[Senate Hearing 109-1127]
[From the U.S. Government Publishing Office]
S. Hrg. 109-1127
S. 1372, THE FAIR RATINGS ACT
=======================================================================
HEARING
before the
COMMITTEE ON COMMERCE,
SCIENCE, AND TRANSPORTATION
UNITED STATES SENATE
ONE HUNDRED NINTH CONGRESS
FIRST SESSION
__________
JULY 27, 2005
__________
Printed for the use of the Committee on Commerce, Science, and
Transportation
U.S. GOVERNMENT PRINTING OFFICE
65-216 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
SENATE COMMITTEE ON COMMERCE, SCIENCE, AND TRANSPORTATION
ONE HUNDRED NINTH CONGRESS
FIRST SESSION
TED STEVENS, Alaska, Chairman
JOHN McCAIN, Arizona DANIEL K. INOUYE, Hawaii, Co-
CONRAD BURNS, Montana Chairman
TRENT LOTT, Mississippi JOHN D. ROCKEFELLER IV, West
KAY BAILEY HUTCHISON, Texas Virginia
OLYMPIA J. SNOWE, Maine JOHN F. KERRY, Massachusetts
GORDON H. SMITH, Oregon BYRON L. DORGAN, North Dakota
JOHN ENSIGN, Nevada BARBARA BOXER, California
GEORGE ALLEN, Virginia BILL NELSON, Florida
JOHN E. SUNUNU, New Hampshire MARIA CANTWELL, Washington
JIM DeMINT, South Carolina FRANK R. LAUTENBERG, New Jersey
DAVID VITTER, Louisiana E. BENJAMIN NELSON, Nebraska
MARK PRYOR, Arkansas
Lisa J. Sutherland, Republican Staff Director
Christine Drager Kurth, Republican Deputy Staff Director
David Russell, Republican Chief Counsel
Margaret L. Cummisky, Democratic Staff Director and Chief Counsel
Samuel E. Whitehorn, Democratic Deputy Staff Director and General
Counsel
Lila Harper Helms, Democratic Policy Director
C O N T E N T S
----------
Page
Hearing held on July 27, 2005.................................... 1
Statement of Senator Burns....................................... 1
Witnesses
Crawford, Kathy, President, Local Broadcast, MindShare........... 21
Prepared statement........................................... 23
Ivie, George, Executive Director/CEO, The Media Rating Council,
Inc............................................................ 3
Prepared statement........................................... 5
Metzger, Gale, Former President, Statistical Research Inc........ 32
Prepared statement........................................... 35
Mullen, Patrick J., President, Tribune Broadcasting Company...... 27
Prepared statement........................................... 29
Shagrin, Ceril, Executive Vice President for Research, Univision. 18
Prepared statement........................................... 20
Whiting, Susan D., President/CEO, Nielsen Media Research......... 13
Prepared statement........................................... 15
Appendix
Inouye, Hon. Daniel K., U.S. Senator from Hawaii, prepared
statement...................................................... 63
Lautenberg, Hon. Frank R., U.S. Senator from New Jersey, prepared
statement...................................................... 63
Letter, dated September 29, 2005, from Kathy Crawford, MindShare,
to Hon. Ted Stevens............................................ 64
Metzger, Gale, supplementary information......................... 68
S. 1372, THE FAIR RATINGS ACT
----------
WEDNESDAY, JULY 27, 2005
U.S. Senate,
Committee on Commerce, Science, and Transportation,
Washington, DC.
The Committee met, pursuant to notice, at 2:30 p.m. in room
SR-253, Russell Senate Office Building, Hon. Conrad Burns,
presiding.
OPENING STATEMENT OF HON. CONRAD BURNS,
U.S. SENATOR FROM MONTANA
Senator Burns. We'll bring the Committee to order.
Around here, when you get to the last week before the
August break, it gets a little compressed. We have to do many
things in one day in order to finish our work before we go home
on the August break. And so, there will be Members in and
Members out. I was advised by the Ranking Member to start the
hearing this morning, and--or this afternoon, and we can do
that.
First of all, I want to thank all the witnesses for coming.
This is a busy time of the year. I look forward to a
stimulating exchange of views this afternoon. Some of us were
gathered here a year ago at a hearing that I convened to hear
about the problems encountered by phasing in of Nielsen's Local
People Meters in several cities around the country. The
questions then were whether the deployment caused minority and
other groups to be undercounted, and whether Nielsen had listed
enough of its customers before rolling out the technology. The
answers appeared to be yes and no, respectively. At the end of
the hearing, Senator Boxer and I told the witnesses that it
would be best if they could all work out the problems among
themselves.
Well, it seems like that process is a work-in-progress. In
this technical area, that usually does not get a lot of
attention, we have had continued controversy around this. And
so, we have a bill. I introduced a bill because I wanted a
solution to the problem and I didn't see that voluntary
industry efforts were making any headway at all. I still think
a voluntary solution would have been best for all concerned.
And I understand the Media Rating Council has come forth
with a voluntary code of conduct, but the Nielsen organization
will not sign on without major changes. I look forward to
hearing more about that today. But I wonder what happens to
voluntary cooperation once things get tough, or once Congress
is not paying attention, as we are today.
This bill is not about Local People Meters. This Nielsen
technology may or may not be the state-of-the-art, but, if it
is, it's better than the diary system. And if Nielsen customers
want it, then so be it.
And I also do not believe it's in the public interest to
worry about whether the given company's ratings go up or down.
That's something for the market to decide. And I think we can
all agree about that.
This bill is about accountability. It's about making sure
that the system is fair and accurate for all Americans. It
would compel Nielsen to come to the table with the auditors at
the Media Rating Council and accept their changes if minimum
accuracy standards are not met. That is what I care about. That
is why I got involved in this debate. And I believe I have
constructed legislation that will make sure this is the case
today and in the future.
Nielsen needs some kind of an effective oversight, because
it is the only game in town. Companies who need TV ratings data
do not have anywhere else to go today, even if there are
serious concerns about the numbers that they're seeing as a
result of Nielsen's samplings. MRC oversight, with meaningful
enforcement power, would remedy this situation in the best
possible way, because the MRC is made up of Nielsen's
customers. The bill would not involve any government agency.
The MRC would retain its independence and responsiveness to the
members as a private-sector expert group.
The industry's self-regulating model has been approved by
Congress many times and in many different sectors of the
economy. I think the MRC model has been working well for the
last 40 years. Maybe it is time for a change. We will find out
only through hearings and gaining more information.
Recent events have showed us that the MRC's lack of power
to enforce its findings, though, is somewhat of a problem. They
need some teeth. And I would hope that maybe we would find we
could give them some.
The bill would also set the stage for a strong MRC role in
guaranteeing the accuracy of the technologies. We have several
systems that may soon be deployed by Nielsen and others that
would capture time-shifting viewing, out-of-home viewing, and
other methods that may not be developed yet.
Of special interest to me, though, is the decision that the
MRC took in March to take another look at the diary system.
This method, unchanged since the 1950s, is still in use in over
150 local television stations around the country, including all
of them, of course, in Montana and many others around the
country. With this decision, as I understand it, the MRC has
asked Nielsen to cooperate in review of the accuracy of the
diary system. I would hate to think that people in rural areas,
in small towns all over the country, are less important than
the people in the big cities where Nielsen is spending the
resources on the people meters. Rural viewers are very
important to me. When I ran a network in television and radio
stations, I had to depend totally on Nielsen data for my
business. So, I hope Nielsen will cooperate with the MRC on
this.
If it does not, in my mind, that is another important
reason that the bill should pass.
Nielsen ratings determine the value of literally billions
of dollars in advertising. Because our TV industry is supported
largely by advertising dollars, Nielsen ratings, in the end,
determine which television shows get aired and which get
canceled; and so, ultimately, determine what kind of content is
distributed on our public airways.
So, television rating systems have extraordinary cultural,
social, and economic implications. Even in the era of the
Internet, television remains our national town hall. It is a
medium that brings Americans together, and it is shared space
that shapes our national experience. And in a very real sense,
the ratings generated by Nielsen determine content that is
available in that shared space. So, the American public has a
clear and compelling interest in ensuring that these ratings
systems are as fair and accurate as possible.
All viewers must be counted. I hope we can agree on that.
And I believe that the FAIR Ratings bill is an important step
in that direction.
I have no one to hand the football off to, so we will start
taking testimony this morning. We want to--or this afternoon--
we want to thank everyone in attendance. I know there's a great
deal of interest in this issue. And I look forward to hearing
the testimony and the dialogue that we may have before it's all
over.
First of all, I'd like to welcome our first panel of
testimony, Mr. George Ivie, Executive Director and CEO of the
Media Rating Council.
Thank you, Mr. Ivie, for coming this morning, and--or this
afternoon, and--I can't get caught up----
[Laughter.]
Senator Burns.--this afternoon, and we look forward to your
testimony.
STATEMENT OF GEORGE IVIE, EXECUTIVE DIRECTOR/CEO, THE MEDIA
RATING COUNCIL, INC.
Mr. Ivie. Senator Burns and distinguished members of the
Committee, my name is George Ivie, and I serve as Executive
Director and CEO of the Media Rating Council. I thank you for
the opportunity, Senator Burns, to testify this morning on
television ratings accuracy and the FAIR Ratings Act.
My written testimony outlines the history and mission of
the MRC and includes descriptions of our administrative and
accreditation procedures, and we believe we have sound
operations and stated policies for the following: voting on and
accrediting research based on standards compliance, limiting
the influence of any one industry sector or member within our
organization, maintaining independence from measurement
services, and, most importantly, ensuring rigorous industry-
driven audit procedures. For example, about independence, our
membership does not include measurement organizations.
We appreciate the Committee's interest in the accuracy of
television ratings and Congress' reaffirmation of the MRC's
role in the form of this FAIR Ratings Act; however, we have
important suggestions for your consideration in both of these
areas.
As you are all aware, a significant MRC concern has been
Nielsen's commercialization of the San Francisco, Washington,
D.C., and Philadelphia LPM markets prior to an MRC audit. As
you know, Nielsen controls the timing of these audits and the
rollout dates. We also have concern about Nielsen's failure to
disclose adequate test data for some of the LPM
implementations. These situations prevent our illumination of
the quality and performance of the new services prior to
commercialization.
We have sought, and received in June, a commitment from
Nielsen to the auditing and impact data closures we requested,
and we hope that Nielsen remains committed to the audits of the
LPMs and their other significant products after the direct
focus of Congress lessens.
Related to the FAIR Ratings Act, our focus is to assure
audits and committee review and impact data disclosures prior
to commercialization of new products. This focus is driven by
the need to illuminate the quality of the rating products to
users so they can make informed usage judgments. We believe
it's important to avoid a situation where non-accredited
products are prohibited from being commercialized through a
blanket rule, but we just as strongly believe that these
products should be audited.
The MRC is not a political organization, and we have not
sought Congressional actions in the form of a ratings bill. The
legislation appears to raise complex issues of antitrust and
liability for the MRC, beyond my particular training and
expertise, but in which we obviously have a great interest. We
remain on course, seeking completion of the initiatives
recommended in our January 27, 2005, letter to the FTC and to
you, Senator Burns, signed by a strong majority of our board.
These items include agreement by Nielsen to the government to
remain in the MRC audit process for the future for its key
products. It is in Nielsen's power today, in this public
government forum, to reaffirm their commitment to the MRC
process for all their significant products, not just LPM and
not just future products.
This seemingly small item is important to assure that
Nielsen continues to respond and dialogue with the industry
about quality and transparency, which, in turn, should instill
greater public confidence.
Most importantly, we intend to gain consensus on, and
adopt, a voluntary code of conduct that was supplied to rating
services for comment several weeks ago. The code is important
because it adds detail and formal structure to how rating
services are expected to act on audit findings and interact
with the MRC. Our members, Arbitron, Media Mark Research, and
other rating services, have expressed support of this approach.
This week, Nielsen communicated their conceptual agreement, and
more dialogue is needed.
We hope you will agree that such a voluntary code of
conduct will do much to promote the vigorous self-regulation
that Congress envisioned in 1964, which is still very much
needed over 40 years later. The final January initiative
entailed establishing a communication linkage between the MRC
and appropriate Congressional and Executive Branch
representatives to call upon when needed. We believe the
voluntary code of conduct is our key solution to the issues we
face, and, when adopted by Nielsen and other rating services,
this code will provide further assurance that measurement
services are meeting the MRC's mandate of accuracy and
transparency.
In closing, the MRC has strived for four decades to be
faithful to the mission Congress defined for us. As always, we
stand ready to work with the Congress in any way that would be
helpful. I very much appreciate the care and thoughtfulness of
you, Senator Burns, and other Members of the Committee, in
considering the issues that significantly impact the media
ratings marketplace.
Whether legislation is required is fundamentally an issue
and a decision for Congress, though we will follow this debate
carefully to ascertain whether such an initiative could affect
our current work. In any event, we believe our key business
priorities are to seek Nielsen's firm and long-term commitment
to the accreditation process and seek adoption of our voluntary
code of conduct by Nielsen and other measurement services.
I would be happy to answer any questions you have.
[The prepared statement of Mr. Ivie follows:]
Prepared Statement of George Ivie, Executive Director/CEO,
The Media Rating Council, Inc.
I. Introduction to the MRC
I am George Ivie, Executive Director and CEO of the Media Rating
Council (MRC), and I am grateful for the opportunity to present our
views on Nielsen's implementation of the local people meter (LPM)
measurement methodology in general-market media research. I would like
to begin by thanking Senator Burns and Ranking Member Inouye for your
leadership in focusing congressional attention on this technical and
important subject.
The MRC is a non-profit organization that reviews and accredits
audience-rating services through the use of rigorous audits. An MRC
audit includes an independent, detailed, and objective examination of
each aspect of the operations of a rating service (including
methodological protocols) through data provided to it by participating
rating services. The central mission of the MRC is to secure for the
media industry, audience measurement services that are valid, reliable,
and effective through an independent evaluation process, without regard
to outcome. The MRC is independent of, and external to, any rating
service and guards its independence zealously.
1. History and Mission of the MRC
During 1963 and 1964, regulation of the TV and Radio industries
including the purpose and accuracy of audience research were the
subjects of extensive public hearings. This process culminated with a
progress report issued to the 89th Congress of the United States (House
Report No. 1212) \1\ in January 1966. These hearings were held by a
Special Subcommittee on Investigations of the House of Representatives
Committee on Interstate and Foreign Commerce and are commonly referred
to as the ``Harris Committee Hearings on Broadcast Ratings.''
---------------------------------------------------------------------------
\1\ House Rpt. No. 1212, 89th Congress (1966).
---------------------------------------------------------------------------
After an extensive investigation and 3 days of testimony, the
Committee determined that Industry self-regulation, including
independent audits of rating services (such as Nielsen Media Research,
Arbitron or MRI) was preferable to government intervention. In its
report, the Committee concluded as follows: ``The enactment, at this
time, of legislation providing for government regulation of broadcast
audience measurement activities is not advisable. The administration of
a statute providing for such regulation would place an unnecessary
burden on the Federal Government, and it is doubtful that more would be
accomplished than can be accomplished by effective industry
regulation.'' \2\
---------------------------------------------------------------------------
\2\ Id. at p. 21.
---------------------------------------------------------------------------
The Harris Committee hearings resulted in the formation of an
Industry-funded organization to review and accredit audience-rating
services called the Broadcast Rating Council (now referred to as the
MRC). At that time, the Broadcast Rating Council's proposed Industry
self-regulation procedures were reviewed by the U.S. Justice Department
and were found not to be in violation of the antitrust laws.\3\
---------------------------------------------------------------------------
\3\ Letter from William Orrick, Jr. Assistant Attorney General,
Antitrust Division, U.S. Department of Justice to Douglas A. Anello,
General Counsel, National Association of Broadcasters (July 16, 1964)
---------------------------------------------------------------------------
Aligned with the actions deemed necessary by the Committee, the
activities of the MRC include, but are not limited to the following:
The establishment and administration of Minimum Standards
for rating operations;
The Accreditation of rating services on the basis of
information submitted by such services; and
Auditing, through independent CPA firms, of the activities
of the rating services.
The MRC's mission as stated in its By-laws is: ``to secure for the
media industry and related users audience measurement services that are
valid, reliable and effective; to evolve and determine minimum
disclosure and ethical criteria for media audience measurement
services; and to provide and administer an audit system designed to
inform users as to whether such audience measurements are conducted in
conformance with the criteria and procedures developed.'' \4\ This
mission was established with the support and guidance of the House
Committee.
---------------------------------------------------------------------------
\4\ MRC By-Laws.--Board of Directors, Media Rating Council,
Effective March 1964, Updated.
---------------------------------------------------------------------------
2. Standards
Consistent with the By-laws of the BRC and its mission, it
developed minimum standards by which media research is to be measured,
which became effective on March 31, 1964 and have been maintained and
updated by the MRC Board of Directors.\5\ The Standards relate to: (a)
ethics and operations, and (b) disclosures. Ethical and Operational
Standards govern the quality and integrity of the entire process by
which ratings are produced. Disclosure Standards specify the detailed
information about a rating service's methodology and each specific
survey, which must be made available to users, the MRC and its CPA
firm, as well as the form in which the information should be made
available.
---------------------------------------------------------------------------
\5\ See Minimum Standards for Media Rating Research, Media Rating
Council, Inc. (last updated = 10/97).
---------------------------------------------------------------------------
3. MRC Accreditation Process
The MRC Accreditation process is completely voluntary and there is
no legal or compulsory requirement that a rating service submit to an
MRC audit. MRC is often compared to similar private industry self-
regulatory organizations such as the Joint Commission on Accreditation
of Healthcare Organizations (JACHO), which is an organization that
audits and accredits participating hospitals for institutional fitness
and high quality patient services. Similarly, the MRC lends its ``seal
of approval'' to rating services that demonstrate compliance with MRC's
standards of media rating research and that make complete
methodological and survey-performance disclosures to their customers
after completing an extensive audit. Over thirty-five rating service
products were submitted to the MRC Accreditation process last year. Of
these thirty-five products, many represented media-types other than
television.
Accreditation is granted by the MRC Board of Directors if a rating
service complies with the MRC's Minimum Standards for Media Rating
Research and makes materially complete methodological and survey-
performance disclosures to their customers.
The MRC has used several nationally known CPA firms throughout the
years to perform these audits. At present, the audits are conducted by
Ernst & Young, under contract to the MRC. Each rating service agrees to
pay MRC assessments to cover their audit cost; the MRC collects no
funds from rating services other than the direct cost of the Ernst &
Young audits. To be clear, the MRC derives no benefit, financially or
otherwise, from the rating service. MRC's revenue is solely derived
from the dues paid to it by its members. In addition, unlike most CPA
firms, Ernst & Young maintains a specialized group of personnel who
have responsibility for auditing rating service operations and
assessing compliance with the MRC's unique Standards. This Ernst &
Young team only works on media rating service audits.
The central element in the monitoring activity of the MRC is its
system of annual external audits of rating service operations. MRC
audits serve these important functions:
They determine whether a rating service merits Accreditation
(or continued Accreditation); the audit report and related
insight provided by the CPA firm is the primary input into the
Accreditation decision,
They provide the MRC with the results of detailed
examinations by CPA auditors which become the basis for quality
improvements in the service, either by voluntary action or
mandated by MRC as a condition for Accreditation, and
They provide a highly beneficial psychological effect on
rating service performance. Knowledge that CPA auditors may
review their work is a powerful spur for quality work by all
field and home-office personnel of the rating service.
The specific methodological approach of the rating service and the
MRC Minimum Standards for Media Rating Research are the primary drivers
of the audit scope for each participating rating service to be executed
by the CPA firm, on behalf of the MRC. Audits are required to be
conducted at least annually. The MRC establishes an audit committee
made up of member organizations that use research of that media-type to
evaluate audit results and recommend a position on ``Accreditation'' to
the Executive Director of the MRC, who then submits such recommendation
to the MRC Board of Directors. Provision is also made for the
suspension or withdrawal of Accreditation and a documented, formal
hearing procedure applies in such instances.
The MRC's audit includes an independent, detailed and objective
examination of each significant aspect of the operations of a rating
service. In the event that a rating service uses outside professional
vendors (for example, for sampling procedures or for editing and
tabulation of data) these sources are also audited and reported upon.
Resulting audit reports are very detailed (typically 150-300
pages); containing many methodological and proprietary details of the
rating service and illumination of the primary strengths and weaknesses
of its operations. The reports are confidential among the MRC members,
who all sign non-disclosure agreements, Ernst & Young and the rating
service. Audit reports include detailed testing and findings for:
Sample design, selection, and recruitment
Sample composition by demographic group
Data collection and fieldwork
Metering, diary or interviewing accuracy
Editing and tabulation procedures
Data processing
Ratings calculations
Assessment of rating service disclosures of methodology and
survey performance
Pursuant to the last bullet above, the MRC mandates that rating
services disclose many methodology and performance measures, which
would be otherwise unknown, for example:
Source of sample frame
Selection method
Respondents by demographic group versus population
Response rates
Existence of special survey treatments for difficult to
recruit respondent groups such as young or ethnic persons
Editing procedures
Minimum reporting requirements for media
Ascription and data adjustment procedures employed
Errors noted in published reports
Data reissue standards and reissue instances
As a result of the disclosures that a rating service must make in
complying with the MRC Accreditation process, specific audit findings
are not disseminated to the public or the press unless waived by the
service, the MRC, and the CPA firm that conducts the audit. Public
disclosure of proprietary techniques can be detrimental to a rating
service's core business, for example endangering patented information,
and the MRC takes very seriously its obligation to keep proprietary
information confidential as well as the audit reports. Recently a
controversy erupted between the MRC and Nielsen Media Research
regarding the apparent leak of information related to the audit of
Nielsen's Los Angeles LPM service to the Los Angeles Times. MRC in no
way endorsed or condones that behavior as it goes directly against its
code of confidentiality. As a result of this incident, the MRC, in
conjunction with its members, have implemented new rules for the
viewing and discussion of draft and final audit reports among its
membership.
What should be made clear, however, is that the MRC can only
publicly comment on its decision to grant, deny, suspend or withdraw
Accreditation without the consent of the rating service and the
independent CPA auditing firm.
Rating services that are awarded MRC Accreditation are given
permission to display the MRC's logo on the audited research product
indicating compliance with our Standards. MRC Standards are publicly
available; more importantly, the extensive methodological and survey
performance disclosures mandated by the MRC are required to be
available to all rating service customers.
II. MRC Membership, Membership Participation and ``Due Process''
1. Membership
Membership in the MRC is completely voluntary and members pay
annual dues of $10,500 (for reference, MRC dues were $7,500 per year in
1964). The dues are universal in the sense that each member pays the
same amount regardless of the overall size of its organization and are
set at a level that allows participation by organizations of all sizes.
The Board of Directors of the MRC is comprised of one appointed
representative, generally a top media research executive, for each
member organization. Currently there are approximately 95 Board members
in total representing television and radio broadcasting, cable, print,
Internet and advertising agency organizations as well as advertisers
and other trade associations.\6\ As indicated by our membership list,
MRC represents a very broad and diverse amalgamation of the media
industry as well as the largest clients of rating services.
Additionally, we have a provision for formal liaison relationships with
the American Association of Advertising Agencies, the Advertising
Research Foundation and the Association of National Advertisers.
Membership is open to any media organization that relies on, or uses
media research and presently includes both general-market media (e.g.,
the ABC, CBS, FOX, NBC networks) and ethnic media organizations (e.g.,
Black Entertainment and Television and Univision). Conversely,
organizations such as Nielsen or Arbitron that produce media ratings
data are not allowed to be members of the MRC.
---------------------------------------------------------------------------
\6\ Full membership list is attached.
---------------------------------------------------------------------------
2. Membership Participation
MRC members play a critical role in the Accreditation process and
provide valuable insight. MRC's ``Television Audit Committee''
comprised of individual representatives from various member
organizations that have an interest in the accuracy and quality of the
rating service's research. The individuals that sit on this committee
are often the top media researchers of their organizations and
generally do not include television executives or representatives of an
organizations' marketing division. It is in this committee, along with
the oversight of the MRC Staff, that true industry oversight of the
quality and accuracy of television audience measurement services is
performed.
As discussed earlier, it is through the MRC Accreditation process
and the use of rigorous and independent audits, that a rating service
gains MRC Accreditation. However, before Accreditation can be achieved,
the Audit Committee has the task of reviewing a draft of the rating
service audit and discussing the results in detail with the auditor
(Ernst & Young) and the staff of the MRC. Additionally, the rating
service has the opportunity to provide its comments, verbatim, in the
audit report or in a separate letter supplied to the audit committee.
This is a confidential process and strict guidelines and procedures are
followed during this review because of the transparency requirement
that a rating service must meet in order to gain MRC Accreditation.
Once a full review of the audit has been completed, the MRC
presents a ``staff recommendation'' to the full committee on whether in
its opinion taking all the available data in front of it; the rating
service should be accredited. This recommendation is prepared to help
guide the committee as it weighs its decision on Accreditation. The
audit committee will then vote on Accreditation, which in turn serves
as a recommendation for the MRC Executive Director to take to the full
MRC Board of Directors for final approval. At this point the Executive
Director will present the recommendation of the audit committee to the
full Board of Directors along with his assessment. The full Board then
has the responsibility and ultimate authority to vote to grant or deny
Accreditation.
3. ``Due Process''
One very important aspect of the voting and approval process is the
controls and safe guards that are in place to assure that a vote of the
audit committee is fair and impartial. The MRC has a formal policy for
membership voting on MRC Accreditation issues that provides stringent
controls and eliminates the potential for outside influence, during and
subsequent to the voting procedure. The policy is not intended to
stifle in any way the thoughtful discussion that takes place in
preparation of the proposals. The policy is designed to insure a more
proper accounting of ballots and to further maintain the
confidentiality of meeting proceedings. Specifically, it:
Verifies that all votes are accounted for
Reduces the likelihood of miscounting votes
Limits the influence of any one member organization, or
collective segments of the Industry
Minimizes the information that can potentially be divulged
to Non-Members, in violation of the signed confidentiality
agreement
Maintains a physical record of the vote
Provides a means for verification
Voting within the MRC can occur at various levels and follows a
pre-established hierarchy. Below is an outline of the levels at which
voting may take place including a summary of the MRC members that are
entitled to participate, and the responsibility of each group.
Sub-committee(s)--
Subcommittees are comprised of a sub-set of individuals from the
MRC Committee(s) responsible for oversight of the measurement service.
Any committee member claiming to have a business or professional
interest in the matter at hand can elect to participate in a
subcommittee. The MRC Staff will work to ensure that the various
segments of the industry are represented in the Sub-committee. The Sub-
committee is responsible for undertaking a detailed review of the
issue. Multiple sub-committee meetings may be held depending on the
complexity of the issue. The Sub-committee vote is designed to make a
recommendation to the Committee(s). A tie vote will necessitate a
detailed review by a larger Sub-committee group or the Committee.
Committees--
MRC Committees are comprised of MRC members who have a business or
professional interest in the medium for which the Committee has
oversight. These committees may be asked to undertake a detailed review
based on the complexity of the issue. The Committee votes whether to
accept the recommendation of the Sub-committee and the Committee vote
is structured to make a recommendation and provide guidance to the
Executive Director. A quorum is required on all voting matters and a
tie vote will necessitate a detailed review by the Board of Directors.
Board of Directors
The Board of Directors represents all active members of the MRC and
vote on the recommendation submitted by the Executive Director. In
addition, the Board is responsible for the final vote on all
Accreditation issues and a quorum is required on all voting matters.
Executive Director
The Executive Director is responsible for making a recommendation
to the Board of Directors and considers the recommendation of the
Committee(s), though he is not required to recommend the Committee(s)
position to the Board. However, the Executive Director must convene a
board meeting to discuss in detail any recommendation whereby the
Executive Director's position differs from that submitted by the
Committee. The Executive Director may take any issue directly to the
Board of Directors for a vote.
Voting Guidelines
All active Board Members are entitled to a vote in the
Accreditation process. A member company designates the
representative(s) to attend meetings and vote. The MRC recommends the
voting representative be a senior ranking individual with knowledge of
the subject matter. When a detailed review of the subject matter is
called for, the voting representative must be in attendance for the
majority of the review meeting. Anyone not in attendance for the full
meeting will be allowed to vote at the discretion of the MRC Executive
Director. A member company representative may participate in-person,
via phone or video-conference and is allowed to represent a maximum of
two votes, for multi-vote organizations. In addition, this
representative is required to submit vote(s) in writing with the
exception of those participating via phone or conference call.
Individuals participating via electronic means (e.g., phone, etc.) have
the option to cast votes via personal call to MRC Staff, fax, or e-
mail. Verbal votes require follow-up written (e.g., fax, e-mail, etc)
confirmation.
Special Circumstances
Special circumstances occur when an MRC member whose company has a
vested interest in the matter being considered. When this occurs, that
member may participate in the review meeting but will not be allowed to
vote. Situations of this nature will be disclosed prior to the start of
the meeting. Any un-anticipated voting conflicts are to be resolved by
the MRC Executive Director.
Voting Results
When a vote takes place the rating service will be advised of the
final outcome as soon as possible and summary-voting results may be
divulged to the Rating Service when deemed appropriate by the Executive
Director. Individual member votes will not be divulged by the MRC and
members are free to state their voting intention prior to the official
vote. However, members may divulge their individual vote outside of the
meeting subject to the policy of the signed Non-Disclosure Agreement on
record at the MRC.
III. Status of LPM Audits--Boston, New York, Los Angeles, Chicago, San
Francisco, Philadelphia, and Washington, D.C.
Nielsen's primary products cover national programs, local programs,
syndication, cable, satellite, as well as dedicated research for
Hispanics and by implication the advertisements for all of these
vehicles. Nielsen also provides several electronic tools and
applications used to deliver ratings to their customers. The MRC
accredits several, but not all, of Nielsen's products.\7\ Nielsen's
National Service based on a people-meter methodology has been MRC-
Accredited since the late 1980s; Nielsen's meter-diary based Local
Service was originally Accredited in the 1960s; Nielsen's National
Hispanic Service (NHTI) has been Accredited since 2000. We believe
these services materially comply with our Standards, although the MRC
does maintain a separate ongoing dialogue with Nielsen regarding
quality issues noted in the audit process in an effort to improve the
quality of research. Other Services such as Nielsen's Hispanic Station
Index (NHSI), and certain other Black and Hispanic Audience Reports are
not currently Accredited or audited.
---------------------------------------------------------------------------
\7\ Complete List of MRC Accredited Services.
---------------------------------------------------------------------------
1. Boston
Nielsen Media Research first ``rolled out'' its local people meter
(LPM) in Boston in 2001. This was Nielsen's first experience with the
LPM in a general-media local market environment. It is our
understanding that Boston was chosen as the first market by Nielsen
because of several factors, including its more homogenous population
and smaller size. While one can argue about this characterization of
the Boston media market, it became clear that Nielsen's assumptions
about easily measuring the market proved to be inaccurate. During
calendar years 2001 and 2002, the MRC audited Nielsen's LPM rollout in
Boston. The audit of the service was extensive and subsequently the MRC
denied its Accreditation to the Boston LPM based on strong concerns
with Nielsen's implementation of the service. However, despite the
concerns raised by the MRC audit and denial of Accreditation, Nielsen
continued a commercial implementation of the Boston LPM. At the same
time, most local broadcasters in Boston did not utilize Nielsen's LPM
services. However, during the ensuing year, Nielsen took extensive
actions to cure the issues raised by MRC's audit. Upon Nielsen making
the recommended changes, MRC gave its Accreditation to the Boston LPM
in the Fall of 2002 approximately 9 months after its initial audit.
After its Boston experience, the MRC Television Committee took the
unusual step of recommending to Nielsen that future LPM implementations
only be commercialized after Accreditation is achieved and that new LPM
sample households not be integrated into Nielsen's National panel prior
to achieving Accreditation.
2. New York
The MRC began its audit process of the New York LPM (NYLPM) during
the early part of 2004. The New York market is arguably the most
difficult market to measure particularly in obtaining the cooperation
of households. The market is highly diverse and represents unique
challenges in compiling accurate and reliable data. Fieldwork began in
this market in April of 2004 and the MRC utilized its full audit scope
and procedures for assessing the service. Ernst & Young conducted the
audit using its standard Nielsen auditing team, which included bi-
lingual personnel. There were many problems identified in the audit,
including race and origin classification errors, excessive and
excessively disproportionate faulting and metering issues. The market's
performance was further complicated by an on-going media campaign in
the New York market, which could have potentially influenced household
participation. Concurrent with the introduction of the LPM, the MRC
closely monitored the existing meter/diary service in New York and
found that this service had degraded.
Subsequently on May 27, the MRC audit committee met to discuss the
audit and the MRC staff recommendation. The audit committee voted to
withhold Accreditation of the NYLPM at that time based on a number of
problems identified in the Ernst and Young audit as well as issues
identified by the MRC staff and the audit committee members. The MRC
sent a letter to Nielsen that communicated detailed areas of concern
and deficiencies with the NYLPM as identified by the audit process and
suggested actions that Nielsen should take to improve the quality of
the service and gain Accreditation. Nielsen commercialized the NYLPM on
June 3, 2004.
On August 26, 2004, the MRC convened a meeting of the audit
committee to assess the results of a re-audit performed by Ernst &
Young to assess performance of certain prior audit issues. Nielsen was
given opportunity to address the Committee during part of the meeting
to share their perspective on the improvement initiatives and the
performance status of the NYLPM. After private deliberations the
committee chose to continue to withhold Accreditation of the NYLPM
service. On August 31, 2004, a letter was sent to Nielsen informing
them of the Committee's decision and outlining the steps necessary to
elevate the Accreditation status of the NYLPM, namely a plan for
updating race information and fault rate stabilization which would be
observed through regular monitoring by the committee.
On October 29, 2004 after review of a credible plan submitted by
Nielsen to address the race classification issues, and observed
improvement in fault rate levels the television committee voted to
grant Conditional Accreditation status to the NYLPM allowing Nielsen to
apply the MRC's Accreditation logo to the New York LPM rating reports.
Since Conditional Accreditation was granted in October 2004, the
television committee has continually monitored the performance of the
NYLPM, including update meetings with Nielsen management and periodic
reviews to reassess the Accreditation status of the NYLPM. As of this
date, the NYLPM service remains Conditionally Accredited.
3. Los Angeles
On July 1, 2004, an MRC audit committee met to review an Ernst &
Young audit of the Los Angeles LPM service (LALPM); at that time
Nielsen had not provided their response to the audit findings, a key
component of the MRC review process. The MRC decided that it was
important to at least conduct a preliminary review of the audit
findings (i.e., absent Nielsen's response) so that it could provide
some illumination of the performance of the LALPM in advance of its
planned commercialization on July 8. To maintain the integrity of the
MRC process, the committee elected not to vote on Accreditation at the
conclusion of this preliminary review until Nielsen submitted their
response for review. The Los Angeles market is a difficult market to
measure due to its ethnic diversity which presents unique challenges in
compiling accurate and reliable data.
Despite the open Accreditation status of the LALPM Service, Nielsen
went ``live'' with the service on July 8, 2004. It was clear through
our experiences in Boston and New York that Nielsen was not yet
implementing LPM services in a manner that is fully compliant with the
MRC's standards.
The audit committee met on July 30, 2004, to conclude the review of
the audit results, including Nielsen's response which was presented in-
person by Nielsen management. After careful consideration the Committee
chose to recommend Conditional Accreditation of the LALPM service
pending Nielsen's submission of an adequate, accepted action-plan to
address: (1) two matters of non-compliance with the MRC's Minimum
Standards for Media Rating Research cited in the audit, and (2) two
performance areas of the Los Angeles LPM Service considered needing
improvement. In addition, an on-going monitoring process was required
by the Television Committee to assure that Nielsen completes the
improvement initiatives specified in its response to the Los Angeles
audit, including the pending action-plan.
On August 19, 2004, upon receipt and acceptance of Nielsen's
action-plan the Conditional Accreditation period began and Nielsen was
authorized to apply the MRC's Accreditation logo to the Los Angeles LPM
rating reports.
Since Conditional Accreditation was granted in August 2004, the
television committee has continually monitored the performance of the
LALPM, including update meetings with Nielsen management and periodic
reviews to reassess the Accreditation status of the LALPM. As of this
date, the LALPM service remains Conditionally Accredited.
4. Chicago
The Chicago LPM (CHLPM) Service was commercialized by Nielsen on
August 5, 2004, prior to an MRC audit. Timing for MRC audits is
controlled by Nielsen and fieldwork was not scheduled to begin until
July 2004, leaving insufficient time for completion of the MRC process
prior to the LPM service going ``live''. The Chicago market contains a
high concentration of minority population groups posing a particular
challenge to measuring accurate and reliable viewing behavior.
An audit committee of the MRC met on September 22, 2004, to review
the findings of the Ernst & Young examination of the CHLPM and based on
the results the audit committee voted to follow the precedence set in
Los Angeles and move to grant Conditional Accreditation to the CHLPM.
The Conditional Accreditation status was scheduled to begin following
receipt and acceptance of an action-plan structured to address specific
audit issues and would also require ongoing monitoring of key
performance metrics for this service. On October 1, 2004, after receipt
of an accepted action-plan, Conditional Accreditation of the CHLPM
began and Nielsen was permitted to apply the MRC Accreditation logo to
the service reports.
Since Conditional Accreditation was granted in October 2004, the
television committee has continually monitored the performance of the
CHLPM service, including update meetings with Nielsen management and
periodic reviews to reassess the Accreditation status of the CHLPM. As
of this date, the CHLPM service remains Conditionally Accredited.
5. San Francisco
The San Francisco LPM (SFLPM) was commercialized on September 30,
2004, prior to an MRC audit and before providing comparative data to
the existing Meter-diary service that would allow the marketplace to
understand the impact of this significant methodological change. MRC
Standards require that measurement services disclose in advance the
estimated impact of a methodological change. The San Francisco market
is racially diverse, containing a high concentration of Asians, and
this diversity presents specific challenges to accurately measure
television viewing behavior.
Fieldwork for the MRC audit began in November 2004, 3 months after
Nielsen commercialized the service. Because of the voluntary nature of
the MRC process, the timing of the audit is controlled by Nielsen.
On March 8, 2005, five months after the SFLPM service was
commercialized by Nielsen, an audit committee of the MRC met to review
the Ernst & Young examination report of the SFLPM and recommended that
the service be granted Conditional Accreditation allowing Nielsen to
apply the MRC Accreditation logo to the SFLPM reports. Nielsen was
informed of specific actions including ongoing monitoring and
performance improvements that would be required for the committee to
consider removal of the conditional aspect of the Accreditation.
On May 6, 2005, the television committee met with Nielsen
management to review the status of the LPM improvement initiatives and
performance metrics and in a private discussion voted to elevate the
status of the SFLPM to full Accreditation.
6. Philadelphia
Nielsen commercialized the Philadelphia LPM on June 30, 2005, prior
to an MRC audit, consequently this service is not Accredited. An MRC
audit is in process for this market with an expected committee review
in October 2005. Because of the voluntary nature of the MRC process,
the timing of the audit is controlled by Nielsen.
7. Washington
Nielsen commercialized the Washington LPM on June 30, 2005, prior
to an MRC audit, consequently this service is not Accredited. An MRC
audit is in process for this market with an expected committee review
in October 2005. Because of the voluntary nature of the MRC process,
the timing of the audit is controlled by Nielsen.
IV. Status of Nielsen Hispanic Measurement Services--National Hispanic
Station Index--Los Angeles and National Hispanic Television
Index
1. Nielsen Hispanic Station Index--Los Angeles (NHSI-LA)
The NHSI-LA Service was audited by MRC during 2000-2001 and,
despite ongoing commercial use of the service, Nielsen chose to not
address the audit issues and terminated the Accreditation process after
two unsuccessful attempts. Nielsen never submitted other NHSI markets
to the Accreditation process.
2. Nielsen Hispanic Television Index (NHTI)
Nielsen's NHTI Service has maintained MRC Accreditation since 2000.
The broadcast television industry members of the MRC, as well as
cable operators and the advertising industry have all voiced their
support for the MRC process in this matter. Central among the
organizations expressing this support are the National Association of
Broadcasters, the Cabletelevision Advertising Bureau, Radio Advertising
Bureau, and the American Association of Advertising Agencies.\8\
---------------------------------------------------------------------------
\8\ Press releases and Organizational statements on the LPM.
---------------------------------------------------------------------------
V. Conclusion
Once again, the MRC would like to thank the Committee for holding
this important hearing on TV Ratings accuracy and the FAIR Ratings Bill
and for allowing the MRC to provide testimony. I continue to believe
that Congress was right in finding that industry self-regulation is
preferable to direct governmental intervention--provided that the
independence and integrity of such an auditing process can be
preserved.
I believe that all of the stakeholders involved in this issue would
agree that the accuracy of Television Ratings is of critical importance
and that the MRC should play a central role in assessing the accuracy
and quality of the new service. *
---------------------------------------------------------------------------
* All the information referred to in the footnotes of this prepared
statement have been retained in Committee files.
Senator Burns. Thank you very much.
Those of you making statements, if we could hold them to
around 5 minutes, that would be great.
He has to be a master. This is his testimony he has handed
in to the Committee, and he got it all in 5 minutes, and I
think that's pretty good. That's probably the standard here.
Thank you very much.
Now we'll hear from Susan Whiting, President and CEO of
Nielsen Media Research.
Thank you for coming today. We appreciate that very much.
STATEMENT OF SUSAN D. WHITING, PRESIDENT/CEO,
NIELSEN MEDIA RESEARCH
Ms. Whiting. Thank you.
Good afternoon. My name is Susan Whiting, and I'm the
President and CEO of Nielsen Media Research.
About a year ago, I first testified before this Committee.
Since then, Nielsen has worked hard to follow your advice and
to make a superior measurement system even better.
Nielsen Media Research is in the truth business, the truth
of what people are actually watching on television and how they
are watching it. Today, for example, the average TV household
has more than----
The Chairman. Could you pull that mike up a little closer?
Ms. Whiting. Is that better?
Today, for example, the average TV household has more than
100 channels. Nielsen has made more advancements and invested
more money in TV audience measurement services than at any
other time in our history. These new investments and
initiatives have resulted in improvements, innovations, and
change. In most cases, changes result in different ratings, but
they also provide a better reflection of viewers' actual
behavior.
As you may have experienced with your voters, it is very
hard to make everyone in your constituency happy. Nielsen is
committed to working with all of our constituents, our
thousands of clients, which include broadcasters, cable
operators, advertising agencies, and advertisers, all with
competing and often conflicting demands on a rating service.
Nielsen must remain independent of conflicting interests
among its diverse client base. For example, on the legislation
we are discussing today two powerful players, Tribune and
Comcast, have taken opposite positions on this bill. Often,
when one party does not agree with our position, they say we
are not listening, that we are arrogant. In fact, we are
listening, not just to one, but to many voices. Given the
progress we have made and the inherent conflicts within the
industry we serve, we do not believe legislation is either
necessary nor advisable. I believe this bill, or any similar
legislation, is both unnecessary and harmful to the long-term
interests of the entire television community.
I think these points were clearly recognize by the FTC in
its March 30, 2005, response to your request that it consider
oversight of TV ratings, where the FTC said that, in Nielsen's
case, ``well-constructed industry self-regulatory efforts can
be more prompt, flexible, and effective than government
regulation.''
The Media Rating Council and Nielsen have established a
strong working relationship that has enabled us to introduce
increasingly more accurate ratings systems. Recently, the MRC
has put forward guidelines for all MRC members and measurement
services, a voluntary code of conduct. We agree, in principle,
with the proposed code of conduct, and we are working with the
MRC on it. For example, we have already agreed that no future
commercial rating service will be launched before it is
audited.
That is why this bill is unnecessary. Here is why it is
harmful. The mandatory accreditation required under this bill
would slow ratings innovation to a crawl. New digital media are
emerging with breathtaking speed. Advertisers and broadcasters
need to know what impact this will have on how audiences watch
TV. If ratings companies have to operate new services without
generating revenue, it is unlikely they will develop or
implement expensive new audience measurement innovations. This
is also a significant barrier to entry into this market by any
competitor.
I think, Senator Burns, that you said it best when you
remarked that the Internet was able to blossom because Congress
didn't know how to regulate it. According to the same
principle, Congress should not regulate television ratings
business. We do not believe it is good policy to transform the
MRC into a vehicle that limits competition from new program
sources, especially from smaller, independent, and minority-
owned stations and networks looking to compete against media
giants. This is why a number of minority-oriented channels have
issued voiced opposition to legislation, including both TV One
and BET.
I should also note that many other clients representing the
advertisers, including the American Association of Advertising
Agencies, the Association of National Advertisers, and the AAF,
whose money this is all about, community groups, and public-
interest organizations have voiced their opposition to
mandatory accreditation and the legislation.
I do want to mention one initiative that came about from
our work with both our task force and the MRC, the creation of
a Special Council for Research Excellence. We created this
council in order to involve the industry in setting the
direction of basic research and development. Nielsen has
committed an additional $2.5 million annually for special
research, as recommended by the council. It is composed of 40
clients representing the entire television industry and chaired
by Mark Kaline, Global Media Manager for Ford Motor Company,
one of the largest buyers of television advertising time in the
United States.
In conclusion, instead of legislation, we need to support
the MRC by agreeing to a new voluntary audit and accreditation
standard that will enable measurement services to respond more
quickly to dynamic changes. Self-regulation dictated through
government mandate has many of the same disadvantages as direct
government oversight, without the protection of formal
rulemaking processes or public accountability.
On behalf of thousands of Nielsen employees in the United
States, across 49 states, I would like to reiterate Nielsen's
commitment to producing the most accurate TV ratings possible,
that we continue to serve a broad and sometimes contentious
client base, and that we are committed to working with the MRC,
our clients, and community leaders to assure transparency and
accuracy in the ratings.
Thank you.
[The prepared statement of Ms. Whiting follows:]
Prepared Statement of Susan D. Whiting, President/CEO,
Nielsen Media Research
Good morning. My name is Susan Whiting and I am President and Chief
Executive Officer of Nielsen Media Research.
It was about a year ago that I first testified before this
committee. Since then my team and I at Nielsen have worked very hard to
follow your advice and to make a superior measurement system even
better. I have met with many Members of the Committee to hear their
concerns and share Nielsen's story, including our vision for the future
of audience measurement technology, and our commitment to working with
all of our clients.
Nielsen Media Research is in the truth business: the truth of what
people are actually watching on television, and how they are watching
it. We all watch television differently today than we did 5 years ago.
Today, for example, the average TV household has more than 100 channels
from which to choose. Consumers are also choosing digital technologies
such as TiVo, Video on Demand and video gaming.
With this diversity of entertainment choices, Nielsen is committed
to providing the entire marketplace with the most accurate TV ratings
possible.
To anyone who has been involved in this industry for the past 5
years, it is clearly apparent that Nielsen has made more advancements
and invested more money in TV audience measurement services that at any
other time in our history.
During the last year, we made significant investments in all
aspects of TV audience measurement--sampling, data collection, data
processing, and data delivery--which we believe will further improve
the accuracy of our ratings. We continue to invest in the leading edge
of measurement technologies and look forward to new systems that will
measure a broad spectrum of digital technologies.
These new investments and initiatives produce change, and different
clients react differently to these changes.
As you may have experienced with your voters, it is very hard to
make everyone in your constituency happy. Nielsen is committed to
working with all of our constituents, our clients, which include
broadcasters, cable operators, advertising agencies and advertisers--
all with competing and often conflicting demands on a ratings service.
A truly independent ratings service, offering the highest quality and
most accurate ratings, is vital for the marketplace to operate
effectively.
Nielsen must remain independent of these conflicting interests. For
example, on the legislation we are discussing today, two powerful
players, the National Association of Broadcasters and Comcast have
taken opposite positions on this legislation.
Unwarranted and Unwise Legislation
Given the progress we have made and the inherent conflicts within
the industry we serve we do not believe legislation is either necessary
or advisable, in fact we feel it is unwarranted and harmful.
I think these points were ably recognized by the Federal Trade
Commission in its March 30, 2005 response to your request that it
consider oversight of TV ratings. As you recall from its response, the
FTC said, that, in Nielsen's case, ``well constructed industry self-
regulatory efforts can be more prompt, flexible and effective than
government regulation.''
I believe S. 1372 is both unnecessary and harmful to the long-term
interests of the entire television community.
First, it is unnecessary. The Media Rating Council and Nielsen
have, over the past 40 years, established a strong working relationship
that has enabled us to introduce increasingly more accurate ratings
systems. Over the past few weeks, for example, the MRC has put forward
guidelines for all MRC members and measurement services--called A
Voluntary Code of Conduct--that would both clarify and strengthen the
MRC's relationship with all measurement services as well as with its
own membership. The MRC recently provided to its members and all
measurement services--television, radio, newspaper and Internet--a
proposed voluntary code of conduct to deal with the rollout of new
measurement technologies in the marketplace. Among the first things we
have already agreed to, for example, is that no future commercial
ratings service will be launched without the transparency of a full
audit having taken place.
Other elements of the Code are under discussion at this time, and
we are confident that, after appropriate give and take, that the
industry will reach agreement with all ratings services on the Code and
we can submit it to the Justice Department and the FTC for a business
review. We believe in principle that the proposed Voluntary Code of
Conduct represents a valid approach to enhancing the MRC process, and
that if it is approved by MRC members and all measurement services, we
intend to adopt it.
In other words, since the free-market, private enterprise system is
working, we do not need a legislative solution to a problem that does
not exist.
That is why S. 1372 is unnecessary. Here is why it is harmful.
The mandatory accreditation required under S. 1372 would slow
ratings innovation to a crawl. Vital new systems for measuring all
forms of digital television could remain idle while MRC members
debated. In an environment that is becoming increasingly governed by
political and economic self-interest, that process could literally take
years. Technology, however, won't wait. Nor will clients. The
transition from analogue to digital television technologies would be
frustrated at the lack of timely measurement.
As you know when you watch television, and from your experience on
the Committee, new digital media are emerging with breathtaking speed,
and audiences are increasingly willing to use devices like DVRs and
Video on Demand to take control of their viewing experiences. The sale
of DVRs is expected to nearly double within 2 years, and advertisers
and broadcasters need to know, as soon as possible, what impact this
will have on how audiences watch TV.
If ratings companies are required to operate new services without
generating revenue for a significant period of time, it is unlikely
they will develop or implement expensive new audience measurement
innovations. Such a prospect also is a significant barrier to entry
into this market by any competitor. Indeed, if technology and
telecommunication firms faced these restrictions, computers, cell
phones and the Internet would still be on the drawing boards.
We do not believe it is good public policy to transform the MRC
into a vehicle that limits competition from new program sources,
especially from smaller, independent, and minority-owned stations and
networks looking to compete against media giants. More precise ratings
technology enhances the voice of minorities by making possible niche
programming on new cable networks and television stations aimed at the
African American, Hispanic, Asian, and Arab-American communities. These
advancements could grind to a halt with mandatory ratings
accreditation. This is why a number of minority competitors had issued
statements in opposition to legislation, including both TV One and BET.
Working With the Task Force
As you know, Nielsen continues to work closely with the Independent
Task Force on Television Measurement. This Task Force was created last
year at the suggestion of Congressman Charles Rangel, for the very
purpose of offsetting the need for Congressional involvement. The Task
Force worked for more than 8 months--and continues to work--and
released a major report to Nielsen, which we shared with the industry,
that included recommendations in the areas of sampling, field
operations, fault rates, diversity and communications.
With your permission, Senator, I would like to submit for the
record a copy of the Task Force's report, Nielsen's response, and the
follow-up report released just last month. Considering the importance
of this Task Force Report and the enormous commitment in time and
effort from people representing a diverse spectrum of Americans,
especially former Representative Mrs. Cardiss Collins who chaired the
Task Force. I should also note that the Task Force has issued a
statement in opposition to S. 1372, and I would also like to submit
those comments for the record.
The Task Force has, indeed, been the focus for many of the very
constructive initiatives that we have been sharing for some time now
with our clients, others in the industry and with Congress. Yet the
full breadth of audience measurement--including sample design, sample
recruiting and maintenance, data collection systems, metering, data
processing and data reporting (all involving hundreds of millions of
dollars in spending by Nielsen)--have improved over the years because
of the painstaking work we have done with our clients through the Media
Rating Council's accreditation process.
I do want to mention just one initiative, and this came about from
our work through the Task Force as well as with the MRC, and that is
the creation of a special Council for Research Excellence, created
earlier this year. We created this Council in order to involve the
industry in setting the direction of basis R&D in the area of
methodological research.
In addition to the tens of millions of dollars we spend each year
on methodological and statistical research, Nielsen has committed an
additional $2.5 million for special research as recommended by the
Council. The Council is composed of 40 clients, including the MRC,
representing the entire television industry. The Council is chaired by
Mark Kaline, global media manager for Ford Motor Company, one of the
largest buyers of television advertising time in the United States.
Responding to a Changing Market
Why would anyone agree to create a Council or serve on a Council
when the MRC, under the bill, would be the final authority over
everything pertaining to the ratings services?
Instead of a new bill we, as an industry, need to support the MRC
by agreeing to a new, voluntary audit and accreditation standards that
will enable measurement services to respond more quickly to dynamic
changes in the television landscape so that digital technologies
including Digital Video Recorders, DVD Recorders, Video on Demand, and
Time Shifting can be included in the measurement of audiences.
Congress has mandated the shift in broadcast television from
analogue to digital. Over the past 12 years, we have supported that
mandate by completely revamping our metering and reporting technology
with investments of over a hundred million dollars. I can only assume
that the underlying assumption behind this mandate is that there would
be no government-imposed barrier to measuring audiences to digital
television. But S. 1372 imposes formidable barriers by mandating that
no ratings service could measure anything without the approval of the
MRC.
Since the last time we were here, we have significantly enhanced
our ability to more accurately measure all television audiences. For
example:
On March 3, 2005, after more than 12 years of R&D, and
hundreds of millions of dollars in spending, Nielsen introduced
a new digital metering system, called the Active/Passive Meter
System, or A/P Meter for short. The A/P Meter is fundamentally
a set meter, but it is also a platform for in-home measurement
of many new digital television devices. In July 2005, Nielsen
began rolling out the new A/P Meter system into the national
and local People Meter samples. Without this system, we would
not be able to measure digital signals and there would be no
viable business model for digital television.
In May we began to implement a program of personal coaching,
performance-based incentives and reminder mailings designed to
reduce overall and differential faulting in Local People Meter
markets. This represents another ongoing multimillion
investment.
In June we delivered a plan for enabling measurement of
Video on Demand programming in our syndicated ratings panels.
DVR measurement has been successfully implemented in our
set-meter and diary markets. We remain on-schedule for
installation of DVR households in the national and local People
Meter samples beginning in January 2006. So far, we have
installed more than 200 DVR households across 47 local markets.
In June 2005, Nielsen completed the translation of all of
its recruitment materials for sample households into Spanish,
developed key recruitment materials in Mandarin, Japanese,
Korean, Vietnamese, and Tagalog. We are also tailoring our
``introductory video'' that is provided to new sample
households for Asian audiences. We also recently added several
training procedures on cultural sensitivity to our 10-week
Field Training program.
Conclusion
To conclude my remarks today, self-regulation dictated through
government mandate has many of the same disadvantages as direct
government oversight, without the protection of formal rulemaking
processes or public accountability.
What is more, it lacks the agility, flexibility and resourcefulness
that come from free market forces. Those qualities have served the
Media Rating Council and its members well for more than four decades,
and they are worth preserving.
I would like to reiterate Nielsen's commitment to producing the
most accurate TV ratings possible; that we serve a broad and sometimes
contentious client base; and that we are committed to working with the
MRC, our clients, and community leaders to assure transparency and
accuracy in the ratings.
Finally we believe in the voluntary MRC accreditation process, and
legislatively mandating this process would be harmful not just to
Nielsen but to everyone.
Thank you.
Senator Burns. Thank you.
Now we'll hear from Ceril Shagrin.
STATEMENT OF CERIL SHAGRIN,
EXECUTIVE VICE PRESIDENT FOR RESEARCH, UNIVISION
Ms. Shagrin. Good morning, Mr. Chairman and members of the
Committee. Actually, I'm as bad as you are, it is afternoon,
isn't it?
[Laughter.]
Senator Burns. It is afternoon.
Ms. Shagrin. My name is Ceril Shagrin. I spent 27 years at
Nielsen Media Research, during which time I interacted with the
Media Ratings Council----
Senator Burns. Pull your microphone--you've got a nice
little soft voice, and we'd like to hear it.
Ms. Shagrin. OK. Is that better?
Senator Burns. You bet.
Ms. Shagrin. I was Nielsen's first quality-assurance
director and the primary contact for the review of the MRC
audit scope and the audit report. While at Nielsen Media, I was
the primary participant in the development and rollout of the
National People Meter Service. I was responsible for the
development and management of Nielsen Hispanic Services, and
involved in the development, testing, and rollout of all new
services.
For the past 6 years, I've been employed by Univision
Communications, where I oversee research needs for all
Univision divisions. Currently, I am the Chairman of the MRC
Television Committee and I am proud to be this year's recipient
of the Malcolm Beville Award for my commitment to the highest
standards in broadcast measurement research.
For the past 33 years, I've had a close relationship with
the MRC, both as a representative of the provider of television
ratings and as a user of those ratings for programming
decisions and for setting value and audience delivery. I have
worked with 6 different MRC Executive Directors. For the past
33 years, I've been driven by the need for quality research and
reliable audience estimates. I believe the MRC has been a major
contributor to achieving that goal.
The television landscape has changed dramatically from a
three-network environment to one of multiple broadcast and
cable choices. At the same time, the United States population
has grown and changed to a multicultural population. In order
to meet the quality standards of television audience
measurement, the samples used to develop audience estimates
must accurately represent the current and changing populations
of Whites, Blacks, Hispanics, and Asians, not just in total
number, but demographically within each of these populations.
While the data-collection instrument must be designed to
accurately collect viewing, no data-collection device can
eliminate the bias of reporting sample that does not accurately
represent the universe being measured. Television rating
services must make decisions on the data-collection tool and
the methodology which best captures viewing within the cost
parameters the individual markets can support.
Installing and maintaining a representative sample is
difficult. It takes properly trained personnel, adherence to
procedures, and continuous testing to search for improvements.
It requires a commitment to standards. The minimum standards of
1975 no longer meet the challenges of audience measurement in
2005. Quality measurement requires constant third-party
monitoring to ensure proper procedures are identified and
followed. The MRC provides that function through continuing
audit and review.
The MRC Television Committee is made up of users of the
audience estimates--broadcast networks, cable networks,
stations, agencies, and advertisers. For the past 6 years, I
think I've attended every one of the MRC meetings related to
television audience measurement. I strongly believe the MRC
audit process has contributed to the continuous improvement of
the quality of TV audience measurement.
Attendees of these meetings invest a significant amount of
time reading and analyzing audit reports. Meetings are long and
detailed. They are well attended. No one is allowed to vote
without the investment of time in the understanding of the
audit issues.
Nielsen received a copy of the audit report prior to
distribution to the Committee, and their comments are included
in the report sent to the Committee and in the discussion and
review of the audit.
New technologies must be audited before being put into
production. New editing rules, processing rules, sample design
and maintenance procedures should be evaluated, and their
impact on audience estimates dimensioned, prior to
implementation to ensure continuation of quality standards.
While my written testimony states that prior to 2004 I can
remember no instance when Nielsen implemented any material
changes in methodology, processing rules, or data-collection
device without prior review and acceptance by the MRC, there
was one exception. That was the Boston LPM market.
Nielsen has said, in recent public statements, that
mandatory accreditation would result in termination of the
Nielsen Hispanic Station Index, the local Hispanic measurement
service. While different sampling procedures are used in some
NHSI markets, there's reason to believe they could not, either
as currently designed or with modifications, meet MRC standards
or that increasing the Hispanic samples in the NSI service
could not provide reliable Hispanic audience estimates. I have
confidence that Nielsen can do that.
For approximately 40 years, Nielsen's local and national
television measurement services have been audited and
accredited. Nielsen has met the MRC quality standards and
continuously strive for improvement. It has made them a better
company, and it has allowed the television industry to grow.
Univision has taken a neutral position on the FAIR Ratings
bill. We take a very strong position on the need for quality
samples and procedures for eliminating bias. We take a positive
position on the need for MRC audits.
I'd like to thank the Committee for the opportunity to
appear here today, and I look forward to answering any
questions.
[The prepared statement of Ms. Shagrin follows:]
Prepared Statement of Ceril Shagrin,
Executive Vice President for Research, Univision
Good morning, Mr. Chairman and members of the Committee.
My name is Ceril Shagrin. I spent 27 years at Nielsen Media
Research during which time I interacted with the Media Rating Council
(MRC) and the Ernst & Young auditors. I was Nielsen's first Quality
Assurance Director and the primary contact for the review of the audit
scope and the audit report. While at Nielsen Media I was a primary
participant in the development and roll out of the National People
Meter Service, responsible for the development and management of
Nielsen Hispanic Services and involved in the development, testing and
rollout of all new services.
For the past 6 years I have been employed by Univision
Communications where I oversee the research needs for all the Univision
divisions.
Currently I am the Chairman of the MRC Television Committee. I am
proud to be this year's recipient of the Malcolm Beville Award for my
commitment to the highest standards in broadcast measurement research.
For the past 33 years I have had a close relationship with the MRC
both as a representative of the provider of television ratings and as a
user of those ratings for programming decisions and for setting value
on audience delivery. I have worked with 6 different MRC Executive
Directors. For the past 33 years I have been driven by the need for
quality research and reliable audience estimates. I believe the MRC has
been a major contributor to achieving that goal.
The television landscape has changed dramatically from a three
network environment to one of multiple broadcast and cable choices. At
the same time the United States population has grown and changed to a
multicultural population. In order to meet the quality standards of
television audience measurement, the samples used to develop audience
estimates must accurately represent the current and changing
populations of Whites, Blacks, Hispanics and Asians not just in total
number but demographically within each of those populations.
While the data collection instrument must be designed to accurately
collect viewing, no data collection device can eliminate the bias of a
reporting sample that does not accurately represent the universe being
measured. Television ratings services must make decisions on the data
collection tool and methodology which best captures viewing within the
cost parameters the individual market can support.
Installing and maintaining a representative sample is difficult. It
takes properly trained personnel, adherence to procedures and
continuous testing to search for improvements. It requires a commitment
to standards. The minimum standards of l975 no longer meet the
challenges of audience measurement in 2005. Quality measurement
requires constant third party monitoring to ensure proper procedures
are identified and followed. The MRC provides that function through
continuing audit and review.
The MRC Television Committee is made up of users of the audience
estimates: broadcast networks, cable networks, stations, agencies and
advertisers. For the past 6 years I have attended all MRC meetings
related to television audience measurement. I strongly believe the MRC
audit process has contributed to the continuous improvement of the
quality of TV audience measurement. Attendees of these meetings invest
a significant amount of time reading and analyzing audit reports.
Meetings are long and detailed. They are well attended. No one is
allowed to vote without the investment of time in the understanding of
the audit issues. Nielsen receives a copy of the audit report prior to
distribution to the committee and their comments are included in the
report sent to the committee and in the discussion and review of the
audit.
New technologies must be audited before being put into production.
New editing rules, processing rules, sample design and maintenance
procedures should be evaluated and their impact on audience estimates
dimensioned prior to implementation to ensure continuation of quality
standards. Prior to 2004, I can remember no instance when Nielsen
implemented any material changes in methodology, processing rules or
data collection device without prior review and acceptance by the MRC.
Nielsen has said in recent public statements that mandatory
accreditation would result in termination of the Nielsen Hispanic
Station Index (NHSI), the local Hispanic measurement service. While
different sampling procedures are used in some NHSI markets, there is
no reason to believe they could not either as currently designed or
with modifications meet MRC standards or that increasing the Hispanic
samples in the NSI service could not provide reliable Hispanic audience
estimates.
For approximately 40 years Nielsen's local and national television
measurement services have been audited and accredited. Nielsen has met
the MRC quality standards and continuously strived for improvement. It
has made them a better company and allowed the television industry to
grow.
Univision has taken a neutral position on the FAIR Ratings Bill. We
take a very strong position on the need for quality samples and
procedures for eliminating bias. We take a positive position on the
need for MRC audits.
I would like to thank the Committee for the opportunity to appear
here today, and I look forward to answering any questions.
Senator Burns. Thank you very much.
Now we'll hear from Ms. Kathy Crawford, President, Local
Broadcast, MindShare Worldwide.
And thank you for coming, Ms. Crawford.
STATEMENT OF KATHY CRAWFORD, PRESIDENT,
LOCAL BROADCAST, MindShare
Ms. Crawford. Good afternoon, Senator Burns.
My name is Kathy Crawford, and I am President of Local
Broadcast at MindShare. In this position, I help our clients
decide what television station to advertise on in 210 local TV
markets.
In the last few years, we have spent billions of dollars in
local TV. Advertising makes local television possible. Almost
all local television revenues come from major companies
employing millions of people trying to reach customers. But the
bill we are discussing today was not written with them in mind.
In fact, I am very concerned that this bill will make it harder
for clients to buy advertising with any confidence that they
are spending their money wisely. As I see it, the bill has
negative implications, not only for the Local People Meter
markets, but also for all local markets. I believe it will make
my clients far less willing to advertise with local television
stations, because we won't have the information we need to
negotiate fair rates with the stations.
Over the past 40 years, the MRC has been a crucial partner
in improving the quality of television ratings, but I am
concerned that, given the MRC member--giving the MRC members
the power to block new technologies and new services will turn
back the clock.
Clearly, the business of television advertising is
changing. When the MRC was created, broadcast was synonymous
with television. But today there are scores of cable networks,
like Oxygen, Spike, Black Entertainment Television, Galavision,
as well as local cable channels appealing to many more
ethnically-diverse portions of the community. We clearly need
new ratings technology to keep up with these changes, but, as
we've seen with LPMs, there's always some resistance to change,
as different methodologies yield different rankings and change
pricing.
The same thing happened when the national broadcast
networks and national advertisers went through the same process
in the 1980s. Back then, no one tried to roll back this process
through legislation.
With the Local People Meter, we've heard a great deal about
fault rates. I believe this has been a red herring, an excuse
to delay accreditation. Indeed, many of the MRC members who now
complain about fault rates in LPMs regularly voted to accredit
meter diary systems in which, quote/unquote, ``fault rates were
even higher.'' If MRC members are so concerned about fault
rates, how did those meter diary markets get accredited? Did
the fact that broadcasters' ratings are higher under the diary
than under LPMs have anything to do with this?
We need an accreditation process that is fast, fair, and
efficient. We need the MRC to serve its traditional role as a
forum for the industry to improve the overall performance of
the measurement services.
To that end, I believe the MRC's voting procedures need to
be seriously overhauled. Currently, a handful of the broadcast
companies can control the MRC, because they have four to five
votes each, through their ownership of cable networks, local
television stations, syndication networks, and national
networks. I don't think that is right. But if you do try to
change the MRC from a voluntary industry group into a
government-mandated regulatory body, I don't think anyone with
a direct stake in the outcome of the vote should be in the MRC
at all, given their incentive to vote their own self-interest.
And if the MRC has different membership structure, who would
choose the members, and who would they report to?
Finally, what would an overhaul like this cost, and who
would pay for it? As we've seen in the past, the television
industry has not wanted to pay for more than one service.
That's too bad, because I, for one, would like to see more
competition in the ratings business. I think it would be good
for all of us, including Nielsen, because it would drive
innovation faster and further. But legislation won't achieve
that goal. To the contrary, this will effectively ensure that
Nielsen never again faces any competition. No company would
invest the vast amount of time, resources, and dollars needed
to start a measurement service if they knew it could remain
idle for a year or more, generating no revenue, as MRC members
debated its fate. What potential Nielsen competitor could
afford that?
This legislation stems from disagreement among private
companies on the accuracy of ratings in certain markets and
among certain audience segments. What we have here is a
complicated technological research dispute over some aspects of
Nielsen's methodology that has been blown out of proportion
into a would-be public-policy issue.
In this regard, are current television ratings adequate?
No. Is there room for improvement? Always. But innovation
cannot be mandated by the government. In fact, we've seen many
times in the past when the government tries to interfere in the
private sector, no matter what its good intentions, it usually
makes the situation worse.
The MRC has played an important role, over the past 40
years, in making television the best-measured medium. Like any
other institution, it can operate better. But I believe the
members of the MRC, as private businesses operating in the
free-enterprise system, can certainly work out for themselves
how to make the organization more effective.
Thank you.
[The prepared statement of Ms. Crawford follows:]
Prepared Statement of Kathy Crawford, President,
Local Broadcast, MindShare
Good afternoon. My name is Kathy Crawford and I am President of
Local Broadcast at MindShare.
A Global Leader in Advertising
MindShare advises some of the world's largest advertisers on what
advertising programs to pursue. We manage all aspects of media
investment, from strategy--including targeting and spending--through
negotiation and placement of advertising.
We advise our clients on the best mix of media to enable them to
reach their target audiences, whether via television, print, digital or
on-line, out-of-home, radio, locally and nationally. Moreover, we
negotiate rates and schedules for our clients, such as the most
effective programming on specific stations or networks; the right
magazines; or the most appropriate websites; so that they reach their
targets at the best price.
In addition, we need to know that our clients are getting what they
pay for. It's not enough to simply place an ad. We also have to be
assured that it has run and that the right audience is being reached.
As President of Local Broadcast at MindShare, I help decide what
television stations our clients advertise on in the 210 local markets.
In the last few years we have spent billions in local broadcasting.
Nothing is more important to them than making sure their money is well-
spent on reaching the right targets.
Advertising makes local television possible. Almost all local
television revenues come from advertisers--the biggest companies in the
world--trying to reach viewers, but the bill we are discussing today
was not written with them in mind.
In fact, I am very concerned that this bill will make it harder for
clients to buy advertising with any confidence that they are spending
their money wisely. As I see it, the bill has negative implications not
only for the Local People Meter markets, but also for all local
markets. I believe it will make my clients far less willing to
advertise with local television stations because we won't have the
information we need to negotiate fair rates with the stations. Let me
explain why.
Nielsen and Television Ratings
Today, Nielsen is the only research service in the U.S. that
measures television audiences, both nationally and locally. There used
to be two systems but the industry decided it only wanted to pay for
one. With just one ratings service now, all of us in the industry are
well aware of Nielsen's shortcomings and its strengths. That is why we
work closely with Nielsen, and with our suppliers, to ensure that the
methodology is sound. We also must be certain that its systems deliver
the most accurate information as quickly as possible, so we can
appropriately recommend to our clients how best to spend their ad
dollars and can negotiate with the stations based upon those ratings.
Because our industry is pro-active in seeking and demanding
improvements in ratings systems, we are constantly working on how to
better understand the changing media landscape.
The Local People Meter is a product of this industry-wide effort.
Nielsen did not decide to offer LPMs in a vacuum. This was a collective
decision reached by the entire television industry because advertisers
were no longer willing to spend billions of dollars on advertising
based on ratings from outmoded systems.
When measurement systems change, there always is some resistance as
different methodologies yield different rankings and possibly change
pricing structures. The national broadcast networks and national
advertisers went through the same process in the 1980s when Nielsen
introduced its National People Meter in response to industry pressures
and the threat of competition from AGB. The difference then was that no
one tried to roll back this progress through legislation.
A Changing Television Landscape
Clearly, the business of television advertising has changed
considerably since the Media Rating Council was established in 1964.
Back then, the only networks on TV were known by their initials--
ABC, CBS and NBC--and local channels were all identified by call
letters. Cable was still a young medium, heavily regulated to keep it
out of the top 100 markets. And the first communication satellite--
Telstar--had just been launched 2 years before.
Moreover, no one in the industry at that time could even have
imagined the concept of time-shifting, when people can watch a
particular show when they want to, not when the networks program it for
them.
In short, when the MRC was created, broadcast was synonymous with
television.
Obviously, that's no longer the case. Today, there are scores of
cable networks with more expressive names like Oxygen, Spike, Black
Entertainment Television, Galavision, as well as local cable channels
appealing to many more ethnically diverse portions of the community.
Each caters to very different audiences, and myriad advertisers and
their agencies work very hard to reach them.
The skies are filled with satellites beaming programming around the
world. And everyone in the industry is familiar with Video on Demand
and Digital Video Recorders such as TiVo.
With the proliferation of cable and satellite, television viewers
have more choices than ever before, and this certainly has affected
ratings. Viewing has not declined but it is now spread among more
program sources, resulting in lower ratings for some broadcaster
networks, and many more cable networks getting some ratings were none
had previously existed. Advances in technology also have given people
many more options beyond television, including the Internet, mobile
phones and video gaming.
The Role of the Media Rating Council
As the television universe expands, my clients--the major
advertisers--demand the most accurate and reliable ratings system
possible.
American television today is the best measured advertising medium
in the world. It is the only U.S. medium that uses electronic meters,
which are far more accurate than the diaries that measure radio or the
surveys that measure newspapers and magazines.
But I am concerned, Senator Burns, that your bill, by giving the
MRC membership the power to block new technologies and new services,
will turn back the clock on the progress we have made in developing an
effective television ratings system.
The MRC staff is composed of business professionals whose job
include making sure that the ratings that are used in the industry meet
the highest standards for accuracy. They do this by insisting on
transparency to all the participants in the market, reviewing
independent audits and working with measurement services to adopt
better technologies and more rigorous procedures.
However, there is no question that when MRC members vote on whether
to approve new technologies, they always, to some degree, consider the
impact on their own bottom line. Ironically, this resistance to change
makes local broadcasting a less attractive option for my clients.
As it's currently constituted, the MRC is still dominated by
broadcasters. In some cases, large broadcasting companies--through
their co-ownership of broadcast networks, cable networks, broadcast
stations, studios and syndicators--have as many as four or five votes
each, while cable operators, advertising agencies and advertisers each
have just one.
This means that just five or six major media companies could work
as a bloc to delay ratings systems that would hurt their bottom line.
This is not speculation. If your bill were in effect today, Nielsen
would have been prevented from introducing Local People Meters in the
country's largest media markets even though the information they are
producing there is much more accurate than the meter/diary systems they
replaced.
There is no question that LPMs are a superior ratings service. They
have larger samples, including more African American and Hispanic
households; they better represent the communities they measure; and
they provide the immediate demographic data that my clients need to
make advertising decisions. Yet in New York, for example, the MRC still
has not accredited the LPM service, even though it was ready to be
introduced almost 15 months ago. How can this be?
Fault Rates Are a Red Herring
Over the past year, we've heard a great deal about LPM fault rates.
I believe this has been a red herring--an excuse to delay
accreditation.
First, fault rates are only one measure of sample quality--and not
the most important one. The composition of the sample and the
acceptance rate are all equally if not more important. However, all
pale in comparison to the sample size and the superior data collection
technology that LPMs use.
But the real reason I believe fault rates are a red herring is that
many of the MRC members who now complain about fault rates in LPMs
regularly voted to accredit meter diary systems in which ``fault''
rates were even higher--not only for the overall market, but for people
of color too.
Indeed, on a comparative set-to-set basis, fault rates have
consistently been higher in metered/diary markets than Local People
Meter markets. What's more, while there are no exact comparisons for
diaries, as many as 14 percent of diaries returned to Nielsen cannot be
used--the equivalent of faulting in diaries. And then of course there
are those diaries that are never returned at all.
If MRC members are so concerned about fault rates, how did those
meter/diary markets get accredited? Did the fact that broadcasters'
ratings are higher under the diary than under LPMs have anything to do
with this?
Like the broadcasters, I too am concerned about fault rates in
Chicago and New York, but I am also concerned about fault rates in
Glendive, Terra Haute, and Tallahassee. If the MRC sets impossibly high
standards for LPM markets, how can they justifying setting lower
standards for non-LPM markets?
An Unworkable Bill
In short, I believe the bill is unworkable.
As I noted earlier, the MRC was created to deal with a television
industry that, essentially, no longer exists. TV audiences today are
becoming more diverse, subdividing into ever-smaller segments. In
response, newer networks and channels are emerging to better serve
these niche markets. Instead of 4-6 channels, the average U.S.
household now receives in excess of 130 channels.
At the same time, the viewer is becoming the boss. Growing numbers
of people are using digital technologies to choose what, when and how
they watch.
In other words, we need an accreditation process that is fast, fair
and efficient. None of which is attainable under the pending
legislation, because the MRC was not created to be an objective
standard-setting organization.
The MRC should however serve its traditional role as a forum for
the industry to improve the overall performance of the measurement
services. To that end I believe the MRC's voting procedures need to be
seriously overhauled. Senator, I know you support the concept of one-
person/one vote and I assume you do not believe it is fair for one
company to have five votes and another to have only one.
If you try to change the MRC from an industry group working to
improve the services offered, and try to make it a more regulatory
body, we would have to consider whether anyone with a stake in the
outcome of the votes should be in the MRC at all, given their incentive
to vote their own self-interest. And if the MRC has a different
membership structure, who would choose the members and who would they
report to?
Finally, what would an overhaul like this cost and who would pay
for it?
In a free market, broadcasters, cable operators, advertisers and
their agencies, among others, should be willing to foot the bill if
they believe such changes are feasible and beneficial. Just as they
should be willing to pay for more than one rating service if they think
that's a practical solution.
Still, as we've seen in the past, the broadcast industry has been
unwilling to help finance more than one service. That's too bad,
because I, for one, would like to see more competition in the ratings
business. I think it would be good for all of us, including Nielsen,
because it would drive innovation faster and further.
But legislation won't achieve that goal. To the contrary, this will
effectively ensure that Nielsen never again ever faces any competition.
No company would invest the vast amount of time, resources and
dollars needed to start a measurement service if they knew it could
remain idle for a year or more--generating no revenue--as MRC members
debated its fate. What potential Nielsen competitor could afford that?
Trying to Legislate a Business-to-Business Issue
I believe this legislation stems from a disagreement on the
accuracy of ratings reporting in certain markets and among certain
audience segments. It proposes external, mandatory regulation of a
system that is in many ways self-regulating, where participants
themselves--buyers, sellers, stations--engage in ongoing dialogue about
the system.
Underlying our dialogue is the agency's responsibility to its
clients, and ultimately the clients' responsibility to its customers
(the same customers the television station is trying to reach with its
programming). Also important is the broadcasters' responsibility toward
dual constituencies: the viewers as well as the advertisers.
Consequently, it is in everyone's best interest that the public is
served.
What we have then is an industry research dispute over whether
Nielsen's methodology accurately reports the size of the audience and
how these procedures might be improved to more accurately report viewer
behavior.
It is not a question of whether a broadcaster is serving the public
interest. Rather, it is a matter of whether the yardstick by which
audiences are measured does so accurately. Not acknowledging the
relationship among advertisers, audiences, and broadcasters would
dampen the dialogue that seeks to improve the system.
The Free Market as the Solution
The founders of the Media Rating Council couldn't have predicted
what television would be like in the 21st century. But they did have
the foresight to explicitly reject government oversight because they
recognized the negative impact on innovation and competition.
That, at least, has not changed in over 40 years.
Attempts to roll back the clock by eliminating systems like Local
People Meters, or by slowing down the development of new technologies
that measure time-shifting and on-demand services, will not stop
change.
Is this regard, are current television ratings fully adequate? No.
Is there room for improvement? Always. But innovation cannot be
mandated by the government. In fact we've seen many times in the past
that when the government tries to interfere in the private sector--no
matter what its good intentions--it usually makes the situation worse.
The MRC has played an important role over the past 40 years in
making television the best measured medium. Like any other institution,
it can operate better. But I believe the members of the MRC--as private
businesses operating in the free enterprise system--can certainly work
out for themselves how to make the organization more effective.
Thank you.
Senator Burns. Thank you.
And we'll hear now from Gale Metzger, former CEO, SMART
Media.
And thank you for coming today. Oops, did I miss somebody?
I did.
Let's call on Mr. Pat Mullen, who is President, Tribune
Broadcasting.
Pull your microphone up, if you would, Mr. Mullen, please.
Mr. Mullen. Certainly.
STATEMENT OF PATRICK J. MULLEN, PRESIDENT,
TRIBUNE BROADCASTING COMPANY
Mr. Mullen. Mr. Chairman, thank you for the opportunity to
appear before you to testify in support of S. 1372, which will
go a long way toward assuring that there is a strong,
independent body to ensure the reliability of television-
audience measurement.
My name is Pat Mullen. Our company, Tribune Broadcasting,
operates 26 major-market television stations located in 15
states from coast to coast, including stations in 8 of the 10
largest markets. Our station's success has depended, in every
case, on accurate count of our audiences. Our stations get a
report card every morning from Nielsen. These ratings determine
which programs remain on the air. We're eager to compete with
our fellow broadcasters and with cable and satellite, but to do
this we must have an honest report card.
Mr. Chairman, I regret to say that the measurement system
that we have today in the largest television markets is not
worthy of the public trust. Congress has repeatedly
acknowledged the importance of a free and robust broadcast
service, which is particularly important in times of crisis. It
deserves a guaranteed minimum standard of accuracy because of
the importance of television news, public affairs, sports, and
entertainment programming to this country's culture and to our
democracy.
We are not here today in an attempt to secure a competitive
advantage over our competitors. Our company welcomes
competition. The problem, Mr. Chairman, is that the keys to our
success, our ratings, are held by a monopoly. When Nielsen had
a competitor, its service and its response to client concerns
were substantially better than they are today. In absence of
competition, we are left to plead for fair treatment and
reliable results. Time and again, Nielsen has turned us away.
We have no choice but to do business with Nielsen. And
despite recent ratings challenges, our company has always had a
good relationship with Nielsen. So, we are here today
reluctantly, but with a sense of urgency.
In 1964, the Media Ratings Council was established, at the
urging of Congress. The MRC's mission is to maintain confidence
in audience research and to secure measurement services that
are valid, reliable and effective. Historically, participation
in the MRC process has been voluntary. The MRC cannot force
anyone to comply with its procedures. The Local People Meter
Service that Nielsen has implemented in New York and Chicago
and other markets has yet to be accredited by the MRC. And
without quoting a myriad of numbers, it is worth nothing that
in New York, on the average day for the week ending July 10,
the viewing choices of nearly one-third of Black and Hispanic
men ages 18-34 in the LPM sample were not reflected in the
ratings. Despite this, Nielsen has just launched the LPM
service in Washington, D.C., and Philadelphia, without MRC
accreditation. It is clear to me that Nielsen submits to the
MRC processes only when it suits its aggressive business
strategies.
In numerous meetings, e-mails, and letters over the past
year, Tribune has pointed out defects in the Nielsen's LPM
sample. Nielsen has acknowledged the difficulties and has
promised to fix these problems. But, despite Nielsen's effort,
it has failed to fix these problems.
For these reasons, in a letter dated May 25, 2005, Tribune
and 17 other broadcast companies urged Nielsen to postpone the
scheduled deployment of the LPM service in Philadelphia and
Washington, D.C., Nielsen refused. In response, the MRC, under
the guidance of Executive Director and CEO George Ivie,
recommended a meeting between Nielsen and either the MRC's
Television Committee or the full board, or that Nielsen
participate in the MRC mediation process. Broadcasters
accepted, with a preference for mediation. Nielsen refused
both.
Finally, on June 28, the MRC Board of Directors approved a
resolution recommending that Nielsen offer LPM service in
additional markets only after completing an MRC audit.
Nielsen--or, excuse me, Tribune then asked Nielsen to accept
the MRC Board's resolution and delay the scheduled launch in
Philadelphia and Washington, D.C. Nielsen's response again was
an immediate no.
Had Nielsen been more responsive to these concerns of
broadcasters and of the MRC, I doubt that we would be here
today.
A promising new measurement service, Arbitron's Portable
People Meters System, is being tested in the Houston market.
This new passive technology measures both television and radio
audiences. Arbitron has licensed this technology in Singapore,
Norway, and Canada. Unfortunately, Nielsen has the contractual
option to form a joint venture with Arbitron to market the PPM
television service in the United States. Because PPMs are an
alternative to Nielsen's proprietary LPM service, it appears
highly unlikely Nielsen will allow the PPM technology to
compete with its LPM service.
Throughout Tribune's long history, we very rarely have
petitioned for Federal intervention in the marketplace;
however, in this case, despite our efforts, we simply do not
have the ability to persuade Nielsen to submit to voluntary MRC
processes. And because Nielsen is a monopoly, we have nowhere
else to turn to get accurate and reliable ratings.
S. 1372, the FAIR Ratings Act, would correct this market
failure. The bill would not impose any undue burden on parties
to the process and would enable the MRC to fulfill its mission.
In my written testimony, I provided examples and written
communications \17\ showing that we are not dealing with a
trivial dispute or sour grapes because our ratings are down.
And after more than a year's experience in New York and
Chicago, the LPM system continues to be embarrassingly
defective.
---------------------------------------------------------------------------
\17\ Information referred to has been retained in Committee files.
---------------------------------------------------------------------------
Clearly, the free market cannot solve this problem, which
is a serious one. Free over-the-air broadcasters, unlike our
cable and satellite competitors, depend on a single revenue
stream, which is derived from advertising. We do not charge a
subscriber fee, and we make our service available free to all.
Accurate and reliable ratings are keys to the health of our
business.
Mr. Chairman, we appreciate your allowing us to take the
time today to express our views and urge the Committee to
favorably report S. 1372.
Thank you.
[The prepared statement of Mr. Mullen follows:]
Prepared Statement of Patrick J. Mullen, President,
Tribune Broadcasting Company
Mr. Chairman and members of the Committee, thank you for the
opportunity to appear before you today to testify in support of S.
1372, which will go a long way toward assuring that there is a strong
independent body to oversee the reliability of television audience
measurement.
My name is Pat Mullen. Our company, Tribune Broadcasting, operates
26 major market television stations located in 15 states from coast to
coast, including stations in 8 of the 10 largest markets.
All of these TV stations are what used to be called ``independent''
stations--local stations that did not have the legacy of a network
identification to hold loyal viewers year after year. Through
innovative local, sports and syndicated programming, Tribune's stations
have provided viewers with an alternative to the ``traditional''
networks, attracting viewers to programs they could not find elsewhere.
Their success has depended in every case on an accurate count of our
audience. New stations, small stations, UHF stations, as well as
broadcast pioneer stations like WGN-TV in Chicago, KTLA in Los Angeles
and WPIX in New York, have found they can compete and succeed if they
provide new and better programming options for viewers.
Our stations get a report card every morning from Nielsen. Those
ratings determine the viability of our business.
They determine the value of our advertising.
This in turn determines how much money can be invested in
new and better programming, and in new digital technology.
And ratings also determine which programs remain on the air,
and which ones will be taken off for apparent lack of viewer
interest.
Today, all but one of Tribune's television stations have affiliated
with the newer networks, The WB and Fox. We are eager to compete with
our fellow broadcasters, and with the ever-increasing number of
networks vying for viewers' attention over cable and satellite. But to
do this we must have an honest report card. A trustworthy measurement
of the size and composition of each competitor's audience.
Mr. Chairman, I regret to say that the measurement system we have
today in the largest television markets is not worthy of public trust.
It does not have the trust of our company or that of more than a dozen
other responsible broadcasters.
Congress has repeatedly acknowledged the importance of a free and
robust broadcast service, which is particularly important in times of
crisis. We believe the system of over-the-air television in America
demands statistically valid and reliable measurement of its audience.
It deserves a guaranteed minimum standard of accuracy because of the
importance of television news, public affairs, sports and entertainment
programming to this country's culture and to our democracy.
In times of crisis, from hurricanes in Florida to fires in
California, when the cable is out and satellite service is interrupted,
broadcasters serve as the first responder on the scene, transmitting
potentially life saving information to our fellow citizens. We are
proud of that record of service. It may prove even more vital if, as in
London and Madrid, terrorist attacks continue to spread beyond the war
zone in the Middle East.
But we are not here today in an attempt to secure an advantage over
our multi-channel competitors or to slow the erosion of our audiences
caused by the growing choices available to viewers. Our company
welcomes competition.
The problem, Mr. Chairman, is that the keys to our success--our
ratings-- are held by a monopoly. When Nielsen had a competitor, its
service and its response to client concerns were substantially better
than they are today. In the absence of competition, we are left to
plead for fair treatment and reliable results. Time and time again,
Nielsen has turned us away.
We have no choice but to do business with Nielsen. Ratings are the
currency on which the advertising business operates. And despite recent
challenges, our company has always had a good relationship with
Nielsen. So we are here today reluctantly, but with a sense of urgency.
In 1964, the Media Rating Council was established at the urging of
Congress. It is a nonprofit organization whose membership includes
representatives of broadcast TV and radio, cable television, print,
advertisers, ad agencies, and now Internet constituencies. The MRC's
mission is to maintain confidence in audience research and secure
measurement services that are valid, reliable and effective. MRC does
this through audits to test the methodology and credibility of research
services, and accreditation to certify services that meet the MRC's
minimum standards. Research services must disclose their data to the
MRC to enable it to validate their measurements.
The Media Rating Council is a classic example of industry self-
regulation. It consumes no tax dollars nor requires government
oversight. It does its job quietly, professionally and efficiently,
with participation by all segments of the industry. In our experience
the MRC has never been used for private gain by one member over
another, or to delay or stop innovation. The very existence of the
MRC's auditing and accreditation processes, and its diverse make-up,
tend to keep participants honest.
Historically, participation in the MRC's processes has been
voluntary. The MRC cannot force anyone to comply with its procedures,
and it cannot require a ratings service to submit to an audit or to
offer only accredited measurement services. Unfortunately, Nielsen has
chosen to ignore the MRC's guidance in deploying Local People Meter
(LPM) service.
The LPM service that Nielsen has implemented in New York and
Chicago has yet to be accredited by the MRC. It is worth noting here
that in New York, on the average day for the week ending July 10, the
viewing choices for nearly one-third of the Black and Hispanic men ages
18-34 in the Nielsen LPM sample were not reflected in the ratings.
(Additional detail is available in the attachments to this
testimony.)**
---------------------------------------------------------------------------
** Attachments retained in Committee files.
---------------------------------------------------------------------------
Despite these kinds of obvious flaws, Nielsen has just launched its
LPM service in Washington, D.C. and Philadelphia--also without MRC
accreditation. It is clear to me that Nielsen submits to MRC processes
only when it suits its aggressive business strategies.
Tribune has tried to work constructively with Nielsen and to
suggest ways to improve audience measurement. In numerous meetings, e-
mails and letters over the past year, Tribune has pointed out defects
in Nielsen's LPM sample and faulting rates. The problems being
presented have led to significant under-reporting of important audience
segments. Nielsen has acknowledged difficulties and has promised to fix
the problems. But despite Nielsen's efforts, it has failed to fix these
problems.
For these reasons, in a letter dated May 25, 2005, Tribune and 17
other broadcast companies embraced the new technology but urged Nielsen
to postpone the scheduled deployment of LPM service in Philadelphia and
Washington until the MRC deemed the system reliable in markets where it
was already being used.
Nielsen responded the following day. It said ``the broadcast group
request for some sort of mandatory, prior MRC accreditation raises
considerable antitrust concerns.'' Nielsen rejected the industry's
proposal and the legitimate concerns detailed in our letter.
In response, the MRC, under the guidance of Executive Director/CEO
George Ivie, recommended a meeting between Nielsen and either the MRC's
Television Committee or the full MRC Board, or that Nielsen participate
in the MRC mediation process. Broadcasters said we would accept either
approach, with a preference for mediation. Nielsen refused both, saying
mediation would be ``unnecessarily cumbersome and time consuming.''
Finally, on June 28, the MRC's Board of Directors approved a
resolution recommending that Nielsen offer LPM service in additional
markets only after completing an MRC audit. Tribune then asked Nielsen
to accept the MRC Board's resolution, delaying the scheduled LPM launch
in Philadelphia and Washington. Nielsen's response was an immediate,
``No.''
So the company continues to ignore the legitimate concerns of its
customers and the MRC.* Its actions are those of the classic
unregulated monopoly, accountable to no one. Had Nielsen been more
responsive to broadcasters' or the MRC's concerns, I doubt we would be
here today.
---------------------------------------------------------------------------
* Correspondence submitted with this testimony documents this
frustrating process. The submitted material has been retained in
Committee files.
---------------------------------------------------------------------------
A promising new measurement service, Arbitron's portable people
meter (PPM), is being tested in the Houston market. This new technology
measures both television and radio signals, and I believe Arbitron
plans to use this system for radio ratings starting in 2006. Arbitron
has licensed this technology in Singapore, Norway and Canada. Although
Arbitron is managing this test, Nielsen has the contractual option to
form a joint venture with Arbitron to market PPM television service
commercially in the United States. Thus, it is our understanding that
Nielsen could effectively control how and when PPM technology will be
deployed for television measurement. Because PPMs are an alternative to
Nielsen's proprietary LPM service, it appears highly unlikely Nielsen
will allow the PPM technology to compete with its LPM service.
We hope this testimony makes clear the need for government
intervention in this critical segment of the U.S. economy. Throughout
Tribune's long history in both print and broadcast journalism, we very
rarely have petitioned for Federal intervention in the marketplace.
Like many of my fellow broadcasters, I personally have spent many days
trying to reach a private solution to this problem with Nielsen. We
simply do not have the ability to persuade Nielsen to submit to MRC
processes and roll out its new measurement systems only after they have
proved reliable to an independent and expert body, the Media Rating
Council.
And because Nielsen is a monopoly, we have nowhere else to turn to
get accurate and reliable ratings.
S. 1372, the FAIR Ratings Act, would correct this market failure by
requiring MRC accreditation before the commercial introduction of any
commercial ratings measurement system. The dispute resolution system
established by the bill would provide ready means to test the
reliability of new measurement systems, and would encourage companies
that design them to vet them thoroughly, ensuring their credibility and
integrity before they are launched commercially. The bill would not
impose any undue burden on parties to the process, and would enable the
Media Ratings Council to fulfill its mission of encouraging the
development of reliable and improved ratings measurement systems, which
we fully support.
The examples included in this report demonstrate that we are not
dealing with a trivial dispute or sour grapes because our ratings are
down. After more than a year's experience in New York and Chicago, the
LPM system continues to be embarrassingly defective.
Sampling issues abound, including problems with response rates, in-
tab representation and fault rates. For example:
New York's LPM response rate averaged 25.3 percent for the
week ending July 3, 2005. This means that 3 out of every 4
households initially designated as sample households refused
installation of a people meter in their home or accepted a
meter but did not contribute any viewing data.
Young men ages 18-34 have been persistently under-
represented in Boston, Chicago, Los Angeles, New York,
Philadelphia and San Francisco. Fault rates for men 18-34
generally are twice as high as those for men ages 55+ in LPM
samples.
Fault rates remain unacceptably high for important audience
segments such as African Americans and Hispanics despite new
coaching initiatives. On the average day in New York for the
week ending July 10, the viewing choices of nearly one-third of
the Black and Hispanic men ages 18-34 in the LPM sample were
not reflected in the ratings.
Chicago sample data for the week ending July 10th show that
almost one-third of the 443 African Americans installed in the
sample were not in tab--meaning their television viewing was
not counted in the ratings.
Households of 5 persons or more have been persistently
under-represented in the total samples in New York, Los
Angeles, Chicago and Boston. In New York, for the week ending
July 10, the viewing choices of more than 1 in 4 of the Black
and Hispanic households of 5 or more persons in the LPM sample
were not reflected in the ratings.
Fault rates for households of 5 or more are generally 2 to 3
times as high as in one-person households.
Clearly, the free market cannot solve this problem, which is a
serious one. Free over-the-air broadcasters, unlike our cable and
satellite competitors, depend on a single revenue stream, which is
derived from advertising. We do not charge a subscriber fee, and we
make our service available free to all. Accurate and reliable ratings
are key to the health of our business. Mr. Chairman, we appreciate your
allowing us the time to make our views known, and urge the Committee to
favorably report S. 1372.
Thank you.
Senator Burns. Thank you very much.
Now we'll have Mr. Gale Metzger, Former President,
Statistical Research Inc.
Thank you for coming today.
STATEMENT OF GALE METZGER, FORMER PRESIDENT, STATISTICAL
RESEARCH INC.
Mr. Metzger. Thank you, Senator Burns. And thank you for
the invitation.
I am Gale Metzger. My entire career has been spent in the
audience-measurement arena. I began at the A.C. Nielsen
Company, and then for 32 years I was President of Statistical
Research, Inc., a company that I founded with Dr. Gerald
Glasser, of New York University. In 1963, I participated in the
Congressional hearings as a Nielsen employee. SRI, our firm,
conducted methodological research for the industry for over 30
years. In the 1990s, we created and operated a ratings
laboratory under the name of SMART. Currently, I am a senior
consultant with Knowledge Networks, Inc.
I appear today representing myself, my own views and
interpretations. There are no lawyers, no public-relations
people or anyone else behind me telling me what to say. I speak
from a lifetime's experience and a deep commitment to the
understanding that research quality makes a difference. Good
information helps markets work better. Bad information
undercuts business performance.
I will briefly address three points:
First, the Media Rating Council. I was present when the
Broadcast Rating Council, now the Media Rating Council, was
formed. I participated in the debates around the operating
protocols. I was the Nielsen person who was responsible for
structuring the first audit of its services. SRI's Syndicated
Audience Measurement Service to the radio industry, RADAR, was
audited by the MRC for over 30 years.
The MRC serves a vital role in our industry. An important
byproduct of its work is to encourage innovations and
improvements in methods. Whether I was working at, or owned, a
service that was being audited, the MRC helped me do a better
job. If audit reports were open and available to all clients,
they would be even more valuable.
Media ratings systems are frail, sometimes more so than we
practitioners like to admit. To use information from these
systems intelligently and effectively, users need to know what
is going on, and they need to know before the data hit the
marketplace, not after. Hence, I agree with the intent of the
proposed legislation, which is that all services providing
marketplace currency be accredited by the MRC.
I understand that Nielsen has expressed objections to the
proposals and stated they would lead to less innovation and
less competition. The opposite would be true. It would be
difficult to have less competition or less innovation than we
have now. And I must insert that I have a totally different
view than was expressed by Ms. Whiting and Ms. Crawford on the
effects on competition and innovation.
It is in Nielsen's and the industry's best interests to
embrace the intent of this legislation. I, further, believe
that complete coverage of all services is what the industry
committed to achieve in testimony before the Congress in 1963.
Changes in the industry structure have made the MRC even
more important today. The networks once dominated national
television. They were permitted to work together on issues
related to methodology. Then, there was a balance of power
between Nielsen and the networks. With a fragmented medium, no
single client or group of clients wield that much influence. In
effect, if Nielsen does not answer to the MRC, it answers to no
one. I believe this explains, in part, Nielsen's new, more
aggressive posture with the MRC.
Nielsen and others may have particular points about the
legislation that warrant discussion. I'm confident details can
be worked out if we have sufficient desire on the part of all
concerned parties to do so. That has apparently not been the
case for the past year, so I think I understand the reasons why
you, Senator Burns, introduced this bill.
Second, why are we here? I submit that we are here because
Nielsen clients feel they are hostages to a company that
controls their basic well-being. Further, Nielsen operations
are deficient, and those deficiencies jeopardize those clients'
businesses. This is not a manufactured controversy. There is a
real problem. When emotions run as high as they currently do
among a large share of clients, something is not right.
Evidence of the industry's effort to bring improvement
include the network support of SRI's methodological research,
the support of CBS and others for the AGB initiative, and, more
recently, the support of 30 networks, advertisers, and agencies
of our SMART Ratings Laboratory.
Nielsen deficiencies are several and significant. Perhaps
the broadest complaint is that Nielsen is not responsive on
data-quality issues and to client concerns unless the threat of
competition is raised.
The people meter was introduced by Nielsen in 1987 only
after a British company, AGB, tried to enter the U.S. market
with a similar meter. Nielsen's new AP meter was announced in
1995, only after the SMART Laboratory was underway with a new
meter in development. When SMART went away, the introduction of
the AP meter was delayed. Ten years after the fact, the AP
meters have just begun to roll out.
In sum, clients will tell you that when the threat of
competition is present, Nielsen is a different company than
when, as now, there is no threat.
A more specific deficiency is Nielsen's metering
technology. It has not kept pace with modern media. We are
moving rapidly into the 21st century with an aged 20th-century
meter platform. The high fault rates in the Nielsen sample is
evidence of their out-of-date technology. Fault rates are high
because the meters are not state-of-the-art. The fault is with
the meters and how they operate. The fault is not with the
homes or the people in them.
Are Nielsen's new systems better than the old? Are the
audience estimates more accurate? The truth is, no one knows.
And that is disconcerting. We do not know the effect of skipped
homes or faulting homes. What is known is that Nielsen is
producing more data and generating more revenue than ever
before.
Data access is a core deficiency of Nielsen. Clients cannot
get to information they need for decisions. Data access is an
important component of quality.
Nielsen weighting procedures are still another point of
contention.
The real problem with each of these and other deficiencies
is that they all affect audience levels. Some reported audience
ratings are higher, and some lower, than is the reality. Some
organizations' bottom lines are improved, and some are made
worse. This observation leads to the last deficiency that I
cite today: Nielsen is almost cavalier about making changes in
procedures. If measurement techniques change, some audiences
will be greater, and some lesser, than before. The real
audience has not changed, but the reported audience change.
Some win, some lose. The only way to prepare for such events is
to communicate planned changes with evidence to the benefits to
the industry and to supply an abundance of data. To force-feed
a change invites disaster. That is what Nielsen effectively
did, and they reaped what they sowed.
Nielsen publicly proclaimed their shock, their dismay and
surprise at the industry's reaction to the LPM. Such reactions,
to me, speak of posturing or a lack of understanding of their
clients' legitimate concerns.
Third, and last, the need for action. At the 1963 hearings
on ratings, Chairman Harris, of the House Committee on
Interstate and Foreign Commerce, referred to the ratings
industry as being in an intolerable situation. Today, many also
feel that situation is intolerable.
I see confusion in the marketplace about the nature of the
problem. More data does not equal better data. Also, it is
unfortunate that a large part of the discussion about the LPM
has focused mainly on minority measurement issues. I do not
believe the problem is associated only with minorities, and not
only with the LPM. The LPM controversy is the tip of the
iceberg.
I know about minority measurements. For over 25 years, SRI
served the National Black Network and Sheridan Broadcasting as
part of our RADAR service. Measurements for minorities should
be judged by the same quality standards, and subject to the
same audit review, as are all other media audience
measurements.
Audience measurements should be inspected for all important
population subgroups, but I do not believe race or ethnicity is
the primary issue with the LPM. I believe the issue is how a
monopolist relates to its clients. The issue is whether these
Nielsen ratings data, when used as currency, are really ``funny
money.'' We just do not know enough about Nielsen research
quality.
Susan looks at Nielsen Media Research and sees a glass
full. I see a gallon jug with a few drops of water in the
bottom, sloshing around.
Whether this proposed legislation goes too far, or not far
enough, I shall leave for others to judge. The MRC is
essential, but it may not be sufficient. Many feel the need for
a joint industry effort to set specifications and award an
industry contract, as occurs in other parts of the world. In
that direction, the Advertising Research Foundation has
structured an audience measurement initiative which is
currently being discussed.
Like most others here, I favor free market solutions for
free markets. Where there are multiple buyers and sellers,
there is seldom cause for the government to become involved.
However, here we have a monopoly. Legislation may be the only
way to get Nielsen to the table, and I think this bill is the
right way to go. And further action may be needed to deal with
the possible industry initiative to further improve the
situation.
Thank you very much.
[The prepared statement of Mr. Metzger follows:]
Prepared Statement of Gale Metzger, Former President,
Statistical Research Inc.
I am Gale Metzger. My first professional job was with the A. C.
Nielsen Company. For 32 years, I was President of Statistical Research,
Inc. a media and marketing research company that I founded with Dr.
Gerald Glasser of New York University. I have been active in the
industry and served as Chairman of the Board of the Advertising
Research Foundation and President of the Radio and Television Research
Council and the Market Research Council.
In 2001, SRI was sold in two parts. Our network radio measurement
service went to ARBITRON and the other operations were sold to
Knowledge Networks, Inc.--a firm I continue to work with as a senior
consultant.
For 48 years, I have been engaged in media research. Over 40 years
ago, I participated in the 1963 Congressional Hearings as a Nielsen
behind-the-scenes overnight supplier of answers to questions posed by
congressional staffers. Fifteen years ago, at the request of key
industry stakeholders, our firm (SRI) conducted an in-depth review of
Nielsen's newly introduced people meter system which resulted in a
seven volume 600-page report. Nielsen called that work ``an outstanding
effort'' and the industry characterized it as a blueprint for progress.
SRI conducted methodological research for the industry for over 30
years. In the 1990s we created and operated a ratings laboratory under
the name of SMART. SMART was an acronym for Systems for Measuring And
Reporting Television. All of that work was dedicated to understanding
and improving measurement methods. The SMART laboratory was successful
in developing new, user-friendly TV meters and in providing audience
data to client desktops along with analytic software to enable use of
ratings information for business decisions on a timely basis. In 1999,
SMART was proposed as a competitive system to Nielsen. The necessary
capital to launch the service, however, was not forthcoming.
In January of this year, I was asked by the Advertising Research
Foundation to provide a historic overview of TV audience measurement in
the United States at a special meeting it convened on the topic of
Accountability of Audience Measurement. I am submitting the paper
provided there as an addendum to my testimony today.
I appear today representing my own views and interpretations of
current events in the television audience measurement business. I have
no lawyers, no public relations people or anyone else behind me telling
me what to say. I speak from a lifetime's experience and a deep
commitment to the understanding that research quality makes a
difference. Good information helps markets work better; bad information
undercuts business performance.
I will briefly address three general points.
First, the role and value of the MRC to the audience
research business.
Second, why we are here? Why is legislation being
considered?
Third, the need for action to enable the television ratings
process to facilitate rather than frustrate the marketplace.
Media Rating Council
I was present when the Broadcast Rating Council, now the Media
Rating Council was formed. I participated in the debates around the
operating rules and helped with drafting the disclosure standards that
are part of the MRC protocol today. I was the person at Nielsen who was
responsible for structuring the first audit of Nielsen. In later years,
SRI provided a syndicated audience measurement service to the radio
industry, RADAR, which service was audited by the MRC for 30 years. I
have deliberated and consulted with MRC executive directors for over 40
years.
The MRC serves a vital role in our industry. By assuring disclosure
of research company methods and by auditing the accuracy and
completeness of disclosure, the MRC enables an informed market. An
important byproduct of is work is to encourage innovations and
improvements in methods. MRC reporting and tracking of key quality
indicators, appropriately and constructively pressures research
companies to rectify weaknesses.
When I was working at or owned a service that was being audited,
the MRC helped me do a better job. Totally independent, it gave
research company management an objective quality control report. When I
was at Nielsen or in my own business, I was paying an audit firm and I
wanted maximum value from that expenditure, just as any other expense.
Hence, there were occasional discussions between the researcher and the
auditor around the audit plans and the most effective use of audit
resources. There was a healthy dialogue, and as a result audit
operations were improved.
I have always believed that audit reports should be open and
available to all clients whether or not the clients were members of the
MRC. I was happy to show my audit reports to my clients. All media
rating systems are frail, sometime more so than we practitioners like
to admit. We manufacture numbers (statistical estimates) that have
broad business and social implications. We use methods that are
subjective and often less than ideal. To use information from these
systems intelligently and effectively, users need to know all. And they
need to know before the data hit the marketplace, not after.
Hence, I agree with the intent of the proposed legislation which is
that all services providing marketplace currency be accredited by the
MRC. I understand that Nielsen has expressed objections to the
proposals and stated that the proposed plan would lead to less
innovation and less competition. First, it would be difficult to have
less competition or less innovation than we have now. Second, it is my
impression that Nielsen has become a reluctant participant and not
permitted select components of their services--new and old--to be
examined by the MRC process.
During the Congressional Hearings in 1963, Nielsen clients were
incensed because they were unaware of some Nielsen procedures disclosed
at the hearings. There is a principle that characterizes all successful
service businesses--keep your clients involved and informed. Never
surprise a client. I believe it is in Nielsen's and the industry's best
interest to embrace the intent of this legislation. I further believe
that complete coverage of all services was what the industry committed
to achieve in testimony before the Congress in 1963.
An important change in the industry structure over the past 20
years has made the MRC industry role even more important today. When
the networks effectively dominated the national television arena and
were permitted to work together on issues related to research
methodology, there was a balance of power between Nielsen and the
networks. With a fragmented medium, no single client or group of
clients wields that much influence. In effect, if Nielsen does not
answer to the MRC, it answers to no one. I believe this explains, in
part, their new, more aggressive posture with the MRC.
Nielsen and others may have particular points about the legislation
that warrant discussion. I am confident that details can be worked out
to the benefit of all, if we have sufficient desire on the part of all
concerned parties to do so. That has apparently not been the case for
the past year, so I think I understand the reasons why Senator Burns
introduced his bill.
Why We Are Here
I submit that we are here because Nielsen clients feel they are
hostages to a company that controls their basic well-being; further,
that Nielsen operations are deficient in important regards and those
deficiencies jeopardize the clients' businesses. This is not a
manufactured controversy; there is a real problem. We are not here
because of normal, expected competitive posturing. I do not defend the
actions of some media companies, but I recognize their actions as a
response to dealing with a monopolist who is unresponsive to the
fundamental issues. When emotions run as high as they currently do
among a large share of the client community, you know something here is
not right.
The industry's natural response should be to work quietly with
Nielsen to improve. Nielsen is the industry's nest and a bird does not
foul its own nest! Agencies do not want to say to advertisers that I am
spending your hundreds of millions of dollars on meaningless numbers;
nor do the media want to say to advertisers that I am taking your
hundreds of millions of dollars on meaningless numbers. So while the
industry has often striven for a constructive response, Nielsen simply
does not react. I believe that Nielsen has been its own worst enemy in
thwarting a constructive dialogue.
Evidence of the industry's efforts to bring improvement include the
networks support of SRI's methodological research, the support of CBS
and others of the AGB initiative and more recently, the support of
thirty networks, advertisers and agencies for the SMART ratings
laboratory.
Nielsen deficiencies are several and significant. Perhaps the
broadest complaint is that Nielsen is not responsive on data quality
issues and to client concerns--unless the threat of competition is
raised. The people meter was introduced by Nielsen in 1987 only after a
British company AGB tried to enter the U.S. market with a similar
meter. Nielsen's new A/P meter was announced in 1995 when the SMART
laboratory was in process. It was noticed by all that when SMART went
away, the introduction of the A/P meter was delayed. Ten years after
the fact, the A/P meters have just begun to roll-out.
The A/P meter involves changes in Nielsen operations. Research
should have been conducted to know how best to proceed. When SMART was
in operation, Nielsen published a copyrighted research plan that was
well framed. After SMART went away, the plan was forgotten and the
industry is now faced with core operating procedures that are
effectively untested and unproven.
In sum, clients will tell you that when competition or the threat
of competition is present, Nielsen is a different company than when, as
now, there is no competition.
A more specific deficiency is Nielsen's metering technology. It has
not kept pace with modern media. That means Nielsen has been unable to
measure many new forms of TV receivers. As a result, homes that are
selected to be in their samples are passed over and other homes with
only old technology replace them. For example, in the Nielsen People
Meter system today, you are not counted if you have a TiVo--or any
other Digital Video Recorder (DVR). Your neighbor who does not have a
TiVo takes your place in representing America's viewing. TiVos have
been around for 6 years. Nielsen says they will meter and report usage
in DVR homes tomorrow. Tomorrow remains elusive. We are moving rapidly
into the 21st century with aging 20th century technology.
The high fault rates in the Nielsen sample is further evidence of
their out-of-date technology. A fault is what the name implies. It
means that some homes that do have meters are not processed, are not
counted, because something is wrong with the data from that home.
Faults have been around forever. Current fault rates in Nielsen Local
People meter samples are high because the meters are not state-of-the-
art. The fault is with the meters and how they operate; the fault is
not with the home or the people in it. Our goal in the SMART laboratory
was to reduce fault rates to 5 percent or less. Though the target level
was not achieved before the lab was dismantled, we were gaining on it.
In fact, we were under 10 percent--and my engineers assured me that we
would get there.
Are Nielsen's new systems better than the old? Are the audience
estimates with the samples omitting bypassed and faulted households
more accurate? The truth is that no one knows and that is
disconcerting. What is known is that Nielsen is producing more data and
generating more revenue than ever before. Perversely, Nielsen reports
they are beginning to measure TiVo households under the old local
meter/diary measurements.
Data access is another core deficiency of Nielsen. Data access is
an important component of quality. The best information is of no value
if you cannot get to it. Clients cannot access the information they
need to make business decisions on a timely basis. Nielsen analysis has
always been slow and expensive. They place a tourniquet on information
flow by their ineptness and cost. SMART showed the way as to how to do
it better. Nielsen has not seen fit to open up the process to allow
effective use of their information.
Nielsen weighting procedures are another point of contention. With
an unweighted sample, all people count the same. Weighting a sample
means that some people count more than others in the statistical
process. There are several good reasons for considering weighting. The
issue here is that Nielsen has changed its attitude toward weighting
which they had touted for 50 years. The old position was that the
sample should not be weighted for demographic characteristics because a
pure probability sampling approach was superior. The new position is a
180 degree shift. Nielsen virtually recommends weighting on every
variable. The problem is that when a statistician advocates weighting,
there is an implication that the resulting data quality are improved.
Nielsen's arguments for weighting are novel, unproven by independent
research and to my knowledge not supported by theory.
My conviction is that they should have introduced some kind of
weighting years ago. Statisticians agree with judicious weighting while
being concerned about the abuse of weighting. It is like putting a new
coat of paint over old wood. The resulting weighted sample may look
better but the non-responding households are still missing.
The real problem with each of these and other deficiencies is that
they all affect audience levels. That means that some audiences'
ratings are higher, and some lower, than is the reality. Some
organization's bottom lines are improved and some made worse.
That observation leads to the last deficiency that I cite today.
Nielsen is almost cavalier about making changes in procedures. I know
from experience that two actions can turn a marketplace on its ear. If
a pricing formula is changed such that some pay more and others pay
less than previously, turmoil will be assured. Similarly, if
measurement techniques are changed, some audiences will be greater and
some lesser than before. The real audience will not have changed but
the reported audience changes. Some win; some lose.
The only way to prepare for such events is to communicate planned
changes with evidence on the benefits to the industry and to supply an
abundance of data (including parallel measurements, if necessary). This
would enable the industry to prepare for an orderly transition from one
operating frame to another. To force feed a change invites disaster.
That is what Nielsen effectively did, and they reaped what they sowed.
Nielsen publicly proclaimed their shock, their dismay and surprise at
the industry reaction to the way in which the LPM was introduced. Such
reactions to me speak of posturing or a lack of understanding of their
clients' legitimate concerns.
Need for Action
In connection with the 1963 hearings on ratings, Chairman Harris of
the House Committee on Interstate and Foreign Commerce referred to the
ratings industry as being in an ``intolerable situation.'' Many today
also feel that the situation is intolerable. Something must be done to
bring balance to the relationship between Nielsen and the industry it
is supposed to serve. As with economic trends, I do not believe this
can go on forever. If an economy is constantly in deficit, that economy
eventually collapses. Some believe that because it is so difficult to
bring improvements into this system that it too will eventually
collapse. The wheels will come off the bus.
I see confusion in the marketplace about the nature of the problem.
More data does not equal better data. Also, it is unfortunate that a
large part of the discussion about the LPM has focused mainly on
minority-measurement issues. I do not believe the problem is associated
only with minorities. The key independent variable with respect to
meter performance is the number of TV sets in the home. More sets
equals more problems and more faults. That is due to an out-of-date
meter platform.
In the late 1960s and early 1970s, SRI did several comparisons of
Nielsen and ARBITRON data. Clients complained that one service favored
their competitor or visa versa. Often, the differences were random and
a function of small sample sizes. Yet people thought they saw patterns.
I believe there is more speculation and political positioning going on
than solid data analyses.
I know about minority measurements. For over 25 years, we served
the National Black Network and Sheridan Broadcasting as part of our
RADAR service. In the early 1970s, we produced special measurements of
Spanish audiences in the New York area which P&G (uncharacteristically)
urged their agencies to use in buying NY Spanish television.
Measurements for minorities should be judged by the same quality
guidelines and subject to the same audit review as are all other media
audience measurements. While I believe audience measurements should be
inspected for all important population subgroups, I do not believe race
or ethnicity is the primary issue with the LPM.
I believe the issue is how a monopolist relates to its clients. The
issue is whether these Nielsen ratings data, when used as the currency,
is really funny money. We just do not know enough today about the
Nielsen research quality.
Nielsen has a difficult task. But syndicated services exist here
and around the world without the acrimony and anger that characterizes
the U.S. television marketplace. I believe Nielsen has the ability--and
needs to find the will--to serve its market proactively. At the end of
the People Meter review in 1988 we recommended that Nielsen concentrate
on three initiatives:
Defined procedures and quality control
Methodological research
Client Involvement
Nielsen seemed to endorse those proposals and I hope they may
reinspect their position today and truly strive to work with the
industry openly and forthrightly.
Whether this proposed legislation goes too far or not far enough I
shall leave for others to judge. Like most others here, I believe a
voluntary industry solution is the best one. But failing that,
legislation may be the only way to get Nielsen to the table, and I
think this bill is the right way to go. I am absolutely convinced that
something must be done.
A disclaimer: some of the points articulated herein (e.g., the
discussion of weighting) are simplified for purpose of clarity.
The author trusts that the thrust of the discussion is clear.
Attachment--History of TV Audience Measurement in the USA
(by Gale D. Metzger)
Introduction
Simon Schama's, Dead Certainties, is a good read. Professor Schama
affirms that the dead, really are dead; everything else--why they died,
how they died--is a matter of interpretation. History is in the eye of
the beholder. For some, his book marked the death of any certainty
about history.
I have few absolutes to offer in my 30-minute history of television
audience research in the USA. You will hear my interpretation of events
and my view of how today contrasts with the past.
For a more complete history of radio and TV measurement prior to
the mid-1980s, you should read Mal Beville's excellent book: Audience
Ratings, published in 1988. Another document that I commend to your
reading is Thirty Years of CONTAM, published in 1995.
I speak from my own personal knowledge of 47 years in the business.
I shall recall the Congressional hearings of 1963 and the associated
fear of government regulation. What was done by Nielsen and the
industry to deal with this threat?
Then I will look at the 1980s and 1990s and how the industry and
Nielsen acted in response to many of the same problems we face today--
albeit on a lesser scale. Last, a review of today's situation and the
lack of resources dedicated to improving ratings quality.
A. C. Nielsen
Sadly, I assure you with dead certainty that the central figure in
the history of TV audience measurement, Arthur Charles Nielsen, is
deceased. Born in 1897, he died in 1980. He left a substantial legacy.
A. C. Nielsen was the son of a Danish immigrant. His father worked
for 40 years in Quaker Oats accounting division. His son inherited his
bean-counter mentality. Witness the label ``audit pioneer'' in the
ARF's 1977 tribute to ``The Founding Fathers of Advertising Research.''
Mr. Nielsen--as he was addressed by most--was a trained engineer and a
scientist. He graduated from the University of Wisconsin's engineering
school with an extraordinary academic record. In 1923, he started the
A. C. Nielsen Company with money borrowed from friends. He was an
excellent businessman, a man of strong character and conviction. I knew
him as a kind and intellectually generous person. For me, he was the
proverbial man with a steel fist in a velvet glove.
The Nielsen Company was a research company that produced good
services and reasonable profits. It reflected Art the scientist/
auditor. He chose a micrometer as a company symbol and a favorite
authority he loved to quote was the scientist, Lord Kelvin, who said:
``If you can measure that of which you speak, and can express
it by a number, you know something of your subject.''
That A. C. Nielsen Company was one I was proud to work for from
1958 to 1969.
The Nielsen business progressed from performance surveys of
industrial machinery in 1923 to retail sales measures in 1933 and to
radio audience measurement in 1942. In 1950, television audiences were
added to the package. He invested 17 years and $15,000,000 before
achieving a break-even financial position in the media business. That
is focusing on the long term. That is a person who did not have to
answer to Wall Street.
His primary objective was to provide accurate and thorough research
to support marketing efficiencies. He knew that good information
lubricated economic gears. He wanted to provide the tools to assess
advertising and promotion options in the marketing process. He wanted
to ``further the science of marketing research.'' So that, while his
name is primarily associated with television research, his personal
goals were broader.
Congressional Hearings of 1963
I shall skip the stories of the early competition in audience
research, well told in Mal's book. I begin in 1963. Hearings were held
before the Committee on Interstate and Foreign Commerce of the House of
Representatives. The subject was ``The Methodology, Accuracy and use of
Ratings in Broadcasting.'' There are several thoughts on how the
hearings came to be. The role of ratings as highlighted in the quiz
show scandals of the late 1950s was one stimulant.
The hearings were to consider the production and use of ratings in
the broadcasters' fulfillment of their statutory obligation to serve
the public interest. If ratings were unreliable, then they were
worthless as a means to that end. Chairman Harris felt there was an
``intolerable situation.'' The question was: is the service of the
public and is television commerce being conducted on funny money?
A year earlier, the prospect of hearings loomed. Congressional
investigators appeared on Nielsen's doorstep and that of many other
companies. The actual hearings commenced on February 9, 1963 and ended
19 months later. The proceedings are memorialized in a 1,932-page
transcript.
The hearings changed audience measurement forever. While Nielsen as
the largest audience research entity had some difficult days during the
hearings, in the end Nielsen was a great benefactor from the process.
Evidence of that is in their market position in the ensuing 40 years.
The Nielsen Company responded well by working with the industry.
Congress tried to determine whether the rating companies were doing
what they said they did. They investigated how rating reports were used
and the effects on both programming and on the sale and purchase of
broadcast time. In brief, they discovered some research companies were
apparently making up the numbers from imaginary surveys. Nielsen and
others were accused of misleading clients as to their procedures and of
not accurately describing the samples on which ratings were based.
Congress was exploring Federal regulation and considering legislation.
The Bureau of Census was asked if it could undertake to provide certain
broadcast data.
The Federal Trade Commission became involved. Several ratings
companies, including Nielsen, Pulse and the parent company of Arbitron,
were ordered ``to cease and desist from misrepresenting the accuracy
and reliability of their measurements, data and reports.'' The FTC
issued guidelines to the media for use of ratings and stated that
audience claims must be ``truthful and not deceptive'' and that the
media must avoid activities intended to ``distort or inflate'' audience
data solely during survey periods.
Those activities led to a hyperactive industry response led
primarily by the NAB and featured broad participation. One NAB
committee included 24 corporate representatives, including delegates
from the AAAA's and the ANA. The three networks formed a Committee on
Nationwide Television Audience Measurement, which became known by the
acronym, CONTAM. The purpose was to improve the quality and
understanding of audience ratings. That purpose would be fulfilled by
three actions:
Require rating services to describe what they do;
Audit rating services to determine if they do what they say;
and
Conduct continuing and effective research to improve rating
methodology.
Work began immediately.
The hearings changed the way the industry operated. They were a
force that improved the audience ratings systems. Arthur Nielsen, Jr.
pledged to the Congressional Committee the Nielsen Company's full
cooperation in achieving the ``broad and worthy'' objectives that were
targeted. Some clients publicly defended Nielsen throughout this
period. In private, all clients were adamant in their conviction that
they would not be blind-sided--not be surprised again by a ratings
service. Full disclosure and auditing of procedures would proceed
forthwith.
Not everyone liked the congressional hearings process. The
investigators were zealous--and at times over the top. During the
hearings, it had been asked of Nielsen if the use of a permanent panel
of sample respondents meant that members might be ferreted out and
influenced to distort the ratings. After the hearings concluded, there
was an incident that raised eyebrows. One of the investigators used
information obtained during the hearings to aid a well-known
entertainer. Some Nielsen homes were contacted to encourage their
viewing of an upcoming special featuring the entertainer. The efforts
were discovered by Nielsen and the culprits were called to account. The
episode was kept under wraps but those who needed to know were informed
and the whole event served to take some of the pressure off of Nielsen
with the client community.
Media Rating Council
As a result of the governmental threat, in less than 1 year, an
entity to audit research procedures was created. It was originally
called the Broadcast Rating Council, which is today's Media Rating
Council. It was to assure ratings companies said what they did and did
what they said. Was any of this anti-competitive in restricting
innovations or options? The Justice Department okayed the actions. The
audits were firmly grounded. The industry had spoken.
To assure consistency in the mechanics, the industry cooperated in
creating standard definitions, standards for reports and disclosures
and standards for presenting sampling error estimates.
The research companies had differing reactions. Nielsen worked with
Ernst and Ernst's Operations Research group to create an effective and
efficient working model. The E&E staff was highly professional and the
early workings were relatively smooth. Over the years the attitude
toward the Audit functions ranged from those who saw it as an asset--an
effective, independent quality control check--to those who sought to
avoid such crosschecks or to treat it as an unnecessary nuisance to be
challenged.
The audit operations have benefited from top quality leadership of
Executive Directors Ken Baker, Mal Beville, John Dimling, Mel Goldberg,
Dick Weinstein and today, George Ivie--who you will hear from later
this afternoon. All of these individuals with their respective Boards
of Directors worked effectively to sort the wheat from chaff and
generally served to be an influence for full and accurate disclosure
and for improvement.
The MRC today is an important and vital resource. I believe the MRC
would be even more valuable if its reports were open to the industry.
The audits are closed to all but MRC directors. The industry only knows
pass or fail. That is insufficient. I believe that open information is
a force for understanding and improvement.
Research companies should have no pretenses. The founder of
Arbitron, Jim Seiler, had a favorite saying. ``The three things you do
not want to see made are sausage, legislation and ratings.'' There is
also the analogy between growing mushrooms and clients. Both should be
kept in the dark and covered with manure. The humor is too close to
reality to be funny. Knowledge is power. Knowing strengths and
weaknesses is key to using data intelligently.
CONTAM
Back to the story--by 1964 the industry was working with standards
and definitions, disclosure and audits. What about the third
commitment, to improve the state-of-the-art? A good share of the early
work was dedicated to defining where we were. How good or bad were the
ratings? What could be said on that subject?
On January 15, 1964, three network executives testified before the
Congressional Committee on the newly formed CONTAM and its early work.
Through extended studies CONTAM demonstrated that sampling theory does
apply to measurement of television viewing behavior. As they said it,
``relatively small samples give good estimates of TV audience size.''
They also committed to tackling other problem areas. CONTAM was joined
in this work by a parallel local committee, COLTAM.
That was the beginning of 35 years of methodological review and
research on the quality of the Nielsen ratings. Until 1999, a
continuous expenditure of funds by a limited number of industry leaders
provided data independent of Nielsen or other syndicated services to
cross-check, evaluate and highlight strengths and weaknesses of the
audience estimates used in the marketplace.
In 1969, a Professor of Business Statistics from New York
University, Gerald Glasser, and I started a research company,
Statistical Research, better known as SRI. Gerry had been a consultant
to the networks and the media industry during the prior tumultuous
years.
Among our first projects was a series of telephone coincidental
studies to measure national audiences. The results were presented at
the 1970 ARF Annual Conference and subsequently published in two
booklets titled ``How Good Are Television Ratings (continued).'' The
work examined the effects of methods on results. The best method
yielded results remarkably similar to Nielsen on TV set usage. In this
and subsequent research we found consistent differences on persons
ratings, where we found more young viewers and fewer older viewers than
Nielsen.
In the 1970s and 1980s, many independent studies were completed.
There was a continuous flow of work. Research on the effects of
weighting and editing rules and of data gathered along with audience
estimates was done. When a new measurement technique was introduced in
Chicago, the resulting disputes were addressed through an independent
study. In 1981, industry sponsored estimates of evolving technologies
were begun. It was a top quality effort to track TV sets and related
hardware in the home. Later that work was extended to cover computer
and telephone devices.
The efficacy of product ratings was explored in 1972. Arbitron had
begun to measure the purchase of packaged goods in their TV diaries.
They produced ratings within product user categories. A widely
sponsored industry study that was reviewed by an ARF Technical
Committee demonstrated that program buying decisions based on the
product data produced inferior results to those based solely on
standard demographic ratings. It was a classic study. It demonstrated
that, in this case, more was less. The added data were not reliable and
were harmful to media buying decisions. Arbitron ceased gathering and
producing the data.
In the 1990s, projects around how people use media were added. They
were designed to look at the media through the eyes of the consumer.
Over a dozen such studies were conducted following an agenda directed
by an industry committee.
There are two ways to assess any data stream. One is independent
research such as the work I just described. The second method is to
track the internal consistency of Nielsen information. For over two
decades, SRI compiled and tracked Nielsen national audiences and sample
statistics for CONTAM. Television usage over 3 dayparts was analyzed on
five bases. The purpose was to detect unexpected variations in HUT
levels based on statistical principles. The data were deseasonalized
and compared to a long-term trend. If something jumped out of line, it
was clear to all. Rather than relying on anecdotal evidence, we had the
full story. Discussions about unexplained variations were cast in a
broad light.
Sample statistics were tracked in the same way. Statistics on
implementation of the Nielsen sample and on tabulation were inspected
on 12 different parameters. For example, ``normal levels'' were defined
for unidentified tuning, meter overflow, and set disconnects. The
abnormal was held up for possible action and resolution.
Over the years, this standard quality control warning system
identified extreme changes in usage levels. After investigating,
sometimes a cause was identified and corrective action taken. On other
occasions, the tracking was used to correct a laxity in operations that
contributed to extra variation.
Another major industry effort, the people meter review, began in
1987. The proposed introduction of people meters by AGB and Nielsen was
a landmark change in the measurement of TV audiences. There were many
questions. Both AGB and Nielsen agreed to participate in a detailed
analysis. However, before the actual work began, AGB withdrew its
proposed service from the U.S. market.
The purpose of the review was to understand and describe the issues
and to identify possible improvements in the system. The product was
seven reports including a data review, exit interviews with former
panelists, sampling and field implementation, household contacts,
processing and editing, an engineering review of the hardware and a
final report.
It asked that Nielsen act on three recommendations:
Defined procedures and quality control,
Ongoing methodological research; and
Client involvement.
Reactions to the effort were enthusiastic. It was called
exhaustive, a new standard of excellence and a blueprint for change.
Nielsen called the work ``an outstanding effort.'' It was a unique
chance for all of the industry to be on the same page.
In 1990, CONTAM published its Principles of Nationwide Television
Audience Measurement, which was in part a derivative of the earlier
People Meter Review.
To this point CONTAM and the industry had focused on checking and
assessing methods and alternatives. It was a reactive role and the
industry was frustrated that promised actions from the people meter
review had not been implemented. Therefore, beginning in the early
1990s, the posture became proactive with S-M-A-R-T. S-M-A-R-T was an
acronym for Systems for Measuring and Reporting Television.
The S-M-A-R-T story is long and involved. By mid-1999, when the
operation ceased, S-M-A-R-T had created a complete laboratory for
future measurement in the digital age. New measurement hardware was
designed and built. The Philadelphia market was the test bed. Clients
were delivered 9 months of Philadelphia audience data to their desktops
on a weekly basis. Software enabled instant analyses at no marginal
cost. In the end there were 30 telecaster, advertiser and agency
sponsors.
Since 1999, independent initiatives to understand and improve the
Nielsen systems have effectively disappeared. The measurement
challenges are greater than ever, but client managements are unwilling
to fund such work. The willingness to attend to the problems,
reactively or proactively, has disappeared. That brings me to
accountability.
Accountability
The 1963 Hearings and the associated fear of regulation was a
powerful attention getting event. The industry and Nielsen became
riveted on taking responsibility for what they did and what they sold.
Consider the industry:
Industry. In the 1950s and 1960s television was dominated by three
players. Each recognized a public responsibility for use of the public
airwaves. Stations were licensed and obligated to ascertain and serve
public interests. The networks studied the medium and its relation to
society.
Two examples. First, Frank Stanton--another of those Founding
Fathers in the ARF compendium--as CEO of CBS, gave money to Paul
Lazerfeld and the Bureau of Applied Research at Columbia to study
television audiences. In 1955, Stanton said ``we owe it to our audience
as well as to ourselves to establish some systematic method of inviting
the public to participate in shaping what we do.'' From that 1955
comment came successive studies published in 1963 under the title of,
The People Look at Television, by Gary Steiner and a later update by
Robert Bower published in 1973.
Second, NBC funded a huge longitudinal study over several years on
the effects of violence on television for children. The resulting book,
Television and Aggression by Stipp, Rubens, Milovsky and Kessler was
published in 1982.
In the 1960s, the three network television research departments
numbered about 100 people each. Then, the research departments had a
budget to do research on the ratings. From the 1960s through the 1990s,
some media entities spent heavily to understand and improve ratings
quality. The S-M-A-R-T initiative was supported largely by selected
networks to the tune of $45,000,000.
The CEO's who approved the S-M-A-R-T expenditures knew that they
were in for the long haul. Payout would be years away. It was okay if
S-M-A-R-T diluted current earnings. But CEO's change and CEO's also
change their minds. The year 1999 was financially crazed and some of
the then-current CEO's saw a greater bottom line by cutting costs and
learning to live with ratings as they were.
Today, with a fragmented industry and each network run as a profit
center, the corporate research department is likely to be ten people
with no independent budget. There is little focus on the long term. In
the 1990s, some of the most successful players sat on the financial
sidelines and reaped the benefits of independent research without
paying the price. Today, all seem to have adopted the sidelines
posture. Evidence of action to test or fulfill a public responsibility
is thin; the Wall Street accountability is clear. Research on research
does not add to today's sales; it dilutes earnings. That is the pox on
our times.
Nielsen. In the 1960s, 1970s and 1980s, Nielsen supported the
industry initiatives to improve the research product. They rightly saw
this work as buttressing the industry's stock of knowledge and
therefore helping their cause. A network research director used early
CONTAM work to tell the ANA that Nielsen data were accurate and
reliable and should be used with confidence in the television
marketplace. Nielsen cooperated with independent studies by supplying
corresponding data for comparisons and participated in analyses. When
disputes arose about methods or measurement changes, those disputes
were sometimes settled by a jointly sponsored study where Nielsen paid
half or by a summit meeting where data were presented and a resolution
planned.
Because of the scope and nature of the People Meter Review,
Statistical Research sought and received legal protection from Nielsen
against any claim arising from the conduct or reporting of the study.
After that review was completed, Nielsen refused to extend that
protection for proposed future work. That refusal and Nielsen's
resistance to the industries more in-depth involvement was one reason
S-M-A-R-T was initiated.
Another episode of inexplicable changes in television usage in 1990
and Nielsen's resistance to investigating and correcting the cause was
also a motivation behind S-M-A-R-T's launch.
The network people will tell you that while S-M-A-R-T was there,
Nielsen was its most responsive self. When S-M-A-R-T happened, a whiff
of competition stirred a reaction. Innovations were talked about and
some things actually changed. Response rates on the national panel
improved. Questions were answered promptly. And corporate contracts
across media entities that Nielsen had refused to consider suddenly
came to be. Nielsen reacted.
Today's Wall Street-thinking also influences Nielsen. Anything
Nielsen does over and above producing rating reports reduces their
earnings. That fact is a powerful deterrent to investing in
improvements.
Current Status
To conclude, I offer my perspective on where we are today. I grant
that the Nielsen task of measuring television audiences is incredibly
difficult. It is far tougher than what was done with S-M-A-R-T. S-M-A-
R-T had the advantage of working in a laboratory, of a fresh start and
of ``off-line'' innovation. Nielsen needs to innovate and introduce new
systems into a service that is valuing the assets of others 365 days
per year.
The scary realities are these. Everything Nielsen does affects the
ratings. Every change in procedures, every change in operations affects
the ratings. The perceived changes as seen through Nielsen data are
additive to any real change in behavior that may occur.
For example, when Nielsen passes over a high tech home, ratings are
changed. When Nielsen alters weighting, or persons prompting sequences,
or the meter itself, or methodology in any way, the ratings are
changed. Some clients are helped and some are hurt.
The methods changes may be planned or unplanned. When meter fault
levels increase, the ratings are changed. When Nielsen decides to
install LPM markets and changes field assignments so that home
maintenance calls are delayed, the ratings are changed. Does anyone
doubt that the male teen and young adult audiences that were lost and
reappeared a year later were due, not to real change in real viewing,
but to changes in Nielsen operations? When methods are changing and
when there are layers of changes, can the media guarantee audience
delivery to their clients? On what set of numbers should the guarantees
be based?
The first and most important recommendation from the 1987 People
Meter Review was that Nielsen institute a more careful approach to its
methodology. It called for defined procedures and quality control. That
recommendation was framed because changing methods change results and
create aberrant data patterns.
Nielsen appears to have been almost cavalier about making changes.
There has been too little attention and too few analyses to support
changes that have occurred. That fact explains, in part, the extended
controversies. Another contributor to the controversies is that Nielsen
has never picked up the quality control monitoring of its own data--or
if Nielsen has, it has not been shared with clients. With today's
technology, Nielsen should have quality control charts available
simultaneous with the release of rating reports. They should cover the
levels and trends by daypart and age groups. They should cover all
aspects of the sample. Without that type of data tracking, sellers or
buyers cannot know how to interpret today's ratings. They act on faith
and hope that the emperor is clothed.
The state of Nielsen metering technology is troubling. One problem
is that there is a potpourri of meters and meter installations. It is a
collage or patchwork quilt of methods. Nielsen bragged some years ago
that it had several hundred different ways of metering a home. That
says there are too many moving parts for the system to work
effectively. What the industry needs is one simple and effective
metering method; one meter that supports reporting of what is tuned to
the 10-second level rather than the average minute. An average minute
rating masks channel changes. The AP meter was around the corner in
1995; it has taken 10 years to get here. It seems the distance between
TV technology today and meter design is ever growing. The hole gets
deeper.
We know that many predesignated sample homes have been omitted
because the meters do not work in complex homes. Reports of high fault
rates means others are also omitted from the tabulated sample. Faults
occur because the meters are not sufficiently robust to deal with the
real world environment. Those homes that fault are a biased subset.
Faulting homes are not randomly selected.
I am omitting discussion of Nielsen reporting systems. A key value
element with data is accessibility in a friendly and affordable manner.
Nielsen clients complain bitterly that they do not have such access.
The ratings cannot be analyzed in a timely and effective way.
So with the methods changes, with the meter frailties and with the
reporting ineptitude, how much confidence should clients or society
place in today's ratings? That question is left hanging. We don't know.
Another recommendation from the People Meter review was that there
should be continuous methodological research. Independent crosschecks
are needed. Nielsen has supplied independent checks in the past. For 40
years, Nielsen had two separate systems. The sum of NSI local ratings
were compared to independently derived NTI national ratings. There was
a distinct and consistent pattern to the differences. That reassured
clients of both services. With the merging of local measurement into
the national, that crosscheck disappears.
In the early 1960s, Nielsen did a noncooperation bias study to
check independently the effects of nonresponse on NSI measurements.
The fact of Nielsen taking responsibility to check its own systems
needs to be reestablished. Some basis for assessing the degree of
confidence associated with Nielsen ratings must exist.
Conclusion
Life is filled with uncertainties. We learn to live with
uncertainty--we can curse the darkness--or we can become proactive. The
1963 hearings gave a jolt to the media audience measurement world, and
the industry was energized for decades. That energy has been
dissipated.
Television audience measurements today are shaky. There seems
little point to ranting or raving at Nielsen or engaging in public
warfare. Such actions only seem to become a distraction. The first step
to solving any problem is to face reality. My view is that the Nielsen
ship has lost its anchor and is adrift, no land in sight.
The industry posture is distressing. The work done pre 2000 was not
for charity. It was work done to protect the media business base. It
was work to provide meaning and context to the audiences being sold.
Can you imagine a network executive going to the ANA today and telling
them to use the Nielsen data with confidence? Those who funded studies
before, now say the industry should do it. What industry? There is no
media industry organization that brings together broadcast and cable to
do this fundamental work. The ARF is a possibility, but finding the
right commitment will be harder than finding the right organization.
Regaining confidence is beyond the client or Nielsen research
community. We have the talent to do better. Art Nielsen's 17 years and
$15,000,000 investment reflects thinking from a different planet. We
lack management leadership. We lack management commitment to invest, to
build and to maintain a modern state-of-the-art system. The clients and
Nielsen both must be willing to dilute current earnings to assure
future earnings.
The last recommendation from the people meter review was for
Nielsen to commit to client involvement. That meant involving clients
in setting priorities. Not everything can be done at once. Clients need
information to permit realistic assessments of the existing data. Legal
clearance is not needed to involve clients in a proactive process. We
simply need a competent and confident Nielsen management that is
willing to engage clients meaningfully and to be candid, warts and all.
The advertisers, the government, someone needs to bring Nielsen and
the clients to a new awakening. Keep in mind the goals of the hearings
and of the People Meter Review. Let's all get on the same page. We work
better when we work together.
There is one thing I feel I can say today with dead certainty. The
current direction of audience measurement cannot go on forever. The TV
currency will become weaker and weaker. It will be funny money. At some
point the wheels come off the bus.
Senator Burns. Thank you, Mr. Metzger.
And I've noticed the attendance today of Representative
Maxine Waters, from California.
And we welcome you. And if you have a statement, why, I'd
be happy to welcome you to the dais. I know you called earlier,
and you're sure welcome here today as we go on with the
questioning.
From the testimony we gathered at the table--there was a
series of questions that came up as we put this hearing
together. There has been controversy with the MRC. What role
does it play? How does it play that role? And do you give it
more power, or teeth, or do you take away some of its powers?
To us, who have to look at problems in the industry, we have to
have someplace to go in which to base decisions, policy
decisions. MRC is the only one we have on this particular
issue. So, I would ask the question that everyone, I guess,
would ask today, If we have nowhere to go for our information
but to the MRC, then where do we go for accurate information
and the information that we need in order to bring some
accountability to the industry that depends on advertising
dollars to survive?
Mr. Mullen, would you like to start with that? And then
I'll ask Ms. Whiting if she'd like to respond to that.
Mr. Mullen. Well, I think the question hits the issue dead
on. The MRC is an organization, historically, that we've been
able to rely upon for independent verification of the processes
that Nielsen uses to gather ratings data. The MRC's charge is
to make sure that the measurement services are valid, reliable,
and effective. If Nielsen had, in fact, followed these
processes voluntarily, we wouldn't be here today with the
problem that we have. The testimony--written testimony that I
have submitted shows a pattern and practice of activities on
Nielsen's part, largely ignoring the concerns and requests of
clients and the actions of the MRC, and rolling out new
technology without accreditation, despite the concerns that we
have clearly pointed out and they have yet to provide any
answers to.
Senator Burns. Ms. Whiting, would you like to respond?
Ms. Whiting. Yes, Senator Burns, I would.
[Laughter.]
Senator Burns. I thought you might.
Ms. Whiting. I'm sure you're not surprised.
First of all, we are committed--let's just go back to
something that we were asked publicly to commit to--we are
committed to the MRC process. I've made that statement hundreds
of times. We are also committed to adopting the voluntary code
of conduct, as long as other research companies, as well, have
reviewed the issues. So, start there.
Second, we've taken every major service we have and
voluntarily applied for accreditation. The issue here, I think,
is about the timing of the accreditation process. And we have,
in fact, applied for accreditation in each of the people meter
markets. You know that some of them are accredited, that the
rest is an ongoing process. But since it is a voluntary
process, and since the process that we participate in does
require an audit, which we are doing, and have done, in each
market, I feel we have, in fact, been cooperating with that
process.
And as to the point of plowing ahead without time, I just
would remind you, we have many clients. We have thousands of
clients. In any given market, we have buyers and sellers. This
all started because advertisers asked us to take the same
technology that we had looked at nationally, in the National
People Meter Service, and bring daily reporting of audience
demographics to the top markets. And that process then began a
conversation with many clients.
This didn't happen overnight. It wasn't pushed down
anyone's throat. There are, as you've seen from a number of the
letters that you've received, supporters on other sides of this
issue. And I think, ultimately, what we were trying to do is
balance buyers' and sellers' requests for improved information,
and the timing sometimes isn't to everyone's liking. But, in
fact, we delayed every major rollout of every people meter
market based on conversations with clients.
Senator Burns. Mr. Ivie, would you like to respond to that,
please?
Mr. Ivie. Yes, sir, I would. Yes, Senator Burns.
I want to try to----
Senator Burns. Pull up the microphone.
Mr. Ivie. Sorry.
Senator Burns. Thank you.
Mr. Ivie. I want to try to give you maybe a little bit more
background. And I know this hearing isn't about LPM,
specifically, but that has been a topic of why we're here, kind
of, what led us here. So, there are a couple of truths on both
sides of this equation that I think you need to know about.
The first truth is that the LPM system, we believe, and
also across a broad spectrum of the industry it's believed,
that the LPM system is, indeed, more accurate than the system
it is replacing. And it has been audited and verified pretty
strongly that it works pretty well. So, when we audited these
LPM systems, at first, we saw that there was a mix of
performance in these new markets. We saw that Nielsen did
certain things very well, and they did certain things not so
well when they implemented this market. But yet our Television
Committee, made up of some 65 organizations from across the
industry, decided that we needed to really do something special
here. We needed to recognize to the marketplace that this was a
superior type of measurement, but we also needed to tell the
marketplace that there were problems with the LPM system that
still needed to be addressed. So, we have this status, and
several of the markets continue in this status, called
``conditional accreditation,'' where we said, ``Nielsen, you
can roll this product out, we believe it's responsible for you
to roll this product out, but we want to recognize that this
product isn't done yet, that there needed to be some very
intense efforts to correct certain things,'' and, for instance,
Mr. Mullen's testimony referenced the faulting, which is one
area that we've been concentrating on with Nielsen. But there
are others.
Now, the other truth on the side of this LPM is that
Nielsen has, in fact, rolled these markets out before audits
have taken place. That is a problem. That's a problem that,
sitting in my seat, I would never like to see happen again. I
don't write bills, I don't do legislation. I wouldn't want to
substitute my experience for your experience in that regard,
but what I can tell you is, we've said that several things need
to be addressed. One is that we don't want it to happen again
that the marketplace is uninformed about the quality of a
product when it's rolled out. Nielsen should commit to that,
and we want that commitment to hold in the future.
We also think that our voluntary code of conduct will help
more clearly define how we interact, make sure Nielsen and
other rating services--because this isn't just about Nielsen--
react to audit findings promptly. And I think the communication
linkage that we're requesting will make sure that if there are
issues like this in the future, if this process has a problem 2
years from now, when your attention is elsewhere, for example,
we can come back to the Executive Branch or Congress and say,
``Hey, we have a problem. We have a rating service that's not
listening to what we're saying.''
So, that's why these two truths need to be considered, that
not everyone here--they're all telling you the truth, probably,
but the idea is that there are multi-sides to this equation.
Senator Burns. Would you like to respond to that, Ms.
Whiting? This is the way I learn things--you put all the
stakeholders at the table, and then we start this dialogue. And
this is very good for me, to be real honest with you. OK?
Ms. Whiting. Thank you, Senator Burns.
I think I said earlier prior to--in fact, I believe it's
true that prior to the introduction of people meters in
individual markets, the process for audits and accreditation
had been that a service would be up and operating--for
instance, we have many metered markets. We have meters and
diaries. The service would be launched, an audit would then be
performed on a live market. That was our procedure for those
kind of markets. When we went to the people meter market, as
Mr. Ivie has described, I think there was a much heightened
sense of review by our clients, in spite of the fact that we
had operated a people meter service for many, many years, about
that service. And we did have people ask us to do audits before
the service, and we replied that the process would have to be
an audit upon a live service. In spite of that, what I've just
indicated earlier, the voluntary code of conduct would propose
that we would do audits before a service was offered. And we've
already committed to doing that to the MRC regardless of
whether the voluntary code of conduct proceeds. So, I think
this issue should be one that we can say we agree upon.
And the other point really is important, which is, we have
many clients. I keep saying this, but the other side of the
house, the advertisers, who want information more quickly, have
been talking about this for years. And their urgency is not
reflected in some of these comments.
Senator Burns. Ms. Crawford?
Ms. Crawford. Thank you, Senator.
I would like to say that, as an advertiser, or an agency
representing advertisers, we almost screamed our way to getting
LPMs in a--faster and faster and faster, because the service
had been available nationally, since 1987, I believe. I believe
that's the date.
All that being said, I think that it bears reminding of
everybody that the LPM, from a fault-rate standpoint--and if
you can use the term ``fault rate'' as it relates to the meter
diary, which is not quite the right terminology, but it's
close--the fault rates in the LPM are better, if there is such
a way to say that, than they are in the meter diary. And yet,
we never came in front of your Committee when the meter/diary
markets were out there. And I would have to ask if that's
because the ratings are higher in the meter diary; and so, they
were acceptable. Now we're here, and the ratings are lower, and
we're now really talking about the diversity of viewership in
all of these 130 channels that the American public is viewing
every single day that is now measurable by an LPM technology
that was not measurable in the meter diary technology, never
mind the diary-only, both of which, by the way, are accredited
by the MRC.
Senator Burns. Tell me about fault rates. Define ``fault
rates'' for those of us who do not understand the term.
Ms. Crawford. Fault rates, in the world of the LPM--I don't
know why I'm the one who's doing that, but----
[Laughter.]
Ms. Crawford. George can do that. I actually--I do know
what they mean, but I'd be a lot happier if George did that.
Senator Burns. OK. Mr. Ivie?
Mr. Ivie. Yes, I'll try to--I want to do two things, if
it's OK. I want to----
Senator Burns. OK.
Mr. Ivie. I'll talk about ``fault rates,'' and define
those, but I also--Ms. Crawford raised the question about the
consistency of our treatment between diary markets and LPM, so
I wanted to spend a minute on that. But let's talk about fault
rates first.
Fault rate is really the situation where the household
that's participating in Nielsen's panel is not interacting
properly, or as designed, with the Nielsen meter somehow. So,
it could be a hardware thing. The household could have
unplugged the meter or moved the meter or it's an electric
problem in the household, or they could--in the case of a
people-meter household, they could be not interacting properly
with the mechanism by which they record their presence in the
room when the television is on. So, if the television is on,
and nobody has entered themselves as a viewer, and they've
changed a channel on the set or something like that, that meter
is smart enough to recognize, ``Hey, there's got to be somebody
there. No one has told me that they're there.'' So, that data
is of suspect quality.
And Nielsen has controls that say, ``If we have enough of
these conditions, where there's suspect quality, we should
remove that house from the ratings for the day.'' And for most
of these faults, it's just removed for a day. It has the
ability to be in the sample the next day, and that condition
would have to reoccur the next day for the household to be
removed. So, when Mr. Mullen said--and I don't have those
statistics in my head, but he said in his testimony that a
third of the African American--or almost a third of the African
American and Hispanic households were not reporting on the
average day, that means that they faulted or didn't interact
properly. Actually, our more recent experience is less than
that, because these fault rates have been getting better as
initiatives have been taking place. But that faulting is a
difficult problem, because you have to coach the households on
how to interact with the meter and not unplug it. But that's
what ``faulting'' is, when they're not interacting properly.
One thing that I want to say, to talk about equality of
treatment, is, when people fault, they're almost certainly
interacting with the television set. It could be that they're
just moving the set, but a lot of times, most times, it's that
they're watching television and they're doing something to the
television that creates the fault. So, we know that people that
fault a lot tend to be watching during that time period. So
the--and you can tell. If you look at households that fault a
lot, they use television more than households that don't fault
a lot. And I don't think anybody would dispute that. We've seen
that hundreds, thousands of times.
That's why there's a difference in treatment from the MRC.
We know that in an LPM household, and in these people meter
households, faulting is directly correlated with when people
are using the television. In the diary service, which I'm not
going to sit in front of you and say is perfect--we have
initiatives to try to work with Nielsen to correct that--we
know that that is not directly linked to viewing, necessarily.
It's linked to other things. ``I'm too busy to complete a
diary. Maybe I can't read or write.'' It's linked to a lot of
other things--literacy, maybe I'm used to working on the
Internet, I'm not used to writing a diary, et cetera.
So, there are a lot of things that come into play, but not
necessarily as direct of a tie to television viewing, as we see
in this LPM. And this stuff needs to be studied. But there is a
difference in treatment, because there is a difference in the
problem. Every problem needs a custom solution, and that's why
we're focusing in two different ways on these services.
Senator Burns. I'm going to go to--I'm going to go to Mr.
Metzger, then Mr. Mullen, and----
Mr. Metzger. Let me describe ``faults'' for you in a
simpler way. It's going to--some households, it's not
processed, something's wrong. That's a fault. Nielsen
designates a sample, that should be measured, to count
everyone. Some of those homes cannot be installed because the
meters will not accommodate modern digital delivery systems and
recording systems. So, you miss those homes. Then you have
homes that fault, so you miss those homes. Now, both of those
homes are not randomly selected, they're particular kinds of
homes. In the case of the faults, we know those occur much more
where there are more sets or more people. Interestingly, in
White households, 11 percent of White households have 5 or more
people. Twenty percent of Black households have 5 or more
people. And almost 30 percent of Hispanic households have 5 or
more people. So, that's the reason minorities should be very
concerned about these systems, whether it's the meter diary or
the A/P meter. They both have the same problem. And part of the
reason those faults are so high and why we're skipping homes is
that the metering technology is still based in the past, it is
not up to snuff with modern technology. You can't expect
people--up to a certain limit--you can put a certain task on
people--you have to make it easy for them. And the task for
them now is too difficult. That's the problem.
Senator Burns. Let me offer an opinion here, because I
don't want to get too far astray on people meters and this type
thing. That's a technology that we can take up in the MRC, and
to say why we can't make accreditation here, or recommend
accreditation, or whatever, I think that's for the industry and
the MRC to work out. And if they find some problems with it,
then I think it's incumbent, or should be incumbent, on the
sampling company, such as Nielsen, to try to work on those
fault rates and to make them whole. That's the real purpose of
this hearing, to be right honest with you. I can't judge, here,
which is best. That's not what this is about. This is about
getting a fairness with the--and doing business with the only
game in town, and how we get these people together.
Mr. Mullen, you want to respond?
Mr. Mullen. Yes, thank you, Mr. Chairman.
The point of fault rates, in bringing that up, is not--we
could debate that all day long. As I say, you can do anything
with numbers, and that's not our purpose. In fact, that's why
the MRC exists, so that we can all share and analyze that data
to determine if Nielsen is, in fact, delivering the product
that they have sold to us in the marketplace.
The point that Mr. Metzger makes is absolutely right. It's
the key audiences of homes with five or more people which tend
to over-index--and in African-American homes and in Hispanic
homes--that are also our key audiences, skew into the younger
demographics, and young homes, that tend to fault on a more
regular basis. When I referenced one-third of African-American
men and Hispanic men in New York in a given week, that is a
year after they've rolled out the service. They've shown no
ability, on a consistent basis, to improve that. That's what
the MRC is looking at. That's why the MRC has not given full
accreditation to the service in New York. And that's our only
concern, is, don't commercialize this product in additional
markets like Washington, D.C., and Philadelphia, until you've
demonstrated to us that you have the ability to measure these
audiences properly in the Nation's largest markets.
Senator Burns. Ms. Whiting?
Ms. Whiting. Yes, please.
I think, because we are talking about broad quality here,
one of the measures of quality is fault rates. But it's not the
only one. It's the sample size. It's the population that you're
measuring. It's the characteristics and the quality. It's the--
it is the technology, which allows you to look at the breadth
of channels, 24 hours a day, 365 days a year. It's the
cooperation rate. And we are lucky, actually, with people
meters, to be able to have the technology that allows us to
know when someone's not doing what we would like them to be
doing, or when a TV set is moved or unplugged. And that has to
be looked at in a relative way. On every broad measure, our
people meter service is better than the service it has
replaced. And there are always going to be issues with
individual days and individual segments. And that's true of
polls, and that's true of any measure and any statistical
sample you take.
But I think, to Ms. Crawford's point, we do need to look at
the fact that we have diary services in some markets, meter
diary service in others, and then people meter service. And
this is an improvement over the current system, and one of many
things we have to continue to improve, and we do that in every
one of our services. They have to have a commitment to
improvement. So, I think it's one of many measures, and it's
really somewhat disingenuous to point out one aspect of what is
a greatly improved system that allows us to report many, many
channels every day with more people of color, with higher
cooperation rates, and with better technology.
And I'd like to add that we do not, anymore, with our A/P
meter, and with the recent addition of our ability to measure
digital video recorders, bypass homes for technology reasons,
which is something Mr. Metzger said.
Senator Burns. Well, I want to--I'm going to throw out
another question here. When you go out and you have to locate
your, whatever it is, a diary or the people meter, you knock on
the door, and you said, ``Would you like to be, or could we ask
you to be, a person that would do this sampling for us?'' how
many times are you turned down? I mean, do you have a turn-down
factor that people--I mean, I'd like to know, when you knock on
the door, who says yes and who says no and--because I know it
has to be a voluntary program, and there has to be consent. I'd
like to know, have you got any kind of a rate on that?
Ms. Whiting. Of course we do. One of the things we have to
report in our different reports, whether they be a diary report
or a meter diary or the people meter, are cooperation rates.
And that's what I think you're asking.
Senator Burns. Yes, I think that's what I'm getting at.
Ms. Whiting. And they're different for different markets,
and they're different for different services, and they're
reported, if it's a daily service, every day, if it's weekly,
every week. In that local people meter service, the average
cooperation rate is about 40 percent, which means 60 percent of
the people turn you down. But if you think about polling and
marketing research, that's actually very high.
Ms. Shagrin. And one of the problems is that the people
that it's hardest to get to say, ``Yes, I'll be in your
sample,'' are the same people that it's hardest to keep giving
you reliable data. So, the point is, 40 percent is good. Ninety
percent giving you usable data is good. But if you're totally
excluding, or almost excluding, certain types of households,
then the resulting audience information is flawed. And I think
the focus on the MRC--at the MRC has been, is the audience--are
the audience estimates biased? Is there a flaw there? And is it
fixable? And I think the pressure has been to say, we think
it's fixable. We think it impacts the quality of the data and
the reliability of the data and causes some differences from
day to day. Because if 20 young men in New York are providing
usable data on Monday, and 20 fewer on Tuesday, your ratings
for adults 18 to 24, or males 18 to 34, change those 2 days, it
has nothing to do with real change. So, we've been focusing, at
the MRC level, on, what are the things that we see in these new
markets that may be causing a bias? What are we pointing out to
Nielsen? And this is why I believe the MRC drives consistent
improvement. What needs to be fixed? And then constantly
reviewing that to say, ``OK, now it is a level that we are very
comfortable with.'' And, as George said earlier, the
conditional accreditation was a very new thing for the MRC, but
the MRC membership agreed to it, because they wanted to send a
clear message that there were a lot of things that were better,
but there were also a lot of things that weren't better and
needed to be addressed.
Senator Burns. OK. Mr. Mullen, you've shown some interest
in this statement? Do you want to respond to that?
Mr. Mullen. No, I certainly agree that the process of
accreditation has been slowed for the LPM, but I think it's
largely because MRC is not yet comfortable that Nielsen is
delivering on the product that it has designed.
Senator Burns. Mr. Ivie?
Mr. Ivie. Yes, I just want to add two points of
clarification. Ms. Whiting quoted a response-rate average of 40
percent. That varies widely sometimes between markets. It could
be higher than 40 percent, or it could be lower. For example,
one of the markets in play that tends to be lower is New York.
That rate is in, sort of, the mid-1920s, is my recollection, in
terms of response rate. So, those response rates vary a lot. If
you go to smaller markets or more rural markets, for example,
those response rates tend to be higher. More urban markets,
where it might be harder to get mail delivered, et cetera,
those response rates can be much more difficult and lower.
I just want to spend a second on--the focus of the MRC is
not--is--when we conduct audits, we look at literally hundreds
of performance areas and standards areas when we look at a
product. Faulting has been discussed here today, but we look at
response rates, sample distribution, you know, the procedures
in the field to install households. And, a lot of times, when
we don't comment on things, it's because they're working good.
And we saw, when we audited these LPMs, that there was a lot
working good. And faulting is one thorn that remains. And
faulting has improved somewhat in some of the markets, but
it's--I mean, those markets remain conditional, because we
still don't believe that faulting performance area meets our
expectations.
Senator Burns. OK. I've got a new question. Let's talk
about the legislation just for a second. Some have said that
this bill would raise barriers to entry. And I know it can cost
a lot of money, $100 million or more in development costs, and
several years to roll out new measurement services. But MRC
audits can cost $100,000 and take a few months. That's 1 or 2
percent, in money terms, and not much more in the terms of
time. I don't see that as a barrier. Tell me if I'm wrong.
Ms. Whiting. Senator Burns, may----
Senator Burns. Yes.
Ms. Whiting.--I reply?
I think the cost is not so much the issue; it's the time.
And it's that there is no clear knowledge, the way the bill is
drafted, of the amount of time that it would take to accredit a
service. And so, as I understand the bill to be written, if you
have to have every change to an accredit service accredited--
and I think that is the way it's written today--before you can
actually offer it, then I do believe, if you are creating a
rating service or putting together innovation or testing a new
technology you want to deploy, you do not know when that
service will be accredited. You have clients who are asking for
it, you have contracts, you have a process, but you could
easily spend a year. And we have had cases where the
accreditation process has taken a year, or more. And if that's
the case, this would not allow any business that I know of to
put that process in place, and I think it would also give pause
to anyone entering the business, if you did not know when you
could actually commercialize your service.
Senator Burns. Mr. Ivie?
Mr. Ivie. I just----
Senator Burns. Everybody should have opinion on this one.
Mr. Ivie. Yes, I----
Senator Burns. But I might----
Mr. Ivie. Just one point of clarification. The audit
process, itself, never--I mean, in my history with the MRC, the
audit process has never taken a year. What takes a year is if
you conduct an audit and you find some issues within the audit,
and ultimately those have to be corrected by the measurement
service. They have to install new initiatives, make changes,
change their sampling, whatever it may be. Those procedures can
take a long time. Then you have to verify that they've been
done. And then, ultimately, you get the decision that it
completely meets the standards, and it becomes accredited. So,
it is valid that we've done audits that have extended long
periods of time, but it's not because of the MRC executing an
audit for a year. Audits don't take a year. Audits take a
month, or they might take something like that. But correcting
issues that are found in audits--and this is the real critical
component--MRC's about process improvement. Improving those
processes can take a long period of time. We have a rating
service that we audit that has been in the audit process,
without full accreditation, for almost a decade, because
they're trying--and this isn't Nielsen; I should clarify that--
because they have certain aspects of their product that do not
meet our standards, and they're figuring out, as they go along,
you know, how to correct that product. And that has taken a
decade, because these things aren't easy. The cooperation rates
and things, trying to get people to cooperate with data-
intensive research--you're trying to gather all sorts of
information about what soap they use or it could be anything.
This stuff is hard to gather, and these process improvements
are very difficult to make. But that is how the industry is
benefited. That product doesn't have accreditation. And what
happens is, they don't bear our logo, the users know that there
are issues. And that----
Senator Burns. Ms. Shagrin?
Ms. Shagrin. If the MRC and the audit process finds a
significant problem, and they go back to the ratings company,
whichever ratings company it is, and says, ``We cannot take a
vote, we cannot give you accreditation because we found this
problem,'' that's how it's protecting the industry. That's how
it's constantly improving the ratings, themselves--and I'm not
limiting it just to television--and getting the best-possible
estimates for this industry to continue to make good decisions
and to grow.
Senator Burns. Now let's hear from Mr. Mullen first, and
then, Ms. Whiting, we'll have all of this accumulated, and you
can take your shot. Yes? And then I'll go to Mr. Metzger.
Mr. Mullen. Yes, Mr. Chairman.
We, first, as a company--and I cannot speak for the entire
broadcast industry, though I think many would feel exactly the
same way--accept accreditation. If a market is accredited, we
accept the ratings, without question, and we compete with those
ratings. The New York market, as an example, has been audited.
There have been problems that have been shown there,
significant problems pointed out to Nielsen by the MRC. Nielsen
has--I give them credit--tried to fix those existing problems.
But, month after month, they have not shown significant
improvement, and, with that, the MRC board, upon evaluating
what they've been doing, has not accredited the market. They've
shown the same problem in other markets. Our largest objection
was, with that type of a track history, they are still rolling
out additional markets without audit and without accreditation.
We have no confidence in the quality of the service and the
ratings that they're providing us. And that does influence the
type of programming that we run and the type of programming
that we will run in the future.
Senator Burns. Ms. Whiting?
Ms. Whiting. Well, I think this discussion actually is a
very good example of the different points of view on why
accreditation should be mandatory or voluntary, and when it
should occur, because we have many clients--and I keep going
back to this--who would say that they're disadvantaged unless
they have the better system that a people meter offers, or a
meter/diary system, or something else. And their point of view
would be that they want this in the commercial world faster.
And that's one reason I believe that we have to look at the
voluntary code of conduct, we have to look at the ability to
have a faster way to market for big changes, and we have to use
the MRC process and not make it mandatory, because if
accreditation--not an audit--can take a year or two, or who
knows how long, what happens to all of the clients who are
disadvantaged in that system?
Senator Burns. Well, let me say, I agree with you in some
parts of your statement, but if voluntary doesn't work----
Ms. Whiting. Well, I think we may disagree over whether it
doesn't work.
Senator Burns. Well, but I'm saying--in some areas, I think
it does, and that's my--I come out of the industry. I like the
idea of--that we put up some sort of a situation where we can
solve our own problems between the broadcaster and the
samplers. But if the voluntary system breaks down, then where
do we go when there's only one company out there? I guess
that's what we look at. We really weren't elected by a
constituency to oversee a monopoly, an unregulated monopoly.
That's what, kind of, causes us concern, because that red flag
goes up, and sometimes we do things that have unintended
consequences, and we don't want to do that in this case,
because I love this industry. And so, once there's a breakdown
or somebody gets up on the wrong side of the breakfast table
one morning and decides not to cooperate----
I want to hear from Mr. Metzger, please. And then I'll come
back to you. You'll get your say. I'll not leave you out.
Mr. Metzger. I said in my testimony that I thought the MRC
aided innovation and competition. The reason I say that is from
my own experience. First, in the context of our RADAR service,
for example, I mean, the MRC helps all of us track where the
fault lines are--not just faults, but any issues we have--and,
because of that industry inspection, it brings a pressure to do
things better. And with our RADAR service--I think it was in
1999, we were looking--our response rates were dropping, also,
but other services--we had experimented on how to get a better
response, and we structured an experiment and went to the MRC
to discuss that experiment before we did it, and got their--to
buy into the process and the sample sizes and so forth. Matter
of fact, at that time, there was a minority network that was
not a part of the MRC, and I insisted they be invited to those
meetings to discuss that. We did the experiment, got it fully
accepted and implemented in a very fast time--in part, because
we had the industry exposure through the MRC to help the
communications process.
And with regard to competition, Michael Porter, of Harvard,
is famous for saying, ``The way to get a good product is to
create value and signal value.'' With our SMART Laboratory, we
created, I think, a better product, clearly. Now, how do you
signal that? Had we got funding, I would have been breaking the
door down of the MRC to get in there and have them work with
us, just like we did with that experiment on improving response
rates, as we rolled out the service. Rolling out the service
was going to take 1 to 2 years. And then, parallel with that,
you can bring the MRC in--I mean, how can I compete with--
Nielsen has 200 salespeople, I've got three or four people out
there. I can go to that one place that the industry is looking
at, work with them in a positive way, and it would help me
break into the market faster, rather than impede me.
Ms. Crawford. So----
Senator Burns. Ms. Crawford?
Ms. Crawford.--what we have here is the fact that the
industry, itself, made the decision not to support a second
rating service. It was the industry that did this. It was not
Nielsen or anybody else that made that decision. It was the
broadcasters, the advertisers, the agencies, and so forth. So,
the bottom line is, is that it's the industry that made that
decision. That's number one.
Number two, I would like to remind everybody that we have
the advertiser, who is paying the ultimate price here. It is
paying for Nielsen's service, or anybody else's, no matter what
the medium is, in the cost-per-spot, as you very well know. As
a result, these advertisers have been waiting 15 years in the
local marketplaces to have a service that told them the next
day how well their spot that ran performed the night before.
This has been a long time in coming. The fault rates are better
in the LPM than they are in the meter/diary markets. And yet,
we're sitting at this table over the LPM. And I'm very
concerned that we are talking about this when we have
accreditation on the meter/diary side and we don't have
accreditation on the LPM side.
Senator Burns. Mr. Mullen?
Mr. Mullen. Mr. Chairman, I would say that when you talk
about paying the ultimate price, ultimately, I think that is
the weight borne by the stations. Not only do we pay the vast
majority of the fees for Nielsen's local services--and Susan
could tell, maybe, better than I, what that percent is, but
certainly a vast majority of it--how our ratings are reported,
based upon that ratings system, ultimately determines our
rates. And if it's not an accurate system, we pay the ultimate
price.
And to the point of accreditation, I know right now that
Arbitron is working with the MRC for accreditation. They are
submitting to the audits and seeking accreditation for a new
service in the Houston market. So, I do not believe that the
ability to go through MRC for accreditation would be a
hindrance for entry into this business at all.
Senator Burns. Ms. Whiting?
Ms. Whiting. Yes, Senator Burns.
Senator Burns. Then I'll turn to Mr. Ivie.
Ms. Whiting. I think stepping back a bit--you asked, what
do we do when we're here at this place, and what could you
expect us to do? And over the last year and a half, starting
with every piece of recommendation that our independent task
force made in a wonderful group of people who worked 9 months
voluntarily, to put together a review of our procedures, made a
set of recommendations that we actually have implemented, are
in the process of doing. In the written testimony I submitted,
there are pages of quality improvements that we've made in the
Local People Meter service voluntarily. And they range--
everything from staffing additions, like 50 people in
significant markets, to incentives, to different coaching
procedures. There have been many, many focused voluntary
improvements in the last year and a half. And you might say,
well, that's because there has been so much attention on this
issue. Of course that's part of it. But it's not all of it. And
I think if you look back at the last 5 years, particularly, of
investment, we have independently, voluntarily invested a lot
of money to improve our service. And I think that speaks to why
we believe that a voluntary code of conduct would allow us the
flexibility to move improvements in the service forward as long
as the rating services--and we are--commit to that fact, that
there have to be audits before the service is commercial.
And I agree with Mr. Mullen that Arbitron is, in fact,
working with the MRC. So are we on many of the things we do
before we do them. But that does not guarantee accreditation.
You can have a review of the process, and if the accreditation
is mandatory, it's a different issue.
Senator Burns. Mr. Ivie?
Mr. Ivie. I just want to address two things. Ms. Whiting
mentioned our voluntary code of conduct again. Something I
didn't communicate to you is, we are working on that with some
urgency. We expect to have sign-off on that voluntary code by
October 15. That's a goal that we've set for ourselves to work
with rating services and our members to have that code
completed and prepared for adoption by the rating services.
The other thing I would say is, all this testimony sounds
nice. I come back to one fact. There were at least three LPM
markets implemented, commercialized, without having an audit.
And Ms. Whiting has agreed to that in front of you, that that
won't happen again. The voluntary code of conduct requires--you
know, states a clear preference, as a voluntary code does, that
that shouldn't happen. Nielsen has a number of products that
are not audited that we believe should be audited, and there
needs to be dialogue about that. But that is really where our
focus is, because that's where we provide value and provide our
service to the marketplace--doing these audits, making sure
Nielsen, with all of the marketplace power that has been
discussed here today, complies with the MRC audit process.
That's where my focus is.
Senator Burns. Mr. Ivie, I've got a question, and I think
this is key to what this hearing is about today, too, also.
There's little or no competition in the field with Nielsen. I
want to start an organization, a research organization. I come
to you, and I present to you a methodology on how we're going
to do it, my technology that I'm using. You look at it, you'd
say, ``I'd like to see a pilot program.'' I facilitate that.
Can I get accreditation from the MRC with a vote after they've
all looked at it, and can I compete in a market with the
Nielsen folks?
Mr. Ivie. Well, you just described a very interactive
scenario. You come to the MRC when you are, sort of, ``baking''
the product. You're designing your product, you come to us, and
you bounce it off of us.
Senator Burns. No, I've got----
Mr. Ivie. We do that.
Senator Burns. Yes.
Mr. Ivie. You then develop a prototype of your meter, you
bring it to us, we can test that in a laboratory. We can go
very far in an audit to conduct that audit and get a lot of
assurance for our members prior to implementation of the first
respondent. But, ultimately, it is a fact that we don't
accredit any research--this is one hallmark of the MRC--until
we see that in operation. So----
Senator Burns. Well, I say I will pay the bill, I'll give
you--you tell me what kind of a pilot program you want me to
run----
Mr. Ivie. Right.
Senator Burns.--and I do that, and I submit it to you.
Mr. Ivie. We have no barriers that would stop you from
coming to us to accredit. We audit all comers. We've never
turned down an audit. If somebody came to us--I mean, one of
the things--I have a stack of legal opinions about two things.
One is, the MRC never discusses anything financial in any of
our meetings; otherwise, we're colluding against the rating
service, financially. That's strictly prohibited. The other
thing I have a stack of legal opinions about historically is,
we cannot turn down audits. We're an equal-access organization.
We don't turn down an audit. You could come to us, we would
execute that audit. And, guess what? The audit timing has
nothing to do with the membership. So, no member can delay an
audit. That's done by the staff, the independent staff of the
MRC, who doesn't work for any broadcaster, cable-caster, or
advertising agency. We work with the rating service. We
establish the audit schedule and timing and when the audit will
be conducted. So, that--the audit process really--there is no
issue of whether that could be delayed, because the audit
process, itself, is controlled by the staff. Now, ultimately,
when that audit is done, we do convene our members to review
that audit, read the audit, go through the pain of meeting with
the independent CPAs that conduct the audit in getting all the
findings. That's when the accreditation decision comes into
play. It either makes it or it doesn't. And if it doesn't,
there's a reason, and that has to be improved. But that's how
the process works. We don't turn any organizations----
Senator Burns. Mr. Metzger?
Mr. Metzger. I think it should be understood that the
accreditation process is not any kind of mystery. Again, I was
audited for 30 years and accredited for 30 years. And if you do
things up to snuff, there's no ambiguity about this, it's
accredited.
Ms. Shagrin. Could I add that the--your scenario, if you
had the pilot audited, your data-collection system and tool
could be accredited, your methodology could be accredited. But
until you rolled it out into a real sample, your sample could
not be accredited, and, therefore, the service would not be
accredited. But going into rollout, you would know that,
assuming you've got a representative sample, that you would be
accredited, because pieces of it would have already passed the
test. But without the sample, without being able to audit and
accredit the sample, the service could not be accredited.
Senator Burns. OK. Mr. Mullen?
Mr. Mullen. Senator, in Ms. Whiting's written testimony and
in her statements today, she talked about Nielsen's willingness
to talk further about the voluntary code of conduct, but in her
written testimony she stated it was applicable to future
commercial ratings, said nothing about existing, and also said
that that would be done with some give-and-take, which we read
to say there are things Nielsen needs to see changed in the
MRC's proposed voluntary code of conduct. That certainly
concerns me.
And then, to take it a step further, again, this is not a
discussion about lower ratings. We have tried for years to work
with Nielsen behind the scenes, professionally, in meetings
face-to-face. Susan and I have been meeting face to face on the
LPM issue now for probably more than 3 years. But we are
dealing with a monopoly that, in my opinion, is acting as one.
Either we allow Nielsen that monopoly to determine the
specifications for their own service, or we depend upon the MRC
for those specifications and oversight. And I think the bill
that you have introduced, S. 1372, helps us accomplish that.
Senator Burns. Well, I'm just going to make a comment now.
I'm going to have to get some more information from the MRC.
We're going to have to sit down and visit about it. Because if
I go out of here and I invest in the technology, and then--and
give them a pilot program any place in the country they want to
take it and test it before I go into the field, before I make
the big investment--because I'd hate to go out here and say,
``OK, now I'm going to sample one area, one ADI,'' and all at
once--and not get accreditation on it. That's a big chance that
I've taken. In other words, there has to be some way that I
know that I'm on solid ground, that I've crossed all the
``t's'' and dotted all the ``i's'' and done everything in
methodology and technology that would merit your accreditation.
Mr. Ivie. One thing that I just should mention. There is
something called a ``pre-audit.'' A pre-audit is--first of all,
it's kind of a misnomer, it's not an audit, but what it is, is
a brief review of a rating service, where we engage--we and a
measurement service that approaches us engages the CPA firm to
go out and walk through the methodology. They don't conduct an
audit, but how does the sampling work, how does the equipment
work, how does the reporting work, all the different aspects
that we generally study. This is very, very cost effective. It
takes about 2 weeks to actually execute a pre-audit. But that
is a service that we offer. And the end product of that service
is a letter that the CPAs produce. Now, these are the same CPAs
that conduct the audit, eventually, in the future.
Senator Burns. OK.
Mr. Ivie. That letter is sent to me, the Director of the
MRC, and the rating service. And that letter says something
like, ``We didn't conduct an audit. All we did was do a walk-
through, but had we been auditing you, these are the things
that we noticed in your process, in your technology, that are
accreditation problems.'' And they communicate that in advance
to the rating service. So, if--you've not spent a penny on a
regular audit yet, you've just had this walk-through, and you
have a list, and I have a list, of the significant deficiencies
that you have. This process is designed just for the exact
thing you were asking a question about. That pre-audit exists,
and it's there.
Senator Burns. OK. Yes, ma'am?
Ms. Whiting. I'd just like to reply to Mr. Mullen. We have
committed publicly to an audit before commercialization of the
next three people-meter markets, and that is outside of the
voluntary code of conduct, for Dallas, Detroit, and Atlanta.
And so, I do think I understand that it's important, and we
have committed to doing that. But, back to the point we were
just discussing, this conversation is why I am concerned about
a mandatory accreditation versus a very strong voluntary code
of conduct, because I think you--here, you could do all the
right steps, and I hope everybody would, in a new service being
offered ahead of time, but you actually, until the final
service or a real sample is up and operating, cannot get full
accreditation. And I think that is an issue for people starting
new services or improving services.
Senator Burns. Mr. Metzger, please. I'll come back to Mr.
Mullen.
Mr. Metzger. Let me first comment, I think it's clear to me
today that your introduction of this legislation has aided the
process of the voluntary code of conduct a great deal.
[Laughter.]
Mr. Metzger. I doubt we'd be this far along at this point.
But, to your specific question about investing and creating a
new meter and taking the jeopardy of doing that, again, the MRC
is not some arbitrary thing sitting out on some other planet.
Senator Burns. Yes.
Mr. Metzger. It is an industry organization. The industry
wants competition. And, in 1997--late 1997, early 1998--we had
our meter up and operating, and Nielsen announced the A/P
meter. And, at that time, the ARF put forth an initiative to do
a parallel audit of those two meters, hiring an independent
engineering laboratory such as--our firm had done that by
auditing Arbitron and Nielsen meters, and hired the Illinois
Institute of Technology to go through those meters and see how
robust they were, where the fault lines were, and so forth.
So, where there's a real option for competition that has
credibility, the industry will respond. I mean, if you handle
this thing right, I think you can get through--I don't think
the issue of mandatory--or voluntary MRC accreditation has any
influence on--frankly, I think--and I said I think it expedites
it, because it brings the power together in one place. You can
go and get a reading and keep people informed.
Senator Burns. Mr. Mullen?
Mr. Mullen. Yes, I agree with Mr. Metzger, in that this
whole process has helped the voluntary code of conduct move
along, but I would want to point out that that voluntary code
of conduct was out and proposed by the MRC well in advance of
Nielsen's decision to roll out both Philadelphia and
Washington, D.C. And the response from Nielsen to audit all
future markets was announced, I believe, the day after they
commercialized the service in those two markets.
Senator Burns. Well, this has been very helpful today.
Mr. Ivie?
Mr. Ivie. I just wanted--can I----
Senator Burns. Yes, sir.
Mr. Ivie.--just address one point? It's different, though.
Senator Burns. I've got to shift gears and go talk about
BSE here in----
Mr. Ivie. All right.
Senator Burns.--a little bit.
Mr. Ivie. Can I just----
Senator Burns. This is a bad venue to bring that up.
Mr. Ivie. All right. I just wanted to make one quick point.
The idea of the membership of the MRC--who is a member and is
that a forum dominated by broadcasters--has been discussed
here, and I didn't get an opportunity to really respond to
that. And I wanted to just put something in the record on that.
The MRC has 95 members today. In 1964, when we were formed,
we had 15. This was at the full knowledge of the Congress when
we were formed. At that time, there were 11 broadcast members,
out of 15. There were two--these are television broadcast
members--there were two radio broadcasters, and there were two
seats held by the American Association of Advertising Agencies.
Today, we have 95 members, there are 27 broadcast seats. So,
that's--if you look at that in terms of diluting the influence
of broadcasters within the MRC, the MRC is a much more diverse
organization today than we ever were in 1964, and that's
reflective of, just as people have said here, the diversity of
the business.
Interestingly enough, we had five seats to one organization
in 1964. That was the NAB. They held 5 of those 11 broadcaster
votes. We've maintained that rule. In other words, no
organization in the MRC can hold more than 5 seats, and they
can only have a seat for a different media identity. So, if you
think about Viacom, they have a radio seat for Infinity, they
have a broadcasting seat, they also have cable----
Senator Burns. Sounds like Rotary.
Mr. Ivie. So, we have a very diverse organization. No one
sector controls it. But it is true that we have issues that
come up that certain sectors are very, very interested in. So,
you'll see sometimes very contentious, hotly-debated issues
among, you know, certain of the things we review. People do
this with passion. So----
Senator Burns. Well, it's like I said, I've got another
commitment here, but I want to thank this panel, the
information and how you've been candid with me and this
Committee--and this will be reviewed by other Senators; we're
just scattered all over the Hill today--and the way you
presented your information. And I appreciate that very much,
being very honest with the Committee. And we'll have to make
some decisions as we go along. Maybe this legislation might
have to be like my idea of going into the sampling business,
maybe.
[Laughter.]
Senator Burns. We might have to change--but we're amenable
to change. I think Congress can sometimes get into areas where
there are unintended consequences, and I don't want to do that.
And the only way we have, bringing this out--I appreciate Mr.
Metzger today a lot because of his experience and what he has
said here today. There's a lot of wisdom there. And I
appreciate it. And I know all of us have different interests.
And I appreciate that.
So, I look forward to working with all of you to solve this
problem. This is an industry that--the broadcast industry--that
I love. I came out of it. I sort of backed into it in the first
place. But it is dear to my heart. And I don't want to do
anything that would damage it, because I think we have a great
responsibility to the American people, to our advertisers, and
to each other to make sure that the industry continues to grow,
and grow in all segments of our society.
I appreciate that. And if you want to extend some--and
you'll get some questions from other Senators. I would ask you
to respond to those Senators and to the Committee. Your full
statement today will be made part of the record. And if you
have more that you would like to add, we would more than
welcome those comments.
Thank you for coming today. This Committee is closed.
[Whereupon, at 4:20 p.m., the hearing was adjourned.]
A P P E N D I X
Prepared Statement of Hon. Daniel K. Inouye, U.S. Senator from Hawaii
Today, the Committee returns to the subject of television ratings.
Previously, in the 108th Congress, we heard from parties concerned
about the decision of Nielsen Media Research to roll out new television
ratings technology in local markets. This technology--often referred to
as ``Local People Meters''--is designed to replace the old, and
admittedly flawed system, which relied on viewers to write down their
viewing choices in paper diaries.
The need to ensure appropriate levels of accuracy and transparency
have long been of interest to members of this committee and date back
to inquires pursued in the early 1960s. Given that the market for
television ratings data currently supports only one television ratings
service--Nielsen--that interest is particularly acute. Moreover, our
interest in this issue is shared by a number of parties who rely on
ratings information. Buyers of television advertising have a clear
interest in knowing the audience make up of the particular shows to bid
on. Similarly, programming distributors--and particularly, free, over-
the-air broadcasters that rely solely on ad-supported revenues--have an
interest in ensuring that ratings data accurately reflects the number
of people watching their programs.
Given these interests, our review of current practices is entirely
appropriate. If there are problems with the current system, we are well
within our right to explore possible improvements. Some of these
improvements may be pursued in the marketplace; others may require
government action. I look forward to reviewing these issues, but hope
that as we go forward, we will be wary of solutions that would give
parties with an economic interest in the outcome, the right to prevent
the roll out of new ratings technologies. In the end, our interests
should focus on promoting accuracy, and should not get sidetracked with
the mediation of commercial disputes.
Accordingly, Mr. Chairman, thank you again for calling for today's
hearing and I look forward to hearing from the witnesses assembled hear
today.
______
Prepared Statement of Hon. Frank R. Lautenberg,
U.S. Senator from New Jersey
Mr. Chairman:
Thank you for calling this hearing to give us an opportunity to
learn more about this issue.
Over the last few decades, television has grown from a couple of
broadcast network affiliate stations in each market to cable systems
with hundreds of channels from which to choose. Keeping track of who is
watching what is obviously extremely complicated.
The Nielsen Corporation has been tracking viewership for decades.
Although it is a private company, its work has enormous
implications for the American people--because viewer-ship not only
determines advertising rates, it ultimately decides what program
choices are available to the American people.
So there is a clear public interest in ensuring that the ratings of
television shows are accurate.
We know that Nielsen has begun using a new system to measure
ratings. It acknowledges that the system isn't perfect, but claims that
the technology is superior and that problems are being corrected.
Some broadcasters have said that the new system isn't getting a
true measure of some viewers--especially minorities. This is a serious
concern to me, and I am disappointed that a year after a Senate hearing
on this same issue, we are learning that the problem has not been
fixed.
Mr. Chairman, as I see it, there are two questions we must explore
today:
First, is Nielsen taking immediate steps to correct any problems
with their system?
And second, is the FAIR Act, which is before us today, the
appropriate Federal legislation to deal with the issue? I believe we
should think long and hard before mandating that a group made up of
Nielsen's clients should have power to decide what technologies Nielsen
can and cannot utilize.
I look forward to hearing from our witnesses. I'm sure they'll do a
good job of explaining both sides of this issue.
______
MindShare
New York, NY, September 29, 2005
Hon. Ted Stevens,
Chairman,
Senate Committee on Commerce, Science, and Transportation,
Washington, DC.
Dear Senator Stevens:
I would like to thank the Senate Committee on Commerce, Science,
and Transportation, particularly Senator Conrad Burns, for inviting me
to testify at the hearing on the FAIR Ratings Act. I appreciated
Senator Burns' courtesy during the hearing, as well as the chance to
express the views held by many of my colleagues in the advertising
community. As you know, advertisers, advertising agencies and
advertising buying companies have all demonstrated concern about any
legislation affecting television ratings and I believe it was important
that their voice be heard.
I would like to take this opportunity to comment for the record on
some of the remarks made by other witnesses.
George Ivie
My observations below are not meant to criticize the Media Rating
Council for what it is today--a voluntary industry organization
operating within the private sector. Instead, I am concerned that the
MRC as currently constituted cannot take on the responsibility for
regulating television audience measurement on the government's behalf.
If the MRC is to be given the authority envisaged by the Fair Rating
Bill, I believe the organization must be completely overhauled so that
research decisions are not determined by special interest voting.
In its new governing role, the MRC must be independent of the
business surrounding the data being regulated. We, in the industry, can
accept a certain level of potential bias today under the MRC's current
advisory role.
Statement: Mr. Ivie said that one of the MRC's stated policies is
to ``limit the influence of any one Industry sector within our
organization.''
Response: If this is an MRC policy, it is a policy that has not
been implemented. The broadcast industry continues to dominate the MRC
today, as it has from the time of its formation. By Mr. Ivie's own
count, 27 of the 65 votes--or almost half of the votes--on the
Television Committee are controlled by broadcasters.
But this does not tell the full story. Large media organizations
can have up to five votes for each of their business segments, so NBC
Universal, for example, has five separate votes: for the NBC Network,
the NBC TV station group, MSNBC, Telemundo and Universal Television. I
also note that Infinity Broadcasting, Viacom's radio network, is one of
five Viacom votes on the Television Committee despite not even being a
consumer of Nielsen ratings.
By contrast, my company, Mindshare, has only one vote. This gives a
handful of companies disproportionate power within the organization--
power that they have used to block accreditation of LPMs in many
markets over the opposition of voters from the advertising and cable
industries. If the broadcast, advertiser and cable companies all had
one vote each, it is possible that the outcome of the accreditation
votes would have been much different.
Statement: Mr. Ivie said that ``Accreditation is granted by the MRC
Board of Directors if a rating service complies with the MRC's Minimum
Standards for Media Rating Research and makes materially complete
methodological and survey-performance disclosures to their customers.''
Response: The accreditation process is not as simple as this
sentence implies. This process has become a lengthy and cumbersome
struggle. The MRC's ``Minimum Standards'' are vague and open to wide
interpretation. Fault rates remain a case in point, demonstrating that
these Minimum Standards are a moving target. A fault rate that was
historically acceptable in a Meter/Diary sample has suddenly become
unacceptable in an LPM sample.
Because of this, delay in accreditation has occurred even though
the collection technique is far superior to the diary, and the sample
is larger and more representative of the market. If the FAIR Ratings
Act were in effect, advertisers and programmers would have been denied
the use of this more accurate information for more than a year.
Statement: Mr. Ivie says, ``Special circumstances occur when an MRC
member whose company has a vested interest in the matter being
considered. When this occurs, that member may participate in the review
meeting but will not be allowed to vote.''
Response: I believe that all MRC members have a vested interest in
the issues before the MRC. Every member stands to profit or lose
through the implementation of new ratings systems. Again, if the FAIR
Ratings bill becomes law, I strongly believe that the MRC membership
would have to be changed so that members would not be in a position to
vote their self-interest and use their power to deny more accurate
ratings systems.
Pat Mullen
Statement: Mr. Mullen says that Nielsen ``does not have the trust
of our company or that of more than a dozen responsible broadcasters.''
Response: Mr. Mullen provides no names to back up this statement;
and although everyone has had their issues with Nielsen over the years,
it is unlikely that he could provide a list of a dozen companies who
would support that statement. Indeed, only a handful of companies have
voiced their public support for the FAIR Ratings bill, compared to
dozens who are opposed, including the Association of National
Advertisers, the American Association of Advertising Agencies, Asian
American Advertising Federation, BET, Azteca American Affiliate Group,
the Latin Business Association, the Urban League and the Rainbow/PUSH
Coalition.
Statement: Mr. Mullen complains several times that Nielsen is a
monopoly, including his statement that ``the keys to our success--our
ratings--are held by a monopoly.''
Response: Nielsen is the only company offering ratings data because
the TV industry itself, led primarily by the broadcasters, decided it
didn't want to pay for more than one service. Indeed, when Gale Metzger
tried to launch SMART as a competitor to Nielsen in the late 1990s the
major broadcast networks made a conscious decision not to fund it.
Nielsen may currently be the only company offering television ratings,
but that does not mean there cannot or will not be competitors in the
future, especially with recent innovations in set top boxes and other
television technology. Moreover, it is worth noting that the Federal
Trade Commission recently wrote that Nielsen has not engaged in
``exclusionary monopoly practices, collusion [or] anticompetitive
mergers.''
Additionally, the FAIR Ratings Act will preclude potential
competitors to Nielsen. What company could afford, what venture
capitalist would fund, and who in the industry would support, a service
that could take years to complete the MRC accreditation process with a
high potential for failure due to previously stated conflict of
interest issues?
Statement: In an apparent reference to fault rates, Mr. Mullen
says, ``It is worth noting that in New York, on the average day for the
week ending July 10, the viewing choices for nearly one-third of the
black and Hispanic men ages 18-34 in the Nielsen LPM sample were not
reflected in the ratings.''
Response: Obviously it is always possible to find one demographic
in one week in one market that appears to show high fault rates, but
looking at the broader picture, it is very clear that fault rates are
lower for all groups in the LPM samples than in the Meter Diary sample.
Mr. Mullen's credibility on this issue would be stronger if he had also
objected to fault rates in Meter/Diary markets.
More to the point, though, it is not true that faulting causes
undercounting. All demographic groups are weighted to make sure they
are represented in ratings to the same extent they are represented in
the overall population.
Statement: Mr. Mullen said that broadcasters ``pay the vast
majority of the fees for Nielsen's local services,'' seeming to imply
that broadcasters should have a greater say in ratings decisions.
Response: I strongly object to Mr. Mullen's assertion that local
broadcasters pay most of the freight for local TV ratings. The money to
support free over-the-air programming at the local level comes almost
exclusively from advertising, so all of a station's operating
expenses--including ratings fees--are ultimately borne by the
advertiser. I might also add that television stations, which operate
under a license granted by the Federal Government, are some of the most
profitable businesses in our economy. Since those high profit margins
are also borne by the advertiser, I think the advertising community has
a right to know that the rates it pays local stations accurately
reflect the number of people watching their programming.
Again, Senator Stevens, it was a great privilege for me to testify
before the Commerce Committee on this important issue, and I truly
appreciate the equitable manner in which Senator Burns' handled
proceedings. I hope Committee Members will keep in mind that the
American Association of Advertising Agencies (AAAA), the American
Advertising Federation (AAF), the Association of National Advertisers
(ANA), the Asian American Advertising Federation (3AF) and the
Association of Hispanic Advertising Agencies (AHAA), in addition to
more than a dozen leading advertising agencies and buying companies,
have all registered their objection to the FAIR Ratings bill. In light
of these concerns, I hope that Senator Burns and the co-sponsors of the
bill will reconsider their support for this legislation.
Sincerely,
Kathy Crawford.
Fault Rate Trends--Chicago
Set Meter vs. LPM Set
------------------------------------------------------------------------
Fault Rate (Set Fault Rate
Week-ending Meter) Week-ending (LPM Set)
------------------------------------------------------------------------
Total Fault Rate Trend
------------------------------------------------------------------------
3/28/ 13.8 3/27/2005 8.8
2004
4/4/2004 12.6 4/3/2005 9.4
4/11/ 12.7 4/10/2005 9.9
2004
4/18/ 11.8 4/17/2005 8.0
2004
4/25/ 12.9 4/24/2005 6.8
2004
5/2/2004 11.5 5/1/2005 7.2
5/9/2004 11.5 5/8/2005 6.4
5/16/ 11.9 5/15/2005 8.2
2004
5/23/ 11.8 5/22/2005 8.5
2004
5/30/ 11.5 5/29/2005 7.9
2004
6/6/2004 11.5 6/5/2005 9.5
6/13/ 12.7 6/12/2005 10.5
2004
6/20/ 12.7 6/19/2005 11.0
2004
6/27/ 11.7 6/26/2005 9.9
2004
7/4/2004 10.6 7/3/2005 10.3
7/11/ 11.3 7/10/2005 11.9
2004
7/18/ 11.1 7/17/2005 10.9
2004
------------------------------------------------------------------------
African-American Fault Rate Trend
------------------------------------------------------------------------
3/28/ 19.8 3/27/2005 13.0
2004
4/4/2004 18.3 4/3/2005 15.0
4/11/ 15.1 4/10/2005 15.2
2004
4/18/ 14.9 4/17/2005 12.3
2004
4/25/ 18.9 4/24/2005 7.7
2004
5/2/2004 15.2 5/1/2005 12.2
5/9/2004 12.1 5/8/2005 10.3
5/16/ 14.1 5/15/2005 12.6
2004
5/23/ 17.6 5/22/2005 15.7
2004
5/30/ 18.9 5/29/2005 15.7
2004
6/6/2004 20.2 6/5/2005 14.6
6/13/ 23.3 6/12/2005 15.7
2004
6/20/ 20.9 6/19/2005 17.0
2004
6/27/ 19.1 6/26/2005 17.3
2004
7/4/2004 19.3 7/3/2005 17.1
7/11/ 20.5 7/10/2005 17.8
2004
7/18/ 18.2 7/17/2005 16.0
2004
------------------------------------------------------------------------
Hispanic Fault Rate Trend
------------------------------------------------------------------------
3/28/ 21.2 3/27/2005 12.1
2004
4/4/2004 15.6 4/3/2005 12.7
4/11/ 23.1 4/10/2005 14.2
2004
4/18/ 18.8 4/17/2005 10.6
2004
4/25/ 15.9 4/24/2005 11.6
2004
5/2/2004 17.2 5/1/2005 9.1
5/9/2004 15.2 5/8/2005 9.2
5/16/ 15.4 5/15/2005 11.8
2004
5/23/ 15.4 5/22/2005 9.9
2004
5/30/ 16.9 5/29/2005 8.8
2004
6/6/2004 16.7 6/5/2005 11.7
6/13/ 18.2 6/12/2005 12.7
2004
6/20/ 18.5 6/19/2005 15.5
2004
6/27/ 17.2 6/26/2005 10.9
2004
7/4/2004 18.5 7/3/2005 9.8
7/11/ 18.8 7/10/2005 13.4
2004
7/18/ 20.6 7/17/2005 13.9
2004
------------------------------------------------------------------------
Fault Rate Trends--New York
Set Meter vs. LPM Set
------------------------------------------------------------------------
Fault Rate (Set Fault Rate
Week-ending Meter) Week-ending (LPM Set)
------------------------------------------------------------------------
Total Fault Rate Trend
------------------------------------------------------------------------
3/28/ 11.3 3/27/2005 9.7
2004
4/4/2004 9.8 4/3/2005 10.3
4/11/ 10.5 4/10/2005 9.9
2004
4/18/ 11.1 4/17/2005 9.7
2004
4/25/ 10.9 4/24/2005 9.5
2004
5/2/2004 12.1 5/1/2005 10.9
5/9/2004 12.7 5/8/2005 9.4
5/16/ 13.2 5/15/2005 7.9
2004
5/23/ 12.7 5/22/2005 7.9
2004
5/30/ 14.2 5/29/2005 7.5
2004
6/6/2004 13.4 6/5/2005 9.5
6/13/ 11.3 6/12/2005 9.8
2004
6/20/ 11.2 6/19/2005 9.4
2004
6/27/ 11.2 6/26/2005 8.4
2004
7/4/2004 9.7 7/3/2005 8.9
7/11/ 8.7 7/10/2005 9.8
2004
7/18/ 8.3 7/17/2005 9.0
2004
------------------------------------------------------------------------
African-American Fault Rate Trend
------------------------------------------------------------------------
3/28/ 20.2 3/27/2005 15.3
2004
4/4/2004 19.8 4/3/2005 17.2
4/11/ 18.0 4/10/2005 15.0
2004
4/18/ 17.2 4/17/2005 14.5
2004
4/25/ 21.2 4/24/2005 14.1
2004
5/2/2004 23.2 5/1/2005 17.4
5/9/2004 21.6 5/8/2005 14.8
5/16/ 22.1 5/15/2005 11.6
2004
5/23/ 23.7 5/22/2005 10.8
2004
5/30/ 22.8 5/29/2005 8.2
2004
6/6/2004 21.1 6/5/2005 11.5
6/13/ 19.1 6/12/2005 14.7
2004
6/20/ 18.2 6/19/2005 13.4
2004
6/27/ 18.2 6/26/2005 12.3
2004
7/4/2004 16.9 7/3/2005 11.5
7/11/ 16.9 7/10/2005 13.8
2004
7/18/ 16.1 7/17/2005 10.6
2004
------------------------------------------------------------------------
Hispanic Fault Rate Trend
------------------------------------------------------------------------
3/28/ 17.4 3/27/2005 12.3
2004
4/4/2004 12.2 4/3/2005 14.2
4/11/ 14.6 4/10/2005 11.6
2004
4/18/ 18.3 4/17/2005 14.2
2004
4/25/ 17.1 4/24/2005 12.9
2004
5/2/2004 19.3 5/1/2005 16.4
5/9/2004 19.8 5/8/2005 13.2
5/16/ 18.8 5/15/2005 10.1
2004
5/23/ 17.9 5/22/2005 10.8
2004
5/30/ 27.5 5/29/2005 11.0
2004
6/6/2004 26.6 6/5/2005 13.1
6/13/ 22.1 6/12/2005 10.6
2004
6/20/ 20.8 6/19/2005 9.3
2004
6/27/ 20.8 6/26/2005 11.8
2004
7/4/2004 10.3 7/3/2005 13.7
7/11/ 12.8 7/10/2005 14.9
2004
7/18/ 14.3 7/17/2005 14.4
2004
------------------------------------------------------------------------
______
Supplementary Information Submitted by Gale Metzger
I appreciated the opportunity to testify before the Senate
Committee on Commerce, Science, and Transportation regarding the FAIR
Ratings Act on July 27. Thank you for allowing me to do so and for
inviting these additional remarks.
As I said at the time, the introduction of legislation has itself
had an apparently positive effect. It seemed to have expedited the
acceptance of the proposed voluntary code of conduct. However, that
prospect does not diminish the need for a mandatory process to assure
that audience measurement services are always committed to the vetting
of their services through the Media Rating Council (MRC).
As explained in my testimony, I comment as one who (A) has no stake
in the outcome, (B) is concerned about core weaknesses in Nielsen
services and (C) believes that the quality of Nielsen research is
important to society and to the industry it serves. We now read that
Nielsen has spent ``more than $4,000,000'' for lobbyists and PR people
to shape the debate on this bill. \1\ They have cast the issues as
government regulation and minority representation. Both
characterizations are misleading.
---------------------------------------------------------------------------
\1\ Nielsen, Long a Gauge of Popularity, Fights to Preserve Its
Own, New York Times, August 8, 2005, Lorne Manly and Raymond Hernandez,
p. C1
---------------------------------------------------------------------------
First, what is proposed is not government regulation but a
government requirement. The requirement mandates a monopoly to work
with the industry and to play by long-standing and long-accepted rules.
To call this bill ``government regulation'' is a smokescreen,
created purposefully to raise a frightening hydra and to conjure a
burden that stifles innovation. Working with the MRC is not working
with the FTC--or any governmental bureaucracy. The MRC is a creation of
the industry itself. Accreditation is held up only if problems arise--
if a ratings service is not doing things right. The MRC system has
worked well for decades and is respected and accepted by all, buyers
and sellers, who have to use audience research numbers. This confidence
is a necessary prerequisite for a media industry that exchanges
billions of dollars on information. That information must be reliable.
In 1963, the industry and Nielsen agreed to work collaboratively to
build and ensure confidence in ratings. The MRC was set-up to avoid
government regulation and now, this legislative proposal only requires
that they live up to that earlier commitment--which in recent times
seems to be deemed as optional by Nielsen. The legislation is
appropriate and will benefit everyone.
Second, this bill does not stifle innovation; monopolies stifle
innovation. As I explained at the hearing, the MRC would expedite
gaining acceptance of new and better approaches to measurement. Had the
SMART initiative been funded, I would have asked the MRC to audit the
rollout in order to signal efficiently the value and accuracy of our
approach.
Nielsen has little or no incentive to do better. During the
discussion section of the hearing, I noted that Nielsen skipped homes
with new digital recording devices, such as TiVo. Ms. Whiting stated
that Nielsen could now meter such homes. Only recently have they
announced that capability, many years after the digital recording
devices entered the marketplace. This lag time between measurement of
time-shifting viewers and actual use exemplifies the ill effect of a
monopolized industry. Further, the industry will have to wait an
additional year, or years, for Nielsen to bring their total national
sample up-to-date and to have the proper share of new high-tech homes
in their sample for complete measurement.
Third, Nielsen says that it is ``in the truth business'' and casts
the dispute as a battle between the buyers and the sellers of the data.
In effect, Nielsen maintains they are above the fray and they can't
satisfy everyone. If Nielsen were to focus on research quality, they
and the television marketplace would be the richer for it.
Instead, the appearance is that they are focused on short-term
profits at the expense of long-term investing to produce a better and
more accurate measurement. The appearance is of doing only what they
have to do to get by. Their history of anomalous and inconsistent
numbers are caused, I believe, by business decisions animated most by
the bottom line. The lack of adequate support staff and the failure to
institute effective quality control procedures over their operations
are illustrations.
The roll-out of an improved meter provides an example of their
slowness to invest for improvement. I noted in my testimony that
Nielsen responds only when a competitor challenges them. In 1998, when
the A/P meter \2\ was being developed and SMART was a real possible
alternative to their services, Nielsen published a well-designed
research plan to create a scientifically-based approach to new A/P
operating rules. Seven years later and much delayed, Nielsen is ready
to roll out this ``A/P meter'' and integrate the technology into the
measurement system. In addition to the inexplicable wait, only on the
eve of the installations of the new meter did Nielsen propose efforts
to fill in the operating-rules gap. The planned research of 1998 had
never been completed and still hasn't. Effectively, Nielsen's new A/P
rules will be set arbitrarily and be modified as experience dictates.
---------------------------------------------------------------------------
\2\ A/P Meter stands for Active/Passive Meter and was designed to
enable identification of programs tuned on metered sets without relying
on station calibration.
---------------------------------------------------------------------------
A champion of truth would have introduced the new meters years ago
and completed the original research plan. Nielsen's slowness to
innovate and to improve damages themselves and the industry.
Fourth, Nielsen's posture as the protector of minority interests
camouflages the true problem. Minority groups are right to be concerned
about the accuracy of Nielsen ratings. As I testified, black and
Hispanic homes have more people in them. That means it is more
difficult to obtain good data from those homes. This is true for every
measurement technique, be it the LPM \3\ or the meter-diary approach.
The MRC is the best assurance minorities have of a measurement service
in which all population segments are appropriately counted.
---------------------------------------------------------------------------
\3\ LPM stands for Local People Meter.
---------------------------------------------------------------------------
Media Rating Council
In my July 27 testimony, I advocated that the MRC audit be an open
process. That is, the audit reports, which are the basis for the
accreditation process, should be available to the entire community.
Transparency would benefit the marketplace by putting all users on an
equal footing. It would bring pressure to improve operations more
quickly when problems are identified.
Effectively, the Statistical Research, Inc. People Meter Review of
1987-88 was a full and complete open audit. The Advertising Research
Foundation has performed audits of media rating services and, as a
matter of policy, their audit results are open to everyone. Open-audit
precedents exist and have served the industry well.
In addition, distribution of the full audit findings takes the onus
off of the current process of accrediting or not accrediting. A pass or
fail grade is too rudimentary for the sophisticated evaluation
required. The market needs to know and understand the accreditation
process. Further, concern articulated about the politicization of
accreditation voting (to the extent it is real) would be ameliorated by
open audits. Everybody could monitor the process and providers would be
accountable from beginning to end.
Conclusion
It would be truly regrettable if this bill did not pass because
some lawmakers don't like government regulation and others fear the
disenfranchisement of minorities--when neither issue is the real
problem. Decoys and diversions created by Nielsen should not triumph.
The real story is that the proposed legislation would reinforce
industry self-regulation and provide better protection that all people,
including minorities, are counted more accurately.
I remain hopeful that Congress will intervene to ensure that the
1963 standards for audience measurement review will persevere and
transcend the political aisles.