 
can be done at this point, a new operation that no one ever 
thought of, a new action that no one had ever thought of 
before. So much is already being done.
    And so that is why what the Bureau needs to focus on going 
forward now that the data collection has started is real time 
metrics of these different operations so that they could see 
very early on, almost like an intel operation, intel on a 
ground operation, which essentially is what the census is, 
making that analogy to the military, what is working, what is 
not, and taking early and direct action to keep the operations 
on track.
    Senator Carper. All right. Thank you. Mr. Zinser.
    Mr. Zinser. Thank you, Senator. I think I would just say a 
couple of things. One is that things are going to go wrong. 
There are going to be problems. There have been problems every 
Decennial, and the trick is the management being able to 
respond to those problems. Especially when you have 600,000 
employees, you are going to have problems. I think it is 
necessary for the Census Bureau and the Department of Commerce, 
our office, GAO, the Congress, to all kind of try to see those 
problems and get on top of them.
    And the second point, I would just go back to Senator 
Burris's concern about the prison populations. I think it is 
difficult to make any changes to the way the Census Bureau does 
things right now, but I am wondering if the focus could be on 
the formula that is used to distribute some of these Federal 
funds, if there couldn't be some algorithm or some factor that 
can be used that takes into consideration how the populations 
of various cities are skewed based on other data concerning 
prison populations and attack the formula rather than try to 
change the way the Census Bureau is doing things. That would be 
my two cents on that issue.
    Senator Carper. OK. Thank you. Dr. Groves.
    Mr. Groves. Well, I want to make a plea to all of us to do 
everything we can over the next few days to tell friends and 
neighbors that this thing called the census is coming, that it 
is a chance for all of us to participate in a building block of 
the democracy, something that the founding fathers envisioned 
and told us to do every 10 years and we have done obediently 
since then.
    And that for those who are worried about the Federal 
deficit and Federal spending, this is the one thing, as you 
noted, Senator, that we can all do to save money. We really 
can. All you have to do is fill out that Census form and return 
it. If you are a private person and you don't want people 
knocking on your door, you can avoid this simply by taking 10 
minutes to fill out this form. All of the benefits of the 
census derive from that simple act, that 10 minutes that we are 
asking people to take, and now is the time, I think, for all 
the leaders of the country in a unified voice to say, come on. 
Let us do this. Let us do it together. Let us count ourselves 
and reapportion the House and get all the benefits from an 
accurate count that we can through our own participation.
    Senator Carper. All right. Thank you.
    Well, I want to conclude again by thanking you for joining 
us today. I want to conclude by thanking each of you for the 
work that the agencies or the entities that you represent have 
put into our efforts to date on this Decennial Census.
    This is certainly important work for our country and one 
that has been important work literally since our founding as a 
Nation. It is in some ways a lot more difficult than it used to 
be. But in some ways, it is easier because we have some better 
tools to enable us to attack these challenges.
    I want to just underline what we see our role as here in 
the Senate and in this Subcommittee, in particular. We have an 
oversight role to make sure that the Executive Branch is doing 
the work that they ought to be doing. We also have an 
obligation to try to find ways that we can be helpful, and to 
the extent that you need help, we have tried to be there to be 
supportive. If it is appropriations, if it is funding, if it is 
other resources that need to be brought to bear, we are trying 
to make sure that those are.
    I want to thank the IG and I want to thank our friends at 
GAO for being critical when it is appropriate, but I think, 
without exception, being constructively critical. There is a 
difference between those two approaches, so thank you for 
always being constructive.
    And for other things that Senator Burris and I and others 
on this Subcommittee and our full Committee that need to be 
doing in the days ahead, we certainly want to do that. We 
certainly have the opportunity through our own public comments 
to encourage people to stand up to be counted and to be proud 
of this opportunity, this constitutional obligation and 
opportunity.
    The hearing record will be open for 2 weeks. My colleagues 
who are not here will have the opportunity to submit their 
questions in writing. We would ask that you respond promptly.
    I want to thank both our majority and our minority staff 
for their work in preparing for this hearing and for everyone 
who has participated in it. Senator Burris.
    Senator Burris. I just think what Dr. Groves just said, 
that we should get that on tape and make a commercial out of 
it. That was a hell of a commercial that you just made for the 
people to fill out the census.
    Senator Carper. That is great.
    All right. With that having been said, this hearing is 
adjourned. Thank you so much.
    [Whereupon, at 4:19 p.m., the Subcommittee was adjourned.]

                            A P P E N D I X

                              ----------