[Senate Hearing 111-635]
[From the U.S. Government Publishing Office]



                                                        S. Hrg. 111-635
 
        2010 CENSUS: A STATUS UPDATE OF KEY DECENNIAL OPERATIONS

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


                                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

                             FIRST SESSION

                               __________

                            OCTOBER 7, 2009

                               __________

       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
53-847                    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 202ï¿½09512ï¿½091800, or 866ï¿½09512ï¿½091800 (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...............................................    12
    Senator Coburn...............................................    15
    Senator McCain...............................................    18
Prepared statements:
    Senator Carper...............................................    33
    Senator McCain...............................................    35

                               WITNESSES
                       Wednesday, October 7, 2009

Hon. Robert Groves, Director, U.S. Census Bureau.................     4
Hon. Todd Zinser, Inspector General, U.S. Department of Commerce.     6
Robert Goldenkoff, Director, Census Issues, Government 
  Accountability Office..........................................     8

                     Alphabetical List of Witnesses

Goldenkoff, Robert:
    Testimony....................................................     8
    Prepared statement...........................................    60
Groves, Hon. Robert:
    Testimony....................................................     4
    Prepared statement...........................................    37
Zinser, Hon. Todd:
    Testimony....................................................     6
    Prepared statement...........................................    50

                                APPENDIX

.................................................................
Questions and responses for the Record from:
    Mr. Groves...................................................    81


        2010 CENSUS: A STATUS UPDATE OF KEY DECENNIAL OPERATIONS

                              ----------                              


                       WEDNESDAY, OCTOBER 7, 2009

                                 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 3:04 p.m., in 
the Dirksen Senate Office Building, room SD-342, Hon. Thomas R. 
Carper, Chairman of the Subcommittee, presiding.
    Present: Senators Carper, Burris, McCain, and Coburn.

              OPENING STATEMENT OF SENATOR CARPER

    Senator Carper. The Subcommittee will come to order. 
Welcome one and all. Today's hearing is, as you may know, a 
continuation of our oversight of the Census Bureau's 
preparation for the 2010 Census.
    The 2010 Census is rapidly approaching us with Census Day 
less than 6 months away. On April 1, April Fool's Day, 2010, 
the Census Bureau will embark upon what many have described as 
the largest peacetime mobilization in American history. With 
nearly a $15 billion budget and an army of some 1.3 million 
census takers, the Census Bureau is responsible for ensuring 
that nearly 300 million residents here in this country are 
correctly counted.
    As my colleagues can probably guess, finding and accurately 
counting nearly 300 million individuals in the correct location 
is, of course, an extremely daunting challenge. Census taking 
has become even more challenging in recent years as our 
Nation's population has steadily grown larger, grown more 
diverse, and grown increasingly difficult to find. For a number 
of reasons, people may also be more reluctant to participate in 
the census.
    Last year, for instance, the Census Bureau encountered 
serious technological challenges that threatened to jeopardize 
the success of the 2010 Census. Since then, the Bureau has put 
forth great effort in putting the census back on track. I am 
told that the Bureau recently completed its address canvassing 
well ahead of schedule, thanks in part to a highly proficient 
staff.
    The handheld computers that have received so much negative 
attention in recent months performed as expected, and the 
Bureau has begun already to open local census offices 
throughout our country.
    Despite these successes, much more work needs to be done by 
the Bureau to put its operational plans in place and to execute 
them effectively. Connecting with young, mobile, and wired 
populations, establishing trust with skeptical populations, and 
integrating the major components of a complex operation are 
just a few of the challenges that lie ahead.
    Investigations conducted both by GAO and the Inspector 
General have revealed serious challenges with the contracting 
and implementation of key information technology systems at the 
Census Bureau. These reports have also noted the unreliability 
of the Census Bureau's cost estimate for the total 2010 
project. In the absence of adequate testing, the processes that 
will be used to follow up with non-responders is a serious 
concern as well.
    Given the sheer magnitude of an undertaking such as the 
decennial Census, a shortcoming in any one area can quickly 
have a domino effect on other operations. For example, a low 
mail response would increase the non-response follow-up 
workload, which in turn would increase the Bureau's staffing 
needs and drive up costs.
    We look forward today to the expert testimony of our 
distinguished panel of witnesses before us. It is my hope that 
today's proceedings will provide us with a clear assessment of 
the complications facing the Census Bureau and how Congress can 
best partner with the Bureau as it works toward achieving its 
goal of an accurate and cost-effective census in 2010.
    In closing, I would like to express my condolences to the 
family of William Sparkman. Mr. Sparkman, you will recall, was 
a census taker who was found in Kentucky a few weeks ago dead. 
Our thoughts and prayers are with him today and with his family 
today.
    I would like to thank all of our hardworking census 
employees who assist us in fulfilling our constitutional 
obligation of conducting this decennial Census and just to say 
to all of them through this panel that we value and appreciate 
their service.
    We have been joined by the Senator from Illinois, Senator 
Burris. I am going to recognize him at this time for any 
comments he might like to make. Welcome. Thank you for joining 
us.
    Senator Burris. Thank you, Mr. Chairman. I want to 
acknowledge our distinguished panel. But, Mr. Chairman, I beat 
my staff over here, so I am going to reserve my remarks because 
I have some deep concern about a package that my daughter 
received in the mail just last week, and my staff is bringing 
that material here.
    I am a former attorney general. If I am suspicious of this 
material, I can imagine what the public would be suspicious of. 
It is a 20-page document. It is a survey, and it says it is 
from the U.S. Census Bureau. And it says it is mailed to an 
address, but it says that you are required by Federal law to 
fill out this survey and to send it back.
    So I am going to have that in a moment when my staff gets 
here with it, and I would love to double-check with our census 
people to see, number one, if they mailed this out. If I am 
suspicious of this document survey, trying to find information 
about individuals, I am just wondering what the general public 
who may receive these in a mass mailing would be curious about 
as to whether or not this is really being put out by the U.S. 
Census Bureau in terms of a survey seeking information this 
early.
    Senator Carper. I am tempted to just go out and ask Mr. 
Groves to respond to this right now, but why don't we wait till 
we see the information, ask you to take a look at it, and then 
I think that would be perfectly appropriate to ask some related 
questions.
    Let me take a moment or two to introduce our first and only 
panel of witnesses. We welcome each of you today before us and 
thank you for joining us.
    Mr. Groves was nominated by President Barack Obama about 8 
months ago. It seemed like long ago, not that long, but he was 
nominated by President Barack Obama to be Director of the 
Census, in April and was confirmed by the U.S. Senate in July. 
Mr. 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 the Associate Director of the 
Census Bureau, Mr. Groves knows how the agency operates and 
what its employees need to successfully implement the decennial 
Census and other programs.
    Mr. Groves and I are today mourning the loss of the Detroit 
Tigers in yesterday's one-game playoff in the American League 
Central Division. We are both big Detroit Tigers fans, so we 
will mourn here together.
    Todd Zinser serves as the Inspector General for the U.S. 
Department of Commerce. As Inspector General, Mr. Zinser leads 
a team of auditors, investigators, attorneys and administrative 
staff responsible for detecting and preventing fraud, waste, 
and abuse in a vast array of business, scientific, economic and 
environmental programs administered by the Department of 
Commerce and its 13 bureaus.
    Mr. Zinser holds a bachelor's degree in political science 
from Northern Kentucky University. Where is that located?
    Mr. Zinser. It is right across from Cincinnati.
    Senator Carper. OK. Near Fort Mitchell?
    Mr. Zinser. Yes.
    Senator Carper. Yes. OK. A master's degree in political 
science from Miami University. Would that be Miami University 
in Oxford, Ohio?
    Mr. Zinser. Yes, sir.
    Senator Carper. Not many people guess that, do they?
    Mr. Zinser. No, sir.
    Senator Carper. All right.
    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 
in political science and master's of public administration 
degree from George Washington University.
    I am going to introduce first for his testimony Mr. Groves, 
and we have, at your request, allocated a bit more time, and 
you have 10 minutes to share your testimony with us. Try to 
stick within that time frame, and we will allow your entire 
testimony, and that of our other witnesses to be made part of 
the record and ask you to summarize as you see fit. Thank you. 
Please proceed.

 TESTIMONY OF THE HON. ROBERT GROVES,\1\ DIRECTOR, U.S. CENSUS 
                             BUREAU

    Mr. Groves. Thank you, Chairman Carper, Ranking Member 
McCain, and other Members of the Subcommittee. I am really 
happy to be here, and thank you for your interest in the 
census.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Groves appears in the Appendix on 
page 37.
---------------------------------------------------------------------------
    I want to begin the way you ended, Senator Carper. The 
census family was diminished recently by a death of one of our 
census interviewers. These are people who knock on doors. They 
are the engine of the data that we produce. And we are saddened 
by this event, and we send our condolences to his family.
    Upon my confirmation, I promised you and I promised 
Secretary Locke that I would spend the first month of my 
service examining the status of the 2010 Census. I did that 
with the help of former Census Bureau Director Kenneth Prewitt, 
and John Thompson, a former principal associate director.
    I consulted folks at the National Academy of Sciences who 
have been on our panels and its technical staff. I then called 
in some academics from around the country with relevant 
technical skills. I talked to my colleagues here several times 
and their staff, and then I reached out to our principal 
contractors and got input from them. And then finally, I had 
tons of briefings with the leadership of the census, and it is 
on that basis that I want to report my findings.
    Let me begin by talking as a survey methodologist and 
imagine that we wrote down the specifications of the 2000 
Census, and then the 2010 Census, and we compared the design 
features. I can tell you that I am firm in my judgment if you 
did that as an exercise that you would prefer the 2010 design. 
I have no doubt in my mind about this. I am sure my colleagues 
around the world would agree, and let me go through the 
reasons.
    The short form only census is a good idea that respects the 
burdensome job of filling out questionnaires and fits with the 
American public's busy life. The bilingual form that is going 
to 13 million households is a good idea. We learned that in 
2000 that it should increase cooperation among Spanish-only 
speakers. A replacement form will go out if you do not return 
your completed questionnaire. We know from years of research 
that this is going to improve cooperation.
    We have two questions on the questionnaire this decade that 
is going to be important because of the doubled-up housing 
problem related to foreclosures, and that is going to help us 
make sure that we have counted everyone in complicated housing 
situations.
    We have kept the address file up-to-date throughout the 
decade. That should make us a stronger file. And we are right 
now in the middle of a new operation to validate the group 
quarters list. This, too, should help us.
    Finally, the stimulus funding allowed us to increase the 
level of partnership in advertising activities. This is a key 
component in the success of a census design that does not have 
an adjustment feature. We have to reach out to the public that 
is difficult to enumerate.
    So on the basis of this, I am very settled in my judgment 
that we have a better design, but the design is not the 
implementation. And I want to speak to two kinds of challenges 
we have, one, an internal challenge and the other external.
    The internal challenges, first, we have a bright, well 
organized senior team, but they do not have the number of 
censuses under their belt that prior decades might have had. 
And to counter that, they have utilized a much more formal and 
structured risk management process, and that is helpful, in my 
judgment. I am going to keep on some of the external advisers 
we have had during this evaluation process to help me and help 
the team in ways that can fill in, and I think that will be 
useful to us.
    The second internal challenge is, as I mentioned in my 
confirmation hearing, the Census Bureau has experienced 
significant retirements in the senior ranks, especially 
mathematical statisticians. We need that kind of talent at some 
point during the census, and I will bring on some outside 
statisticians to help on this.
    Third, the movement from the handheld computer use to non-
response follow-up had an effect on the need to develop 
software for the paper-based operations during non-response 
follow-up. GAO, my colleague down the panel, has recognized 
that the testing of that system is a critical path on the 
census. We agree. I have reviewed the testing procedures. I 
have brought in outside folks to ask the question, Are we 
testing the right core subsystems of that? And I have been 
pleased with the judgment on that score. We have a big test 
coming up around Thanksgiving time, which will be a load test 
on that, which will be an important milestone that we are all 
watching on.
    We also established an internal review team that has 
already paid off. It contains our chief technology officer 
(CIO) from the Department of Commerce and other experts. They 
have already proposed three changes. We implemented them 
immediately. We have brought in IT security specialists as part 
of the software development. We have built a bridge between the 
internal Census Bureau developments and the outside contractors 
for integrative purposes, and we are adding other testing into 
the process. So we are on this problem. It is a tight schedule. 
We have to come through on this, and it is an internal 
challenge.
    The fourth risk we will soon know about, and that is how 
good is our master address file. Senator Carper mentioned that 
we just finished walking every street of this country building 
the master address file. We are right now in the middle of 
evaluating that. In a few weeks, we will know how that looks, 
and it is an important step for us.
    Then finally, and something I am really quite personally 
concerned about, it concerns cost estimation and control. We 
need better cost estimation and better cost control at the 
Census Bureau. As you may know, the address canvassing 
operation had an overrun attached to it. The overrun, I 
believe, is related to weaknesses in the cost modeling process. 
We are completely redoing the cost modeling for the non-
response follow-up stage, which is the biggest next thing we 
do, and I am happy to talk about that, if you wish, during this 
hearing.
    We also have some external challenges that I am concerned 
about. The biggest is the uncertainty about the mail return 
rate. The behavior of the American public in March and April of 
this year is a big uncertainty with regard to that. Scores of 
millions of dollars will be spent following up for houses that 
did not return the mail questionnaire. It is important to hit 
that target, to estimate that target well.
    There are problems in doing this. One, the vacancy rate of 
American households is greater now than in prior decades. This 
is a concern for us because this has cost implications. Two, 
more and more families are doubled up in single-family houses. 
This is a concern to us because the burden of filling out a 
questionnaire for more people is larger and may affect return 
rates. Three, the rate of homelessness is greater, and there is 
a new class of homelessness in this country that we are new to 
as a data collection organization. And then finally, the public 
debate and the tension over immigration issues is ongoing, and 
this may affect the mail return rate among new immigrants.
    We are examining all the data we have in house to try to 
estimate this. As you may know, we continue to see declining 
participation rates in our sample surveys. This is a danger 
signal to us with regard to how people will behave come March 
and April.
    The second external challenge is the new media environment. 
The blogosphere causes us problems in getting our facts out. 
And third, we are worried about computer phishing that may 
affect people's understanding of what is going on.
    I made a variety of changes to the design that I am happy 
to talk about. We have altered the design of the coverage 
measurement survey. We have added a master trace project. We 
will have an Internet test in this census, and we will do a 
post hoc administrative record census.
    We have some things upcoming. I want to warn the 
Subcommittee--and you can see this in my full testimony--that 
we have a variety of operations in the fall that are tightly 
scheduled. Some of these may have bumps. We will be transparent 
about those bumps. We will tell you when things are not going 
well. We are hopeful that things will go well. As Senator 
Carper mentioned, this is a sequential process, and each 
successive stage needs to go well for the overall census to go 
well.
    Thank you.
    Senator Carper. Right on the money. Thank you.
    Mr. Zinser, you are recognized. Please proceed.

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

    Mr. Zinser. Thank you, Mr. Chairman, Senator Coburn, and 
Senator Burris. I appreciate the opportunity to testify today 
on the Census Bureau's readiness for next year's decennial 
Census. Oversight of the 2010 Census has been a top priority of 
my office since 2002. Today I would like to summarize my 
testimony by making three points.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Zinser appears in the Appendix on 
page 50.
---------------------------------------------------------------------------
    First, the dye is cast on the decennial. No real 
opportunity remains to affect the 2010 decennial planning or 
make major course corrections. My office is focusing our 
oversight on the status of high-risk areas to see whether the 
implementation of existing planning, system development, and 
operations are on track. Key areas include the automated paper-
based operations control system, validation of the address list 
for group quarters, the communications contract and partnership 
program, and various enumeration activities.
    My second point, Mr. Chairman, is that the decennial Census 
is the ultimate schedule-driven program with all of the 
inherent risks and consequences that you get with large, 
complex, schedule-driven projects. At issue is the continuing 
potential for rushed and incomplete requirements; time pressure 
to cut corners in program design, development and testing; 
uncontrolled cost growth and increased operational risks and 
quality risks.
    For example, while handheld computers were able to support 
address canvassing, the Census Bureau discovered that the 
handhelds could not support canvassing of large blocks. This 
caused the Bureau to implement a quickly developed contingency 
operation. And in an upcoming report, we will discuss 
limitations that prevented using the handheld computers to 
correct addresses filed late in the quality control process to 
add missed housing units or delete duplicates. These problems 
can reduce the accuracy of the address file.
    Moreover, the switch to paper for non-response follow-up, 
while a necessary decision, introduced schedule challenges of 
its own. Most problematic is the development of the paper-based 
operation control system which must now be developed, tested 
and deployed in one-third of the desired time frame.
    My third point, Mr. Chairman, is that there is one thing we 
can have confidence in, and that is the dedication of the rank 
and file census workforce. If you visit the census staff or a 
local census office, the dedication of the staff is clearly 
evident. In our view, a key factor in overcoming the setbacks 
and management lapses experienced is the focus and commitment 
of the census workforce, and this perhaps more than anything 
else increases our confidence in the success of the decennial 
Census.
    I would also like to commend the Subcommittee for its 
prompt action in confirming the director. The presence of a 
permanent director during the final period leading up to census 
day has immeasurable benefits. But no matter how successful the 
outcome, the many continuing risks associated with getting to 
the finish line for 2010, the unacceptable cost growth, and the 
unknown toll on census staff working to cobble together a 
massive operation this late in the decade make it imperative 
that the experience of the 2010 Census not be repeated.
    To this end and given where we are with the 2010 decennial, 
many of the recommendations that we have made represent lessons 
learned and look ahead to the 2020 decennial, which the Census 
Bureau has already started working on.
    For example, the budget process for 2012, which begins next 
year, already is critical for laying the foundation for better 
cost controls and fundamental improvements, including a more 
robust research program.
    In summary, the Census Bureau is well on its way to 
completing the count next year, yet with major operations still 
to come and an immovable deadline, ongoing oversight remains 
critical. We commend the Subcommittee for its oversight, and we 
commend the efforts of all those in the Bureau, the Department 
of Commerce, and Congress who are working to make the decennial 
Census a success.
    Mr. Chairman, that is my summary of my testimony. I would 
be happy to answer any questions.
    Senator Carper. Thanks, Mr. Zinser. Mr. Goldenkoff, please 
proceed.

  TESTIMONY OF ROBERT GOLDENKOFF,\1\ DIRECTOR, CENSUS ISSUES, 
                GOVERNMENT ACCOUNTABILITY OFFICE

    Mr. Goldenkoff. Mr. Chairman, Members of the Subcommittee, 
I am pleased to be here today to provide a progress report on 
the U.S. Census Bureau's implementation of the 2010 Census.
---------------------------------------------------------------------------
    \1\ The prepared statement of Mr. Goldenkoff appears in the 
Appendix on page 60.
---------------------------------------------------------------------------
    As in March 2008, we designated the 2010 Census a high-risk 
area, citing a number of longstanding and emerging challenges, 
including weaknesses in the Bureau's IT management, problems 
with the performance of handheld computers that are used to 
collect data, and uncertainty over the final cost of the 
census, which is now estimated at around $14.7 billion.
    Overarching all of these concerns was the lack of a full-
dress rehearsal which limited the Bureau's ability to 
demonstrate critical enumeration activities under near census-
like conditions as well as the lack of time to complete 
remaining work. Collectively, these issues raised questions 
about the Bureau's readiness for the 2010 Census.
    As requested, my remarks today will focus on the Bureau's 
preparedness for the head count, paying particular attention 
to, first, the rollout of key IT systems; second, our 
preliminary findings on the results of address canvassing; and, 
third, the Bureau's progress in improving its cost estimation 
abilities.
    My main point today is that the Bureau continues to make 
noteworthy progress in mitigating risks and in keeping the 
decennial on track. Further, we are encouraged by the seating 
of a presidentially-appointed Census director this past July as 
well as by the experienced advisers he has put in place to 
assist him.
    That said, a number of challenges and uncertainties still 
need to be addressed. The bottom line is that while there have 
been a number of positive developments over the last few 
months, the 2010 Census remains a high-risk area because of the 
amount of work that still needs to be completed under a very 
tight time frame, as well as for the inherent uncertainties in 
managing such a complex enterprise.
    With respect to IT, the Bureau has made important strides 
in improving its oversight of this critical function. For 
example, the Bureau named an official to monitor the testing of 
various IT systems and has strengthened other aspects of its IT 
management.
    Still, we found that in some cases, requirements, and 
testing plans have not been finalized, and it is difficult to 
gauge progress because of vague metrics.
    Senator Carper. I am sorry. Because of what?
    Mr. Goldenkoff. Vague metrics. Further, several areas such 
as the control system that will be used to manage complex field 
operations face aggressive testing and implementation time 
frames. If any significant problems are identified during 
upcoming tests, there generally will be little time to resolve 
them before the system needs to be deployed.
    Address canvassing, an operation where temporary workers 
known as listers go door-to-door to verify and update address 
data, finished ahead of schedule, but, as was already 
mentioned, was over budget. The Bureau estimated that address 
canvassing would cost around $356 million, but estimates of the 
actual cost are $444 million, an overrun of $88 million or 25 
percent.
    A key reason for the overrun is that the Bureau did not 
update its cost estimates to reflect the changes to the address 
canvassing workload. Further, the Bureau did not follow its 
staffing strategy and hired too many listers.
    The Bureau's efforts to fingerprint employees, which is 
required as part of a criminal background check, did not 
proceed smoothly, in part because of training issues. As a 
result, over 35,000 temporary census workers, over a fifth of 
the address canvassing workforce, were hired despite the fact 
that their fingerprints could not be processed and they were 
not fully screened for employment eligibility. The Bureau is 
refining its instruction manuals and taking other steps to 
improve the fingerprinting process for future operations.
    With respect to costs, we are unable to verify the accuracy 
of the $14.7 billion cost of the 2010 Census because key 
details and assumptions are unavailable. The Bureau is taking 
steps to improve its cost estimation process for 2020, 
including training its staff in cost estimation skills.
    In summary, while the Bureau has taken a number of actions 
to mitigate risks and its overall readiness for 2010 has 
improved, considerable work remains to be completed under very 
tight time frames. Although the Bureau is to be commended for 
its efforts to get the census back on track, a successful 
census is a daunting challenge and even a small setback or a 
misstep could potentially derail the census.
    In light of this challenging operational environment, high 
levels of public participation and continued Bureau and 
congressional attention to stewardship, performance and 
accountability will be key to success.
    Chairman Carper, Members of the Subcommittee, this 
concludes my remarks, and I will be happy to answer any 
questions that you might have.
    Senator Carper. Thank you, Mr. Goldenkoff.
    We have been joined by Senator Coburn. Senator Coburn, both 
Senator Burris and I made brief remarks to open. Would you like 
to do that before we go to questions?
    Senator Coburn. I will wait.
    Senator Carper. OK. Fair enough. Thank you.
    Let's just start off by asking a question of Mr. Groves. 
This turns out a pretty timely question.
    But there is an amendment on the floor today, on the 
Appropriations Bill which involves the Commerce Department, 
that would prohibit the use of census funds for programs that 
do not include questions regarding U.S. citizenship and 
immigration status. And while I have some concerns about how 
this amendment might impact public participation in the 
decennial Census, I just want to ask Mr. Groves what thoughts 
you may have about the amendment. I think a copy of it has been 
provided to you today to consider; is that correct?
    Mr. Groves. I have not seen the amendment.
    Senator Carper. OK.
    Mr. Groves. I have not seen the amendment, but let me 
comment----
    Senator Carper. No, it is short. Let me just read it to 
you, all right?
    Mr. Groves. OK.
    Senator Carper. On page 110, line 7, ``Strike the word 
activities and insert activities provided further that none of 
the funds provided in this Act, or in any other act, for any 
fiscal year, may be used for collection of census data that 
does not include questions regarding U.S. citizenship and 
immigration status.''
    That is the thrust of the amendment.
    Mr. Groves. I would open with several observations. In 
March 1790, the Census Act was passed that specified that all 
residents, citizens or not, would be counted in decennial 
censuses at the place where they usually reside. And every 
decade since then, we have done this.
    Senator Carper. I am sorry. Every year since when have we 
done this?
    Mr. Groves. Since 1790.
    Senator Carper. OK.
    Mr. Groves. So it is indeed the case that if you read the 
history of this, of proposals to change that arise from time to 
time; this is not the first such proposal.
    There are logistical issues. It is now October 7, 2009. We 
have printed over 100 million forms; that is, the 
questionnaires approved by Capitol Hill. It does not have this 
question in it, and we are implementing that both because of 
this philosophy to minimize the burden on the American public 
and in concert with the tradition in the law of counting 
residents whether they are citizens or not.
    Senator Carper. All right. A more general question.
    You have been in your position now for, what, about 2 
months?
    Mr. Groves. Something like that.
    Senator Carper. About 2 months.
    Mr. Groves. It seems longer.
    Senator Carper. Yes, I am sure. I know you have been busy. 
And would you say, given what you have seen, are you encouraged 
about the ability of the Census Bureau to do the job that is 
presented to you by next year?
    Mr. Groves. Yes. One way to say this in a crisp way that is 
more memorable, I think if we knew right now two things that we 
do not know, but we will know soon, then I would have full 
confidence. I could even say something like it is sort of up to 
the American public now. That, we cannot control. And the two 
things are this software development that is going on that we 
are looking at very carefully. It must work, where it is very 
tightly scheduled.
    Senator Carper. What kind of software development, for what 
purpose?
    Mr. Groves. As a result of not using the handhelds for the 
non-response follow-up stage, we are going to use paper. We are 
writing software to keep track of all these pieces of paper and 
make sure we go only to the houses that have not completed the 
questionnaire. That is a set of software systems.
    Since that was a late change, the software development is 
late. The work is going on night and day, I can tell you. So 
far, so good, but it is very tightly scheduled. We will know 
more around Thanksgiving time because we have a big load test 
that we are doing that has been carefully laid out to answer 
the question is it up to snuff? And then we will have some 
other releases.
    Will that paper-based operation control system software 
work as desired?
    Senator Carper. And roughly, when do you expect we will 
know the answer to that question?
    Mr. Groves. Thanksgiving is a big date for us.
    Senator Carper. OK.
    Mr. Groves. We will know more there. Then the first release 
will be in early January. One good thing is we will have some 
early tests of it on a small level to find some bugs.
    The second thing is the quality of the master address file. 
This is the list of addresses that will produce the mail-out 
questionnaires. A lot of the quality of the 2010 Census will 
depend on the quality of that file. We have the file. So far, 
so good. It looks pretty good in terms of its overall size, but 
we are doing all sorts of other diagnostics on it. And in a 
matter of weeks, I think by the end of this month, we will have 
those diagnostics in hand and I will feel better, or worse, 
about how that looks.
    Those two things in place, those are the key uncertainties 
internally. And the biggest uncertainty is something that we 
all worry about, and that is how will our fellow residents of 
the country respond when we send this questionnaire out. That 
determines a lot of subsequent operations.
    Senator Carper. OK. Another follow-up question, if I can. 
This is with respect to IT systems testing. And you certainly 
have spoken to this, but I want to pursue this line of 
questioning just a little bit further, if I could.
    I believe since 2005, the GAO has reported on weaknesses in 
the Bureau's management of its IT acquisitions and management 
in the tasking of key decennial systems. Given the numerous IT 
management weaknesses that GAO and the Inspector General have 
found in the Bureau's IT acquisitions, what steps are you 
taking agencywide, Dr. Groves, to improve oversight and control 
of the Bureau's IT management?
    Mr. Groves. So I take it this question goes beyond just the 
decennial Census. It is really an organizational thing.
    Senator Carper. Yes.
    Mr. Groves. We have a new CIO, Brian McGrath, and we are 
meeting often about the structure of IT within the Census 
Bureau. We have also brought him in--he is the chair of this 
group, this internal oversight group of this software 
development. It is a wonderful integrative tool for a new 
leader of IT because he has seen these operations as they go 
on.
    There are various things that I think need to be put in 
place. One has to do with the moving picture of IT security 
during software development. The IT security regulatory 
environment is changing almost daily as the threats to IT 
security are changing. And we need to get better at this. We 
need to build in IT security as part of the software 
development, and indeed, that was one of the first changes we 
did to this paper-based operation software development.
    The other has to do with uniformity in IT operations, and 
those are the kinds of things we are discussing; can we 
structure IT operations and software development more 
carefully, more efficiently? Can we save money? Can we do the 
same things for less money?
    Then I think the third area has to do with the use of 
outside contractors for software development; how do we do that 
better than we have. When we have software development that is 
going to be done outside of the Census Bureau, what is the 
talent pool that we need inside the Census Bureau to do that 
well? It is not merely sending money to an outside contractor. 
You need talent inside to produce good software done by people 
outside.
    So those are the kinds of things we are focusing on.
    Senator Carper. All right, thanks. My time has expired. Dr. 
Coburn, please.
    Senator Coburn. I will yield to Senator Burris.
    Senator Carper. OK. Fair enough. Senator Burris.

              OPENING STATEMENT OF SENATOR BURRIS

    Senator Burris. Thank you, Mr. Chairman.
    I have since received my material, and, Dr. Groves, I also 
learned that this is something, evidently, that is sent out by 
the U.S. Department of Commerce; that it arrived at my 
daughter's address in Evanston, Illinois just as a no-name. But 
it is a form saying that this information comes from the 
American Community Survey. And naturally, being a former 
attorney general, I have dealt with a lot of mail fraud and 
identify theft, and I was not so sure that this----
    Senator Carper. You have done it or you have investigated 
it?
    Senator Burris. I have investigated it. [Laughter.]
    Senator Carper. For the record, I wanted to make that 
clear.
    Senator Burris. Thank you, Mr. Chairman.
    This is a survey of about 20 pages. I heard all this 
information about single-page census form. And so my daughter 
brought this over the weekend, and actually, when I was reading 
my briefing last night, I called her and said, ``Would you fax 
me a copy of that thing that you supposedly received from the 
Census Bureau?'' And she sent us that, and then I had my staff 
to try to run this down.
    But this thing is a survey. It says you have to fill it out 
by law. It is not census information. It is a survey that is 
done randomly, and it is mailed out randomly.
    Are you familiar with this, Mr. Groves?
    Mr. Groves. Yes, I could----
    Senator Burris. And I call myself being halfway 
knowledgeable. My daughter who is a Ph.D. did not know what 
this was, and you are mailing this out randomly without any 
knowledge to the public.
    How do they know what they are supposed to do with this 
thing? And they tell you that if you do not fill it out and 
send it back, you are violating the Federal law.
    You have to do a better job, wouldn't you, in getting 
information out to the public that this thing is coming 
randomly to your home and your address and that--you should 
really have a better explanation for it. I thought it really 
was a fraudulent document until my staff did the research and 
talked to somebody at the Census Bureau, because it just was 
not adding up. In all my knowledge dealing with the census, I 
have never seen one of these things, and no one else that I 
know of have seen one of these. And so do you-all expect people 
to send these back?
    Mr. Groves. Yes. I could give you the background on this 
particular survey. Maybe the first thing to say is that our 
little organization, in addition to doing a census every year, 
every decade, does thousands of sample surveys. We do the 
unemployment, the current population survey, which produces the 
monthly unemployment rate that the Bureau of Labor Statistics 
releases. We do retail trade surveys. We do all sorts of 
surveys. So we are actively engaged in measurement in all sorts 
of ways.
    On this particular survey, actually, this Subcommittee 
played a role in this survey in a real way in prior years. This 
is related to the decennial Census in its origin. In 2000, in 
addition to what we call the short form, which is now this 10-
question questionnaire, about a sixth of the households got a 
questionnaire that sort of looks like that one. And in Census 
Bureau jargon, that was called the long form.
    Senator Burris. This is long. It is about 20 pages.
    Mr. Groves. That is right. And one thing we discovered over 
the decades is that the rate of people responding to that long 
form during the decennial Census was much lower than the rate 
responding to the short form, but every one of those questions 
in that long questionnaire has a statute underlying it that 
requires the Census Bureau to collect those data in order that 
Federal programs be administered.
    So instead of doing this once a decade, this American 
Community Survey was invented on a sample basis rolling 
throughout the decade. And hopefully, your daughter received a 
letter in the mail that said you have been selected as part of 
this survey, this is the purpose of the survey, and look 
forward to getting this questionnaire. It is going to come in a 
few days. And then she would have gotten the questionnaire. And 
hopefully, the questionnaire and the letter has, 800 numbers if 
she was worried about the legitimacy of it to get more 
information.
    Senator Burris. My reaction to the 800 number was--I did 
not call it, but maybe I should have. You would call it and you 
would get somebody saying yes, this is a survey and it would 
all be part of the scam.
    Mr. Groves. So the other thing, the letter that she would 
have gotten has the Census Bureau letterhead on it, and she 
could call for the information number.
    Senator Burris. And you are putting that burden on a 
taxpayer to go through all of this without some type of 
promotional advertisement information----
    Mr. Groves. Yes, let me speak to that. The promotion side 
is something that we have as a challenge throughout all of 
this. In fact, every one of our surveys faces this issue, how 
do we alert samples of people. About 2 million households get 
that a year out of the 134 million households. It is a small 
sample that gets it; how do we effectively advertise when a 
small sample gets the questionnaire, and that is a constant 
challenge for us. I accept that, criticism that we need to do 
better there.
    Senator Burris. How about it is just an address? I mean, 
how can you hold somebody legally liable if you just send 
something to the address, not directed to any party? So who is 
going to be violating the statute if it is not addressed to a 
party?
    Mr. Groves. Let me tell you how this works. We actually do 
not know the names of people in the country. We do have a list 
of the addresses. So we mail this to the address. If a 
household did not return that questionnaire, we will attempt to 
contact them by telephone in order to take the data. A lot of 
people mistaken or misplace the questionnaire and we follow up 
by phone. And if we cannot reach them by phone, a sample of 
them, we will actually knock on the doors of those households 
and seek to collect the information from them that way.
    At the end of this process, about 98 percent of the 
households are measured that fall in this sample. So if you ask 
the question do people do this, about 98 percent of them do 
this.
    Senator Burris. Has anyone ever been prosecuted for failure 
to do this?
    Mr. Groves. Not on the American Community Survey, no, not 
to my knowledge.
    Senator Burris. So would you take a look at this in terms 
of how we better can inform the public of this? I mean, if you 
have to go through those steps, the people knowing what this is 
or knowing that it is valid, it would appear to me there would 
be a better effort on the part of the Census Bureau to send out 
or either to----
    Mr. Groves. We can always----
    Senator Burris [continuing]. Come up with some system that 
would alert people. I mean, I do not consider ourselves as the 
brightest people in the world, but I do not think my daughter 
and I are the dumbest people in the world. And I was taken 
aback because I was looking for a Census Bureau short form. It 
is coming so close to the census. And I said, well, the short 
form should not be coming out yet. I mean, it is too early. So 
it was coming at the time that the census is coming out, and 
this form is coming out. I said, well, somebody must be playing 
some games, so let me look into this. And I did not know I was 
going to be at this hearing today. But, Mr. Chairman, it surely 
worked out to give me the proper information in reference to--I 
guarantee you now she is going to fill it out and send it back 
because----
    Mr. Groves. I would appreciate that.
    Senator Burris. Now that you know that she received one, 
she might be the one to get prosecuted for failure to return 
it.
    Mr. Groves. I hear what you are saying, and I appreciate 
it. Thank you.
    Senator Burris. Thank you, Mr. Chairman.
    Senator Carper. Thank you, Senator Burris.
    We have been joined by Ranking Republican Member, Senator 
McCain. Welcome. We are delighted that you could come, and I am 
not sure whether I should yield to you or Dr. Coburn who is----
    Senator McCain. Well, the doctor is in, so why don't we 
yield to him, Mr. Chairman.
    Senator Carper. All right.
    Senator McCain. Thank you.
    Senator Carper. Let's do that. Thanks.
    Senator Coburn. Mr. Groves, what did it cost to print 100 
million copies of the short form?
    Mr. Groves. I do not know the full printing----
    Senator Coburn. Does anybody in your staff know that 
answer?
    Mr. Groves. We can get this answer. How much?
    Mr. Mesenbourg. All I know is $22 million under the 
original estimate.
    Mr. Groves. The original estimate was $22 million. The 
current printing contracts are coming below estimated costs 
right now.

              OPENING STATEMENT OF SENATOR COBURN

    Senator Coburn. Well, first of all, I am very glad you are 
where you are. Thank you for serving. We had a great 
conversation in my office prior to your confirmation.
    We are now at $48 per person and growing to count people in 
this country. Just so my colleague from Illinois knows, I got 
one of those forms last year and I refused to fill out half of 
it, because I found it offensive and I found no basis in law 
for them to force me to answer those questions. I did return 
it, and I returned the pertinent information that the 
government should have an interest in, but I did not answer all 
of the questions.
    It is called the long form, and there is nothing in the 
Constitution, in the first article, that gives the power of the 
census the ability to do that. It does give the ability of the 
Federal Government to do enumeration but not to go into the 
detail of which the long form goes. So therefore, I stand 
guilty of not filling out completely the long form and will 
defend anybody that does not want to do that. It is a totally 
different matter when it comes to the short form, however, 
because we do have a responsibility to have a count.
    I want that number, and the reason I want that number is I 
am a co-sponsor of that amendment, Senator Carper. When the 
census was started, it is for resident citizens. It is not 
residents. The count is to be for citizens. The amendment that 
is going to be offered, which I am sure will be defeated, has a 
great impact and what the goal of everything you do is about, 
which is how do we apportion representative government in this 
country.
    I want to enter into the record, actually based on what we 
know statistically, in California, if we do not just count 
citizens, California will have five extra seats that they, in 
fact, do not deserve on resident citizens. Illinois will have 
an extra seat. Indiana will lose a seat. Iowa will lose a seat. 
Louisiana will lose a seat. Michigan will lose a seat. 
Mississippi will lose a seat. New York will gain a seat. North 
Carolina will lose a seat. Oregon will lose a seat. 
Pennsylvania will lose a seat. South Carolina will lose a seat, 
and Texas will gain two seats.
    So there are two parts to this hearing. One is how are you 
doing, and I sleep a whole lot better at night knowing you are 
there. The second part is to do the intended aspect. I want to 
make one point. In your long form that I fill out, it asks the 
very question that Senator Vitter wants to have asked on the 
census, are you a citizen of the United States. I mean, you ask 
that in the long form. Every one that gets mailed out gets 
asked that question.
    I believe you would answer that affirmatively, correct?
    Mr. Groves. That is in the American Community Survey.
    Senator Coburn. Yes, that is in the long form. So we do not 
have any problem asking it a million times a year in the 
country, but when it comes to the very purpose for which we do 
a count, which is to apportion the States, we have conveniently 
decided we are not going to ask that question. It is not about 
partisan issues, and it is really not about State. It is about 
doing what our Constitution says.
    I understand the concerns on the other side of the aisle, 
and I understand how that is. But States are going to apportion 
their representation based on what this gentleman and his 
department does. It ought to accurately reflect the true intent 
of the founders as is reflected in the long form. We are so far 
over budget on this census. We are $8 billion. We are 120 
percent greater than what the last census is now, and what do 
you think it is going to end up costing? Do you think the $14.6 
billion is right?
    Mr. Groves. I can tell you what my aspiration is.
    Senator Coburn. What is your aspiration?
    Mr. Groves. My aspiration is that the American public 
returns this questionnaire at higher rates than we have ever 
seen before and I give back and my associates give back 
hundreds of millions of dollars to the Treasury that we have in 
contingency funds, not knowing that number. That would be 
wonderful.
    Senator Coburn. OK. Well, I think there is a great way for 
us to help you do that. I think one is to get this amendment in 
so that we reestablish confidence in the government. And then 
two is to enroll us as Members of Congress to tell people how 
important it is to fill out this form because not only is it 
their duty as a citizen so we get an accurate count, but it is 
their duty to help us save money by returning the form.
    You were not present when Senator Carper and I had most of 
the discussions years ago on this issue. You and I had 
conversations about how we should plan for 2020 and the fact 
that we have it online. If you think about it, we are at $48 a 
person right now and counting. You can give everybody in the 
country a $25 stipend if they will just fill it out online for 
you, and we would still be less money than what we are going to 
spend right now.
    So what I want us to do is to think forward as we learn 
from the mistakes that the GAO and the IG have done a great job 
in terms of trying to direct this.
    The software development programs you have ongoing now, 
which are critical to you carrying out a successful--I believe 
everybody agrees with that. That is a critical piece right now. 
Is that a cost-plus contract?
    Mr. Groves. These are Census Bureau employees who are doing 
this work and----
    Senator Coburn. None of that is outside?
    Mr. Groves. Yes, the big change in the re-plan because of 
the handheld problem was to bring this inside, and so these are 
people who are literally about 30 feet from me. I see them 
every day.
    Senator Coburn. OK. So we do not have any outside vendors 
now doing any of this integration?
    Mr. Groves. Well, we still are relying on outside 
contractors for pieces of software that were part of the 
original agreement.
    Senator Coburn. And were those cost-plus contracts?
    Mr. Groves. These were--these have incentive schemes in 
some of them and that----
    Senator Coburn. But they are not true cost-plus contracts? 
They are a fixed-price contract with an incentive?
    Mr. Groves. The key ones are like that, yes.
    Senator Coburn. OK. All right.
    So do you agree with the IG and the GAO, the real problem 
in the $88 million overage, do you agree with their assessment 
in that the reason we came in 25 percent over budget was the 
two critical factors, one, planning, and two is employing 
people that you did not need?
    Mr. Groves. I would say it is----
    Senator Coburn. And I may have heard that wrong.
    Did I hear that wrong?
    Mr. Goldenkoff. No, I think that covers it.
    Mr. Groves. I think we agree, but I would say it slightly 
differently because there is another component that I find 
interesting and more diagnostic about what we should do to 
clean things up.
    The amount of work was larger than anticipated. You could 
say that should have been anticipated. You could debate that. 
The composition of the workload was different. In what way? 
Well, there were more deletes. We went out to a place, and 
there was not a house there. OK? More than anticipated. And 
when you diagnose that, so why didn't we hit that right, that 
is about $30 million of the----
    Senator Coburn. $38 million.
    Mr. Groves. I put the blame on cost modeling strategies, 
and I think of this as sort of a top-down cost model versus a 
bottom-up cost model. That scared me because I am worried that 
non-response follow-up may be subject to similar logical 
errors.
    We have a new team in there. We have a new top-down cost 
modeling for non-response follow-up, and we are building a 
bottom-up cost model. And we are going to look at how they 
agree or disagree. I do not anticipate that they will agree. 
But what usually happens is when you see the nature of the 
disagreement, you learn something about the assumptions of one 
or another of the procedures. We are doing that now because we 
have to hit that right, and I am worried about that.
    Senator Coburn. Well, I am well over my time. Thank you. 
Are we going to have another round?
    Senator Carper. Yes, we are.
    Senator Coburn. All right. Thank you.
    Senator Carper. Senator McCain.

              OPENING STATEMENT OF SENATOR MCCAIN

    Senator McCain. Thank you, Mr. Chairman, and thank you for 
holding this hearing.
    Dr. Groves, I thank you for the good work you do, and I 
enjoyed our meeting that we had in my office, and I am 
especially appreciative of your candor.
    I think you made a step in the right direction by 
terminating the Bureau's partnership with ACORN. I cannot tell 
you how many of my constituents contacted me about that issue. 
And as you noted in your termination letter to ACORN, the 
Bureau's partnership with ACORN had ``indeed become a 
distraction from our mission and may even become a 
discouragement to public cooperation, negatively impacting the 
2010 Census efforts.''
    Doctor, is the partnership terminated permanently or just 
for the census?
    Mr. Groves. I am not sure. We do not have partnership----
    Senator McCain. That last over the----
    Mr. Groves. Yes, facilities in other ways.
    Senator McCain. When an organization is terminated, is 
there a review process they have to go through if they want to 
become eligible to partner with the Census Bureau again?
    Mr. Groves. Every partnership proposal is reviewed 
internally by our staff on all of the dimensions that you just 
mentioned.
    Senator McCain. I was somewhat surprised that you have 
close to 80,000 partnership agreements with national and local 
groups.
    Mr. Groves. Our aspiration is to get that way over 100,000.
    Senator McCain. And that is wonderful, but how do you 
monitor that many partnerships?
    Mr. Groves. We have over 3,000 partnership specialists in 
local census areas. Their job actually is to identify areas 
where we need to get the word out in ways that could be 
effectively done with local neighborhood, city level 
organizations that are trusted voices. So they do that outreach 
and working with them, collaborating with them.
    Senator McCain. So you are confident that there is 
sufficient oversight of the partnerships that you have, that 
money is not misspent and there is not abuses?
    Mr. Groves. The partners do not receive money from us. The 
partners agree to get the word out about the census for their 
particular group.
    Senator McCain. Let me say this. So you are confident that 
they do not misuse their partnership with you?
    Mr. Groves. This is part of our oversight of those 
activities.
    Senator McCain. I am told by the doctor that you do pay for 
response follow-up?
    Mr. Groves. Sure. Do you mean when you do not return the 
form?
    Senator McCain. Yes.
    Mr. Groves. We then hire--these are different than 
partnership people. Partners do not do this work. These are 
census Federal employees who go through all the screening of 
all Federal employees.
    Senator McCain. Mr. Zinser, you made in your opening 
statement that the initial life cycle cost was 11.5 years but 
it is now 14.7 years.
    What is your degree of confidence that 14.7 years is the 
final number?
    Mr. Zinser. Thank you, Mr. McCain. I think a lot of the 
cost factors are still unknown, exactly how high that is going 
to go. Based on the variables that Dr. Groves talked about, we 
do not know what the response rate is going to be. We do not 
know if there are problems with this paper-based operation 
control system that require contingencies. That could increase 
the cost of the census. So I think at this point it is just 
unknown, sir.
    Senator McCain. Mr. Goldenkoff.
    Mr. Goldenkoff. We would agree with that, and there are two 
issues there, an internal one and an external one. Internal is 
the quality of the Bureau's cost estimate itself and the 
Bureau's ability to do adequate cost modeling. What we found is 
that with the cost estimates, they lacked a sensitivity 
analysis. There was very little in the way of documentation 
that we could look at, so it was very difficult for us to 
verify how good that number was because there was not a whole 
lot to look at.
    Senator McCain. So your confidence level about there not 
being further cost overruns is not strong?
    Mr. Goldenkoff. No, not at all. And then externally, of 
course, as was said, the key is the response rate. If they get 
a higher than expected response rate, obviously, that will 
bring costs down.
    But another key unknown, part of the way the Bureau hopes 
to save some money on non-response follow-up is by removing 
late mail returns. If you send your questionnaire in late after 
the Bureau makes that initial cut of non-response follow-up, 
rather than send an enumerator out and then find out that you 
have already mailed it in, they will actually cross your name 
off the list. Well, that approach has never been tested before, 
and so that is an unknown. If that does not work, then there 
could be a lot of unnecessary visits to housing units.
    Senator McCain. And you are obviously aware of this 
somewhat bleak assessment, Dr. Groves?
    Mr. Groves. Yes, I am aware of that bleak assessment.
    Senator McCain. But perhaps, Doctor, you could----
    Senator Carper. Just repeat your response.
    Mr. Groves. I am aware of that bleak assessment.
    Senator McCain. Perhaps you could for the record maybe 
submit to the Subcommittee some of the actions that you are 
going to take since it is clear that these issues are before 
us, that we need to at least salvage what we can. I mean, these 
are tough times in America and to already have at least a $3 
billion cost overrun and more to come, I think we deserve 
tighter controls or better estimates, one of the two.
    Senator Coburn. Would the gentleman yield for a second?
    Senator McCain. Sure.
    Senator Coburn. I want to make sure that Dr. Groves is not 
held accountable because most of the consequences of the cost 
overruns we are seeing today, his hands are not on. And matter 
of fact, I would just tell my colleague that because he is 
there, it is probably going to cost less than what it would 
have had he not been there.
    So I think he should try to respond to us, but he still 
needs to be recognized that his fingerprints are not on the 
problems that he inherited.
    Senator McCain. Well, as I mentioned before, I appreciate 
Dr. Groves not only for his willingness to serve but his candor 
about the difficulties he faces. But it still does not relieve 
us of the requirement to ask you to give us a plan as to how 
you can minimize the damage, a lot of which was inflicted 
before you came.
    Mr. Groves. I would be happy to do that and can do that 
orally or in writing, whatever way you want.
    Senator McCain. I think perhaps in writing, if it is 
agreeable to you. I know you are incredibly busy, or if you 
would rather do it orally, whichever. I would rather make it 
easier for you.
    The other thing I do not understand--help me out here, 
Doctor. We now have means of communications that were unheard 
of as short a time ago as the last census, OK? I mean, we have 
these devices with us everywhere. Everyone has them. We can e-
mail instantaneously at no cost, literally, certainly not the 
cost of a long distance phone call or even when we only had 
access to wired telephones. Every business, every industry in 
America that has adopted these new technologies have 
experienced dramatic cost savings. That is why they are so 
popular.
    Why in the world wouldn't we have adopted some of these 
technologies, which give us the ability to not only communicate 
initially. If I e-mail somebody, they are going to e-mail me 
directly back. If I send them a letter and ask them to fill out 
a form, it is going to be a week, 2 weeks, a month, whatever, 
not to mention all the associated costs with it.
    Why couldn't we and why can't we employ technologies which 
allow us to communicate in breathtaking fashion with our 
citizenry in the conduct of this census?
    Mr. Groves. There were decisions made before I arrived in 
this town that are relevant to our challenge that are too late 
to undo, in my belief. We are adding a test of Internet use as 
an experimental component because I agree with your premise.
    I can say that looking forward to 2020, and I have said 
this, that I cannot imagine--it is not--I do not have the 
capacity to imagine a 2020 Census without utilizing this 
technology. In the same breath, though, we all have to 
acknowledge that none of us know what the Internet of 2020 is 
going to look like. And therein lies a request for all of us. 
We need a 2020 developmental process that has faster cycle 
time. We cannot lock in to technology so early in the decade 
that it is old by the time----
    Senator McCain. Is it really too late, since we are 
conducting a census in 2010, to not employ some of these 
technologies?
    Mr. Groves. In my judgment, the answer unfortunately is 
yes, it is too late for some of these things, depending on how 
they are used.
    Senator McCain. Would you set up some test programs?
    Mr. Groves. We do have a test implanted, I am happy to say, 
that will be mounted in August 2010 to examine how people 
respond on a Web survey version of the short form versus a 
paper version. It is a critical component on this. But to get 
in place an Internet option, which I know several of you are 
interested in, in my judgment would be so risky now that it 
could hurt other things.
    Senator McCain. You cannot make it up, that we have this 
technology and have had it for a number of years that 
incredible communicating capabilities, and we are still mailing 
people letters. I notice you are nodding your head, Mr. 
Goldenkoff.
    Mr. Goldenkoff. Well, the census that we are taking today 
is essentially the same census that we took back in 1970 in 
terms of we mail out a bunch of forms and the population mails 
the forms back.
    But I would agree with Dr. Groves that it really is too 
late to employ an Internet option now mainly because it is a 
lot more complicated than just putting a version of the 
questionnaire up on the Internet, a digital version of a paper-
based questionnaire.
    Senator McCain. Despite the fact that people pay their 
bills, conduct all their lives over the Internet----
    Mr. Goldenkoff. No, that is correct. I think a key question 
for the Census Bureau, is why can't the Census Bureau do what 
IRS already does, what Amazon already does and thousands of 
other organizations----
    Senator McCain. Or even voting in some places.
    Mr. Goldenkoff. And people send very sensitive information 
across the Internet. But to do it at this late date because, 
for example, as the information needs to be kept confidential, 
so how do you protect the data? How do you archive it for over 
70 years? How do you authenticate the people who are 
responding? How do you know that if I am responding to a 
questionnaire, that I am an actual household member?
    So those are things that really need to be worked out and 
probably not something we want to do in just the few months 
remaining. I think the Bureau has enough on its plate at this 
point, but it is certainly something for continued testing for 
2020----
    Senator McCain. Did you want to add something?
    Mr. Zinser. Senator, I would. We recommended in 2006 that 
they take college dormitories and test responses over the 
Internet with college dormitories. That remains an 
unimplemented, open recommendation, and you might have to put 
it in the law, sir. You might have to pass a law to get them to 
do it.
    Senator McCain. Well, maybe we ought to consider that, Mr. 
Chairman, seriously.
    And, Dr. Groves, if you would like to make a contribution 
to posterity, and I mean this in all sincerity, look forward 
and give us an outline and some plans for how we can utilize 
existing technologies in 2020, much less the technologies that 
we think will be available in the future.
    So it is just kind of a shame that we are not going to have 
as reliable a census as we could have had if we had used 
existing technology to carry out, really, one of the 
fundamental requirements of democracy. And so, Doctor, I am 
sure you have a lot to do, but I hope that you will start 
thinking about that as well. I may not be around. I am afraid 
Dr. Coburn will be, but anyway, I hope----
    Senator Coburn. It is a race at the end.
    Senator McCain [continuing]. We can work it out. There you 
go. Do you have any comment?
    Mr. Groves. Well, I can promise you, Senator, that this is 
foremost on my mind because, as was stated earlier, the use of 
these technologies is not just quality related, it is cost 
related. And the inflation of the costs of the decennial Census 
is of paramount concern to me. So I can give you a solemn 
promise that we are on this one. But for 2010, unfortunately, 
it is too late, in my judgment, to do something.
    Senator McCain. Thank you very much.
    Senator Carper. The line of questioning that you have just 
pursued, Senator, reminds me of a line of questioning that Dr. 
Coburn and I pursued over a span of several years with Dr. 
Groves' predecessor. And it is a source of real frustration. 
And we just heard literally the line of questioning all over 
again. I am encouraged that the----
    Senator McCain. Mr. Chairman, maybe Dr. Groves' 
recommendation, maybe we should write it in law in some way if 
we can figure out a way to frame it so that it does not 
restrict the use of future technologies. That might be the 
challenge.
    Senator Carper. One of the questions I had to ask, Dr. 
Groves, during your confirmation process was really a question 
just along the line that you are asking, to gauge for myself 
his interest in making sure that we do the next decennial in 
2020; we get the number as accurate as we can, but we do it in 
a cost effective way. And I think we will all agree that the 
technology that we are using for the 2010 is not the most cost 
effective way.
    I am encouraged that we have at least an embedded test, a 
demonstration that we can build on. The important thing is that 
we build on it and you let us know what help you need from our 
end in order to make sure that we do use the technology that is 
available the next time out.
    I think it would be great if we could not only have a 
census in 2020 that we could rely on in terms of the numbers of 
people we are counting but also actually be coming in not at $3 
or $4 billion more than the last census but actually a couple 
billion dollars less. And speaking of aspirations, that would 
be a good one for all of us.
    In terms of cost drivers, Dr. Groves, my sense is one of 
the cost drivers--I almost said Dr. McCain but Senator McCain 
was asking about, certainty or assurance that we feel about, is 
it really going to be the number--what is it, $14.7 billion? 
How confident do you feel about it?
    One of the real cost drivers here is the--correct me if I 
am wrong--but it is the number of households who actually 
respond to the mail-out. That is the key. And can you give us 
some idea, for every percentage point, extra percentage point, 
that we have to go out and go door-to-door with enumerators and 
question the people who have not responded to the written 
survey--there is a cost associated. And I have heard the cost 
before. But can you tell us what that might be?
    Mr. Groves. Well, the current number--this is a number I am 
scrubbing, too. But the current number is between $80 and $90 
million for every percentage point decline. So why does it cost 
that much? Well, you have to hire more people to knock on more 
doors, to travel more miles, and it is human costs on that.
    Senator Carper. All right. Thank you.
    A question, if I could, for Mr. Zinser and Mr. Goldenkoff, 
please. And it is similar to one we have already asked Dr. 
Groves.
    But, gentlemen, given that there is very little time 
remaining to rest and develop the paper-based operation control 
system, what kind of problems are we talking about having if 
the system development and testings run late?
    Mr. Zinser. If they run late, the first problem, I think, 
is that it shortens the time necessary to actually train the 
people that are going to be using the system. So the first 
problem you have is if they run late, they may get the system 
in place, but people have trouble using it. If it is later than 
that, then you have half a million or more enumerators out 
trying to conduct non-response follow-up operations, and the 
Census Bureau does not have a contingency plan right now in 
place on how they are going to manage that work force without 
this automated system.
    Senator Carper. All right. Mr. Goldenkoff.
    Mr. Goldenkoff. Right. I would echo that 100 percent. If 
the software development runs late, it will affect the time 
that managers would have to familiarize themselves with the 
operation and using the software. If any changes need to take 
place in terms of the system or how to operate the system, that 
means more cost in terms of sending out errata sheets and the 
training manuals. So there is this sort of ripple effect for 
the downstream operations.
    Also, it could affect the need for employees to do work-
arounds if those instruction manuals are not updated in time. 
The census employees start doing their own thing. So it could 
have both operational impact and cost impact.
    Senator Carper. All right. Another question, if I could, 
for you, Dr. Groves. We talked already a little bit about non-
response follow-up. But let's talk about for a minute or two 
about contingency planning, and what are the Census Bureau's 
contingency plans for addressing, we will say, a much lower 
than anticipated mail response rate next spring?
    Will the Census Bureau be prepared to increase the number 
of census takers substantially if that is needed? Would you 
modify your basic media campaign in order to get to target 
areas of the country where the response was especially low?
    Mr. Groves. There are, I think, a couple of things to know 
about this. We have the return rates from 2000 day by day that 
we can track. So even though the mail return rate in 2000 was 
about 67 percent, on April 1 of 2000, it was 57 percent. So it 
is an interesting fact that people tend to turn in their forms 
early.
    We are going to have this day by day. We will know if we 
are falling behind at that point. And we have held back some 
money in the paid media campaign as kind of contingency money 
to target it. We will actually know where, what areas are 
coming up shorter than other areas. We can re-target money 
pretty quickly. Now, it will be late. People will have forms in 
their house that they are not filling out, but that would be 
one thing.
    The second thing to note is that this is a very different 
labor market in 2010 or 2009 than it was in 2000. The 
unemployment rate, although creating great suffering for our 
country, has a benefit to the Census Bureau. The quantity and 
quality of applicants that we have seen both in address 
canvassing and in this operation going on now is unprecedented. 
These are people that are highly educated, highly skilled, 
highly motivated. They are putting in more hours than we got in 
2000.
    So I think various things would happen if we had a lower 
mail return rate, and the input of new advertising did not help 
it. We would first utilize the existing staff and get more 
hours out of them. Second, we would hire more. The hiring would 
probably have some of our experienced people become supervisors 
rather than enumerators. We would kind of develop a more 
hierarchical structure to take the supervision in place. And 
the important thing, I think, we will know this early. We will 
know this by April 1 certainly, if we are in trouble.
    We start interviewing May 1, the non-response follow-up, 
and we will have a month, right, as the work-around month. 
April will be a big month for us if that happens.
    Senator Carper. All right. Let me yield to Senator Burris.
    Senator Burris. Thank you, Mr. Chairman.
    Dr. Groves, I need to follow up on what Senator McCain just 
asked you. What was the Census Bureau's relation with ACORN?
    Mr. Groves. ACORN was one of these 80,000 partners.
    Senator Burris. And what do you mean by partners? What did 
that consist of?
    Mr. Groves. Yes. You could go to our website, and everyone 
who is interested in becoming a partner fills out a form. The 
form--the whole purpose of the partnership is to find trusted 
voices in diverse communities around the country to get the 
word out. There are various things that partners can do. This 
is not a contractual relationship. This is a voluntary 
relationship on the part of the partnership organization.
    Senator Burris. So who made the decision to sever the 
relationship with ACORN and on what basis did you make that 
relationship?
    Mr. Groves. I made that decision.
    Senator Burris. Based on what?
    Mr. Groves. I made the decision because we are kind of in 
constant contact with our regional offices, and we were 
learning that the recruitment of other partners was inhibited 
by our partnership with ACORN. We were having trouble getting 
other trusted voices in communities because of this. The 
distraction that ACORN was causing in our own operations was 
sufficient to say that it was hurting the----
    Senator Burris. Because of the erroneous news report that 
there were a couple of ACORN people who may have violated the 
law?
    Mr. Groves. Those news reports were part of that----
    Senator Burris. Based on that so-called Fox sting that was 
perpetrated, which ACORN now is suing some of those people 
about.
    Mr. Groves. That was part of it. There were other----
    Senator Burris. Did you give ACORN a hearing? Did you call 
anybody in and question that relationship at all?
    Mr. Groves. We talked to ACORN about this matter on the day 
we made the decision. We informed them before going public.
    Senator Burris. You informed them, but did you question 
them about their relationship or you just informed them that 
you had severed the relationship based on news reports?
    Mr. Groves. Now, I want to say again that we are in contact 
with our regional offices whose job it is to reach out to other 
partners, recruit other partners to help us in this endeavor--
--
    Senator Burris. Dr. Groves, you said you canceled the 
relationship. I am not talking about your other partners.
    Mr. Groves. That is correct.
    Senator Burris. You advised me that you canceled the 
relationship.
    Mr. Groves. That is correct.
    Senator Burris. So did you have any type of conference with 
them before you canceled it?
    Mr. Groves. We had a conversation with them before we 
canceled it.
    Senator Burris. But you had made up your mind that you 
would cancel it based on news reports. That is what I assume. 
That is what I hear you say.
    Mr. Groves. No, that is incorrect, Senator. Let me restate 
it.
    We get information from Census Bureau regional offices 
throughout the country who are now seeking and recruiting other 
partner organizations. We are in contact with them, and we are 
asking them how things are going. We had reports that 
recruiting other partners because of the distraction that ACORN 
was causing was causing them----
    Senator Burris. Describe the distraction for me, sir.
    Mr. Groves. These are reports when they are reaching out to 
form other partnerships that people said, gee, I am not sure I 
want to partner with Census because of the----
    Senator Burris. Do you have any documentation on that?
    Mr. Groves. What, documentation on what?
    Senator Burris. These other partners notifying you that 
they do not want to partner with the Census because the Census 
was partnering with ACORN.
    Mr. Groves. These were telephone conversations among Census 
Bureau staff. I doubt if we have documentation, written 
documentation on that.
    Senator Burris. So what you are telling me is you get a 
couple people calling in--so I can take an organization--I can 
get 20 people to call into your organization and say that the 
XYZ group that you are partnering with, I do not want to 
partner with them. And, therefore, you would then turn around 
and cancel some other contract based on that?
    Is that what you are telling me, Doctor?
    Mr. Groves. If a partner organization inhibits us from 
continuing to get the word out about census nationally, yes.
    Senator Burris. OK. I am not one of those that support your 
canceling the arrangement with ACORN because I think ACORN is 
getting a very unfair assessment based on some agendas that 
have nothing to do with service to the community and probably 
the very best grassroots organization you could have helping 
you in these difficult communities. Especially in the minority 
and undercounted communities would be an organization with the 
reach of an ACORN. And I hope that you would look at that 
again, and I will be following up with you and probably would 
like to sit down with you individually and with some ACORN 
members to find out why this action has been taken based on a 
television news report that had another agenda. I am concerned 
about that, Mr. Groves, and I do not think that is a fair 
indication of any type of action for the Census Bureau to cut 
out an organization that has its roots in the--because I am 
concerned about the undercount in the African American and the 
Hispanic communities or even the poor communities. And the best 
group that you can get in those communities is someone who has 
their action or their feet on the ground or their ear to the 
ground in those communities is an organization such as ACORN.
    Mr. Groves. The people that ACORN reaches out to and serves 
are very important for the quality of the census. We need the 
participation of all of those constituents, and we are striving 
to form partnerships in every locale that are trusted voices 
for those communities. And I would be happy to sit down with 
you.
    Senator Burris. Yes, I think we should. And I will have my 
staff to follow up with you because I am deeply concerned about 
this one-sided attack on an organization that really is at the 
heart of trying to deal with the problem that we are dealing 
with. And I am very familiar with ACORN in Illinois.
    I know any organization has some rotten apples in the 
barrel. I do not want to give my whole speech here about some 
of these government contracts and some of these other 
organizations, that cause somebody in the military personnel to 
be killed over in the war zone because of shoddy contracts. But 
they were paid billions of dollars of taxpayers' money. We get 
an organization that in 15 years gets $53 million, $2 million a 
year, from the Federal Government, and because of one TV 
report, they end up being just ostracized and criticized and 
turned out by government agencies. And I do not agree with my 
colleagues on the Senate vote to cut off the funds because I 
voted in opposition to that amendment to cut off the funds, and 
I do not agree with that.
    I have got to sit down with you and just find out what 
basis we are putting out an organization that can be very 
helpful in those communities in dealing with an undercount in 
getting into those households and reaching out for those 
communities that can give us what we need to make sure we get 
our population counted. Thank you, sir.
    Mr. Groves. [Nods affirmatively.]
    Senator Burris. Thank you, sir.
    Senator Carper. Thank you, Senator Burris. And I appreciate 
your willingness to sit down with Senator Burris and have some 
further discussion. That would be, I think, a good idea. Thank 
you.
    I think this is going to be the first census in which the 
Bureau will fingerprint our temporary Census employees. 
Previously, they were only subject to, I think, a name 
background check.
    Can you provide an overview for us of the screening process 
that the Bureau uses to examine the employment suitability of 
temporary fieldworkers?
    Mr. Groves. I would be happy to, Senator. It starts with a 
so-called name check, where every applicant's name is submitted 
to FBI data resources, and we check whether there are any 
problems connected with that name. After that point, those who 
pass that name check and are in the applicant pool, and let's 
say all of the other aspects of their application, gets them to 
training.
    On the first day of training, they are fingerprinted. And 
as you noted, this is a new step. We did not do this in 2000, 
and it has upped the scrutiny of the applicant pool for 
criminal background.
    There are two fingerprints that are done by two different 
finger printers. So they put these on cards, and so each 
applicant has two cards independently done by two different 
folks. Those are then scanned in our national processing center 
in Jeffersonville, Indiana and forwarded electronically to the 
FBI for checks.
    We did this in address canvassing, and there are various 
key quality issues that arise in this. One is how many of those 
fingerprints or what proportion of the fingerprints that we 
take can be read electronically at the backend by the FBI. We 
went into this thinking that would be about 70 percent or you 
could say 30 percent failed reads. The address canvassing 
experience was that that was about 22 percent of those that 
could not be read.
    The FBI processed those fingerprints on average in about 22 
hours, and for those that had a negative report, it took about 
8 days to get that. So that is a process that gives us on 
average kind of an 8-day time between taking the fingerprints 
and seeing whether there are any problems. And we are now doing 
this in operation and group quarters validation.
    We found various things that we did not like about that. We 
do not like the 22 percent rate. We would like that to be 
lower. There are procedures, several procedures, that are going 
on. The FBI--there are various things that are just logistical. 
Some people have drier fingers than other people. If you have 
real dry hands, fingerprints do not take as well. So we are 
using right now some lotion that the FBI suggested to pick up 
the prints better to try to reduce that rate. We have changed 
the training procedures for people taking the fingerprints. All 
of these things are driven towards trying to get that rate 
lower and trying to be more efficient in this.
    Senator Carper. Let me just follow up, if I can. I think a 
little more than one out of five for address canvassing 
employees had, as we say, unclassified fingerprints and were 
only subject to name background check. The largest operation is 
non-response follow-up.
    In looking forward, let me maybe ask Mr. Goldenkoff. In 
looking forward, does GAO have any concern about solely relying 
on name check for people with unclassifiable prints?
    Mr. Goldenkoff. Yes, because as was observed during address 
canvassing, name check alone failed to identify people who are 
unsuitable for Census employment because of a criminal record. 
There were 1,800 people actually who passed the name check but 
were caught using the fingerprinting approach. And of those 
1,800 people, about 785, roughly, were disqualified for Census 
employment because their criminal background made them 
unsuitable. It was serious enough.
    Senator Carper. Give us some examples of why someone might 
have been deemed unsuitable.
    Mr. Goldenkoff. Manslaughter, rape. So certain crimes if 
they happened a long time ago, if they were minor, that is OK, 
but, some of these were very serious crimes. And so if you take 
the percentage of those people who were unsuitable for Census 
employment out of those who were not caught by name check 
alone, it is possible that--we estimate that around 200 
people--of those individuals whose fingerprints could not be 
classified, it is possible that 200 of those had a criminal 
background that would have otherwise made them unsuitable for 
Census employment.
    Of course, because non-response follow-up is a much larger 
operation, until the Bureau figures out how best to deal with 
people with unclassified fingerprints, then you are going to 
have that many more people who would be unsuitable for Census 
employment working non-response follow-up.
    Senator Carper. OK. I want to shift the focus just a little 
bit before we conclude by coming back to the issue of the 
undercount. In May, our Subcommittee held a field hearing up in 
Philadelphia on how the Census Bureau plans to address the 
challenges of obtaining a complete and an accurate counting, 
particularly in urban communities in the 2010 decennial. At 
that hearing, I learned that in my very own home state of 
Delaware, nearly 12,000 residents were missed in the 2000 
Census.
    What makes these areas particularly concerning is their 
differential impacts on various subgroups. Minorities, renters 
and children, for example, are more likely to be undercounted 
by the census while more affluent groups such as people with 
vacation homes, families with kids in college out of State are 
more likely to be enumerated more than once.
    Dr. Groves, let me just ask specifically what strategies 
does the Bureau plan to implement or is trying to implement in 
the months ahead to maximize participation within historically 
hard to count populations living in urban communities?
    Mr. Groves. Well, as you know already, this is one of our 
central concerns, and one of the things I have been doing over 
the last few weeks after I did this risk assessment is to visit 
the regional offices. And there are interesting things 
happening there under the direction of this new design, and 
they are very small area targeting operations. So we have these 
things called census tracts that you can think of as about 
4,000 units in a tracts. And every tract in this country is 
being scrutinized by partnership specialists and operations 
specialists, and there is a special plan for every tract.
    Now, this is relatively new because we have targeting data 
now. We have this wonderful database at the tract level. We did 
not have this in the past. So for every tract in the country, 
there is a plan. The plan in the best of worlds is tailored, 
customized to that neighborhood. Sometimes it involves reaching 
out to certain neighborhood community groups. Sometimes it is 
an observation that there is a newspaper, a local newspaper, 
that hits a language group that populates that tract.
    So every plan is slightly different. The difference on this 
is that we are trying to tailor our methods to the problems at 
hand and doing different things in different places. I am 
hopeful about this. It is something new for us. It sounds 
right. It fits the notion of what we have learned over past 
decades, and that is a big effort that we are engaged in.
    Senator Carper. The downturn in our economy has resulted, 
as in an increase in foreclosures and vacant housing units as 
well as persons that are doubling in households--I think you 
might have said that earlier today--and living in group 
quarters.
    What steps are being taken by the Census Bureau to try to 
better ensure that it fully counts individuals and families who 
have been maybe uprooted during the course of the recession and 
because of the foreclosure crises?
    Mr. Groves. I think there are two components of that 
population that we are concerned about. One are the doubled-up 
houses that you mentioned and the newly homeless because of 
this problem. So let me talk about those separately.
    On the doubled-up thing, part of the communications, I 
think, effort that we are targeting tries to remind people that 
we count them where they usually reside, and we want people to 
know in a double-up house that we want the original residents 
and the new residents there when they do not have a usual 
residence somewhere else. So that message needs to be 
reinforced given what has happened as a society.
    On the newly homeless, especially on the West Coast, we are 
worried about folks who were actually quite well off a couple 
of years ago who are now homeless, and we are very interested 
in learning how to approach them. Many of them are living in 
cars, the only possession that they have retained. And we want 
to make sure that we are sensitive to their needs and sensitive 
to their concerns when we approach them, and that is a new 
population we have to worry about. So we are talking to people 
about how to deal with it.
    Senator Carper. OK. Thanks. Maybe one or two more and then 
I am done.
    What are the Bureau's plan for counting Hurricane Katrina 
evacuees, those that are living involuntarily outside of the 
Gulf Coast because their neighborhoods still, after all these 
years, have not been rebuilt?
    Mr. Groves. I visited New Orleans a few weeks ago to meet 
with local officials there who are very concerned about this. I 
had been there a year earlier, and New Orleans, I can say, has 
a sense of optimism that was not there a year earlier.
    Senator Carper. Well, that is encouraging.
    Mr. Groves. And building is happening, and they look 
forward to a bigger city. We, as Senator, count people where 
they usually reside. Some of the people who aspire to be back 
in New Orleans in the coming months are not there now. We are 
doing something extraordinary. We have changed our methods in 
several areas in the Gulf Coast. This affects Saint Bernard, 
Plaquemines, and Orleans Parish around New Orleans where we 
will hand deliver forms. We want to hand deliver those forms 
because we want to make sure that addresses that were 
unoccupied in summer or unoccupied in fall but occupied in 
April will get a form and can report their residency there.
    For those who are not in those areas on April, we must say 
by law the census is a snapshot of where people are usually 
living around April of the census year. We must rely on this.
    I have talked to officials there about the need in the Gulf 
Coast and other areas to make sure we update our population 
estimates throughout the decade to reflect their success in 
rebuilding these cities, and I think that is an important 
obligation as a statistician. And the country needs to follow 
the progress of those rebuilding efforts, and we have, as 
something called a special census. We can redo the census in an 
area. We have done over 300 special censuses in this decade, 
and we can do that if funds are forthcoming. We also have a 
population estimates program that updates population size.
    Senator Carper. Thank you. I am going to call on each of 
our witnesses. I will probably start with Mr. Goldenkoff. I 
will make you a doctor yet.
    Mr. Goldenkoff. I have an honorary degree. Thank you.
    Senator Carper. I will ask you if you have any closing 
thoughts given what we have talked about here, sort of the 
interaction, the responses that the questions we have asked and 
the responses you have given; if there is anything else you 
want to add before we conclude.
    Before I do that, Dr. Groves, I mentioned that in the State 
of Delaware, my own State, we believe that about 12,000 
residents were missed in the 2000 Census. It is sort of ironic, 
I think, maybe a cruel irony, that it turns out there is an 
overcount in a census and there is an undercount in a census. 
And the overcount tends to fall amongst--and I used our own 
family as an example. We have two sons that are in college out 
of State. They may get counted in our own home in Delaware, and 
they may be counted in their respective colleges sort of where 
they attend.
    There are a number of people in Delaware who--a number of 
folks who come to our beaches to spend part of the year, maybe 
the summer or come out throughout the year. They own homes at 
Bethany Beach and Rehoboth and Dewey and Fenwick Island and 
Lewes, Delaware, and they live in Maryland or Pennsylvania or 
New Jersey or Virginia. But they are counted as residents in 
their own States, and then they are counted as residents in 
some cases in Delaware where they have a second home.
    So the irony of it is we have an overcount that occurs, but 
a lot of times it is among more affluent in our society. And 
the undercount that occurs--and I think it is probably in 
several million in number. But the undercount occurs among, if 
you will, the least of these and the people that are usually 
the lowest income and minorities, folks that are renters and in 
some cases, children.
    Are you able to help us with the national numbers? I know 
the number for Delaware is 12,000. But if you go back maybe to 
the 2000 Census, the undercount and the overcount, and maybe 
the net. Can you help us with that?
    Mr. Groves. One of the methods we use to judge the quality 
of the census, as you know better than most, is a large sample 
survey. This decade it is called the Census Coverage 
Measurement Operation. And for the first time at State levels, 
we will estimate components, the overcount and the undercount 
separately and some other components. We will not have enough 
sample to do that at lower levels, but I think at the State 
level, you will get your wish this decade to decompose those.
    This is new for us. It depends on the quality of the sample 
survey. One of the interventions I have taken here is to try to 
beef up the measurement quality of that, but we should have 
these figures. I cannot wait to see them, actually.
    Senator Carper. Good. Going back to 2000, the 2000 
decennial, do we have any idea what the undercount and the 
overcount were and maybe the net difference?
    Mr. Groves. We do not. That sample survey called ACS was 
not designed for these components. Now, one could go if you had 
the data and compute these, but this was not part of the 
official process.
    Senator Carper. All right. Thank you.
    Mr. Goldenkoff, any closing thoughts you would like to 
share with us?
    Mr. Goldenkoff. Well, I think I would like to commend the 
Bureau for the progress that it has made since 2008 when we 
first put the decennial Census on our high-risk list. 
Certainly, it is still a high-risk area, but we are encouraged 
by a lot of the signs that we are seeing. We are much more 
optimistic now than we were in previous months. I think a lot 
of that credit goes to Dr. Groves.
    Senator Carper. Mr. Groves.
    Mr. Goldenkoff. Dr. Robert Groves, yes. The tone starts at 
the top, and I have been involved with the census since 1997. 
And I will say this, there is a culture of transparency and 
openness that was not there before. Dr. Groves and I get 
together about once a month to talk about census issues, what 
is working, what is not. He has been very responsive to us.
    One of the things that is so important if you want to 
address problems, the first step is acknowledging that you have 
one. And as you heard today, there is a litany of things. I 
mean, some of the challenges that Dr. Groves mentioned, and 
that attitude was not always there in past years. We had a lot 
of pushback on our recommendations. We have been making these 
same recommendations about the IT, the acquisition management, 
some of the other operational issues, for years now, going back 
as far as 2004, and we have often experienced pushback. But now 
there is, I think, a willingness to listen to us, to embrace 
oversight from us, from Commerce IG, and I think that is very 
refreshing.
    Senator Carper. Dr. Groves, are you going to sit there and 
take that?
    Mr. Groves. This brutal criticism is really----
    Senator Carper. Thank you for those comments. Mr. Zinser.
    Mr. Zinser. Yes, sir. I think I would agree with that, but 
I also would caution that type of transparency does not come 
natural to a bureaucracy to begin with. And I do not think it 
comes natural to the Census Bureau. I will give you one 
example.
    This paper-based operation control system that they are 
working on, the approach they are taking, they call it a ``just 
in time approach.'' Now, for a businessman, that might be a 
good way to run your business with your inventory, but for 
software development, just in time is not very comforting. But 
they use these labels, or they tend to use these labels, and I 
would just encourage the Census Bureau as we go into these 
final months to be transparent about their problems. And I 
think with Dr. Groves at the helm, I think there is a good 
chance of that.
    Senator Carper. All right. Thanks.
    Any closing comments, Mr. Groves.
    Mr. Groves. I mourn the Detroit Tigers today.
    Senator Carper. So do I. We have a long time to opening 
game of next spring.
    We appreciate very much your being here today. We 
appreciate really the stewardship that each of you have brought 
to this challenge, counting the people in this country, trying 
to do it accurately, trying to do it in a cost-effective way, 
and your own responsibilities, whether it is with GAO or 
Inspector General or previously with the Census Bureau, now as 
its leader. I am encouraged, and I think my colleagues are as 
well, with what we are hearing here today.
    Not any time to just kind of sit back and rest on our 
laurels; there is plenty of work to do, obviously. But we are 
mindful of the progress that being made, mindful that a lot of 
good people are working very hard to help us achieve our goals, 
and confident that if we give it our very best efforts, we will 
succeed in doing so but also mindful that everything we do, we 
can do better. And I do not want to be sitting here 10 years 
from now and with Senator McCain, Senator Coburn, Senator 
Burris, and others and have to say why aren't we using 
technology that others have been using for not only 15, but 20 
years; why aren't we using that technology to better count the 
people in this country? We have got to figure that out and do 
that right.
    Thank you so much. And with that, this hearing is--before 
we adjourn--you may receive some questions for the record from 
those who were here or were not here. And if you get those 
questions, just please respond promptly, if you would. And with 
that, this hearing is adjourned. Thanks so much.
    [Whereupon, at 4:51 p.m., the Subcommittee was adjourned.]



                            A P P E N D I X

                              ----------                              

[GRAPHIC] [TIFF OMITTED] T3847.001

[GRAPHIC] [TIFF OMITTED] T3847.002

[GRAPHIC] [TIFF OMITTED] T3847.003

[GRAPHIC] [TIFF OMITTED] T3847.004

[GRAPHIC] [TIFF OMITTED] T3847.005

[GRAPHIC] [TIFF OMITTED] T3847.006

[GRAPHIC] [TIFF OMITTED] T3847.007

[GRAPHIC] [TIFF OMITTED] T3847.008

[GRAPHIC] [TIFF OMITTED] T3847.009

[GRAPHIC] [TIFF OMITTED] T3847.010

[GRAPHIC] [TIFF OMITTED] T3847.011

[GRAPHIC] [TIFF OMITTED] T3847.012

[GRAPHIC] [TIFF OMITTED] T3847.013

[GRAPHIC] [TIFF OMITTED] T3847.014

[GRAPHIC] [TIFF OMITTED] T3847.015

[GRAPHIC] [TIFF OMITTED] T3847.016

[GRAPHIC] [TIFF OMITTED] T3847.017

[GRAPHIC] [TIFF OMITTED] T3847.018

[GRAPHIC] [TIFF OMITTED] T3847.019

[GRAPHIC] [TIFF OMITTED] T3847.020

[GRAPHIC] [TIFF OMITTED] T3847.021

[GRAPHIC] [TIFF OMITTED] T3847.022

[GRAPHIC] [TIFF OMITTED] T3847.023

[GRAPHIC] [TIFF OMITTED] T3847.024

[GRAPHIC] [TIFF OMITTED] T3847.025

[GRAPHIC] [TIFF OMITTED] T3847.026

[GRAPHIC] [TIFF OMITTED] T3847.027

[GRAPHIC] [TIFF OMITTED] T3847.028

[GRAPHIC] [TIFF OMITTED] T3847.029

[GRAPHIC] [TIFF OMITTED] T3847.030

[GRAPHIC] [TIFF OMITTED] T3847.031

[GRAPHIC] [TIFF OMITTED] T3847.032

[GRAPHIC] [TIFF OMITTED] T3847.033

[GRAPHIC] [TIFF OMITTED] T3847.034

[GRAPHIC] [TIFF OMITTED] T3847.035

[GRAPHIC] [TIFF OMITTED] T3847.036

[GRAPHIC] [TIFF OMITTED] T3847.037

[GRAPHIC] [TIFF OMITTED] T3847.038

[GRAPHIC] [TIFF OMITTED] T3847.039

[GRAPHIC] [TIFF OMITTED] T3847.040

[GRAPHIC] [TIFF OMITTED] T3847.041

[GRAPHIC] [TIFF OMITTED] T3847.042

[GRAPHIC] [TIFF OMITTED] T3847.043

[GRAPHIC] [TIFF OMITTED] T3847.044

[GRAPHIC] [TIFF OMITTED] T3847.045

[GRAPHIC] [TIFF OMITTED] T3847.046

[GRAPHIC] [TIFF OMITTED] T3847.047

[GRAPHIC] [TIFF OMITTED] T3847.048

[GRAPHIC] [TIFF OMITTED] T3847.049

[GRAPHIC] [TIFF OMITTED] T3847.050

[GRAPHIC] [TIFF OMITTED] T3847.051

[GRAPHIC] [TIFF OMITTED] T3847.052

                                 
