[Senate Hearing 112-244]
[From the U.S. Government Publishing Office]
S. Hrg. 112-244
CENSUS: LEARNING LESSONS FROM 2010, PLANNING FOR 2020
=======================================================================
HEARING
before the
FEDERAL FINANCIAL MANAGEMENT, GOVERNMENT
INFORMATION, FEDERAL SERVICES, AND
INTERNATIONAL SECURITY SUBCOMMITTEE
of the
COMMITTEE ON
HOMELAND SECURITY AND GOVERNMENTAL AFFAIRS
UNITED STATES SENATE
ONE HUNDRED TWELFTH CONGRESS
FIRST SESSION
__________
APRIL 6, 2011
__________
Available via the World Wide Web: http://www.fdsys.gov
Printed for the use of the
Committee on Homeland Security and Governmental Affairs
_____
U.S. GOVERNMENT PRINTING OFFICE
67-126 PDF WASHINGTON : 2011
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing
Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800; DC
area (202) 512-1800 Fax: (202) 512-2104 Mail: Stop IDCC, Washington, DC
20402-0001
COMMITTEE ON HOMELAND SECURITY AND GOVERNMENTAL AFFAIRS
JOSEPH I. LIEBERMAN, Connecticut, Chairman
CARL LEVIN, Michigan SUSAN M. COLLINS, Maine
DANIEL K. AKAKA, Hawaii TOM COBURN, Oklahoma
THOMAS R. CARPER, Delaware SCOTT P. BROWN, Massachusetts
MARK L. PRYOR, Arkansas JOHN McCAIN, Arizona
MARY L. LANDRIEU, Louisiana RON JOHNSON, Wisconsin
CLAIRE McCASKILL, Missouri JOHN ENSIGN, Nevada
JON TESTER, Montana ROB PORTMAN, Ohio
MARK BEGICH, Alaska RAND PAUL, Kentucky
Michael L. Alexander, Staff Director
Nicholas A. Rossi, Minority Staff Director
Trina Driessnack Tyrer, Chief Clerk
Joyce Ward, Publications Clerk and GPO Detailee
------
SUBCOMMITTEE ON FEDERAL FINANCIAL MANAGEMENT, GOVERNMENT INFORMATION,
FEDERAL SERVICES, AND INTERNATIONAL SECURITY
THOMAS R. CARPER, Delaware, Chairman
CARL LEVIN, Michigan SCOTT P. BROWN, Massachusetts
DANIEL K. AKAKA, Hawaii TOM COBURN, Oklahoma
MARK L. PRYOR, Arkansas JOHN McCAIN, Arizona
CLAIRE McCASKILL, Missouri RON JOHNSON, Wisconsin
MARK BEGICH, Alaska ROB PORTMAN, Ohio
John Kilvington, Staff Director
William Wright, Minority Staff Director
Deirdre G. Armstrong, Chief Clerk
C O N T E N T S
------
Opening statements:
Page
Senator Carper............................................... 1
Senator Brown................................................ 3
Prepared statements:
Senator Carper............................................... 41
Senator Brown................................................ 44
WITNESSES
WEDNESDAY, APRIL 6, 2011
Hon. Robert M. Groves, Director, U.S. Census Bureau, U.S.
Department of Commerce......................................... 6
Hon. Todd J. Zinser, Inspector General, U.S. Department of
Commerce....................................................... 8
Robert Goldenkoff, Director, Strategic Issues, U.S. Government
Accountability Office.......................................... 9
Daniel Castro, Senior Analyst, Information Technology and
Innovation Foundation.......................................... 25
Thomas M. Cook, Ph.D., Co-Chair, National Research Council Panel
to Review the 2010 Census...................................... 27
Arturo Vargas, Executive Director, National Association of Latino
Elected and Appointed Officials (NALEO) Educational Fund....... 29
Alphabetical List of Witnesses
Castro, Daniel:
Testimony.................................................... 25
Prepared statement........................................... 103
Cook, Thomas M. Ph.D.:
Testimony.................................................... 27
Prepared statement........................................... 117
Goldenkoff, Robert:
Testimony.................................................... 9
Prepared statement........................................... 72
Groves, Hon. Robert M.:
Testimony.................................................... 6
Prepared statement........................................... 46
Vargas, Arturo.:
Testimony.................................................... 29
Prepared statement........................................... 120
Zinser, Hon. Todd J.:
Testimony.................................................... 8
Prepared statement........................................... 56
APPENDIX
Chart submitted by Senator Brown................................. 147
Prepared Statement submitted for the Record by Wade Henderson,
President and CEO, The Leadership Conference on Civil and Human
Rights......................................................... 148
CENSUS: LEARNING LESSONS FROM 2010, PLANNING FOR 2020
----------
WEDNESDAY, APRIL 6, 2011
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 1:30 p.m., in
Room SD-342, Dirksen Senate Office Building, Hon. Thomas R.
Carper, Chairman of the Subcommittee, presiding.
Present: Senators Carper, Brown, and Coburn.
OPENING STATEMENT OF SENATOR CARPER
Senator Carper. The hearing will come to order. Welcome one
and all. Senator Brown and I are happy to see you all. Thank
you for joining us today on an important hearing, a very
important hearing.
I am going to make a brief statement and then turn it over
to Senator Brown, and then we will introduce our witnesses and
get this show on the road.
Today's hearing will examine the lessons learned from the
2010 Census while identifying initiatives that show promise for
producing an even more accurate and more cost effective census
in 2020.
I want to begin by congratulating Dr. Groves; his
predecessor, Dr. Murdock, who sat in this seat once or twice
himself when he was our Census Director; and the career
professionals at the Census Bureau who did an excellent job in
carrying out the 2010 Census. As a result of their hard work,
the Census Bureau was able to overcome a number of operational
and organizational challenges, including shortcomings with
critical information technology systems.
The Bureau completed key operations on schedule, hired
nearly 900,000 temporary workers, obtained an acceptable
participation rate ultimately of 74 percent, and managed to
report its population figures in time to support redistricting
so that we would know in Delaware we still would have only one
at-large U.S. Representative.
Despite these achievements, the 2010 Census was the most
expensive in our Nation's history by far, even taking inflation
into account. The total cost of decennial operations escalated
from an initial estimate of $11.3 billion to close to $13
billion. Even more disturbing is the fact that with all the
modern scientific improvements and technological advancements
that have been made over the years, the framework for
conducting the 2010 Census was based off of a model that I
believe was used in the 1970s.
Although the methodological basics of the census have
remained the same over the past 40 years, the cost of the
census decidedly has not. The average cost per household was
$98 in 2010, compared to $70 in 2010, compared to $16 in 1970.
I have been told that the total cost of the 2020 Census could
rise to as much as $30 billion if we keep going on this track.
In my view, that is not acceptable any more than budget
deficits of $1.2 trillion are acceptable. It is especially not
acceptable at a time when we are struggling to find solutions
to the most serious deficit problems and the debt crises that
our country is currently facing.
We have spoken at previous hearings here about the need for
us to look in every nook and cranny of the Federal Government--
domestic, defense, entitlement spending, along with tax
expenditures--and ask this question: Is it possible to get
better results for less money? The hard truth is that many
programs' funding levels will be reduced. They need to be
reduced. Even some of the most popular and worthwhile programs
out there will likely be asked to do more with less, or at
least to do more without a whole lot more money. The Census
Bureau, despite the vital and constitutionally mandated nature
of its work, cannot be immune from this sort of examination.
While most Americans want us to reduce the deficit,
determining the best path forward will not be easy. Many
believe that those of us who have been sent here to Washington
are not capable of doing the hard work and making the hard
decisions that we were hired to do--effectively managing the
Federal dollars, their tax dollars that they have entrusted us
with. They look at our spending decisions that we have made in
recent years and question whether the culture here is broken.
They question whether we are capable of making the kind of
tough decisions that they and their families have to make on an
almost daily or weekly basis with their own budgets. And I do
not blame them for being skeptical, and I am afraid that their
skepticism proved to be well founded when looking at the kind
of avoidable management failures that contributed to the growth
in cost of carrying out the 2010 Census.
Today we will look at the Bureau's planning efforts for the
2020 decennial, and although it is 9 years away, it is never
too early to start thinking about ways to reduce costs and
improve quality through more efficient data collection. More
importantly, we need to make certain that the issues that lead
to the failures and cost overruns that we saw in recent years
have been addressed and will not reoccur. Taxpayers should not
be expected to pick up the tab for them again.
Looking ahead, the Bureau's research should focus on how
existing technology can be incorporated into the 2020 design.
Obviously, the Internet is here to stay, at least for my
lifetime, and according to the experts, an Internet response
option could have saved the Bureau tens of millions of dollars
in processing costs in 2010. Future research should not only
focus on how to implement Internet data collection but also how
to reap the benefits--financial and otherwise--of it and other
technologies the next time around. We also need to make certain
that the people who make up our growing and changing country
are comfortable enough with the security of the data collection
methods we use to allow for an accurate census.
Moreover, steady leadership will also be critical in
reversing a trend of decennial Censuses marked by poor planning
and escalating costs. The 2010 Census experienced several
changes in leadership and vast spans of time with acting or
interim Directors, further putting the operation at risk. In
the 27 months leading up to Census Day, the Bureau had, count
them, not one, not two, but three different Directors. I plan
to introduce legislation this year that would, among other
things, make the Director of the Census Bureau a Presidential
term appointment of 5 years. A fixed term would help avoid
leadership gaps during critical decennial Census planning
stages and facilitate the longer-term planning so vital to
decennial Censuses.
Senator Coburn and I introduced legislation last year to
establish a term appointment for the Census Director and to
make a number of other changes at the Bureau aimed at
preventing serious problems in the future. It passed the Senate
unanimously but failed to be taken up in the House. And I would
like to work with you, Dr. Groves, if I can, if we can, to make
whatever changes are necessary to put together something that
addresses the lessons learned from 2010 and can enjoy
bipartisan support as our proposal did in the last Congress.
We look forward to hearing from our witnesses today who
will help us to identify ways to best balance the need for an
accurate census with the need to ensure a reasonable cost for
this endeavor.
Senator Brown, 10 years from today, I suspect you will
still be here, but I am not sure that I will be. I might, but I
would not want to bet on that. But whoever does sit in the
seats where you and I sit, I do not want them to be saying,
``How do we end up spending twice as much for the census in
2020, as we spent in 2010? How did we do that?'' That is what
we did from 2000 to 2010, and we have done it again. I just do
not want the folks in this Committee to go through that. I do
not want the Senate to go through that. I do not want the
people of our country to go through that. And I know the
groundwork is already being laid today this year to make sure
that we do not see history repeat itself. And we are anxious to
learn how we can help to make sure that we end up in 9 years
from now that we have a better count, a more accurate census,
and we have done it not for twice as much money but maybe, if
we are smart, the same amount of money.
All right. Scott.
OPENING STATEMENT OF SENATOR BROWN
Senator Brown. Thank you, Mr. Chairman. Just to note I am
bouncing back and forth to the hearing, so I will only be able
to stay for the first panel, depending on the time. But I do
appreciate the opportunity to come. Sir, it is good to see you
again.
As required by our Constitution, as you noted, our country
has conducted a census every year since, obviously, 1790, and
it is a vital undertaking. The results, are utilized to
apportion seats in the House, for redistricting, and
determining the annual distribution of more than $400 billion
in Federal and State funds. And while we must strive to ensure
that every person is counted, we cannot afford to have the out-
of-control spending that seems to be potentially going--well,
going on and getting worse continually.
The cost of counting each housing unit has escalated from
around $16 in 1970 to $98 in 2010, and I have learned in my
brief tenure here that we cannot simply continue to do things
the way we have always done them. We have to think outside the
box, modernize, get up with the times. I feel sometimes that I
am--I know we used to have records, the little needle--I tell
my kids and younger people, I say, ``I used to listen to
records.'' They look at me like I have three heads. And
sometimes--I mean, you all know what I am talking about, what a
record is. But you look around, and you see how we do stuff in
the Federal Government, and it is like I feel like I am back in
the 1970s, talking about records, whether it is the Arlington
National Cemetery and they keep wounded--our fallen heroes on
cue cards, index cards, or--I just do not get it, with the
amount of money that we spend on these things. So we have to
find a way to do it better, to get a better bang for our buck.
And for the most part the basic model of census taking has
not changed since the 1970s, and we need to update, we need to
streamline, consolidate, do it better. And we are relying on
the old-school way of doing things, and it is just not, I do
not think, working just based on the costs that we are seeing
and we will be talking about.
With an array of Internet-based technologies, you have
Facebook, Twitter, IMs, the whole range of ways that we can do
it better, and I am hoping that we can kind of, with your
leadership, sir, as we talked about, do it better.
We are the world leader in inventing and commercializing
technology and technological innovation, and it is something,
being from Massachusetts, and Cambridge in particular--that is
where it all begins. That is where the think tanks, many of
them, are in our great country. And yet it seems like we are
lagging behind a country like Canada, for example, in
integrating the Internet into the census.
I am convinced that we can break this cycle and do it
better and be more cost effective, and I am excited to have the
opportunity to discuss that with you. And while I expect the
Census Bureau to say the right things about reforming the
process in 2020, I have been here long enough, a little over a
year, to know that the taxpayers and Congress have the right to
remain skeptical based on past performance, not necessarily of
this organization but of what we see throughout government
today. And I am going to work with the Chairman, as we do on
many, many things, to try to find a way to bring it out and
potentially offer solutions, suggestions, find out how we can
help through legislative or other types of either making
regulation or eliminating regulation, and, finally, how we can
get our tax dollars to be spent in a more efficient manner.
I look forward to the witnesses speaking, Mr. Chairman, and
I want to thank you once again for holding this hearing.
Senator Carper. You bet. I am glad we could do it together.
I am going to go ahead and introduce our panel. I think
maybe--I might be mistaken, but I think you have all been here
before, maybe a couple times. It is a good thing, Senator
Brown, that we are not paying them on a per appearance basis.
That would get pretty expensive.
Let me just welcome, first of all, Dr. Groves, nominated by
President Barack Obama to be Director of the Census in April
and confirmed by the Senate last July. Dr. Groves is an expert
in survey methodology and has spent decades working to
strengthen the Federal statistical system, improve its staffing
through training programs, and keep it committed to the highest
scientific principles of accuracy and efficiency. Having once
served as Associate Director of the Census Bureau, Dr. Groves
knows how the agency operates and what its employees need to
successfully implement the decennial Census and other programs.
Welcome. Nice to see you again.
Todd Zinser, also known as the Honorable Todd Zinser--and I
was kidding him earlier when I came out here, Senator Brown, to
say hello. I do not know if this ever happens. Do you know that
every now and then we get phone calls at home from people, and
we have these Do Not Call lists, and still people call. And if
they were, like, calling from, like, the University of Michigan
or Ohio State or someplace like that, and one day I got this
call from a fellow at the other end of the phone, and he said,
``Is Hon there?'' And I said, ``Pardon me?'' And he said, ``Is
Hon there?'' And I was trying to think, ``Who could he be
calling for?'' And then I was thinking, ``Oh, Hon. H-O-N
period, short for `Honorable.' '' And so I said, ``This is
Hon.'' [Laughter.]
And he said, ``Oh, Hon, how are you doing?'' I forget where
he was calling from. But he said, ``I am calling from so-and-
so, and you have been great to support our charity or trust
before. I just wanted to call and see if you could do it
again.'' And I said--so he made his pitch, and I said, ``Hon
have no money.'' And he gives me about another 30 seconds, and
I said a little more strongly, ``Hon have no money.'' And he
comes back to me a third time and gave me his pitch, and I
said, ``Hon have no money. Call Hon Castle.'' That is Mike
Castle, our Congressman. ``He has the money.'' And so he said
good-bye and he never called back. [Laughter.]
But Hon. Todd Zinser, welcome. Todd 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 and scientific and economic and environmental programs
that are administered by the Department of Commerce and its 13
Bureaus. Mr. Zinser holds a bachelor's degree in political
science from Northern Kentucky University and a master's degree
in political science from Miami University. Is that Miami
University in Oxford, Ohio?
Mr. Zinser. Yes, Senator.
Senator Carper. All right. Home of the Bobcats? Is that
what they are there? Ohio University? I think so. We are
Buckeyes at Ohio State.
Robert Goldenkoff is the Director of Strategic Issues at
the Government Accountability Office (GAO) where he is
responsible for reviewing the 2010 Census and governmentwide
human capital reforms. Mr. Goldenkoff has also performed
research on issues involving transportation security, human
trafficking, and Federal statistical programs. He received his
Bachelor of Arts in Political Science and his Master's of
Public Administration degree from The George Washington
University. All right.
Your entire statements will be made part of the record, and
once you have concluded, I am going to ask Senator Brown to
take the first questions, and then I will take my nap while he
is doing--no, I will not do that. But I will be listening
intently to the questions and the answers.
But, Dr. Groves, it is great to see you. Thanks for taking
on this job. You are recognized. If you go a whole lot over 5
minutes, I have to rein you in, so just keep that in mind.
Thanks so much. Please proceed.
STATEMENT OF HON. ROBERT M. GROVES,\1\ DIRECTOR, U.S. CENSUS
BUREAU, U.S. DEPARTMENT OF COMMERCE
Mr. Groves. Mr. Chairman, Senator Brown, I am happy to be
here and thank you for the invitation.
---------------------------------------------------------------------------
\1\ The prepared statement of Mr. Groves appears in the appendix on
page 46.
---------------------------------------------------------------------------
Although the Census Bureau has a formal program of
evaluations and assessments on the 2010 Census, those are not
yet complete. But I do have information I can share on
preliminary quality indicators.
My testimony in written form is really in three pieces:
Evaluation of the census, our organizational change endeavors
at the Census Bureau, and then lessons learned. I am going to
concentrate on the third part. But I can note that the
preliminary findings on the quality of the 2010 Census are
positive in the majority and show improvements over the 2000
effort. I would be happy to expand on that.
I want the Committee to know that we have also been engaged
in a variety of organizational change initiatives that we care
deeply about. We have basically concluded that our business
model of collecting social and economic data faces severe
challenges over the long run. We know we must innovate in order
to remain useful and relevant to the country. Further, we know
that this innovation is not likely to be funded by added
resources. We must become more efficient and fund innovations
from cost-saving measures, and that is what these programs are
about. I want to mention three specifically.
First, we have mounted a program that is seeking proposals
from throughout all the employee groups for cost efficiencies.
It was heart-warming to see last year that we received over 650
proposals from folk throughout the Census Bureau on how to make
what they do more efficient, and we are pursuing a lot of the
good ideas and saving money already.
Second, we have partnered with other Federal agencies who
sponsor surveys that we collect data for in order to find out
ways that we can save money for them. This will have ripple
effects to other agencies.
Third, we are vigorously trying to tear down the boundaries
among the silos of the Census Bureau. We are trying to seek
organization-wide solutions. Let me rattle off a few of those.
We have instituted a corporate hiring program for new
statisticians to assure that they will move across the
organization in the early years of their career spreading
innovation across the silos.
We are moving aggressively on enterprise architecture
solutions on the information technology (IT) front. This means
a greater emphasis on the Internet and cloud computing, a
consolidation of data storage systems that is already saving
money. We have built the Technology Innovation Center to do
quick prototyping of new solutions.
We have greatly expanded our Internet data collection, soon
to cover 60 of our sample surveys, allowing approximately
900,000 respondents the opportunity to respond online. And I
want to note that increasingly people are using the Internet
options we are providing on hand-held devices like iPhones and
Droids and iPads.
These changes together, in my belief, will make us a more
unified, integrated organization, one that is ready to mount a
successful 2020 Census, and that is what I want to turn to now.
I want to go through eight lessons that I have learned
personally, each of which has generated a principle for the
organization of the development plans for 2020.
Lesson one, the multi-decade cost increase of the decennial
Census must be halted. Hence, we are attempting to design a
2020 Census that costs less per housing unit than the 2010
Census while maintaining the quality of the results.
Lesson two, the traditional non-response follow-up
procedures that we have used over past decades are inefficient
and costly. We want to make the census convenient to diverse
groups using multiple modes of data collection. This means the
traditional mail, but also phone, multiple Internet options,
face-to-face, and other modes as they emerge.
Lesson three, systems development that requires first-use
perfection must be abandoned. We need end-to-end tests of
production systems, ideally within real survey production
environments.
Lesson four, too few of the system and procedure
developments of the 2010 Census were designed to benefit the
entire institution. Thus, the fourth principle is that we want
to develop systems within the survey production environments of
the Census Bureau. We plan to use the American Community Survey
as a chief test bed for the 2020 Census systems development.
Let me skip to lesson six. We have concluded that a small
number of large test censuses create intolerable risks for the
Census Bureau. We want to do many small tests. We feel that the
evidence of updating the Master Address List was partially--
that partial updating in the last decade was successful. We
want to build on that success.
Let me sum up. Overall, we know of no single method of
collecting census data that is optimal for all the diverse
subpopulations of the United States. Some residents have told
us they do not want people visiting their home. Some residents
told us that information they have already provided in other
government forms ought to be used. Some residents want to use
the Internet at any time of the day on any device they favor to
fit their lifestyle. And some want to speak by telephone to
someone who speaks their language and understands their
subculture.
By making the census more convenient, we hope to reduce the
size of the expensive field follow-up activities. This is the
most important and most expensive part of the data collection.
We are concentrating our efforts there to achieve a quality
census.
Those are my oral remarks. I would be happy to answer
questions.
Senator Carper. Well, thanks for those oral remarks. We
look forward to those questions and answers.
Mr. Zinser, please proceed.
STATEMENT OF HON. TODD J. ZINSER,\1\ INSPECTOR GENERAL, U.S.
DEPARTMENT OF COMMERCE
Mr. Zinser. Thank you, Mr. Chairman, Senator Brown. Thank
you for inviting us to testify today about lessons learned from
the 2010 Census and how we might apply those lessons to the
decennial Census in 2020.
---------------------------------------------------------------------------
\1\ The prepared statement of Mr. Zinser appears in the appendix on
page 56.
---------------------------------------------------------------------------
The 2010 decennial was an enormous undertaking with a
current cost estimate of nearly $13 billion. It required more
than a decade of planning, testing, and implementing dozens of
operations as well as hundreds of thousands of employees, to
accomplish.
My testimony today is based on oversight we provided over
the last decade to both the planning and execution of the
decennial. Our oversight sent over 100 Office of Inspector
General (OIG) staff to every State and the District of
Columbia. We provided feedback to stakeholders on headquarters
activities and from the field through reports, testimony, and
real-time communications back to the Census Bureau.
While the census has successfully completed its 2010
operations, this decennial carried with it a high cost and a
level of risk that should not be repeated. Factoring in trends
in population and cost growth, GAO recently estimated that the
current design model could mean a 2020 decennial cost as high
as $30 billion. Such cost growth is simply unsustainable.
To achieve a quality count with much greater cost
containment, Census must fundamentally change the design,
implementation, and management of the decennial Census, and it
must start now. My testimony today covers seven challenges for
the Census Bureau to address for the 2020 Census.
First, Census must revamp their cost estimation and budget
processes to increase accuracy, flexibility, and transparency.
Second, Census should use the Internet and administrative
records to contain costs and improve accuracy. There are
already numerous Federal agencies that collect similar
information about U.S. households at significant duplicated
costs. Use of existing administrative records could greatly
assist Census in reducing the cost of many of its operations.
It is a complex issue but not insurmountable, and a solid
commitment to use the Internet for 2020 is imperative.
Third, Census should implement a more effective decennial
test program. Site tests for 2010 were scheduled at 2-year
intervals. Each test transpired over 3 years of planning,
implementation, and evaluation. The tests overlapped, which
made it difficult to apply the results from one test to the
next. Census now plans to conduct a larger number of smaller
tests and more closely align its research with its testing
program.
Fourth, Census should effectively automate field data
collection. Census tried to maximize the use of automation for
the 2010 decennial but fell short, and as a result, costs and
risks increased substantially. Census must shore up its IT
processes early in the decade to prepare for successfully
implementing automated data collection.
Fifth, we recommended that Census avoid a massive end-of-
decade field operation through continuous updating of address
lists and maps. Instead of the large end-of-decade address
canvassing operation, which cost $444 million and experienced a
25 percent cost overrun, Census is planning to update its
address lists and maps continuously throughout the decade and
is considering other options to meet its address and map
requirements.
Sixth, the Bureau should implement improved project
planning and management techniques early in the decade. For the
2010 decennial, Census tracked more than 9,000 activities over
several years for 44 different operations. We have made
recommendations aimed at strengthening project and risk
management.
Finally, a Census Bureau Director position should be
established to span Administrations. For the life cycle of the
2010 decennial, we counted six Directors and Acting Directors.
Census would benefit from greater leadership continuity.
Census has already embarked on its plans; however, it will
need continued focus, engagement, and resources throughout the
decade from the Department of Commerce, the Office of
Management and Budget (OMB), and Congress to help ensure that
the 2020 Census fulfills the promise of better technology,
methods, and operations.
That concludes my summary, Mr. Chairman, and I would be
happy to answer any questions you or other Members have.
Senator Carper. Good. And I will just telegraph an early
pitch, Dr. Groves. When the questions come to me, one of the
questions I am going to be asking is for you to walk through
that list of seven recommendations from Mr. Zinser, and I want
you to be prepared to comment on those, please.
Mr. Goldenkoff, please proceed.
STATEMENT OF ROBERT GOLDENKOFF,\1\ DIRECTOR, STRATEGIC ISSUES,
U.S. GOVERNMENT ACCOUNTABILITY OFFICE
Mr. Goldenkoff. Mr. Chairman, Senator Brown, I would like
to thank you for the opportunity to be here today to discuss
lessons learned from the 2010 Census and initiatives that show
promise for delivering a more cost effective enumeration in
2020.
---------------------------------------------------------------------------
\1\ The prepared statement of Mr. Goldenkoff appears in the
appendix on page 72.
---------------------------------------------------------------------------
The 2010 Census was an operational success in that the
Census Bureau generally completed its peak data collection
activities consistent with its plans and released the data used
to apportion and redistrict Congress several days ahead of
legally mandated deadlines. Nevertheless, in gearing up for the
enumeration, the Bureau had to overcome a series of hurdles
that jeopardized a complete count.
First, internal issues, including longstanding weaknesses
in its IT management procedures, threatened the Bureau's
readiness for the enumeration and led us to add the 2010 Census
to GAO's list of high-risk Federal programs.
At the same time, external societal trends, such as an
increasingly diverse population, have made a cost-effective
head count inherently difficult. Much like going up a down
escalator, over the past 40 years the Bureau has been investing
substantially more resources each decade in order to secure a
complete count.
For example, as Senator Brown noted earlier, in constant
2010 dollars the cost of enumerating each household has
escalated from around $16 in 1970 to around $98 in 2010, an
increase of over 500 percent. This trend is unsustainable.
Meanwhile, the 2010 Census, with a total cost of around $13
billion, was the most expensive head count in our Nation's
history.
Simply put, the singular challenge facing the U.S. Census
Bureau is how to control the cost of the 2020 Census while
maintaining its accuracy. In this regard, my remarks today will
focus on four key lessons learned from 2010 that will be
important for the Bureau to address as it continues its
planning efforts for 2020.
The first lesson learned is the importance of fundamentally
re-examining the Nation's approach to taking the census. This
is critical because simply refining current methods, some of
which have been in place for decades, will not bring about the
reforms needed to obtain acceptable results given ongoing and
newly emerging societal trends. A fundamental re-examination
means rethinking the Bureau's approach to planning, testing,
implementing, monitoring, and evaluating the census. Potential
focus areas include making better use of administrative
records, such as driver's licenses, as well as social media,
such as the Internet.
The second lesson learned is the importance of tailoring
key census operations to specific locations and population
groups. The Bureau plans to complete over 70 studies of the
2010 Census. As this research is completed, it will be critical
for the Bureau to assess the costs and benefits of each
operation so it can allocate its resources more efficiently in
2020.
The third lesson learned centers on institutionalizing
efforts to address those areas that made the 2010 Census a
high-risk area. This includes incorporating best practice for
IT acquisition management, developing more reliable cost
estimates, and ensuring key operations are fully tested under
operational conditions.
The fourth lesson learned involves ensuring that the
Bureau's organizational culture and structure as well as its
approach to strategic planning, human capital management, and
other internal functions are aligned towards producing more
cost-effective outcomes. These actions are needed because some
of the operational problems that occurred during the 2010 and
prior censuses were symptomatic of deeper organizational
issues, such as inadequate human capital planning.
Importantly, the Bureau has launched an ambitious planning
program for 2020, taking such measures as reforming aspects of
its IT management. As these actions gain momentum, it will be
important that they enhance the Bureau's capacity to conduct an
accurate count, control costs, manage risks, and be more nimble
in adapting to the social, demographic, technical, and other
changes that can be expected in the future.
In closing, the Bureau goes to great lengths each decade to
improve specific census-taking activities, but these
incremental modifications have not kept pace with societal and
technological changes. The Bureau is well aware of this and has
wasted no time in launching the planning efforts needed for a
more cost-effective enumeration in 2020.
It will also be important for Congress to continue its
strong oversight of the census, and we look forward to
supporting the Subcommittee in this regard.
Mr. Chairman, Senator Brown, this concludes my remarks, and
I will be pleased to respond to any questions that you might
have.
Senator Carper. Mr. Goldenkoff, thank you very, very much.
First of all, let me just ask you, if I could, Mr. Zinser
and Mr. Goldenkoff, you heard Dr. Groves testify here today,
and you have heard him testify a number of times before and
have worked with him to help ensure that we get a better count
going forward for less money. What did you hear from Dr. Groves
today that you were especially pleased to hear? And what were
the things that you did not hear that you wish you might have?
Do you want to go first, Mr. Zinser.
Mr. Zinser. Yes, I also had the opportunity to review Dr.
Groves's testimony before we came up today, and I have to say
that I think that Dr. Groves's observations, the observations
from my office, and Mr. Goldenkoff's observations are all right
on the same page. I think that we are pretty much in agreement
with what Dr. Groves has laid out. And I think that what we
would want to see more of is more attention paid to some of the
nuts and bolts management issues for budgeting and project
management.
Senator Carper. All right. I am going to suspend right
there. I said I wanted to ask Senator Brown to lead off because
he has another hearing to run off to. I will come back and pick
that up where we started.
Scott, I am sorry. You go right ahead. Thank you.
Senator Brown. Thank you, Mr. Chairman.
So, Mr. Groves, the growth of completing the census has
been unsustainable, as I talked about in the beginning, and
where the measures of cost of counting each household, which
has grown from $16 to $98 in 2010, from $4.1 billion to over
$12 billion from 1990 to 2010, and we cannot continue on. In
your opinion, what has caused the explosive growth?
Mr. Groves. I think if you look over the decades, there are
several drivers of it. One has to do--most of the drivers could
focus in on the non-response follow-up procedures. People are
sent mail questionnaires over the past decade--
Senator Brown. How much does the mail actually cost? Like,
what does one of the mailings--because I know I got about 30 of
them. That was after I sent it in.
Mr. Groves. Well, what we said throughout the census this
year, last year, was to return the mail questionnaire costs
about 42 cents. To call on your household costs us about $57.
So that is the ratio that is so important in addressing your
question.
Senator Brown. So it is still better, more effective, to do
it via mail.
Mr. Groves. Yes. If mail worked 100 percent, it would be a
very cheap census. It is a technique that works when it works.
The problem is those rates are going down. As your chart\1\
shows, the black line is--the cost and the line are related to
one another.
---------------------------------------------------------------------------
\1\ The chart submitted by Senator Brown appears in the appendix on
page 147.
---------------------------------------------------------------------------
When the returns do not come in, then we go out and knock
on doors. We knocked on 47 million household--
Senator Brown. So you are saying that this--and people just
looking, so that chart above the heads of everybody is the
reduction, the 63 percent, which is the mail response rate and
the money, the $98 million projected. That is not just mail.
That is the follow-up of the phone calls, the door knocking.
That is the whole shebang after.
Mr. Groves. Absolutely. And so this decade, we knocked on
47 million household doors, and that cost a lot of money. So if
you say how do you stop that trend, we are focusing on that
follow-up procedure. What is driving those costs? And how do we
reduce the number of households that require that expensive
personal visit?
Senator Brown. And I know we had this conversation, so
everyone who is listening is clear. So this is just people in
households. This has nothing to do with people that are here
legally or illegally. It is just people, period, right?
Mr. Groves. Our mandate under the Constitution, under the
Census Act of 1790 that has been renewed, is that we count all
residents.
Senator Brown. Whether they are here legally or not.
Mr. Groves. Correct.
Senator Brown. So do we have an accurate count of how many
U.S. citizens are here?
Mr. Groves. The decennial Census does not have a question
about citizenship.
Senator Brown. Isn't that a little unusual? We are trying
to find out, like, who is here, and we are giving monies to
States and we are trying to make determinations as to who is
representing who in Congress, and we do not even know how many
U.S. citizens are in the States?
Mr. Groves. For purposes of the decennial Census under the
law--
Senator Brown. Yes, but from--
Mr. Groves [continuing]. We count all--
Senator Brown. I know, but does it seem unusual that we
would not do it that way as well and find out, okay, we have
households and--and, by the way, we need to find out how many
people are here who are United States citizens so we can divvy
up the funds properly. Does that make sense, or am I missing
something?
Mr. Groves. There is a wonderful phrase in the
Constitution, Article I, Section 2, that notes that Congress
shall by law direct how the census is done.
Senator Brown. Right.
Mr. Groves. And I believe Congress has the power to change
that.
Senator Brown. And I believe that is what we kind of talked
about. We had a little bit of a go-round, and I appreciate you
being consistent in making that recommendation. I am not sure
if there is an effort to make that determination and give
congressional guidance or change to do that.
What performance measures should Congress track to ensure
that the census keeps its promise to lower costs per housing
unit in the 2010 and 2020 Census in the future?
Mr. Groves. I think there are various things that could be
done, and they all go under the rubric of watching us over the
early years of this decade. That is going to be key, to attend
to our progress. We have constructed an integrated set of
research steps that answer key questions, and every one of the
questions is related both to cost and quality of the census. So
we are going to be producing those answers over the coming
years, if we are funded to do this. You should hear those
answers, and you should be satisfied with those answers that we
are moving in the right direction to keep that focus. It is
critical.
Senator Brown. Could you do your job with half the money?
Mr. Groves. I do not know the answer to that. I think it is
unlikely. I need to--
Senator Brown. Can you do it with $98 million for the next
go-round in 2020?
Mr. Groves. For $98 million or for--
Senator Brown. Or $98 per household. Do you--
Mr. Groves. I see. Per household. Well, I can tell you our
goal is to reduce that red--
Senator Brown. Yes, what is the goal? Is it--to what? Do
you have a number you are trying to shoot for?
Mr. Groves. We do not have a number. Let me tell you how we
are addressing the cost estimation because this is relevant to
Robert's comments. We are doing modeling of different cost
outcomes based on different scenarios, different assumptions.
Our research is basically going to tell us as the months go by
which of those assumptions are correct. So we will narrow in on
the cost. But every research question we are addressing has
cost impacts, and we want to share with you those answers to
keep us honest on cost reduction.
Senator Brown. I want to apologize. I know the numbers we
went over the other day and obviously earlier. I know it is
about $12 billion to actually do what you did, and this is per
household, $98. That is why I have these, and I do not use them
enough, so I apologize.
The 2000 Census included an Internet response option, yet
the 2010 Census did not as the census, again, relied on the
same kind of mail-out, mail-back method used for decades.
Meanwhile, the cost has escalated. Why decades into the
Internet revolution did it not contain that option?
Mr. Groves. This decision was made in the middle of the
decade between 2000 and 2010. The reasons, I have been told,
that led to that decision were concerns about security, IT
security issues.
At this point those are not valid. We are doing large
numbers of sample surveys using the Internet. We have conquered
the IT challenges on this quite successfully. As your chart
shows, there are other countries that have been doing this for
some time. We can do this. We are doing it.
Senator Coburn. Do not give him credit for my chart.
Mr. Groves. Oh, I am sorry. Sorry.
Senator Brown. I am not. I am just reading down the order
they gave me the questions here. [Laughter.]
On that note I am not going to get the big guy mad, so
here, I am all done.
Senator Carper. Dr. Coburn, why don't you jump in here.
Reclaim your chart.
Senator Brown. I have seen his wrath.
Senator Coburn. First of all, I want to say publicly how
enthused I am that we have very super competent leadership at
the Census, and I have great faith in Dr. Groves. I have seen
what he took on, how he accomplished his mission, and his
commitment to using science to make his organization more
efficient. I am one of your big backers. I told you that in my
office, and I appreciate the job that you have done and the
people under you that have helped you accomplish that.
How much do we spend on the American Community Survey (ACS)
every year?
Mr. Groves. It is roughly $200 million.
Senator Coburn. $200 million, and do you have plans to put
the American Community Survey online?
Mr. Groves. We are actually in the middle of an Internet
test on the ACS. It is a bigger challenge, I need to tell you,
than the short form--
Senator Coburn. Well, I understand.
Mr. Groves. But we are testing it right now.
Senator Coburn. Just for history, the reason it was not on
the Internet is there was a contract between Lockheed Martin
and the Census to do an online test, and they came up with a
garbage excuse that they could not manage the security when 72
percent of the income tax that is paid in this country, is
online.
This chart comes from England. I saw this in the paper last
week, and I said I have to bring this and show this to Dr.
Groves. The fact is that they are advertising, and they are
saying get it done. Lockheed did it. The very contract we
turned down they did for Great Britain, and it is working
wonderfully over there.
So we know it is possible I will not go through all of the
questions that I have on Internet, but I think it is important.
I know you are committed to bringing us up to speed, and we are
going to save hundreds of millions of dollars annually if, in
fact, we accomplish this task.
What are the main management and operational challenges
that you really faced during the 2010--I do not want you to
take a long time with it, but what were your two big
challenges? And how did you address them?
Mr. Groves. Well, we had a fantastic team, I want you to
know. The folks that followed up on the replanning efforts
produced a lot of saved operations. The chief challenges were
software challenges. We had a system that monitored the work
flow that was not working properly for about 3 weeks. That was
a scary time. We got it working, and it actually really purred
along at the end. But the first few weeks were kind of scary.
We were--well, let me stop at that. That was the chief
management threat that we had.
Senator Coburn. OK. For our GAO and IG witnesses, have
either of you done any estimates on what you think the cost
savings could be if we utilized the Internet in the census?
Mr. Zinser. We have not done an estimate like that, sir.
Senator Coburn. OK. GAO?
Mr. Goldenkoff. We have not either. You should know that
there are some large up-front costs getting the system up and
running, and those costs would need to be offset by the higher
response rate. But we have not done any estimates as of yet.
Senator Coburn. It is important that we go to the IRS and
say, ``What are the problems you had in getting this going?''
In other words, we learn from our experience rather than try to
do it again. I hope that we are going to be doing that in terms
of good correlation with their experiences and how they got
this up and running and got the security going. We do not have
to reinvent it every time we do something in the Government in
terms of IT.
Dr. Groves, let me go to one other question. Senator Brown
asked you, Do you have the power to change the questions on the
census?
Mr. Groves. On the decennial Census?
Senator Coburn. Yes.
Mr. Groves. The process by which the decennial Census
questions are arrived at is a laborious one that brings in a
whole lot of stakeholders. We then submit the questionnaire to
Congress in the year that ends in 6, I think, and again in 7
for your review. So it is truly a collaborative process.
Senator Coburn. Do we actually act on that?
Mr. Groves. I think that has varied over decades, Senator,
on how Congress has reacted to that.
Senator Coburn. Following up a little bit on Senator Brown,
we could have a question in the decennial Census that asked:
Are you a U.S. citizen? Are you a legal resident? Are you
other?
Mr. Groves. That is a possible census--
Senator Coburn. There is nothing that precludes us from
asking that?
Mr. Groves. Not the way I understand it.
Senator Coburn. OK. All right. That is what I wanted to
make sure.
The other thing is we had testimony by the IG. Why is it
important to have a Census Director that spans Administrations?
Mr. Goldenkoff. Yes. What it comes down to is stewardship.
The life cycle of a census spans the course of the decade, and
several Presidential Administrations. To implement change, as
you well know, can take years. And so what has happened in the
run-up to the 2010 Census, there was a lot of turnover among
the Census Directors. If you look back, since 1969 the average
tenure is about 3 years for the Census Director. The longest
tenure was 5 years.
Senator Coburn. Yes. So my question to Dr. Groves: Are we
going to get to keep you?
Mr. Groves. I do not believe I can answer that question.
Senator Coburn. Well, if you were invited, are we going to
get to keep you?
Mr. Groves. I do not know, Senator.
Senator Coburn. I am saying it somewhat in humor, but it is
not humorous. Continuity in agencies like this is really
important. When we get great leadership, we should do
everything to keep that leadership and to make sure that
continuity and the management plan that goes with it is carried
out. My hope and my wish would be that you, in fact--I will
work on my side to make sure you get asked. You work on your
side with your wife to make sure you can. [Laughter.]
Senator Carper. Just to follow up on Dr. Coburn's last
comment there, Senator Brown and I hosted a hearing in this
room a couple of weeks ago with folks from the Department of
Defense, GAO, and a couple of other witnesses, and the thing we
focused on was major weapon systems cost overruns, which have
grown from $42 billion in 2000 to $402 billion last year,
almost a ten-fold increase over 10 years.
One of the things that we have learned, as Senator Coburn
and I earlier drilled down in this stuff, is that it turns out
that the folks in the Department of Defense at the senior level
in charge of overseeing acquisition, development and
acquisition of these major weapons systems, have huge turnover,
an extended period of time where there is basically at the
Assistant Secretary level nobody there. A lot of the direct
reports are not in position, and no wonder we are just chasing
our tail and not doing a very good job at it.
So it is not just the census, but that is just--it is not
uncommon, whether the President is George W. Bush or Barack
Obama, to have something that looks like administrative Swiss
cheese and Executive Branch Swiss cheese, and we have too many
vacancies. One of the things that we have been working on--and
Senator Schumer and Senator Alexander I think are providing
good leadership here--is to reduce by about a third the number
of positions that require confirmation. And we would love to
not only do that, but also to be able to say that whoever is
going to serve as our Census Director--and I hope it will be
you--will serve for a 5-year term with the opportunity to go
beyond that if there is interest in doing that.
All right. I want to go back to the questions that I was
asking of Mr. Zinser and Mr. Goldenkoff. What I was asking is
what you heard from Dr. Groves in his testimony that you are
very pleased with, and I think what you are saying is that the
three of you, the entities that you represent, appear to be on
the same page, which is nice to hear. And I will come back to
you say what were maybe one or two things that you did not hear
that you would like to have heard. But, Mr. Goldenkoff, let me
ask of you first, what did you hear that you especially liked?
And maybe mention a thing or two that you think that you would
like to have heard.
Mr. Goldenkoff. Well, I think it is important to recognize
that the Census Bureau sees that there are really two
components to the challenges that they face going forward. The
first component includes the need to refine and improve
existing operations, in some cases develop new and innovative
techniques, and bring on new technology, like the Internet.
The second component, of course, is the internal management
piece, things like human capital management, their
organizational structure, and from what Dr. Groves said, he is
addressing that as well. So it is important that they combat
the issue of a cost-effective census from these two
perspectives.
What I would like to hear more about is a governance
structure. The Census Bureau has a lot of tests; they have a
lot of things in place, a lot of pieces of the puzzle. The big
challenge going forward then is how is all this going to come
together and how is it going to coalesce into a path to a more
cost-effective census in 2020.
Senator Carper. Mr. Zinser, anything that comes to mind
that you did not hear that you would like to have?
Mr. Zinser. Yes, I think the things that we have pointed
out in some of our reports deal with those kinds of nuts and
bolts that Robert just talked about: Better budgeting, better
project management. For example, with the number of activities
and operations that make up a census, they need integration in
their budgeting and project management documentation.
I think risk management is an area where greater effort is
called for. And I think if they can focus on those kinds of
issues, eventually that will result in a more effective
operation.
Senator Carper. OK, thanks. And on page 3 of your written
testimony, Mr. Zinser, you mention top management challenges
for the 2020 Census, and I think you mention maybe seven of
them.
Mr. Zinser. Yes, sir.
Senator Carper. Of those seven, just pick out one or two of
what you think are the most critical challenges, and then I am
going to ask Dr. Groves to comment on those, please.
Mr. Zinser. I think the most critical challenge that we
identify is addressing the issue of the use of administrative
records to help supplement the enumeration process. It is an
area where--
Senator Carper. For example? Give us an example of that.
Mr. Zinser. Well, there are numerous Federal agencies that
collect information about U.S. households, whether it is the
Veterans Administration or the Social Security Administration,
and there is a lot of data out there that other agencies have
collected that the Census Bureau actually does use for some of
its mission.
I think that there are plans and exercises underway to try
to figure out how to use that type of information, those
administrative records, for the decennial. And I think if that
type of information was used, we could reduce costs for many of
the Census Bureau's operations.
Senator Carper. OK. Would you comment on those points, Dr.
Groves, please?
Mr. Groves. On the administrative records, let me frame the
issue. When we examine our non-response follow-up outcomes, one
negative sign in the 2010 Census is that 22 percent of the
people where we knocked on their doors in a follow-up act, we
never reached. We did not have data from them. And then under
our rules, we seek information from a building manager or
neighbor to determine the count of people inside those houses.
Senator Carper. Would you just pause for a second? Did you
say 22 percent of the people that you tried to follow up with
because you had not heard from them initially, 22 percent never
provided--
Mr. Groves. Right. And in--
Senator Carper. What percent would that be overall?
Mr. Groves. It is 22 percent of roughly 24 percent. So it
ends up being a single-digit number in the--
Senator Carper. About 4 or 5 percent?
Mr. Groves. Yes, something like that.
Senator Carper. So roughly we heard from 95 percent of the
people in the country, households or residents, and roughly--
Mr. Groves. Right, and that 22-percent figure should be
compared in 2000 to 17 percent, so that is a move in the wrong
direction.
Now, I have also received e-mails of people saying, ``Why
are you asking me these questions? Because I have given you the
answers already.'' Now--
Senator Carper. In other formats?
Mr. Groves. Yes. They did not actually give them to us.
They gave them to another Government agency. And they are
right. And under our current procedures--under the old
procedures we would not use those data in any way. So what the
Inspector General is noting is that is a missed opportunity. We
have people who would prefer us to use those data and not
bother them again. But for a variety of reasons, we are not
doing that. Some of them have to do with agreements with other
agencies.
Now, as a statistician, I think our first obligation is to
answer the question, Could we get good data? What kinds of
people are covered that way? What kinds of people are not? And
we know that the records are inadequate for some subpopulation,
so you would not want to use it that way. That could harm the
quality of the census. And we need to check how the attributes
of people are reported there.
So you may recall, when I first testified in front of this
Subcommittee after my confirmation, I noted that we added a
test into the 2010 effort to see whether administrative records
could cover the population. Well, we are in the middle of that
test now, and that would be the first kind of technical answer.
But I would hope Congress would talk about this because this is
a change and we have to make sure everyone is comfortable with
the change.
Senator Carper. Good. Well, we are pretty good at talking
about things. We will certainly talk about that, too. Senator
Brown.
Senator Brown. Thank you, Mr. Chairman.
Mr. Groves, the Commerce Department IG reported that the
Census Bureau hired for the 2010 Census more than 140,000
temporary field employees who received training but worked 40
or fewer non-training production hours, costing the Bureau in
excess of $80 million. What can be done in 2020 to avoid this
waste of taxpayer money?
Mr. Groves. Well, we went back and diagnosed some of that.
First of all, the IG's figures I do not doubt. A lot of that
occurred in the early operations. Let me tell you what
happened. We actually underestimated the ability to recruit,
hire, and train people that did a good job and stayed with us.
We used production models from the 2000 cycle where the
unemployment rate was much lower than it was in the 2010 cycle.
We were able to hire people who really wanted the work. They
put in a lot of hours. They were very good. They finished the
work faster than we thought.
One of the problems is, looking forward, getting good
estimates of productivity next time that takes into account
what the labor market conditions are at the time. We undershot
what the production actually was.
The second thing is sort of risk management on that. It is
a very common tendency in production processes to make sure you
produce on time, on schedule, and one way to reduce the risk as
a manager is to overhire, and then you complete your test. We
need to manage that process better, and we are talking about
how to do that.
Senator Brown. Well, I know you also had some inquiries for
folks that actually were not doing it the right way, and I know
you and I talked about that a little bit. You seem to be the--
Friday night you get a call from somebody saying, ``Hey, by the
way, did you know that this census worker did A, B, C, or D?''
On one occasion, I guess, brought his dog to work with him,
then he was told not to, and, in fact, then got--
Senator Carper. We actually have Senators who do that,
don't we?
Senator Brown. Yes. [Laughter.]
Senator Carper. And they do not always behave well in the
halls.
Senator Brown. These do. Nice try.
Could you explain a little bit about those situations and
how you handle them?
Mr. Groves. Well, there were a lot of situations. When you
have 600,000 people out on the street knocking on 47 million
household doors, a lot of things happen. Some of them are
wonderful things. Some of our enumerators actually saved lives
because they knocked on a door where someone was in the middle
of a heart attack and they--
Senator Brown. Probably because you guys were coming, that
is why they had the heart attack. [Laughter.]
Mr. Groves. Others were bad things, so there were 700
incidents. About 35 percent of them against our enumerators,
where people drew weapons on our enumerators. So it is a very
complicated process. You have to watch it every day. We have
wonderful people who jump on these incidents very quickly and
manage them. We have pretty strict termination rules, so these
are temporary employees--
Senator Brown. How many did you actually terminate then?
Mr. Groves. I actually do not know. I could get you this. I
would be happy to do so. But there were a lot of terminations
because there is not a lot of working with folk--
Senator Brown. If you could let the Chairman and I know how
many folks were terminated during this last census for
inappropriate behavior or just failure to do their jobs, that
would be appreciated. You can just pick up the phone and call
us.
Mr. Groves. Sure, I would be happy to do that.
Senator Brown. Do not reinvent the wheel. I do not want to
do that.
Just to explore a little bit what Dr. Coburn said, if you
are going to use the Internet like that, what are the fraud
prevention mechanisms in place in something like that?
Mr. Groves. Well, a lot has to do with IT security,
encryption procedures--
Senator Brown. I mean just on the individual. How do you
know the individual is--forget the illegal/legal issue, but
what if it is somebody just visiting out of the country?
Mr. Groves. I think the key quality control procedures are
similar on the Internet as they would be on paper. The same
thing can happen on paper, and so we have reinterviewing
procedures to double-check things. We have a lot of statistical
techniques to look at outliers, data that do not look right,
and we follow up on those cases.
Senator Brown. What about the availability of private
industry technology such as mapping and address database
systems? It seems like this group that you all--not you per se
but the Census Bureau actually reinvents the wheel every 10
years. Is there any way to kind of incorporate everything that
other people have been doing for generations now?
Mr. Groves. Well, on the mapping side, we are--
Senator Brown. Computer generations I mean.
Mr. Groves. Yes. All of these things we are pretty actively
partnering in and reaching out to private industry. This is
especially true on the mapping and geographical systems. We are
doing a lot of work with a variety of companies. We are
planning. Our great hope is to save the country money in about
2019 by continuously updating the address file, and we think
that can be done with a lot of new partnerships. So if we can
do that, you will see even more of that, hopefully.
Senator Brown. Great. Well, I appreciate it. Mr. Chairman,
I have to get to the next hearing.
Senator Carper. You bet. I just want to say before you head
for the next hearing, Dr. Coburn and I were here when the
presentation was not as good and the news was not as good.
Senator Coburn. In the 1960s.
Senator Carper. No, not in the 1960s. I think 6 years ago--
not even that, 4 years ago. This is a lot better. I like to say
if it is not perfect, make it better. We have room for
improvement. What does Johnny Collins say about me? He says I
am one of those people who believes in every pile of horse
manure is a pony. [Laughter.]
That is one of many things he says about me.
All right. If I could, maybe a couple questions for Dr.
Groves and then maybe one for Mr. Goldenkoff, maybe even one
for Mr. Zinser, and then we will turn it over to our next
panel.
Dr. Groves, what is the Census doing to ensure that its
plans for an Internet response option will succeed in 2020
given our experience from 2010?
Mr. Groves. Well, we are doing a variety of things, and
maybe the watch word on this is integration. I believe that the
Internet operations we are using on other sample surveys are
relevant to Internet usage in the 2020 decennial Census. We
want to learn lessons from those.
Second, the tricky thing for us this decade will be to do
enough testing of the Internet that will stay nimble on
devices. So the devices that will access the Internet in 2020
will be multi-fold. Some of them have not been conceived of
yet. We want a modern set of alternative tools, devices to
access the Internet, because we think that is the way to
achieve this higher convenience. So we need a lot of tests of
Internet, little, small tests, in order to learn incrementally
and to stay fresh. We cannot lock into device-specific
solutions.
So the way we are avoiding that, we are spending a lot of
time right now getting the base architecture straight. So these
early years ought to get the infrastructure both technically
and procedurally articulated, but allow the device-specific
solutions to be unspecified at that time. Get the architecture
right, then go forward, and at the last moment fix the device
types.
Senator Carper. All right. What are the risks that the
Census anticipated for employing an Internet response option?
And what actions are planned to mitigate these risks?
Mr. Groves. Well, I have talked already about the IT
security side, and the mitigation on that is actually the
things we are going through now in production sample surveys.
So I am pretty sure--I am confident that our IT security group
is staying current with all of the threats that we have on IT
systems, and they need to stay current, and they need to pay
attention to our Internet tests on that.
I think the other unknown will be the reaction of the
American public, especially groups that are traditionally hard
to enumerate, to Internet options as the decade goes by. As
broadband access disperses throughout the different income
groups, we need to watch in order to predict carefully how they
are adapting to Internet use. And so our studies have to be
wise on that so that we can estimate the costs, which will be
related to what proportion choose the Internet for the 2020.
Senator Carper. Mr. Goldenkoff, we are going to come back
and talk a little bit more in this question with you about the
Internet. We see from Dr. Coburn's poster over here that--is
this Canada?
Senator Coburn. England.
Senator Carper. England. We see that in England they have
been using the Internet, and as it turns out, I do not think
they are the only country that has been using it. Some have
done so with some success. Others have done so with failure.
First of all, I do not know if you can mention a couple of
countries that you think might be pretty good role models for
us to look at and see what they are doing right, maybe a couple
to look at to see what they did wrong. But how do we engage the
assistance of other countries that have succeeded--my question
here says how do we engage other counties, but I think it is
how do we engage other countries to see where they have
succeeded and where they have failed. The National Governors
Association (NGA) has something called the Center for Best
Practices, and it is an opportunity for Governors from States
across the country to share what is working and to help other
States that would like to learn from them.
I do not know that we have a Center for Best Practices for
nations like ours that want to learn how to do a census,
conduct a census every 10 years and do it more accurately and
most cost effectively, but it would be nice if we had something
like that. So point us in the right direction. How do we engage
the assistance of other countries that have done well doing
this and have not done well?
Mr. Goldenkoff. On the Internet alone or--
Senator Carper. Yes.
Mr. Goldenkoff [continuing]. Just a general census?
Senator Carper. No. Internet.
Mr. Goldenkoff. Well, Canada has been using the Internet,
some other countries as well. I believe Brazil has used the
Internet. I guess the Census Bureau--I do not want to speak for
it, but are there liaisons or folks who liaison with other
countries?
Senator Carper. Go ahead, Dr. Groves. Feel free.
Mr. Groves. Robert is right. We have an ongoing interchange
with Statistics Canada that has actually been quite aggressive.
We may have people up there right now. They are preparing for
theirs. And we have gone back and forth.
Brazil was a very interesting census this last year because
they used hand-held devices for the entire country, so we sent
a delegation down there. We are watching the U.K. It turns out
that there is a small family of census people around the world
who keep in touch.
Senator Carper. Isn't that nice.
Mr. Groves. They are nice people. [Laughter.]
Senator Carper. All right. Well, that is good to hear. That
is good to hear.
Maybe one for Mr. Zinser. The Census Bureau has a variety
of ongoing evaluations in place to measure the overall
effectiveness of the 2010 design. What steps should it take to
ensure its researching and testing results drive decisions for
future decennial operations?
Mr. Zinser. Well, I think the evaluations that they have
underway right now are the best opportunity we have to know
whether or not the census was of quality. If you ask the
question right now--was the census a success--I do not think
you can actually answer that until you get the results of their
evaluations.
I think Robert is right that the operations were a success
and that the counts were delivered on time. But in terms of the
overall quality of the census data, I think we need to wait for
those evaluations, and then that will inform you in terms of
how good the census actually was.
Senator Carper. OK. All right. And, Dr. Groves, back to you
for another one, if I could. I think there has been some
mention here that the Census is looking, I think it was said,
at six different design options for the 2020 Census. Give us
some idea when the Census will decide on a final design and
what criteria will it use to make its final decision?
Mr. Groves. We anticipate that late 2015 into the 2016
period we would have enough of the findings that the outlines
of the design could be articulated. We are looking at right now
six different alternatives, and I will not go through all of
them, but they vary on how we keep up--how the address list
works, how we keep it up; how we enumerate people, different
modes at different sequences; and then how we organize the
management of the census, how decentralized it is--remember, we
had about 500 different local census offices this time--versus
how centralized could it be. And that will determine
infrastructure costs.
So we are looking at all three of those dimensions, and we
are narrowing things as each month goes by. As we get research
findings, we will be able to drop options, and we would love to
keep you up to date with our progress on that and tell you our
decision process and our recommendations.
Senator Carper. All right. Good. Thank you.
The last question is a question I ask panelists but on a
wide range of issues, and that is, what should we be doing in
this branch of government, the Legislative Branch of
Government, to help make sure--in this case, how do we make
sure that we get a more accurate census 9 years down the road
and we get it in a more cost-effective way, better results,
less money? And we are going to introduce legislation that is
very similar to what Senator Coburn and I introduced last year
that passed unanimously in the Senate. We are going to
introduce legislation probably--when would you say? Tomorrow
afternoon? Tomorrow morning? Maybe not that soon. But we would
like to have your advice on what ought to be in there, and
maybe what ought not to be in there. We would welcome that. I
would ask you to give us that for the record. Folks on our
Subcommittee have 2 weeks to submit questions, and if you would
just respond promptly, we would be grateful. But one of the
questions we will be submitting in writing is as we go forward
with this legislative process, look at the legislation we
offered last time, what is good about it, what should be
changed, maybe what should be dropped, and we would appreciate
your constructive criticism.
And my sense, Dr. Groves, is that you are warming to your
job, and it sounds like you have a good team around you. A
friend of mine is a basketball coach. He has been coaching high
school basketball in Delaware for about 25 years. I ran into
him a couple months ago at the Special Olympics basketball
tournament, which is hosted by the University of Delaware. It
was a great day, a great weekend. And I walked into the Bob
Carpenter Center there at the University of Delaware where the
basketball tournament was going to take place. And while
walking in with my basketball coach friend, we were talking
about the lessons that we learned for life from athletic
competition, all kinds of lessons we learned by virtue of
playing sports. And he talked to me--I mention this as kind of
timely coming right at the end of March Madness, but he said,
``In basketball the best players are not just the ones who
shoot the best. They are not necessarily the ones who rebound
best or dribble best or pass best.'' He said, ``The best
basketball players are the ones who make everybody else on the
team better.''
Think about that. The best basketball players are those who
make everybody else on the team better.
Part of what GAO does and what our IGs are doing is trying
to make sure that everybody on the team that you lead is
better. And I think we are seeing improvement. Clearly we need
to see more, but I am encouraged by the direction that we are
taking.
We thank each of you, one, for your leadership and, two,
for your persistence in this goal to get better results for
less money. Thank you.
With that, we will dismiss this panel and invite the second
panel forward. Thank you so much.
Gentlemen, welcome. Good to see you.
Dr. Thomas Cook, right now the State of Delaware, Governor
Jack Markell, has a secretary of finance, and guess what his
name is?
Dr. Cook. Tom Cook.
Senator Carper. Thomas Cook. We call him ``Tommy.''
Nice to see you, Mr. Castro. We have one doctor, and should
I call you Dr. Castro or is it just Mr. Castro?
Mr. Castro. Mister.
Senator Carper. OK. And how about Mr. Vargas? Is it mister?
Mr. Vargas. Mister.
Senator Carper. OK. So it is. Two misters and a doctor. All
right. I am going to just give a short introduction for each of
you. We are happy that you are here. We appreciate your
presence and your testimony before us.
Daniel Castro is a Senior Analyst with the Information
Technology and Innovation Foundation, specializing in
information technology policy. His research interests include
health IT--that is one of mine, too--data privacy, e-commerce,
e-government, electronic voting, information security and
accessibility. He has experience in the private, nonprofit, and
government sectors. Before joining the Information Technology
and Innovation Foundation, Mr. Castro worked as an IT analyst
at GAO--is that right?
Mr. Castro. That is right.
Senator Carper. All right--where he audited IT security and
management controls at various government agencies. He has a
bachelor's degree in foreign service from Georgetown University
and a master's degree in information security technology and
management from Carnegie Mellon University, two very fine
universities.
Dr. Cook, Thomas M. Cook, is co-chair of the National
Research Council's Panel to Review the 2010 Census. He was
elected to the National Academy of Engineering in 1995 for
leadership in advancing operations research within the
transportation industry, and he has served as President of the
Institute of Management Sciences and the Institute for
Operations Research and the Management Sciences (INFORMS).
Pretty clever. He holds a master's degree in business
administration from Southern Methodist University--does that
make you a mustang?
Dr. Cook. It does, yes.
Senator Carper. Yes, a mustang--and a Ph.D. in operations
research from the University of Texas. A longhorn, indeed. Hook
'em, horns.
Finally, Mr. Vargas, Arturo Vargas, is the Executive
Director of the National Association of Latino Elected and
Appointed Officials, a national membership organization of
Latino policymakers and their supporters. Prior to joining the
National Association of Latino Elected and Appointed Officials,
Mr. Vargas was Vice President for community education and
public policy at the Mexican American Legal Defense and
Education Fund, where he supervised and directed the
organization's community education and leadership development
programs. Mr. Vargas is nationally recognized as an expert in
Latino demographic trends, electoral participation, voting
rights, the census, and redistricting. That is a pretty good
portfolio.
All right, gentlemen. We are glad you are here. We
appreciate your preparation for today's hearing, and your
entire statements will be made part of the record. If you would
like to summarize, that would be fine. If you go over 5
minutes, that is all right. If you go way over 5 minutes, that
is not all right, so I will rein you back in. But why don't you
lead us off, Mr. Castro. Again, welcome. Thank you.
STATEMENT OF DANIEL CASTRO,\1\ SENIOR ANALYST, INFORMATION
TECHNOLOGY AND INNOVATION FOUNDATION
Mr. Castro. Thank you. Mr. Chairman, I commend you for
exploring ways to use information technology to improve the
census. As we heard today, the 2010 Census cost approximately
$13 billion, more than any census before it, and in my view did
not use IT efficiently or cost-effectively. This afternoon I
would like to discuss a few specific recommendations for how
the Census Bureau can better use IT in 2020.
---------------------------------------------------------------------------
\1\ The prepared statement of Mr. Castro appears in the appendix on
page 103.
---------------------------------------------------------------------------
First, Congress should require the Census Bureau to allow
individuals to submit their census form online. Worldwide, more
than 30 countries are providing or experimenting with an
Internet response option for their census, including Canada,
Singapore, Norway, and Australia. Allowing individuals to
submit their form online would increase convenience,
accessibility, and usability for citizens and improve accuracy,
reduce costs, and increase security for the Census Bureau.
For citizens, online forms can be made more user friendly
than a paper form by providing contextual help and multilingual
support.
Some people with disabilities find an online form is easier
to complete and return than a paper form because of the
accessibility features available on computers, such as large
text and screen readers.
Collecting data online can also improve data accuracy over
paper-based methods by better handling atypical responses,
using automated error checking, and eliminating the errors that
can occur during the scanning, decoding, and transcribing
process.
And perhaps most importantly, of course, using the Internet
to collect census data can help reduce the cost of data
collection by reducing the mailback costs, processing costs,
and then the follow-up cost.
In addition to allowing individuals to submit their census
form line, the Census Bureau should incorporate current
technology trends into its planning and operations. I want to
briefly discuss three trends.
First, cloud computing. Cloud computing is a term that
refers to the practice of selling information technology as a
service. Essentially, cloud computing allows organizations to
rent computing power on an as-needed basis. An organization can
scale up or down its IT usage according to demand.
Organizations benefit from the flexibility that cloud computing
offers them as they do not have to make long-term commitments
or have fixed costs. Government agencies, for example, can
better align cost with output by only paying for their actual
use of IT rather than having to overbuild capacity based on
potential demand.
The concepts behind cloud computing--on-demand, scalable,
and pay-per-use--make it ideal for applications such as the
census, which have variable demand for resources. The computing
resources needed by the Census Bureau peaked sharply during the
rather short period of time when individuals and census workers
are submitting responses but go unused at other times. This
means that if the Census Bureau or its contractor use cloud
computing, they would not need to invest in a large amount of
IT infrastructure but could instead only pay for the actual
resources used, and this can, of course, help eliminate
government waste.
The second technology trend that the Census Bureau should
take into account is the proliferation of low-cost, high-
performance mobile devices, such as smart phones and tablet PCs
that access the Internet. Using a mobile device for data
collection and address canvassing can allow census workers to
enter data more accurately and efficiently. Rather than
developing proprietary and expensive hand-held devices, as the
Bureau chose to do in 2010, in the future it should use low-
cost, off-the-shelf equipment, similar to what Brazil did.
By developing platform-neutral mobile apps that run in the
cloud, the Census Bureau can build data collection tools for
census workers that will work on tomorrow's mobile devices. In
addition, if the Census Bureau uses off-the-shelf products in
2020--for example, a consumer-grade tablet PC--it could then
donate these computers to low-income schools after the census
is complete.
Third, the Census Bureau should more actively engage with
individuals who use social networks and mobile devices, which
is an increasingly large share of the U.S. population. This can
help achieve higher response rates and reduce the need for non-
response follow-up, one of the most costly aspects of the
census.
In 2020, individuals will increasingly access the Internet
on mobile devices. Therefore, the Census Bureau should be sure
to incorporate tools to make it easier for individuals to
complete the census using these devices. For example,
technology like QR codes, which are kind of matrix barcodes,
could give individuals the ability to point the camera of a
smart phone at the census form and automatically be directed to
their personal census form online.
In short, the Census Bureau should use IT in the 2020
Census to not only improve existing operations but to find
innovative ways to use technology to deliver more value to
citizens. For example, the Census Bureau or even Congress may
eventually decide that collecting data every decade no longer
makes sense in a world that demands real-time intelligence, and
instead turn to population registers or other sources for this
information. Given the rising cost of conducting the decennial
Census, the Census Bureau should welcome the opportunity to use
IT to reduce costs and improve quality. Certainly, technology
is not a panacea, but it can help organizations like the Census
Bureau achieve their mission more efficiently and effectively.
Thank you.
Senator Carper. Thanks. That was really good. That was one
of the best explanations of cloud computing that I have heard.
Mr. Castro. Thank you.
Senator Carper. And even I could understand that, so that
is good.
Dr. Cook, please proceed.
STATEMENT OF THOMAS M. COOK,\1\ PH.D., CO-CHAIR, NATIONAL
RESEARCH COUNCIL PANEL TO REVIEW THE 2010 CENSUS
Dr. Cook. Mr. Chairman, thank you for asking me to testify
before you today. I am Tom Cook, co-chair of the National
Research Council's Panel to Review the 2010 Census. As such, I
am pleased to be able to summarize the panel's recently
released interim report, ``Change in the 2020 Census: Not
Whether But How.''
---------------------------------------------------------------------------
\1\ The prepared statement of Dr. Cook appears in the appendix on
page 117.
---------------------------------------------------------------------------
I also speak in the capacity in which I accepted the panel
chairmanship in 2009--as an experienced systems engineer
viewing the challenges of the decennial Census from anew. I
trust that you understand that I speak on the panel's behalf
and the National Academies' behalf when commenting on the
panel's interim report, but that, particularly when answering
any questions you may have, my opinions are strictly my own and
should not be construed as formal guidance from the panel or
the Academies.
The Panel to Review the 2010 Census is charged to provide
an independent evaluation of the 2010 Census with an eye toward
suggesting research and development for a more cost-effective
2020 Census. In support of that work, the panel held five
public meetings during the first year of operation, but many of
our panel's impressions were formed through the extensive
series of 58 site visits conducted during 2010 to local
offices, regional census centers, data capture sites, and other
census support facilities.
Our panel is not yet in a position to provide a thorough
evaluation of the 2010 Census; much remains to be learned from
the Bureau's Census Coverage Measurement program and its
procedural evaluations. But I think it is safe to note some
broad outlines as a prelude to 2020 planning. Through our site
visits, we were uniformly impressed by the dedication of the
local and regional census staff--a workforce of exceptionally
high quality. Yet the great paradox of the 2010 Census is that
this high-quality workforce was made to execute plans and
procedures that largely follow the scripts of the 1970 census.
Moreover, in several key respects--including the failed attempt
to fully develop the handheld computers, and the 2006 decision
not to permit Internet response--the 2010 Census was arguably
more hindered than enabled by technology.
From our 18 months of work, I think that the panel is
convinced that it is possible to make the 2020 Census much more
efficient and cost-effective than its predecessors. However,
the central premise of the report is that these significant
efficiencies are possible if, and only if, there is a major
transformation from the 40-year-old, paper-driven processes to
processes that are facilitated using today's technology.
Successfully executing that major transformation will require:
One, a senior management commitment to change that is
publicly announced early in the process and continuously
communicated throughout that transformation process;
Two, continued and frequent involvement and oversight in
the planning process from senior management representing key
departments, including the field organization, not just
headquarters;
Three, adequate early investment in the research and
planning phases of the transformation process;
And, finally, external help for all phases of the
transformation process from research and planning through
development, testing, and implementation. I think this last
point is really important.
In the report, the panel's core recommendations are
attitudinal in nature. We suggest that the Census Bureau needs
to put some stakes in the ground that should not be subject to
debate, once agreed upon. As our ``Not Whether But How''
subtitle suggests, we think the Census Bureau should explore
possible changes as real, viable options but not as purely
hypothetical ideas. As has been observed in the past, increased
use of administrative records data has been thought of as the
``next big thing'' for the next census, for at least the past
three decennials. Until the question changes from simply
whether a change could be made to precisely how and to what
degree a change could be made, promising innovations will
remain as merely hypothetical.
In our report, we explicitly recommend that the Census
Bureau set clear and publicly announced goals. We argue that
the Bureau should commit to significantly reducing, not just
containing, the per housing unit cost of the census, while
limiting the extent of census error. Our experience with
successful reengineering projects like the one we are
anticipating in both the public and private sector is that
setting bold goals is essential to underscore the need and the
importance of that reengineering--again, to avoid it being a
purely hypothetical exercise.
The panel report identifies four high-priority topic areas
for research and development for 2020 planning:
First, the application of operations engineering to census
field data collection operations;
Second, emphasizing multiple modes of response to the
census, including response via the Internet;
Third, the use of administrative records-based information
to supplement a variety of operations;
And, fourth, the continuous improvement and updating of the
Bureau's geographic resources.
A point to emphasize is that the Census Bureau should not
reinvent the wheel but should build on the work from external
experiences. It should learn from other countries, like we
discussed earlier. We spent a lot of time in Canada, at least
we spent a couple days in Canada--not a lot of time but a
couple of days in Canada.
Senator Carper. Did it seem like a lot of time?
Dr. Cook. Yes, it seemed like a lot of time. But it was
very good time spent because they made huge progress with the
Internet, but not only with the Internet but with the field
automation as well.
Senator Carper. Those Canadians are clever, aren't they?
Dr. Cook. Yes, and we can learn a lot from them.
Senator Carper. Yes.
Dr. Cook. The use of administrative records is an area
where ``not whether but how'' is particularly salient. As our
report states, the idea of records as a wholesale substitution
for the census is no longer the most interesting question, if
it ever was. What is interesting or important to study is the
extent to which records might be used throughout the census
process--for updating the address list and inventories of group
quarters facilities, a substitute to asking neighbors or
landlords in ``last resort'' or proxy enumeration or, more
critically, for possible cost reduction--as a possible
supplement to non-response follow-up.
To be sure, there are thorny legal and practical issues
that must be worked through regarding the use of records, but
the existence of those challenges should not stymie active
exploration of the cost-quality trade-offs involved in using
them.
Thank you again for the opportunity to testify. I think I
am out of time.
Senator Carper. Your time has expired. All right. Thanks so
much, Dr. Cook, for that testimony.
Mr. Vargas, please proceed.
STATEMENT OF ARTURO VARGAS,\1\ EXECUTIVE DIRECTOR, NATIONAL
ASSOCIATION OF LATINO ELECTED AND APPOINTED OFFICIALS (NALEO)
EDUCATIONAL FUND
Mr. Vargas. Good afternoon, Chairman Carper. Thank you for
the opportunity to appear before you again today--
---------------------------------------------------------------------------
\1\ The prepared statement of Mr. Vargas appears in the appendix on
page 120.
---------------------------------------------------------------------------
Senator Carper. It is our pleasure.
Mr. Vargas. --on behalf of the NALEO Educational Fund, this
time to discuss the results of the 2010 Census and planning for
2020.
We are one of the leading organizations in census policy
development and public education. In 2010, we led the largest
and most comprehensive privately funded census outreach program
targeting the Nation's more than 50 million Latinos. This
effort included the participation of thousands of elected
officials, community leaders, national and local organizations,
schools, churches, businesses, and a partnership with the
Spanish language media companies Univision, ImpreMedia, and
Entravision.
We believe the 2010 Census was generally a success in
counting every single person living in the United States on
April 1, 2010, as is constitutionally required, and we commend
the Bureau for its undertaking. However, we believe that there
was not a full count of the Latino population because of
significant barriers, many of which are relevant to Census
2020. So we offer recommendations that we hope the Census
Bureau will take into account.
First, as has been previously discussed, we support the
independence of the Census Bureau Directorship, making it a 5-
year appointment, not coterminous with the Presidential
Administration, and we will support provisions of law that you
will introduce on that.
There were several improvements in the 2010 Census
questionnaire. The Bureau redesigned the format and wording of
the questionnaire on race and Hispanic origin to obtain more
accurate responses. However, confusion regarding the
differences between the race and Hispanic origin questions
persist. We encourage the Bureau to continue testing these
questions to secure better data.
The Bureau mailed out bilingual English and Spanish
language forms directly to certain households for the first
time. The Bureau's own analysis shows this strategy led to
higher mail response rates and, thus, cost savings. However, we
experienced difficulties in obtaining information about the
dissemination of these questionnaires. We urge the Bureau to
consider how to make the information on distribution of
bilingual questionnaires more useful to its partners and
examine data in large centers of Latino population to determine
where to distribute the bilingual form in the future.
We applaud the Bureau for disseminating regular information
on response rates. The Bureau provided this information in real
time on its Web site, which is critical for local outreach
efforts. However, we note that the Bureau's Spanish language
Web site was not as comprehensive as it was in English.
The Bureau and its outreach partners experienced
significant challenges in Texas' colonia areas. Much of this
was as a result of miscommunication between the Census Bureau
personnel and the local community. In essence, the local
community was not adequately informed by the Bureau of the
strategies that the Bureau would use to count in the colonias,
that they would not be receiving the form in the mail. This
resulted in extreme confusion and mistrust of the Census, and
as a result, local leaders lack such confidence in the 2020
Census that they are challenging the count.
Senator Carper. Say that again? 2020?
Mr. Vargas. 2010. I am sorry.
Senator Carper. Good. Thank you.
Mr. Vargas. We commend the Census Bureau for seeking the
input of stakeholders such as ourselves and acting on several
of our recommendations with regard to its media plan. However,
despite our urging, the Bureau did not implement a significant
communication strategy targeting native English-language-
speaking Latinos. The Bureau and its contractors failed to
recognize that a communications strategy aimed only at Spanish-
dominant Latinos will not reach all of the Latino population.
The Bureau needs to have a dual strategy of reaching both
English-language-dominant and Spanish-dominant Latinos.
We believe in a robust Partnership Program in making the
census a success, and there were numerous cases where the
vitality of local partnerships played a role in the success of
initial local outreach. We recommend that the partnership
specialists continue their efforts throughout NRFU operations.
Many partner organizations had otherwise no avenues of contact
with the Bureau when the mail-it-out/mail-it-back process
ended. The Bureau should enhance its Partnership Program and
make it an ongoing component of its outreach efforts on all
census activities and between decennials.
Now, with regard to promoting trust in confidentiality, we
recommend that the Federal Government establish an interagency
task force to educate all Federal agencies about the importance
of promoting the Census 2020 and provide guidance on activities
that promote public trust in the confidentiality of the census.
The task force should develop best practices for State and
local governments so that the public receives a consistent
message regarding confidentiality.
Before the census began, there was significant doubt among
many whether the PATRIOT Act superseded other Federal law
guaranteeing privacy in the census. The Department of Justice
issued a letter only a month before Census Day stating that the
PATRIOT Act did not override Title 13 of the U.S. Code. If the
PATRIOT Act is renewed by the Congress, the Department of
Justice should reissue a statement regarding the supremacy of
census confidentiality well in advance of 2020, and the Bureau
should actively publicize this fact.
We also found a need for better communication and
coordination between the Bureau's national office and regional
and local operations. At times national policies were not
communicated effectively to local offices, and national
headquarters was not aware of problems in the field. There were
often inconsistent interpretation and implementation of
practices between local offices.
We believe that the Census Bureau's Advisory Committees
played an important role in guiding and monitoring critical
census policies for 2010 and other census operations. The
charter of the Decennial Advisory Committee has expired, and we
appreciate Director Groves' having worked with the committee to
solicit recommendations for future advisory committees. We urge
the Bureau to implement the recommendations so that a new
advisory committee can provide input in the earliest stages of
Census 2020 planning, including such discussions that we are
having today about Internet responses.
Finally, the Congress is considering a continuing
resolution (CR) for fiscal year 2011. The House-passed version
of the continuing resolution would appropriate 15 percent less
than the President's request. This proposed funding would have
a detrimental impact on several important census activities,
including planning for 2020. We urge the Senate to reject the
House-passed version of the CR and to ensure that the Census
Bureau has the resources needed to conduct the ACS and Census
2020 planning in a cost-effective manner.
Thank you, sir, for the opportunity to testify before you.
Senator Carper. Good. Thank you so much.
I am going to come right back to you, Mr. Vargas, if I
could, and I will ask each of you the same question, just a
very brief question, and I would just ask for a very brief
response. Then we will come back and follow up.
Mr. Vargas, do you think with respect to the census it is
realistic for us to try to achieve in 2020 a better result,
maybe a more accurate result for less money?
Mr. Vargas. Oh, absolutely, and I think some of the
strategies we are discussing here about Internet responses are
an important topic. However--
Senator Carper. You can just stop right there. I will come
back. We will come back.
Mr. Vargas. OK.
Senator Carper. Same question, Dr. Cook.
Dr. Cook. Absolutely, but I think that you should guard
against overly conservative budgets. You should have a bold
budget, an aggressive budget.
Senator Carper. OK. Thank you. Mr. Castro.
Mr. Castro. Yes.
Senator Carper. That was pretty short. Good. Right to the
point.
All right. Mr. Vargas, you are sitting on that side of the
table. If you were sitting on this side of the table, what
would you be doing as a member of the Legislative Branch to try
to ensure that in 2020 we actually achieve a better result, a
more accurate result for less money? So if you were sitting
over here, if you were Senator Vargas, the Honorable Vargas.
Mr. Vargas. The Honorable Vargas, not Hon. Vargas.
[Laughter.]
What I would do is set up certain milestones for the Census
Bureau to reach between now and 2015 when the Bureau decides on
the design of the 2020 Census.
The one thing I was going to say with regard to Internet
responses or any new way of taking the census, what we need to
ensure is that we have the cooperation of the public. And
however the census is conducted, it must be done in a way that
the public actually trusts the confidentiality of the census.
Senator Carper. OK. Thank you.
Dr. Cook. Senator Cook. We actually had a Senator Cook for
many years, the mother of secretary of finance Tom Cook.
Dr. Cook. What I would do is make sure that immediately
``or very soon'' you get a third-party, objective opinion, and
maybe another one, of what the planning process is all about.
That is one thing.
The second thing is I would make sure that the planning
process is adequately funded. That is where the cost will be
driven, by the quality of that process.
Senator Carper. All right. Thank you.
Dr. Cook. And I would again suggest that it is a major
overhaul, it is a blank-sheet-of-paper approach.
Senator Carper. All right. Thank you, sir. Mr. Castro.
Mr. Castro. I think the biggest challenge that I see in
2020--which would be a repeat of 2010 and 2000--is that the
Census Bureau and many Government organizations are very risk
averse to applying new technology and to have innovation, to
have--
Senator Carper. Why do you suppose that is?
Mr. Castro. Well, because they are judged mainly on
performance, not cost savings, and Dr. Groves alluded to this
when he talked about overstaffing and personnel. People
overstaff on technology as well. They use--but they know they
will not get in trouble for--not what is innovative and new and
exciting, and that is a real risk.
Senator Carper. That is a great point. All right.
Mr. Vargas, back to you. A similar kind of question, but
instead of having you put on a Senator hat and sitting on this
side of the dais, put your hat on where Dr. Groves was sitting
and you are the person who is the Director of the Census
Bureau. What would you be focusing on to make sure we get
better results for less money?
Mr. Vargas. Two things. One is I would continue to promote
the Partnership Program so that the relationships that the
Bureau established with community organizations and with
community leaders are sustained as he suggested in his written
testimony, so that come 2018, 2019, we are not investing large
amounts of money to promote Census 2020.
Number two, I would also make sure that I work with the
Congress to ensure that every household has access to the
Internet and to broadband. There is a disparity right now
between African American and Latino households and white and
Asian households in terms of access to the Internet.
Now, responding to the census via the Internet would be
terrific if you have the kind of capacity and accessibility
that everyone else has.
Senator Carper. Give us some idea what the difference is
between Internet access between families, say, of different
ethnic origin.
Mr. Vargas. Well, certainly it is a cost factor.
Senator Carper. No. I am looking for percentages, like 50
percent, 60 percent, 70 percent. Do you know that? Do you have
any idea?
Mr. Vargas. I do not know off the top of my head.
Senator Carper. We will just ask you to respond for the
record.
Mr. Vargas. I would be happy to.
Senator Carper. OK. Thanks.
information for the record
According to the National Telecommunications and
Information Administration (NTIA), 68.8 percent of Asian, 68.3
percent of White non-Hispanic, 49.9 percent of Black, 46.1
percent of Native American and American Indian, and 45.2
percent of Hispanic households use broadband in the home.\1\
---------------------------------------------------------------------------
\1\ Gomez, Anna. June 23, 2011. http://www.naleo.org/2011--Images/
Anna%20Gomez%20Broa dband%20Session.pdf (accessed August 15, 2011).
Senator Carper. Dr. Cook, same question, please.
Dr. Cook. If I were Dr. Groves, I would immerse myself in
this planning process and make it the number one priority for
the next several years.
Number two, I would try to get some external help because I
do not think all the resources required exist at the Census
Bureau.
And number three, I would make sure we had adequate funding
for that planning process.
Senator Carper. All right. Thank you. You are on message.
In my business we say that if you repeat the same thing over
and over again, you are on message. That is good. Mr. Castro.
Mr. Castro. I would look to using technologies,
specifically the Internet and mobile devices, for the
communications side. We talked about the cost, 42 cents to send
out a mail piece. It is a fraction of a cent to send out an e-
mail or online notice on Facebook. In 10 years everyone is
online, even the distributions with different demographics
based on race. It is much higher when you talk about mobile
phones and how different demographics use mobile phones for
Internet access. It is very easy to communicate, and it is a
lot cheaper. You might not get 100, but if you can do 90
percent at a fraction of a cent, that is a lot cheaper than 42
cents.
Senator Carper. Good. Dr. Cook.
Dr. Cook. One other thought. If I were Dr. Groves, I
would--
Senator Carper. You would get rid of that tie, wouldn't
you?
Dr. Cook. Yes, first of all. But I would put some stakes in
the ground. I would say we are going to do the Internet, we are
going to automate the field, and we are going to look at other
census bureaus for best practices.
Senator Carper. All right. Good. Thanks.
Let me just ask maybe a couple of follow-up questions, and
then we will call it an afternoon. I have to go over and start
voting in a little bit.
Let me come back, if I could, Mr. Vargas, to you for this
next question. Much of the success of the 2010 decennial can be
attributed to partnerships--you have alluded to that--with
community-based organizations. Could you just describe for us
the value of the Partnership Program and assess for us, if you
will, its overall effectiveness in ensuring fuller
participation of hard-to-count groups? And what should the
Census Bureau be doing in 2020, between now and 2020, to keep
stakeholders better informed?
Mr. Vargas. Well, the value of the Partnership Program is
that you have a staff of outreach workers who are developing
relationships with trusted messengers in local communities.
Individuals like myself who are willing to stand up and tell
the people who believe me, telling them to believe the Census
Bureau when they say that the census is safe and confidential.
So the Partnership Program is absolutely key in that.
But another thing that also worked extremely well in both
2000 and 2010 was the paid advertising campaign, and that is
something that I think the Bureau needs to continue investment
in.
What was important about the role of nonprofits
organizations, though, this time around is that there were very
little resources provided by State or local governments as
there were in 2000--or in 2010 for organizations such as my
own.
Senator Carper. Say that again? There was less?
Mr. Vargas. There was less. For example, California in 2000
spent $24 million to promote the census within California.
Senator Carper. In what year?
Mr. Vargas. 2000. And in 2010, the amount was less than $1
million.
Senator Carper. Did it have anything to do with their
financial situation?
Mr. Vargas. It had everything to do with the recession,
which is why the role of private foundations was so important.
So I would also encourage the Bureau to maintain relationships
with those foundations so that they are primed to be able to
fund independent efforts come 2018 and 2019.
Senator Carper. OK. Good. Thank you.
One for Dr. Cook and for Mr. Castro. In your statements, I
believe you mentioned that other countries have used--in fact,
I think each of you mentioned other countries have used the
Internet to collect census data, and we have heard that from
others, Senators as well as witnesses. But what has their
experience been like, to the extent you can comment on it? Do
you have any estimates of the range of the savings, the
magnitude of the savings that could be expected, reasonably
expected by using the Internet? What steps should the Bureau
take to minimize security risks? Three parts.
Dr. Cook. As I said, we spent some time in Canada, and they
did not go on the Internet for an initial response because of a
cost-savings motivation. It was mostly because they basically
thought it was the right thing to do. They did not know what
the cost consequences would be before they went. They found
that--well, what they say is a 30-percent take-up rate is their
breakeven; if they get more than 30 percent, they start making
lots of money on the Internet. But another interesting thing
they found was that the quality of the response was
significantly better on the Internet, and, therefore, they did
not have to redo them.
When they calculated the cost savings of the Internet, I am
not sure they captured all the cost savings because some of
those savings are probably hidden. For example, if you are on
the Internet, you have real-time information of who has
responded and who has not, which would avoid the necessity of
somebody knocking on the door two or three times.
The same thing is true with the field operations being
automated. If that information is real-time and you can say do
not go to that next house because we just received an Internet
form, those things, that real-time response, I do not think
they have even tried to measure the cost-effectiveness of that.
Senator Carper. Yes, that is a good point.
Dr. Cook. I think that is a big one.
Senator Carper. That is a good point.
Mr. Castro, do you want to take a shot at those couple
questions? Do you want me to repeat them or are you okay?
Mr. Castro. I think I have them.
Senator Carper. All right. Good.
Mr. Castro. I will focus on what we can learn from other
countries. In Canada, I think what was really interesting in
2011 this year, as they prepare to conduct their census, is
that they are not spending a lot of money to rebuild their
program. They are using what they did in 2006 with a small
upgrade. That is a huge savings right there. Once you do it
once, you do not have to keep doing it again and again. And as
we see, once you do it for ACS or another survey program, you
do not have to reinvent the wheel every time.
The second big savings that I think we can see in a country
like Singapore, what they have done is they have promoted the
Internet response option as a cost-saving measure. So what that
means is first you are given a mailing that says do it online;
then you maybe are given another mailing saying do it online.
Then you are given the option to do it by telephone in an
automated manner, then through mail, and then through--so you
get to the most expensive ways last, and that is another way of
really driving costs in the right direction.
Senator Carper. All right. Good. Maybe one more, if I
could, for you, Mr. Castro. You spoke in your testimony about
cloud computing, but I want to come back and focus on it just a
little bit more. But help us understand how the census can use
cloud computing. And what specific applications do you see,
since this is contracting for--you mentioned the needs for the
technology, for the computer technologies, goes up and down. It
is variable over the life of the census. And the Census Bureau
would seem to be kind of an ideal candidate for this sort of
thing. But just help us a little bit understand better how the
census can use cloud computing. What specific applications do
you see the census contracting for?
Mr. Castro. Sure. So I guess in 2007, I did a model looking
at the numbers that we had for the 2010 Census to look at the
cost savings that were there, and one of the challenges that
the Census Bureau cited at the time, the reason that they said
an Internet response option might not have cost savings, is
because it is so difficult for them to predict the response. So
one of the great things about using cloud computing is you do
not have to know this kind of intelligence ahead of time about
what the actual response will be online. You do not need to
know if everyone will go at 6:00 p.m. when they get home the
day the forms arrive or if they take 2 weeks to respond. You
can just buy the capacity and buy the bandwidth and the server
space and the processing power, and you will pay a fixed cost
for that. It does not matter how much you use or when you use
it because there is that much capacity available. And it is
really important, I think, when we are talking about this that
we talk about the option. There are different types of cloud
computing. There are public clouds and there are private
clouds. Right now most governments have been operating on the
private cloud, which is basically spending a lot of the money
and sharing it among government people. So, you are sharing
resources within government. It is a lot cheaper when you share
resources among everyone, including private companies and,
public cloud offerings. So I think it is very important, when
we look at cloud computing and how the Census Bureau can use it
to drive savings, that public clouds are definitely on the
table.
Senator Carper. Thank you very much.
Sometimes I like to at the end of a hearing just ask you
all if you would like to--I will not ask you to give the
benediction, but I would like for you just to share with us a
closing thought or two. And then I will offer a thought or two,
and we will call it a day.
Mr. Vargas, do you want to lead us in the benediction? Any
closing thought? Maybe something that just pops up given the
discussion we had with this panel, maybe looking back at the
earlier panel, or just something that has been triggered by
virtue of this conversation.
Mr. Vargas. I guess my final thought would be that I would
express my appreciation to Director Groves, who stepped into a
role that I do not think many people would have really been
delighted to do, but he did so in an admirable fashion, and I
think his leadership was critical at the time to make sure that
this census was executed as well as it was, given everything
that he inherited.
And I would also express appreciation to the thousands and
thousands and thousands of Americans across the country who
partnered with the Census Bureau to pull it off.
Senator Carper. Thanks. Dr. Cook.
Dr. Cook. I guess my one thought is that the leadership of
the Census Bureau ought to make the design of the 2020 Census
the number one priority now.
Senator Carper. Really?
Dr. Cook. Yes.
Senator Carper. OK. Thanks.
Mr. Castro, any closing thought?
Mr. Castro. I will be a little more specific on mine. I
think it is interesting that the census, part of the Commerce
Department, the Commerce Department is releasing a National
Strategy for Trusted Identities in Cyberspace on Friday, and
that is something that was not mentioned today, but I think
that has huge implications for how the census can be done in
2020. I would just encourage you to look at that as well.
Senator Carper. OK. Thanks.
A couple questions for Mr. Castro. I think I noted in your
biography that you worked at GAO for a while. When were you
there?
Mr. Castro. In 2006.
Senator Carper. For one year?
Mr. Castro. For one year.
Senator Carper. OK. Was it a pretty good year?
Mr. Castro. It was a good year, got a lot done.
Senator Carper. What was your job then?
Mr. Castro. I was an information security analyst.
Senator Carper. OK. I thought each of you did just a very
nice job with your testimonies, and I thought you did an
especially nice job taking some fairly complex concepts and
making them, even for guys like me, understandable, which is no
small gift.
Dr. Cook, in looking at your background, I think you went
to graduate school, maybe got your Ph.D. at the University of
Texas?
Dr. Cook. Yes.
Senator Carper. They have great athletic teams as well as
good academics, and I am trying to remember last--was it last
night when we had the women's NCAA playoff?
Dr. Cook. That was A&M, yes.
Senator Carper. Yes, it was A&M, but A&M and Notre Dame,
and Texas A&M won.
Now, I know from some of my friends who have gone to A&M
and others who have gone to Texas that they do not always see
eye to eye and there is like a friendly rivalry, kind of like
Ohio State and Michigan.
Dr. Cook. Yes.
Senator Carper. Is there still that rivalry?
Dr. Cook. Oh, yes, big time.
Senator Carper. When Texas A&M, the Aggie women basketball
team, take it all and win the NCAA, how do they feel at the
University of Texas about this, the home of--
Dr. Cook. I have no idea. [Laughter.]
I was in graduate school there and had no time for
athletics.
Senator Carper. I understand. All right.
I think the last thing I want to mention here is sort of
putting all this in context--and I will go back to where I
started off. We face huge budget deficits. We also have a
growing population and a need to count us well, accurately, and
so the next time that we try to figure out how many U.S.
Representatives are going to go to particular States, large and
small, what should they get? We want to be able to ensure that
when city councils or county councils are apportioned that they
actually get the numbers right and the apportionment right. One
of the reasons why that is because we want people to sort of
trust and believe in their political system. Democracy is a
hard enough system anyway, but it is important that we get the
numbers right and the apportionment right so that the right
number of people get the right amount of representation.
The other thing is that a fair amount of money is
apportioned or distributed based on population, and we want to
ensure that we do the best job that we can. So sort of putting
this in context, why is it important for us to have an accurate
census? Well, because there is a lot of money that flows from
the census, either to the right places or, frankly, not the
right places. And the other thing is sort of the basic bedrock
of our democratic society is making sure we know how many
people live in a particular place so they get the appropriate
representation, at least the numbers of representation,
hopefully the appropriate representation, too.
And, last, to put it in context, we are having this battle
not over the long-term budget. We are having what I call a
skirmish now in terms of what are we going to pass for a
spending plan for the next less than 6 months to fund the
government to the end of this fiscal year, the end of
September. Those are important issues, but the really big
issues of what are we going to do for the next 6 years or the
next 16 years to ratchet down our deficit and get it back in
control. And we need to look in every nook and cranny of our
Federal Government in order to find ways to do a lot of the
traditional things that we have done better, more accurately,
and more cost-effectively. And some of the new things that come
along, to be able to use--really to put in place what I call a
culture of thrift in all aspects of our government.
You all have been very helpful in your testimony today in
helping us to look through this lens just a little differently,
maybe look through a little different lens, to get us to that
outcome, which we all seek, and that is better results for less
money, or at least better results for not much more money.
Dr. Cook. Less money.
Senator Carper. Maybe less, that would be good. That is
what we are shooting for.
All righty. Again, Members of our Subcommittee will have a
couple of weeks to submit questions to you, and we would ask
that if you get any questions, respond promptly. One of the
questions you will probably get from me is the legislation that
Senator Coburn and I sought to move last time, we are going to
try to move similar legislation this time. We would welcome any
thoughts that you have as to how we might amend our earlier
version from the last Congress and add an aspect or two or
maybe take something away or amend something. So we would
welcome your input there.
Again, you all have done a very nice job, and we are
grateful to you for your time and for your input.
With that, this hearing is--and I also want to say to our
staffs, too, to our Democratic staff, to our Republican
colleagues, how much we appreciate the work that they have done
in preparing for this day and for the follow-up that will flow
from it.
All right. With that having been said, this hearing is
adjourned. Thanks so much.
[Whereupon, at 3:33 p.m., the Subcommittee was adjourned.]
A P P E N D I X
----------