[House Hearing, 111 Congress]
[From the U.S. Government Publishing Office]





   THE 2010 CENSUS INTEGRATED COMMUNICATIONS CAMPAIGN; CRITERIA FOR 
                IMPLEMENTATION: MEASUREMENTS FOR SUCCESS

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

                                HEARING

                               before the

                  SUBCOMMITTEE ON INFORMATION POLICY,
                     CENSUS, AND NATIONAL ARCHIVES

                                 of the

                         COMMITTEE ON OVERSIGHT
                         AND GOVERNMENT REFORM

                        HOUSE OF REPRESENTATIVES

                     ONE HUNDRED ELEVENTH CONGRESS

                             FIRST SESSION

                               __________

                           SEPTEMBER 22, 2009

                               __________

                           Serial No. 111-27

                               __________

Printed for the use of the Committee on Oversight and Government Reform


  Available via the World Wide Web: http://www.gpoaccess.gov/congress/
                               index.html
                     http://www.oversight.house.gov


                  U.S. GOVERNMENT PRINTING OFFICE
54-383 PDF                WASHINGTON : 2009
-----------------------------------------------------------------------
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 OVERSIGHT AND GOVERNMENT REFORM

                   EDOLPHUS TOWNS, New York, Chairman
PAUL E. KANJORSKI, Pennsylvania      DARRELL E. ISSA, California
CAROLYN B. MALONEY, New York         DAN BURTON, Indiana
ELIJAH E. CUMMINGS, Maryland         JOHN L. MICA, Florida
DENNIS J. KUCINICH, Ohio             MARK E. SOUDER, Indiana
JOHN F. TIERNEY, Massachusetts       JOHN J. DUNCAN, Jr., Tennessee
WM. LACY CLAY, Missouri              MICHAEL R. TURNER, Ohio
DIANE E. WATSON, California          LYNN A. WESTMORELAND, Georgia
STEPHEN F. LYNCH, Massachusetts      PATRICK T. McHENRY, North Carolina
JIM COOPER, Tennessee                BRIAN P. BILBRAY, California
GERRY E. CONNOLLY, Virginia          JIM JORDAN, Ohio
MIKE QUIGLEY, Illinois               JEFF FLAKE, Arizona
MARCY KAPTUR, Ohio                   JEFF FORTENBERRY, Nebraska
ELEANOR HOLMES NORTON, District of   JASON CHAFFETZ, Utah
    Columbia                         AARON SCHOCK, Illinois
PATRICK J. KENNEDY, Rhode Island     BLAINE LUETKEMEYER, Missouri
DANNY K. DAVIS, Illinois             ------ ------
CHRIS VAN HOLLEN, Maryland
HENRY CUELLAR, Texas
PAUL W. HODES, New Hampshire
CHRISTOPHER S. MURPHY, Connecticut
PETER WELCH, Vermont
BILL FOSTER, Illinois
JACKIE SPEIER, California
STEVE DRIEHAUS, Ohio
------ ------

                      Ron Stroman, Staff Director
                Michael McCarthy, Deputy Staff Director
                      Carla Hultberg, Chief Clerk
                  Larry Brady, Minority Staff Director

   Subcommittee on Information Policy, Census, and National Archives

                   WM. LACY CLAY, Missouri, Chairman
PAUL E. KANJORSKI, Pennsylvania      PATRICK T. McHENRY, North Carolina
CAROLYN B. MALONEY, New York         LYNN A. WESTMORELAND, Georgia
ELEANOR HOLMES NORTON, District of   JOHN L. MICA, Florida
    Columbia                         JASON CHAFFETZ, Utah
DANNY K. DAVIS, Illinois
STEVE DRIEHAUS, Ohio
DIANE E. WATSON, California
                     Darryl Piggee, Staff Director











                            C O N T E N T S

                              ----------                              
                                                                   Page
Hearing held on September 22, 2009...............................     1
Statement of:
    Gordon, Judith J., Principal Assistant Inspector General for 
      Audit and Evaluation, U.S. Department of Commerce; and Jeff 
      Tarakajian, chairman and CEO, Draftfcb.....................    97
        Gordon, Judith J.........................................    97
        Tarakajian, Jeff.........................................   107
    Groves, Robert, Director, Census Bureau......................     6
Letters, statements, etc., submitted for the record by:
    Clay, Hon. Wm. Lacy, a Representative in Congress from the 
      State of Missouri, prepared statement of...................     3
    Gordon, Judith J., Principal Assistant Inspector General for 
      Audit and Evaluation, U.S. Department of Commerce, prepared 
      statement of...............................................    99
    Groves, Robert, Director, Census Bureau:
        Letter dated September 30, 2009..........................    61
        Prepared statement of....................................    12
    McHenry, Hon. Patrick T., a Representative in Congress from 
      the State of North Carolina, prepared statement of.........   189
    Tarakajian, Jeff, chairman and CEO, Draftfcb, prepared 
      statement of...............................................   109

 
   THE 2010 CENSUS INTEGRATED COMMUNICATIONS CAMPAIGN; CRITERIA FOR 
                IMPLEMENTATION: MEASUREMENTS FOR SUCCESS

                              ----------                              


                      TUESDAY, SEPTEMBER 22, 2009

                  House of Representatives,
   Subcommittee on Information Policy, Census, and 
                                 National Archives,
              Committee on Oversight and Government Reform,
                                                    Washington, DC.
    The subcommittee met, pursuant to notice, at 2:01 p.m., in 
room 2154, Rayburn House Office Building, Hon. Wm. Lacy Clay 
(chairman of the subcommittee) presiding.
    Present: Representatives Clay, McHenry, and Westmoreland.
    Staff present: Darryl Piggee, staff director/counsel; Jean 
Gosa, clerk; Frank Davis, professional staff member; Yvette 
Cravins, counsel; Charisma Williams, staff assistant; Leneal 
Scott, information systems manager (full committee); Adam 
Hodge, deputy press secretary (full committee); John Cuaderes, 
minority deputy staff director; Dan Blankenburg, minority 
director of outreach/senior advisor; Adam Fromm, minority chief 
clerk/Member liaison; and Chapin Fay, minority counsel.
    Mr. Clay. The Information Policy, Census, and National 
Archives Subcommittee will now come to order.
    Good afternoon, and welcome to today's hearing entitled, 
``The 2010 Census Integrated Communications Campaign; Criteria 
for Implementation: Measurements for Success.''
    Today's hearing has a twofold purpose. We will begin the 
hearing with an update of Census operations from Dr. Groves, 
our new Census Director. This is Dr. Groves' first appearance 
before the Information Policy Subcommittee, so welcome, Dr. 
Groves. After Dr. Groves' presentation, I will have questions 
for Dr. Groves, along with the ranking minority member.
    In the second part of the hearing, we will hear testimony 
regarding the 2010 census integrated communications plan from 
our entire panel and proceed with questions from Members in the 
usual format.
    And, without objection, the Chair and the ranking minority 
member will have 5 minutes to make opening statements after Dr. 
Groves's update of Census operations. All other Members seeking 
recognition will hold their opening statements until the second 
part of the hearing, where they can make opening statements not 
to exceed 3 minutes.
    Without objection, Members and witnesses may have 5 
legislative days to submit a written statement or extraneous 
materials for the record.
    Let me start with an introduction of our new Census 
Director, Dr. Robert Groves.
    President Barack Obama nominated Robert M. Groves for 
Director of the U.S. Census Bureau on April 2, 2009, and Dr. 
Groves was confirmed by the Senate on July 13, 2009. Dr. Groves 
began his tenure as Director on July 13, 2009.
    Dr. Groves had been director of the University of Michigan 
Survey Research Center and research professor at the Joint 
Program in Survey Methodology at the University of Maryland.
    Dr. Groves was elected a fellow of the American Statistical 
Association in 1982, elected a member of the International 
Statistical Institute in 1994, and named a national associate 
of the National Research Council, National Academy of Sciences, 
in 2004.
    Dr. Groves was the Census Bureau's Associate Director for 
Statistical Design, Methodology, and Standards from 1990 to 
1992. In 2008, Dr. Groves became a recipient of the prestigious 
Julius Shiskin Memorial Award in recognition for contributions 
in the development of economic statistics.
    Dr. Groves has authored or coauthored seven books and more 
than 50 articles. Dr. Groves's 1989 book, ``Survey Errors and 
Survey Costs,'' was named 1 of the 50 most influential books in 
survey research by the American Association of Public Opinion 
Research. His book, ``Nonresponse in Household Interview 
Surveys,'' with Mick Couper, written during his time at the 
Bureau, received the 2008 AAPOR Book Award.
    Dr. Groves has a Bachelor's Degree from Dartmouth College 
and a Master's Degree in statistics and sociology from the 
University of Michigan and also a Doctorate at the University 
of Michigan.
    Again, welcome, Dr. Groves.
    [The prepared statement of Hon. Wm. Lacy Clay follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Mr. Clay. And it is the policy of the Oversight and 
Government Reform Committee to swear in our witnesses before 
they testify. Would you please stand and raise your right hand?
    [Witness sworn.]
    Mr. Clay. And let the record reflect that the witness 
answered in the affirmative.
    And, Dr. Groves, would you please proceed?

      STATEMENT OF ROBERT GROVES, DIRECTOR, CENSUS BUREAU

    Dr. Groves. Thank you, Mr. Chairman.
    Chairman Clay, Ranking Member McHenry, and other members of 
the subcommittee, I appreciate this opportunity for being here.
    Upon my confirmation, I promised Congress and Secretary 
Locke that I would spend the first month of my directorship 
evaluating the key components of the 2010 census. I have done 
that.
    The reason for this, as you know, as this committee knows 
well, is that the difficulties with the hand-held computer 
development in the middle of the decade required a major 
replanning. And many things have happened since those events in 
2008, but I needed to take time to make my own professional 
assessment.
    Let me give you a sense of how I did this. Before I 
arrived, plans were in development to bring on two consultants, 
former Census Bureau Director Ken Pruitt and former Principal 
Associate Director John Thompson. They are in place, they were 
in place when I arrived, and I have used them greatly to help 
me on this risk assessment. I have also consulted with members 
of the National Academy of Sciences Panels of the Census. I 
have reached out to a lot of key academic scientists around the 
country and, actually, around the world with relevant technical 
skills. I have met with the staff of GAO, of OMB, of the Office 
of the Inspector General in Commerce. I have talked to project 
leaders of all our major contractors. I am meeting twice weekly 
with MITRE Corp., contractors who offer independent evaluations 
of major Census activities. And then I have had just tons of 
productive meetings with the administrative and technical 
leadership within Census. This has given me the basis of what I 
will report today.
    I have four different kinds of comments. I want to tell you 
my assessment of the 2010 census as a survey methodologist, the 
design on paper, as it were. I will go through some external 
challenges I see facing the 2010 census. I will go through 
internal challenges. And then I want to report on changes I 
have made to Census experimental programs.
    First, let's look at the design of the 2010 census. I can 
say with absolute assurance as a professional survey 
methodologist that if I wrote down the design features of the 
2000 census next to the design features of the 2010 census, I 
would take the 2010 census in a flash. This is a better design, 
and I am sure most of my colleagues around the world would 
agree with this.
    Why do I say this? Using only the short form of the 
questionnaire is a good idea. This should help encourage public 
participation. Sending bilingual questionnaires to 13 million 
households, a first for the Census, is a good idea. We have 
known for decades that supplying people replacement 
questionnaires if they don't return the first one is a good 
idea. This should increase cooperation.
    And you know, I am sure this committee knows, that there 
are two new questions in the short-form questionnaire specially 
designed to improve the coverage properties of the census. 
These are good ideas. You also know that the master address 
file was updated throughout the decade, and that should give us 
a better set of addresses from which we do our mailing.
    A new operation called Group Quarters Validation that is 
going to go on in just a few weeks should improve the quality 
of our listings on crucial kinds of houses that are hard to 
cover: dormitories, multi-unit structures, and so on. And then, 
as this committee knows better than I, the additional funding 
provided by the American Reinvestment and Recovery Act is 
making a difference for our partnership and outreach activities 
in a major way.
    So, with this kind of assessment, I say again it is an easy 
judgment that most professionals would prefer the 2010 design. 
But a superior design doesn't make, necessarily, for a superior 
product, and so I want to speak to a set of challenges that I 
see remaining for my and my colleagues at the Census Bureau. 
And I will start with a set of internal challenges.
    First, the replan of the census in 2008 brought on a new 
leadership team with fewer censuses under their belt than we 
have seen in past decades. This weakness, however, in my 
judgment, is countered by a much more formal and open and 
transparent risk management process that was adopted during the 
replanning.
    And to bolster this further, I have decided to continue 
vigorous use of external advisers, both through existing 
contracts and with John Thompson and Ken Pruitt. Further, I am 
extremely fortunate to have Deputy Under Secretary for Economic 
Affairs Nancy Potok, who was recently appointed, who is a 
former principal associate director with whom I enjoy 
consultative relationships.
    Second, the second internal challenge, like a lot of 
Federal agencies, the Census Bureau has experienced significant 
retirements in senior ranks. I am especially concerned about 
this at the senior mathematical statistical ranks. While we are 
trying to aggressively recruit new talent, I will attempt to 
bring in some outside talent of that nature.
    Third, as you know, because of the replanning of the census 
away from the handhelds for the nonresponse followup stage, we 
are using paper-based operations. The control system for those 
operations is being written, as we speak, with a talented group 
of programmers in Suitland. But this phase of development is 
very tightly scheduled, and it is worth concern.
    As you know, a recent GAO report called for a complete end-
to-end test of this paper-based operation control system. And I 
have examined that recommendation and met with a lot of people 
about the testing process for this system.
    The current plan within the Census Bureau is to have an 
integrated test of core subsystems of the overall design. I 
asked for an outside review of the definition of what that 
``core'' meant, and that review satisfied me that the 
definition does, indeed, represent what should be tested.
    There will also be a large load test of this control system 
around Thanksgiving, which should attempt to simulate the full 
operational load. I have also asked that this test include real 
users at the skill levels of the users of the system during 
production. I have asked that the testing design include 
sequential testing of each of the planned releases of the 
software.
    We have at Census two simulated local Census Offices, one 
at Suitland and one in Stockton, CA. They will be key 
components of the test to make the test realistic, and I 
support that design.
    I should also note that we have created an internal review 
team over this software development, led by our new CIO, Brian 
McGrath. It also contains the Chief Technical Officer of 
Commerce and other experts. They have already provided value 
added, in my belief. Three changes have been made based on 
their input that should improve the overall design and 
implementation of the software, and I look forward to other 
changes from the group.
    The fourth internal challenge is that, at this day, on this 
day, we do not yet know the quality of the master address file. 
We are going to know that in a matter of weeks. When we know 
that, we will have, in my hope, greater assurance that we have 
a master address file that will serve us well in the following 
stages. I would be happy to report to you on those findings 
when we have them.
    Fifth, I believe there is a current challenge regarding 
cost estimation and cost control within the decennial census 
operations and the Census Bureau more broadly. In my belief, we 
need better cost estimation and control at the Bureau. One 
finding of the review of the address canvassing operation that 
you may know about was that the cost models used to guide the 
work didn't forecast the total costs completely well. We have 
to strengthen the cost information and control system within 
the Bureau. We have already intervened in processes to tighten 
that up for nonresponse followup, which is a very large 
activity that will take place in the summer of next year.
    So these are the five principal internal challenges, in my 
belief. There are four external challenges, I believe.
    One, and the most important for this committee and for me 
and for all the leadership of the country, is estimating the 
mail return rate. What will the American public do when we send 
them out these forms? This is a very difficult thing to 
estimate. This is something I have spent my life trying to 
estimate, so I know the difficulty.
    The reason it is difficult to estimate is that the 
population has changed in the last 10 years. In this recession, 
the vacancy rate of households is much higher than it was in 
2000. More and more families are doubling up in houses, due to 
foreclosures and other events. The rate of people experiencing 
homelessness is higher. And, at the same time, we have a public 
debate and attention over immigration issues. And then, five, 
in other surveys that we have been doing, the response rates 
are declining throughout the decade.
    All of these things point to some difficulty in estimating 
what is going to happen when we mail out forms to the American 
public. Will they return them? That is a very important thing, 
as this committee knows, because for each 1 percentage point 
misestimation of that, large sums of money are involved in 
sending people out to followup. So we have to get that right. 
That is a big external challenge.
    Second, we are in a new media environment relative to 2000. 
You know this well. More and more people get the news from 
nontraditional sources. We are doing all we can to learn about 
the blogosphere and how it is going to affect the image of the 
Census Bureau and the behavior of the American public. We have 
launched a media response team that is meeting every Wednesday 
morning to help us get the facts out about census in a way that 
may benefit the return rate.
    Third, there is a challenge for which I need your help. I 
am asking Members of Congress and all census stakeholders to 
work with us to ensure that the census is not tainted by 
intense political debates driving the news media. I can't 
stress this point strongly enough. If the public believes that 
the census data are slanted by partisan influence of one side 
or the other, the credibility of the statistics is destroyed. 
Once destroyed, the public trust can't be easily or quickly 
restored, and we are in deep trouble, both as a Census Bureau, 
as a census, and as a country, in my belief.
    The fourth external challenge is that we live in a digital 
environment that raises the threat of Internet scams and cyber 
crimes like phishing and the widespread use of the Census 
Bureau logo and the brand. I have directed the Chief 
Information Officer of the Census Bureau to establish a team 
that unites our IT security officials with experts from the 
private sector. And I would be happy to report on this in the 
near future, about how we are going to swoop in on fake Web 
sites that appear to be Census Bureau Web sites during this 
census.
    Those are the internal and external challenges. Let me tell 
you four things that I have done to change features of the 
experimental program in the Census.
    No. 1, the first concerns the census coverage measurement 
program, which is used to measure differential undercount. As 
you know, this design has come under some criticism by the 
National Academy of Sciences, and that has to do with the very 
late interviewing start. This is the mechanism by which we 
measure the quality of the census. I am concerned about the 
quality of the recall of where people were on April 1st, when 
they fall into this sample. I am concerned about the quality of 
the matching operations.
    I have brought together a group of statisticians from 
around the country to give us advice on how beef up the quality 
of the measurement of the census coverage program at the risk 
of the sample size of this. This is a tradeoff decision, but, 
in the professional judgment of the statisticians that I have 
been consulting, it appears that we can build a better quality 
estimate of the census if we cut some of the sample and put 
more money into the quality of the measurement.
    Second, we will development a master trace project that 
will follow cases through the census cycle. This will be a 
research tool to understand the tradeoff of operations and the 
quality impacts.
    Third, we will mount an Internet measurement, a reinterview 
study for the census that will focus on how people behave 
differently when they fill out a Web questionnaire versus a 
paper questionnaire. This will be a critical component of 
looking forward about how we use Internet measurement.
    And then, fourth, we will mount a post-hoc administrative 
record census using administrative record data systems we have 
within the Census Bureau, micro-linking them to returns in the 
census to ask the question: If we had done an administrative 
census in 2010, what kinds of people would have been included? 
What kinds of people had been missed? And how are the data 
reported? How are the attributes of people reported versus 
their self-reports, or comparing their self-reports to 
administrative records?
    So I have gone through internal and external challenges and 
also have given you four changes I have made. The internal 
challenges, the uncertainty that I am most concerned about are 
the programming tasks on the paper-based operations control 
system and the not-yet-known quality of the master address 
file.
    But I want to emphasize this as strongly as I can: These 
uncertainties, Mr. Chairman, are swamped by the uncertainties 
about how the American public are going to respond when we send 
out this questionnaire. And it is this that we should focus on, 
I think, as the leadership of this country, because this is the 
single most important thing we can coalesce around to improve 
the quality of the census.
    My clock is not working, and I don't know how I am doing on 
time.
    Mr. Clay. You are doing just fine, Mr. Director.
    Dr. Groves. OK. Let me say a few things about the 
communications plan, and let me know if I am going too long 
here.
    I want to turn to the integrated communications plan 
because I know you are interested in this. You know why this is 
important: Because it is a chief tool to improve the mail 
response rate, to address differential undercount, and to 
assure at that last stage when we send out enumerators to knock 
on doors that people will understand why they are there and 
will cooperate with them.
    Now, I understand before I came on board this subcommittee 
had a briefing on this program, I think it was last spring. So 
I won't go over the entire program, but I would be happy to 
give you a more formal briefing later if you want.
    I directed Associate Director Steve Jost to do a complete 
scrub of the communications campaign when they came in. He was 
there a little before I was there. And the goals of this 
program articulated at that point were to target traditionally 
hard-to-count and linguistically isolated groups to improve 
their mail response rates, but also to help increase the 
overall mail response rate in order to reduce the workload on 
the nonresponse followup operation--a nonresponse followup 
operation that I remind us is now a paper-based, pretty labor-
intensive operation.
    In addition, there was in place an academic assessment 
panel, and we used them to give us guidance and feedback and 
ideas to improve the communications campaign. And then, as I 
have already mentioned, through the ARRA money, $100 million 
was added to advertising activities. We can increase our paid 
media efforts with that extra money, including $43 million 
specifically for local advertising buys.
    The balance of those stimulus funds will be directed to 
partnership support, to public relations, to the census in the 
schools program that I think you have been briefed on already, 
and to the implementation of a 2010 census road tour. With that 
additional funding, in adjusted dollars, we are now exceeding 
the scope of the 2000 census communications campaign.
    Moreover, because funds are available in fiscal year 2009 
which can be used for the advanced purchases of advertising 
time, these funds are expected to provide greater exposure of 
the Census Bureau's message than in 2000. And there is a reason 
for this: Proportionately more of the money is targeted to low-
response areas than was true in the 2000 effort that was more 
nationally targeted.
    At this point, the nonresponse followup media buys haven't 
been fully planned, negotiated, or bought, but our target 
frequency is more than five contacts over the course of the 
nonresponse followup campaign. If our estimates are right on 
this, the American public will see the Census Bureau image and 
get the message many times throughout this campaign. It is 
multitargeted, multimedia, multilingual, and, to my joy, 
research-based.
    One part of the plan already in place will allow us to 
assess and respond to potential issues stemming from the 
replan, and that is a continuous tracking and monitoring 
system. So this will be an advertising campaign for which we 
will have ongoing, near real-time data of how things are going. 
Money has been held back to retarget if we need to focus on 
areas that are showing unexpected results.
    Let me give you kind of a hit parade of the things that are 
the features. We have expanded the number of languages in the 
paid advertising campaign from 14 to 28. We have revamped the 
Web site that will actually go live in a few weeks. I urge you 
and your staffs to look at this when it is live. It should be 
kind of cool.
    We have upgraded the census in the schools program, 
expanding it from K to 8 to K to 12, in 28 different languages. 
We have expanded the plans and the scope of the census road 
tour, something that was quite successful in 2000. We have 
doubled the sampling of the National Partnership Office, and 
they are working together with their colleagues in other 
functional areas at the Bureau. And then we have expanded our 
language assistance programs in a variety of ways, using an 
advance letter and other tools.
    Mr. Chairman, Ranking Member McHenry, you asked in a recent 
letter, I think it was September 9th, that you sent to me for 
updated budget estimates for advertising among the specific 
population groups. We are in the middle of setting of seeking 
RFP responses and trying to achieve those targets. We are 
reviewing and finalizing the creative decisions for use of the 
ARRA money. And we are launching media negotiations for 
national and local advertising buys. We think by late October, 
early November, we will be in shape to give you all of the 
details that you deserve and you would like to know and be 
happy to meet with you at that time.
    Those are my remarks. I am happy to be here with you, and I 
thank you for your interest. And I would be happy to answer 
your questions.
    [The prepared statement of Dr. Groves follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Mr. Clay. Thank you so much, Dr. Groves.
    Let me go to the ranking member now. We will each ask a 
series of questions, and we will then call the rest of the 
panel forward. So I will recognize Mr. McHenry.
    Mr. McHenry. Well, thank you, Mr. Chairman. And thank you 
for holding this hearing. It is nice that we can have a 
bipartisan hearing, and we are all equally interested in what 
the Census Bureau is doing in terms of communications and 
preparing for census day, which is just over--well, just 6 
months away, less than 6 months away.
    Dr. Groves, congratulations on your appointment and getting 
through the process.
    Dr. Groves. Thank you, sir.
    Mr. McHenry. We are very happy you are at the Bureau, and 
we welcome you before the committee. Thank you.
    And I also do want to take a moment and thank you for your 
decisive leadership when it came to the issue of ACORN. This 
was an issue that not all of us have raised, but I, in 
particular, raised it repeatedly before your appointment. And 
my concern was what you mentioned in your testimony about other 
things, is the credibility of the Bureau and the brand, the 
Census Brand, and its use by other organizations. And so I 
commend you for your decisive action there. I know it hasn't 
won wide acclaim, but I think it is important to the integrity 
of the census. So thank you so much for that.
    Well, my staff and the chairman's staff, as well as the 
subcommittee's staff, have heard from the Bureau; they were 
briefed by the Bureau on what happened with the address 
canvassing and how successful that was. They said at that time 
that there would likely be a report in early November. Is that 
still the case?
    Dr. Groves. I know we are doing the analysis and processing 
on that file right now. We have increased the kind of 
diagnostics we are seeking out of the file. Whether that date 
is exactly the right date, we will certainly have information 
about that time and would be happy to share with you as soon as 
we have it. This is a very important component, as you know. 
We've got to get this thing right.
    Mr. McHenry. Now, you mentioned in your testimony what that 
master address file looks like, how valid it is and how strong 
it is. Does that relate back to the canvassing results?
    Dr. Groves. Yeah. As you know, the process we go through is 
a pretty open one. We seek input from local areas for addresses 
that they know and want to add to that file. So we went out 
and, believe it or not, visited 145 million different addresses 
in the country. So every address in the country, basically, was 
visited.
    On 98 percent of those, we took GPS coordinates to help 
locate them. And we verified whether we could find the unit, 
forward it. And when we had trouble locating the unit or 
couldn't find the unit or it was missed in space, it was mis-
geocoded, as we say, then we made a note of that.
    And that process will identify--some of those could not be 
found. A common reason for that is that there is a plan to 
build a subdivision, a small subdivision. Building permits may 
have been let, and, in this recession, the houses weren't yet 
built, but they are planned to be built. And in those kinds of 
cases, we would mark those for potential delete. We actually 
keep the records on the file, but it wouldn't be part of the 
mailing operation.
    Mr. McHenry. So that process is pretty massive. But it is 
likely your target date is sometime in November to, sort of, 
have this?
    Dr. Groves. I could get you an exact date on this. I 
promise, we'll get back to you.
    Mr. McHenry. Would you be willing to come back before the 
end of the year and testify about it?
    Dr. Groves. This is the kind of information that your 
committee deserves, and we would be happy to share it.
    Mr. McHenry. Well, thank you. I appreciate it. And we'd 
hope to have you back, you know, a time or two before March of 
next year, as well, to make sure that we are, sort of, up to 
date, we know what needs to be communicated, and what other 
Members of Congress can help communicate for the Bureau, as 
well.
    Dr. Groves. We'd love to do that. In fact, knowing and 
getting the word out about what are the next operations that we 
are doing--for example, we are mounting this big Group Quarters 
Validation Operation in just a couple weeks. We are out there 
listing for the coverage measurement study. And the more we get 
the word out through you and receive your questions, I think 
the better the whole country is for the census.
    Mr. McHenry. Uh-huh. And I'll have additional questions on 
the communications effort. But there's one story that a 
constituent told me. He said that he was working in his yard 
one day and a gentleman came by, just was walking down the 
street, and had a handheld and said, ``Is this your house?'' He 
says, ``Who are you?'' He says, ``I'm with the Census Bureau, 
and we just need to confirm your address.'' And he said, ``How 
dare he.'' And I thought, ``well, actually, he's trying to save 
you money, so he can just mail you something and you can 
respond back.''
    And so the communications effort is going to be very 
important so that when that guy is working in his yard and 
somebody from the Census comes by, he goes, ``Oh, I didn't mail 
in my form,'' so there's some awareness there. And I certainly 
appreciate that. And I think the committee, as well as 
Congress, on a bipartisan basis, wants to make sure that we 
have the money there necessary, the resources there necessary, 
so that we can get the message out and communicate effectively 
across every community in this country, as the Constitution 
mandates.
    And in your testimony, your written testimony says, and 
what you in essence said, ``One of the findings in our review 
of the address canvassing operation was that the cost models 
used to guide the work did not forecast correctly total costs, 
and we experienced a cost overrun in components of that 
operation. We need to strengthen our cost information 
management structures within the Census Bureau.''
    Can you go into further detail about the amounts and maybe 
the components that experienced cost overruns?
    Dr. Groves. Well, I would be happy to brief you on the 
exact numbers, but let me give you just rough orders of 
magnitude.
    One of the things that was discovered I don't view as a 
misestimation of the cost model, but an unexpectedly large 
workload. So the size of the number of addresses that we went 
out with was larger than we thought.
    Now, why did that happen? Well, this was the first decade 
where we had continuous updates, and so we were receiving from 
the Postal Service routinely through the decade more and more 
addresses. And this was, kind of, the first opportunity to go 
out there and check all of those. Estimating that is a hard 
thing to do, and so roughly half of the overrun is higher 
workload.
    The most troubling part of the overrun for me, from my 
perspective, is about a $30 million component of the overrun 
that had to do with a component of work that occurred if we 
found one of those addresses as a potential delete; you 
couldn't find the address.
    Then, appropriately, and I think to the benefit of the 
Census Bureau, there were quality control procedures to 
followup to make sure that really was something that should 
have been deleted. And those were costly operations.
    So part of the misestimation had to do with not 
anticipating so many deletes out of the file, because in 2000 
there weren't as many, proportionally. So that was the kind of 
misestimation.
    What we are doing right now because of that--you know, that 
operation is over. We can't save the money that was spent. But 
we can put in place procedures to try to prevent such overruns 
from happening in the future.
    And there are, kind of, two things happening now for 
nonresponse followup cost estimation. We are doing a big scrub 
of assumptions at a high level, and that will produce a new 
estimate. And then we are going to bring in folks at the 
operations level and build what some people term a ``bottom-
up'' cost model. We are actually going to have two cost 
estimation procedures, and when they don't agree, we are going 
to fight about what's the right assumption. And I think that 
fighting is really a healthy kind of thing to zero in on the 
costs.
    Mr. McHenry. Uh-huh. And so this is basically what you are 
going to do for the estimated $15 billion, you know, the 
billions that are going to be spent next year, you are taking 
this model----
    Dr. Groves. Well, this will be about a $2 billion component 
related to nonresponse followup. This is the May-June-July big 
push.
    Mr. McHenry. OK. And so there are some lessons out of that. 
One, you mentioned before that you had--well, that Mr. 
Mesenbourg mentioned in his testimony, which was that you had 
highly qualified applicants, and you didn't have to have those 
additional interviews. And so you could foresee some savings 
next year on not having to have multiple interviews and, in 
essence, people not wanting to stay with the job or dropping 
out.
    Dr. Groves. Let me tell you some of that. I have been going 
across the regions now, and in every region that I have been to 
the story is the same.
    This horrible recession we are going through has a benefit 
for us, and the benefit is more applicants of better quality. 
And, once they are hired, they don't quit. They work as many 
hours as they can possibly get. This is all very good for us.
    That lesson of address canvassing we got. And it is 
adjusting targeting for hiring of nonresponse followup. I 
guarantee you that.
    Mr. McHenry. OK. So those assumptions are--and you are 
going to have some more estimates going forward for this 
committee?
    Dr. Groves. Absolutely. We'd be happy to share things with 
you.
    You know, this is tough work. I don't even want to say it's 
easy, because you are making predictions about future behaviors 
that you can't really observe. But we are going to do it 
honestly. We are going to use multiple methods. And we'll see--
--
    Mr. McHenry. I thought that was a career you have chosen 
for yourself.
    Dr. Groves. Well, I know, I know, I know. It is a career.
    Mr. McHenry. Estimating, yeah.
    You know, as a component of this, are there any specifics, 
any specific ideas for controlling these costs that you can 
give us as an example?
    Dr. Groves. Yeah. As a real easy example on the thing we 
were just talking about, every survey organization, every 
census around the world hires more people than they think they 
need to do the job, right? And we don't need to do as much of 
that as we thought we did. So we can reduce those hiring and 
training costs in a major way.
    Mr. McHenry. Thank you.
    And I was just informed very politely by the chairman that 
little red light does mean something. So I yield back the no 
time that I have remaining.
    Thank you, Mr. Chairman.
    Mr. Clay. Thank you, Mr. McHenry. And your colleague, Mr. 
Westmoreland, kind of bluntly reminded you.
    Let me also followup with how Mr. McHenry started and say 
that, in a private phone conversation, I commended the Director 
for his prompt action that he took with ACORN. Let me publicly 
state that I agree with the Director's position, as far as 
removing ACORN from the 2010 census.
    You took prompt action. And they had become a distraction. 
So, very good. And I support you, support your actions.
    Let me ask you, what actions do you plan on taking for the 
nonresponse followup operation to avoid similar cost overruns 
to those experienced doing the address canvassing operation? 
Anything different?
    Dr. Groves. Well, I think the first thing to note, and I 
haven't said it yet today, about nonresponse followup, it's a 
tougher job. It's a job that is done in the evenings and on 
weekends. You have to go to an address when people are at home. 
And so the activities of address canvassing are only partially 
informative to us about what's going to happen in nonresponse 
followup. That is just a caution about how difficult it is.
    But some of the things we are putting in place are those 
that will give us more information at a lower level of detail 
about how, in different local Census Offices, how different 
assignment areas are responding. Do we need more resources in 
one place or another? Are local Census Offices that are using 
one particular approach or calling at a certain rate more 
effective than others?
    What we hope to do is both at the national and regional and 
local level, to have more integrated information, near real-
time, to check and monitor progress so that we can deploy 
resources where they are needed as efficiently as possible.
    Mr. Clay. What steps are you taking to ensure that the 
temporary field staffs follow proper procedures for succeeding 
operations, to avoid the problems reported by the IG during 
address canvassing?
    Dr. Groves. Right. We saw the IG report; I looked at it. 
And we accept and appreciate what they are doing. I think I 
need to say that publicly, because I believe it. What we did 
with that was to act on that information as soon as we could to 
intervene.
    You asked a slightly different question, and that is, what 
can you do at a system level to assure that all the troops at 
low levels are doing what they are trained to do? In addition 
to good training, we have in place, as you know supervisory and 
evaluative criteria that, if we see workers, especially at this 
very compressed, nonresponse followup stage, if we see workers 
violating the training guidelines, we can intervene and 
terminate them very quickly. And we have those management 
procedures in place.
    Hiring such a large group of people that we will do is not 
a simple task. It will be quite likely that one of those people 
is not following--at least one of those people is not following 
training guidelines. We can't fully prevent this. We can, 
however, have management structures in place to intervene as 
quickly as possible, and we do.
    Mr. Clay. Also, you are now in the process of opening an 
additional 344 local Census Offices, for a total of 494. How is 
that process going?
    Dr. Groves. Oh, pretty good. We are on target on signing--
you know, this is a massive operation. It's just incredible, 
looking at it. So there are leases involved where we need the 
partnership with GSA. You have to build out these places; then 
you have to get equipment and furniture in all of them. And 
this is like a huge logistical operation.
    I thought a clever thing was done on the initial local 
Census Offices. As you know, one of our subcontractors, Harris, 
comes in and sets up computer networks. They did a few of the 
LCOs, and they sort of stopped and said, ``OK, how are we 
doing? What are we doing wrong? How could we do this better?'' 
and retooled slightly. And then they are rolling that out for 
others.
    So, so far, so good. We had a few glitches in a certain 
area with leases, but those are getting cleaned up. So we are 
optimistic on this one.
    Mr. Clay. Good.
    The need to comply with Federal legislation associated with 
FBI background checks is of significant importance to me. And I 
would like to ask you about the Bureau's plan to fingerprint, 
using ink and paper, hundreds of thousands of enumerators 
needed for our decennial census.
    Here is my concern. The use of ink and paper to capture and 
process fingerprints is highly prone to error and rejection. I 
have heard up to 40 percent of all fingerprints taken by 
trained personnel can be rejected, causing delays and, most 
importantly, the inability to comply with Federal legislation 
governing successful passage of an FBI background 
investigation.
    Has the Bureau considered using electronic fingerprinting 
as an alternate method to capture the fingerprints for 
processing and comply with the law?
    Dr. Groves. Well, I wasn't there, obviously, Mr. Chairman, 
but I have been briefed on a review. There was, indeed--and I 
could get you a report on this, if you want--an attempt to 
evaluate the purchase of electronic measurement devices. The 
costs of those for an effort as massive of the hiring that we 
are going through was judged to be not worth the quality 
enhancement or the speed enhancement.
    And so, in a way, the address canvassing operation was a 
test of this paper-based fingerprinting, and let's evaluate the 
test. So I am told, going into this, instead of the 40 percent 
figure you just cited about rejects, we were anticipating about 
30 percent. The actual number was 22 percent. And when we 
diagnosed, so why should we put up with 22 percent? Can we do 
better? There are people working on improved training for the 
people taking the fingerprints on this--two separate cards are 
taken per person--to see if we can get better at that.
    The other part of the cost has to do with FBI processing of 
these things. So we're concerned, can the FBI handle the big 
load--we are going to have a lot of fingerprints going through 
the FBI process--in a timely fashion in order to be compliant 
with the law? And we are doing a big load test on that in a few 
weeks, mid-October, to basically simulate the full workload of 
the Census Fingerprintings. We are going to shove that much 
through the system and then get an FBI judgment that they can 
or cannot come through on that. So we'll see mid-October.
    Mr. Clay. Well, Director Groves, please share with this 
committee your documentation and comparisons of the 
fingerprinting. Because I have contrary information that says 
it would be a savings of $10 million to $20 million on the part 
of the Bureau if you used electronic fingerprints. So let's 
share the documentation with this subcommittee.
    Dr. Groves. Yeah, yeah, no, I'd be happy to do that. And, 
in fact, if you have some really much, much cheaper method that 
has the quality we are after, I would love to hear about it.
    Mr. Clay. Yeah, well, let's do some comparison shopping 
here----
    Dr. Groves. OK. Great, great.
    [The information referred to follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Mr. Clay [continuing]. And determine if we can save the 
taxpayers some money.
    In response to Mr. McHenry's inquiry about future hearings, 
we do plan on inviting the Director back for updates, as well 
as other subject matter, in particular for a future hearing. 
It's my understanding that there has been a political thawing 
about sampling that we may need to explore in a bipartisan 
manner. So that could be a future hearing, too.
    Mr. McHenry. Are you smiling as you say that?
    Mr. Clay. Yes, I am, but not facetiously.
    Let me now call the two additional witnesses up to the 
table.
    Mr. Westmoreland. Mr. Chairman, could I have about 5 
minutes just to ask a couple? Mr. McHenry said we were going to 
take 15 minutes a side, and I think he took 10.
    Mr. Clay. That was not the format, but I'll tell you what, 
I'll give you 5 minutes.
    Mr. Westmoreland. OK. If you will do that, Mr. Chairman----
    Mr. Clay. You are going to take 5 of his.
    Mr. Westmoreland. Well, if you don't mind me doing that.
    Mr. Clay. Mr. Westmoreland, you may proceed.
    Mr. Westmoreland. Thank you. And just a couple quick 
comments, Mr. Chairman.
    But, Dr. Groves, I want to tell you that I appreciate your 
sincerity in the meetings that we've had. I think you are very 
sincere about giving the American people the best count 
possible. And, Mr. Jackson, the same thing in the meetings that 
we have had.
    And you were talking about the people under you. I think 
people that work for somebody or work for a corporate group 
look at the sincerity of the people that are leading them and 
want to do that same type job. So I commend you for that.
    Dr. Groves. Thank you. Appreciate that.
    Mr. Westmoreland. And I also want to say that you are 
exactly right on the credibility. We have to make sure that the 
public understands that we are going to count everybody, we are 
going to do the best job that we can counting everybody.
    I want to encourage you to look at letting Members of 
Congress do some PSAs telling people to fill out their form. 
Every meeting I have now, I tell people, ``Please fill out your 
census and send it in. And if you do that, we'll have a good, 
accurate count.''
    Dr. Groves. And we'll save the taxpayer money.
    Mr. Westmoreland. And we'll save the taxpayers money. 
That's correct.
    A quick question. I know that earlier this month our staffs 
were briefed by the Census Bureau on the address canvassing, as 
Mr. McHenry mentioned. They said that the handhelds worked 
well. Do you agree with that, I mean, from everything that you 
have heard?
    Dr. Groves. The way I see it is that they worked well 
enough for that task. You know, we trimmed the task a little. 
We took the large blocks and we didn't use the handhelds for 
the large blocks, because we knew they were having trouble with 
the large blocks.
    So I think the way I'd prefer to think of it: The way we 
used them, they worked well.
    Mr. Westmoreland. OK. Well, I know that they were used for 
this GPS, you know, the address. And what's the problem--if 
they work well in the environment that you say, could they--
there's no way they could have been used to get the responses 
for these 10 questions of people that go out for a nonresponse 
followup?
    Because I know that, at least in the estimations, I 
believe, in the nonresponse followup, if we were able to use 
these handhelds, it would have saved a little over $1 billion 
of taxpayer money.
    Dr. Groves. No, I see the appeal of this. No, I'm with you 
on the logic of your question. And the disappointing answer, I 
think, is that, although they were useful for address 
canvassing, the questionnaire use of those things is another 
software leap, and they weren't ready for that. And, indeed, 
the replanning was motivated by that knowledge. So that 
programming was stopped. You know, they are just not ready for 
that.
    They are also not the kind of GPS devices that you and I 
may have in our car that say, you know, ``turn left'' and so 
on. They allow us to put spots on maps and code those 
coordinates, but they are not really navigational devices in 
the way that you could imagine being used in nonresponse 
followup.
    And then the killer final point is there aren't enough of 
them. Even if those two things were solved that I just 
mentioned, we don't have enough of these devices to run 
nonresponse followup if we wanted to.
    So, you know, it is regrettable, but it is a matter of fact 
that using those in nonresponse followup is not a prudent 
thing.
    Mr. Westmoreland. And that's a real shame, because 
taxpayers spent a lot of money buying those things. And it 
would have been great if they could have been used for the 
nonresponse followup and those simple answers.
    But, with that, Mr. Chairman, if I have any time left, I 
will yield that to Mr. McHenry. If not, thank you for the time.
    Mr. Clay. Thank you, Mr. Westmoreland.
    Now we will proceed under the normal operations of the 
hearing and ask the two witnesses to come forward. We will 
now--and as I stated earlier and without objection, the Chair 
and ranking minority member will have 5 minutes to make opening 
statements, followed by opening statements not to exceed 3 
minutes by any other Member who seeks recognition. And without 
objection, Members and witnesses may have 5 legislative days to 
submit a written statement of extraneous material for the 
record.
    I will now introduce the rest of our panel. We have heard 
from Dr. Groves, who I have already introduced.
    Our next witness will be Judith J. Gordon, who has served 
in Executive Leadership positions within the Office of 
Inspector General since 1994 and became responsible for audit 
and evaluation in June 2008. In her current position, she is 
responsible for the Audit, Evaluation, and Oversight of 
Department of Commerce Program, Organization, Operations, and 
Management, as well as external activities funded by the 
Commerce through contracts or financial assistance such as 
loan, grants, and cooperative agreements.
    Prior to this appointment, Ms. Gordon served 14 years as 
Assistant Inspector General for Systems Evaluation, where she 
led a staff responsible for the Review and Oversight of 
Commerce Information Technology Systems, Policies, Programs, 
and Contracts. Ms. Gordon also served as the Director of OIG 
Systems Evaluation Division from 1991 to 1994. Ms. Gordon 
received a B.A. in economics and a master's degree in public 
policy from the University of Michigan and completed the course 
work for the doctoral program in economics at American 
University.
    Our third witness will be Mr. Jeff Tarakajian--I got it 
right--Tarakajian, executive VP of DraftFCB, the prior 
contractor of the 2010 census integrated communications 
campaign.
    I want to welcome our entire panel to this hearing; and, as 
is the policy of the committee, we swear in all witnesses. 
Would you please stand and raise your right hands.
    [Witnesses sworn.]
    Mr. Clay. Thank you. You may be seated, and let the record 
reflect that all witnesses answered in the affirmative.
    Each of you will have 5 minutes to make an opening 
statement. Your complete written testimony will be included in 
the hearing record. The yellow light will indicate it is time 
to sum up, and the red light will indicate that your time has 
expired.
    Dr. Groves, do you have any additional statements you want 
to make at this time?
    Dr. Groves. I am happy to hear the testimony of my 
colleagues.
    Mr. Clay. That will be good.
    Ms. Gordon, you may proceed with your opening statement.

 STATEMENTS OF JUDITH J. GORDON, PRINCIPAL ASSISTANT INSPECTOR 
GENERAL FOR AUDIT AND EVALUATION, U.S. DEPARTMENT OF COMMERCE; 
        AND JEFF TARAKAJIAN, CHAIRMAN AND CEO, DRAFTFCB

                 STATEMENT OF JUDITH J. GORDON

    Ms. Gordon. Chairman Clay, Ranking Member McHenry----
    Mr. Clay. Make sure your mic is on. Push that button. There 
you go. Thank you.
    Ms. Gordon. Chairman Clay, Ranking Member McHenry, and 
members of the subcommittee, I am pleased to be here today to 
discuss the Census Bureau's management of next year's----
    Mr. Clay. Ms. Gordon, would you pull the mic closer to you? 
There you go.
    Ms. Gordon [continuing]. Of next year's decennial census 
and communications campaign. We recognize the challenges faced 
by the Bureau in this enormous yet critical task. Oversight of 
the 2010 census has been a top priority of the Office of 
Inspector General.
    Today, I will focus my remarks on three areas: one, the 
significant issues we have identified over the past decade in 
key Census operations; two, the problems discussed in our first 
quarterly report to Congress last month; and, three, our 
ongoing review of the Bureau's communications campaign and 
partnership program.
    Over the past decade, we have found critical shortcomings 
in such areas as contracting, address lists, systems 
development, and enumerating hard-to-count populations. These 
challenges remain to this day. Our audit of award fee and 
contract type on field data collection automation resulted in 
several improvements when the contract was renegotiated. This 
occurred after the well-publicized decision to abandon use of 
handheld computers for nonresponse followup.
    We have focused considerable attention on address 
canvassing, as this is key to a successful census. In our 
observations nationwide, we found a central procedure is not 
followed. The Bureau quickly directed the field to correct the 
problem, but at that point over half of the operation had been 
completed.
    Quality control is critical to identifying and correcting 
errors when address listers do not follow procedures. While our 
review is not yet complete, we found that quality control 
employees were unable to make certain address list corrections 
when needed.
    Our first quarterly report to Congress examined the 
Bureau's program management limitations. While risk management 
has improved over census 2000, specific limitations in program 
management systems and data hamper its ability to plan and 
manage the census. Examples include the lack of integrated 
objective measures of cost schedule and progress, unreliable 
cost estimates, and late risk management activities.
    Further, Census stopped reporting the risks associated with 
its handheld computers as a key issue in its monthly status 
reports to Congress, the Department, and OMB, even though the 
issue had not been resolved. This lack of transparency cast 
doubt on overall reporting accuracy.
    Finally, we have been monitoring the Bureau's 
communications campaign, including its contract as well as the 
partnership program. While we continue to assess the 
challenges, the Bureau's management appears to be going well. 
We have, however, noted some delays in getting promotional 
materials to local offices.
    The partnership program is a related component of the 
communications campaign. Census used $120 million in Recovery 
Act funding to hire an additional 2,000 individuals to increase 
partnerships in hard-to-count communities. We will be looking 
at how well Census uses its vastly increased partnership staff.
    In conclusion, the Bureau is taking positive steps to 
increase the mail response rate and participation of hard-to-
count populations.
    With the limitations in its project management systems, 
Census faces significant challenges in assessing progress and 
forecasting cost and schedule overruns for the duration of the 
decennial. Major areas we intend to watch going forward include 
the quality of the master address file, the use of the 
communications campaign and partnership staff, the Bureau's 
progress in developing automation for nonresponse followup on a 
highly compressed schedule and components of the enumeration 
process.
    Mr. Chairman, this completes my summary; and I would be 
happy to respond to questions.
    Mr. Clay. Thank you, Ms. Gordon. You--don't worry about the 
clock. You did fine. You did well. You were under 5 minutes.
    [The prepared statement of Ms. Gordon follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    
    Mr. Clay. Mr. Tarakajian, you are next.

                  STATEMENT OF JEFF TARAKAJIAN

    Mr. Tarakajian. Mr. Chairman, Ranking Member McHenry, 
members of the subcommittee, Team Census 2010 thanks you for 
the opportunity to speak with you today about the integrated 
communications campaign for the 2010 census.
    I am joined today by a colleague at DraftFCB, Mark Hall, 
but I also want to recognize the contributions of our 
subcontractor partners, without whom this campaign would be 
impossible to execute. A few of them are GlobalHue, D Exposito, 
GlobalHue Latino, IW Group, Allied Media, and G&G Advertising, 
who have all tirelessly worked on behalf of the campaign.
    The topic you requested for today's testimony is Criteria 
for Implementation: Measurement for Success. From the very 
beginning of the contract, literally all of the activities that 
the team has been focused around have to do with achieving 
successfully the three goals of the campaign; and, just to 
remind us, they are: increasing the overall mail response, 
delivering an accurate census and reducing the differential 
undercount and, finally, encouraging cooperation with 
enumerators.
    Everything that we are doing has begun aligned with those 
goals, and everything we continue to do remains aligned. 
Absolutely everything we have done is research based. Our 
approach to ensuring success has been to listen and learn from 
others and incorporate that learning into the campaign, for 
example, learning from the consumer through very extensive 
quantitative and qualitative research.
    We have had two phases of communication strategy testing, 
two phases of creative concept testing. The Census Bureau's own 
segmentation analysis, our Census barriers, attitudes, and 
motivator studies have all contributed to a vast amount of 
consumer knowledge. Learning from the opinions and knowledge of 
stakeholders, that of advisory committees, that of oversight, 
including members of this committee, Congress, Senate, and the 
Department of Commerce. Learning from analyzing the 2000 census 
program and its achievements. Learning from the Census Bureau 
itself. The fresh perspective of new leadership at the Bureau, 
as well as field headquarters, regions, and local offices.
    We have had extensive learning from third-party sources, 
and I will just name a few of these. From Simmons Market 
Research, Pew Research, Competitrack, and Yankelovich and 
learning from the recent academic assessment panel report and 
its recommendations.
    And, finally, there is the learning from each other and our 
own professional experience in developing and implementing 
integration communications plans.
    So we will continue as a team to do whatever it takes to 
listen, to learn, and incorporate into the campaign what we 
need to make it successful so that when the final comprehensive 
evaluation of the campaign is done by NORC, it will be apparent 
how the campaign has driven the successful achievements of its 
goals.
    Today, I am pleased to report that we remain on track to 
deliver the campaign to the marketplace in January. We are on 
budget, and we thank the government for the infusion of 
Recovery Act funds which, as the Director has outlined, has 
helped in many ways to expand this campaign and make it more 
pervasive.
    And also we are on track to deliver the very aggressive 
small business subcontracting goals of the contract. We have 
just completed the second and final round of creative concept 
testing and look forward this fall to finalizing all of the 
media buys--the up front media buy, the national media buy, and 
the local media buy--and producing all the creative and giving 
our stakeholders one additional chance to see work in progress 
materials and plans before implementation.
    So, today, we look forward to your questions, your 
observations, and advice that you may have about this 
extraordinary effort and are willing to answer any questions 
that you may have. Thank you.
    [The prepared statement of Mr. Tarakajian follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

    Mr. Clay. Thank you so much for those presentations.
    We will now go to the question and answer period, and we 
will start with Mr. McHenry who will get 10 minutes, and each 
subsequent question he will get 10 minutes. Mr. McHenry, you 
are recognized.
    Mr. McHenry. Thank you, Mr. Chairman.
    Thank you all for testifying. I certainly appreciate it.
    The overall concerns about the communication effort I 
raised with the Bureau back in July, I believe. We wrote a 
letter about the contract and making sure that we have some 
reasonable congressional oversight over this process. I mean, 
it is hundreds of millions of dollars. It has had a substantial 
increase and I think--I believe rightfully so. I think we had 
about $175 million spent on advertising or communications in 
2000; and we are going to have about $260,000 $270,000--I am 
sorry--$260 million this time around. Is that approximately 
right, Dr. Groves?
    Dr. Groves. Well, prior to the Recovery Act money, the sum 
is closer to sort of $200 million. The $100 million infusion 
added to that. So, for the advertising itself, the number of 
320 or 322 is probably the best one to use.
    Mr. McHenry. OK. So certainly we are talking about a 
substantial increase which, as you note in your testimony, is 
according to what the Bureau thinks is necessary and proper and 
will have an effect on lowering nonresponse followup, the need 
for a nonresponse followup.
    So, you know, that was my question before. I had asked 
about the DraftFCB and the GlobalHue contract, and we got back 
basically a quarter of what you sent back. The contract, a 
quarter it of is redacted. We basically have blank sheets of 
paper here that are just greyed out. And that is about 25 
percent of what you sent back; and it appears, based on some of 
the few words that are actually here, it is DraftFCB's small 
business contracting plan.
    My question is about how Congress is able to provide 
oversight over this when we can't even get a document that 
isn't in full sections being redacted. Would you be willing to 
work with us to provide us with this information?
    Dr. Groves. As you know, Congressman, I believe the cause 
of the redaction has to do with the proposer labeling as 
proprietary some of the information within the proposal. 
Working within those constraints, we would be happy to do 
whenever possible.
    Mr. McHenry. OK. And in your letter back to me, you 
reference the Freedom of Information Act, which, you know, 
means any citizen can request this information. We have 
actually--Congress has appropriated the money. We are providing 
oversight. And you reference the Freedom of Information Act in 
multiple places saying that you have already, you know--you 
have released this information under the Freedom of Information 
Act, and you are basically forwarding me that.
    That wasn't my request. And I can understand certain 
sections being redacted of proprietary information but not 25 
pages worth of greyed-out material. I have seen intelligence 
reports not this greyed out. And I am not trying to minimize 
this. I certainly think it is important. But I would like to 
have some cooperation so we can provide some oversight and 
transparency here.
    Dr. Groves. I would be happy to talk to you about that.
    Mr. McHenry. OK. I would really appreciate that.
    You have been showing every sign of willingness to work 
with all interested parties, and I do appreciate your 
leadership. And I am not really here to sort of beat up on you 
on this. I just would like to have some knowledge beyond what 
was sent here. It is almost laughable, the number of redactions 
we have here.
    And even, furthermore, you know, our committee outlines 
what we are requesting as best practices. And this is something 
that the committee puts together. It is not a Republican or a 
Democrat thing. But it is documents responsive to the request 
should not be destroyed, modified, removed, transferred, or 
otherwise made inaccessible to the subcommittee. And neither 
the procurement integrity provisions of title 41 nor the Trade 
Secrets Act of title 18, section 1905, which is part of what 
you reference, prevent Congress from receiving proprietary or 
procurement-sensitive information. So, you know, I think 
providing us with that information is reasonable.
    Additionally, DraftFCB and GlobalHue and their parent 
company and their public group have a pretty troublesome legal 
history. GlobalHue is accused of overbilling the Bermuda 
government $1.8 million; and, among other things, it is alleged 
that GlobalHue failed to keep invoices and billing records and 
charged commissions of up to 181 percent on media bias. What 
was the process to contract with them?
    Dr. Groves. Well, Jeff may want to respond a little on the 
subcontracting side.
    But, as I think this committee was briefed, the process by 
which the original contract was led followed all of the Federal 
guidelines for acquisition of these kinds of services. And 
there are in place, as you know, Congressman, the kind of 
oversight--financial and service delivery oversight--that is 
specified by Federal acquisitions. So those things are in 
place.
    The reference you are making I believe is to an earlier 
behavior on the part of one of the subcontractors.
    Mr. McHenry. Yes. And, also, DraftFCB lost a contract with 
Wal-Mart over allegations of overbilling; and Interpublic was 
fined $12 million by the FCC for accounting fraud. Mr. 
Tarakajian, would you like to respond to that?
    Mr. Tarakajian. Let me answer your GlobalHue question 
first, and then I will subsequently answer the other questions.
    Mr. McHenry. Sure.
    Mr. Tarakajian. GlobalHue as well as all the other 
subcontractors who are part of this contract went through a 
process before we were awarded this contract to identify their 
expertise, their willingness to work with us, their personnel, 
the backgrounds of their personnel, their skill set, their 
financial acumen, their past experience working on the census 
campaign, which was important for a variety of the 
subcontractors. We took into account a whole number of factors 
in putting together a list.
    The other thing to realize is that there were a number of 
other firms like ourselves who were in the marketplace to team 
up with other subcontractors at the time. So we faced a 
competitive environment as well as did everybody else in that 
many of the subcontractors teamed up with other players and 
therefore were unwilling to team up with us or vice versa. So I 
want to reassure you we did go through a very rigorous process.
    The Bermuda situation is something that just came about. It 
was not part of the background when this contract was awarded 
or when the contract was being put together.
    As for ourselves, just to set the record straight on Wal-
Mart, there was a solicitation by Wal-Mart. It came out of our 
Chicago office. We run this out of our New York office. And our 
parent company did a thorough investigation of that, and their 
investigation showed that there was no illegal activities, no 
improprietary activities on behalf of our company relative to 
the Wal-Mart contract.
    Mr. McHenry. OK. I certainly appreciate you addressing 
that.
    I guess the question for you, in general, some of the stuff 
I don't see in the contract and one standard part is a media 
buying fee. There is certainly a percentage for the purchase of 
media. What is that percentage you are charging?
    Mr. Tarakajian. Actually, by contract, there is no media 
commission. Media is handled 100 percent on a pass-through 
basis. So, therefore, the only cost connected with media buying 
is the actual labor involved with making the actual buy. But 
there is no media commission in the contract.
    Mr. McHenry. OK. So it is zero. OK. And there is just 
simply a handling fee, in essence, for labor?
    Mr. Tarakajian. That is correct.
    Mr. McHenry. OK. All right. OK. Well, thank you so much for 
your testimony; and thank you for addressing those issues as 
well. I appreciate you taking the opportunity to fully put 
those things to rest. Thank you.
    Mr. Clay. Thank you, Mr. McHenry.
    Mr. Tarakajian, let me ask you about the--could you discuss 
with the subcommittee the tradeoff between the cost and 
benefits of paid versus earned media? Do you have any opinion 
about that?
    Mr. Tarakajian. Yes. Generally speaking, a key difference 
between paid media and earned media is that in paid media you 
completely control the message and you control not only the 
message itself but where that message is placed, what time, 
etc. Earned media is kind of the opposite. You place a message 
out into the marketplace, and you try to direct it in a certain 
channel in the hopes that you do end up with the message 
expressed the way you would like the message to be expressed in 
the channels in which you would like them to be seen.
    The value of earned media is that it has credibility that 
paid media does not have with target audiences because earned 
media is viewed as coming from trusted voices as opposed to 
paid media where the population knows that the advertiser 
actually pays to have that message put into programming.
    Mr. Clay. Thank you for that response.
    How do you respond to members of your Academic Assessment 
Panel suggesting that the grassroot efforts needs to be 
enhanced as opposed to the paid television media plan? Why not 
use less expensive media that may be more appropriate in 
reaching specific groups, especially those hard-to-count 
populations?
    Mr. Tarakajian. One of the messages or one of the 
recommendations that the Academic Assessment Panel was to step 
back and take a look at the division of spending, the 
allocation of money between paid media and the partnership 
efforts. And, as I think many of you know, in the Recovery Act 
funds, there is an increase to the partnership effort or rather 
a sizable increase to the partnership effort resulting from the 
Recovery Act spending. We do believe that we are reaching the 
right allocation between partnership support materials, which 
is what our role is in this, and the paid media spending.
    Even the paid media spending is dramatically skewed toward 
ethnic populations, multicultural population, the hard-to-count 
population. In fact, in order of magnitude, there is--
approximately where we are heading is that roughly 20 percent 
of the population speaks another language other than English in 
terms of their consumption of media. But that is actually where 
close to 60 percent of the dollars are likely to be channeled. 
Whereas only 40 percent of the dollars are likely to be 
channeled against 80 percent of the population that consumes 
media in English.
    So I think the combination of partnership in activities and 
what is being done in the way the paid media effort is being 
planned are together surrounding the hard-to-count populations 
and motivating them, hopefully, to participate.
    Mr. Clay. Will that include--will those lopsided amounts be 
included in the American Recovery and Reinvestment Act money 
also that you all received?
    Mr. Tarakajian. The numbers that I just quoted reflect the 
total amount of the base plan plus the Recovery Act together. 
And what we were able to do so far is take the Recovery Act 
money and skew it disproportionately toward the ethnic, 
multicultural, and hard-to-count audiences to arrive at those 
numbers.
    Mr. Clay. You know, this is a real reversal from a couple 
of years ago. I mean, when things were tighter with the Census 
budget and different groups started weighing in with this 
subcommittee, they were all fearful that they were losing 
ground in the communications campaign. So you are here today to 
tell me that it has been reversed in that it is now geared 
toward those hard-to-count communities?
    Mr. Tarakajian. Yes. As I said--I will repeat--in fact, 
what I have said, that the majority of the paid media 
allocation is targeted to the ethnic in language, in culture, 
hard-to-count populations. And you see that reflected really as 
a result of the Recovery Act money, the increase in the number 
of languages in the campaign from 14 now to 28 languages. Those 
kinds of things have really, you know, enabled us to make the 
kind of change that a lot of our stakeholders were urging us to 
do early on.
    Mr. Clay. The OIG responded that the Census Bureau has 
been, in their words, diligent in monitoring the Integrated 
Communications Campaign, but there have been delays in delivery 
of initial plan and promotional items. From your perspective, 
what caused these delays and have they been eliminated and will 
these problems be corrected before the decennial?
    Mr. Tarakajian. In the written testimony, I outlined where 
we are as of today and where we will be shortly by the end of 
October in terms of delivery of promotional materials and 
items. And when we had the meeting with the Inspector General 
back in April, we talked about that deficiency and moved very 
aggressively with our team to get out a lot of promotional 
materials into the field, which we have done to the point where 
today I believe there are roughly 11 million pieces that are 
physical pieces, plus a lot of pieces that are on the Web site 
that people can download. So there is a full assortment of 
items that are out there.
    In response to your question specifically about what caused 
the delay and what has changed since then, I think there are a 
number of factors. One is that requirements were difficult to 
get out of the Bureau at the time, and we went back and forth 
on requirement setting and I think landed in a place where it 
was very clear to us what needed to happen. I think our team 
was not as fast as it could be in addressing some of those 
requirement changes.
    And I think, third, the review process that was then in 
place that has now changed dramatically, led to the cycle time. 
Where we are today is the Bureau has implemented a review 
process that is much more streamlined. They have subject matter 
experts that are assigned to each batch of materials, depending 
upon what the topic of those materials are. And what that has 
done is make sure that the right content is reviewed by the 
right person and we get the kind of feedback that we need on a 
more timely basis.
    So I think we have caught up, but we continue to push very 
aggressively on this front to make sure that deadlines are not 
missed and that we meet the expectations in the field.
    Mr. Clay. Very good response.
    And a final question for either you or Dr. Groves. What is 
the Census plan to reach the single, unattached, mobile person? 
Is there a nondigital system in place to reach this group in 
2010? What is the compelling message for this segment of the 
population? And is there a mechanism in place to monitor the 
Internet in respect to the 2010 census? Either one of you.
    Mr. Tarakajian. Let me start with your last point, the 
monitoring. I think Dr. Groves talked a little bit about 
learning about the blogosphere and monitoring there. We have in 
the communications contract a continuous tracking study that 
has an Internet monitor component to it. There will be a base 
wave that will be done this fall, and then we will have 
continuous tracking while the campaign is in the marketplace 
next year.
    One of the recommendations of the AAP, the Academic 
Assessment Panel, was to take a look at the single, unattached, 
mobile segment and ensure that there is more than Internet 
advertising to reach that segment. The answer is that is 
something that we are looking at right now.
    We had in the plan additional things. They will be exposed 
to television, they will be exposed to radio, they will be 
exposed to all the multimedia that everybody else is. The road 
tour and all the other elements will reach them. The challenge, 
though, is that they tend to be, on an index basis, less 
exposed to those media than many of their counterparts. So one 
of the things in the replanning that we are looking at is 
exactly how to fine tune that plan against that group; and when 
we share that with the Bureau in October, that can be part of 
what we ultimately share again with stakeholders later on this 
fall.
    Mr. Clay. It has to be like a high wire act to figure out 
how you are going to touch this segment of the population when 
they know there is an overreliance on texting, cell phones, and 
other new gadgets that are coming out it seems like on a 
monthly basis.
    Perhaps you have something to add to it, Dr. Groves. If 
not, I do understand.
    Dr. Groves. I do think it is actually implied in Jeff's 
point. But on the electronic communication with this 
population, clearly, we have an opportunity to engage our 
hundreds of thousands of partner operations. Because many of 
them have their own electronic communication with their 
constituents, if you will. And so if we can be smart about this 
so that the content we might prepare actually migrates to their 
Web sites in various ways that might be an effective tool.
    Mr. Clay. Thank you for your response.
    Mr. McHenry.
    Mr. McHenry. Yes. And, actually, it is a nice transition 
point. Because, Dr. Groves, you mentioned this. The Internet 
measurement you mentioned in terms of the effectiveness as sort 
of a measurement tool afterwards to make the next census 
better. Can you delve into that? Because I would like to hear 
some of your ideas on ways that we can integrate this, if not 
this census, then in the future.
    Dr. Groves. Well, I think the Census has a plan that must 
go forward the way it is. So what we have added is really a 
little experimental component to answer an important but only 
one of the questions that are related to looking forward in how 
the Internet might be used. That is, do we as humans react to 
questions on a Web questionnaire in different ways? And this is 
part of a much larger research agenda that is going on around 
the world. There is all sorts of work going on on how you can 
portray what measurement effectively so that people answer as 
carefully and as well as they can. That is the focus of this 
particular test.
    But, looking forward, I think it is safe to say and I think 
I--this is a unanimous opinion in the field, that we can't 
imagine a 2020 census without an Internet component. This is 
actually, I think, a very easy judgment to make.
    The harder judgment is how do you best integrate it, and 
that is hard because you and I don't know what the Internet of 
2020 will look like. It will not look like the Internet of 
2010. I think that is a safe bet. So all of these new gismos we 
have will be old-fashioned by 2020.
    And the wisdom that we all require I think is choosing a 
course of planning and cycle testing so that we have a use of 
an Internet in 2020 that is the optimal use of that Internet of 
2020. This is hard.
    I think we have a wonderful vehicle at the Census Bureau 
now, the American Community Survey, which could indeed be used 
more or less as the space shuttle is used, to add on little 
experiments throughout the decade, to inquire when a new gismo 
is created within the Internet. Is that going to be useful for 
us to measure the American society one way or the other? And if 
we are good about this, we can choose the right role of the 
Internet.
    One thing that I think is important to note, the findings 
of survey methodological research on the world or on Internet 
use are not particularly wonderful in terms of whether the 
response rate increases greatly with an Internet option. It is 
the biggest disappointment to my profession right now. We had 
great hopes that if I offered you an Internet option versus a 
paper questionnaire, you would go immediately to the Internet 
option. People are not behaving the way we thought. This is a 
problem for us.
    So this will not be a panacea for 2020. It should be a 
useful tool and an armament of other tools. But, by itself, at 
least at this point, the Internet of 2010 is not that tool.
    Mr. McHenry. Is Internet advertising a component of the 
plan as it now stands?
    Mr. Tarakajian. Internet advertising, paid Internet 
advertising, social media, getting our presence on other 
people's Web sites, anyway you look at it, having a strong 
presence on the Internet, including our own Web site, is a key 
component of where we are headed.
    Mr. McHenry. Very good. Very good.
    You mentioned the American Community Survey. Can you take a 
moment to explain the American Community Survey and whether or 
not you think it effective and important.
    Dr. Groves. Well, as you know better than most American 
citizens, Congressman, the American community survey had as its 
seed the long form of the census. So a wonderful--as I 
mentioned in my testimony, a wonderful property of the 2010 
census is that we are asking Americans to do a much shorter, 
simpler task than before.
    Yet, at the same time, this Congress and earlier Congresses 
have passed many laws that require the measurement of certain 
attributes of the population in order to redistribute funds. 
Every question in that roughly 69 question questionnaire has a 
law sitting behind it that you and your colleagues have passed, 
and that is the tool that allows us to administer--allow 
different agencies to administer those laws.
    Now, it has one other benefit and that is for small 
business owners, for small town mayors, they were cheated in a 
way in past census designs. They had wonderfully rich data once 
every 10 years but only once every 10 years. And now we are 
supplying those towns and those decisionmakers at the very 
small levels of geography more up-to-date information. This is 
a wonderful, grand new thing that the society is going to get. 
It is going to require a lot of education of local people on 
how to use it wisely. So we have a big education task ahead of 
us. But once it soaks into the society--this is a wonderful 
thing for all of us.
    Mr. McHenry. All right. Thank you for touching on that. 
There has been some discussion and debate about the need for 
it, and I certainly agree with you that it is preferable to the 
long form. I think it--in terms of the average American's 
experience and the response rates we have seen, it is going to 
be very fascinating, and I am sure you are interested to see 
the outcome of response rates without the long form as a----
    Dr. Groves. It is going to be fascinating.
    Mr. McHenry. Well, thank you for your testimony. Thank you 
for addressing those two questions that are just of interest to 
me. I appreciate your willingness for being here today. And 
thank you, Mr. Chairman, for your leadership.
    Mr. Clay. Thank you so much, Mr. McHenry; and let me thank 
the panel for their testimony today. Again, Dr. Groves, it is 
so good to have you here in this initial hearing. Believe you 
me, there will be other invitations to come back; and we look 
forward to you coming back. We certainly look forward to the 
sharing of information between the Bureau and the subcommittee 
on the fingerprinting issue.
    Without objection, I will submit an opening statement for 
the record and any other Members' opening statements for the 
record.
    Without objection, the hearing stands adjourned.
    [Whereupon, at 3:38 p.m., the subcommittee was adjourned.]
    [The prepared statement of Hon. Patrick T. McHenry 
follows:]

    [GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]


                                 
