Citation
Conceptualizing and measuring the effects of brand equity on television program ratings performance

Material Information

Title:
Conceptualizing and measuring the effects of brand equity on television program ratings performance
Creator:
McDowell, Walter
Publication Date:
Language:
English
Physical Description:
vii, 95 leaves : ; 29 cm.

Subjects

Subjects / Keywords:
Brand loyalty ( jstor )
Brands ( jstor )
Broadcasting ( jstor )
Broadcasting industry ( jstor )
Consumer research ( jstor )
Marketing ( jstor )
Marketing mixes ( jstor )
News content ( jstor )
Television programs ( jstor )
Viewers ( jstor )
Dissertations, Academic -- Mass Communication -- UF ( lcsh )
Mass Communication thesis, Ph. D ( lcsh )
City of Orlando ( local )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1998.
Bibliography:
Includes bibliographical references (leaves 88-94).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Walter S. McDowell.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
0029545392 ( ALEPH )
41360856 ( OCLC )

Downloads

This item has the following downloads:


Full Text











CONCEPTUALIZING AND MEASURING THE EFFECTS OF BRAND EQUITY ON
TELEVISION PROGRAM RATINGS PERFORMANCE









By

WALTER S. McDOWELL














A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1998














ACKNOWLEDGMENTS

The author thanks Mr. Skip Skiffington, Program Director of Station A, and Ms. Jenny Butler of Nielsen Media Research for providing proprietary overnight Nielsen ratings data.














TABLE OF CONTENTS

Paqe

ACKNOW LEDGMENTS ........................................................................................ ii

L IS T O F T A B L E S .................................................................................................... v

A B S T R A C T .............................................................................................................. v i

CHAPTERS

1 IN T R O D U C T IO N ............................................................................ 1

2 REVIEW OF LITERATURE ........................................................... 4

Competition Changes Everything .............................................. 4
Media Available versus Media Used ............................. 7
Rethinking Program Content .......................................... 9
Birth of a Buzzword ............................................................ 10
Branding Local News ....................................................... 12
S u m m a ry ............................................................................. 1 3
Brands and Brand Equity ............................................................. 14
The Emergence of Brand Equity ..................................... 16
Product Brand and Product Categories ........................ 17
Conceptual Definitions of Brand Equity ..................................... 17
Consumer Attitudes versus Consumer Behavior ....... 20
Definitions From the Private Sector ................................ 21
Definitions From Academia ............................................. 23
In e rtia ................................................................................... 2 7
Brand Equity According To Keller ................................... 28
Operation Definitions of Brand Equity ....................................... 29
Inheritance Effects ......................................................................... 33
Overview of Research ....................................................... 33
Viewer Motivation and Loyalty .......................................... 35
Applying Brand Equity to Television Program Performance.. 37
Before the Buzz, There W as Audience Loyalty .............. 37
Network versus Station versus Program Equity ........... 38
Adapting Brand Terminology ........................................... 39
Controlling the Marketing Mix ........................................... 40

ifi








Inertia, Momentum, and Equity .................................................... 43
Interfacing Momentum With Keller's
Conceptualizations ........................................................ 45
Equity and Performance ................................................... 46
The Nielsens: Measuring Program Performance ................... 47

3 RESEARCH HYPOTHESES ......................................................... 49

4 RESEARCH METHODOLOGY ..................................................... 51

Controlling The Marketing Mix ...................................................... 53
Establishing a Comparative Equity Benchmark ....................... 56
Benchmark #1: Nielsen 1 0-Year Audience Trend Analysis ... 57
Benchmark #2: Marshall Marketing Loyalty Survey .................. 58

5 R E S U LT S ........................................................................................ 6 1

Hypotheses One, Two, and Three .............................................. 61
Hypotheses Four and Five ........................................................... 63
Hypotheses Six and Seven .......................................................... 68

6 D IS C U S S IO N .................................................................................. 7 1

S tudy L im itatio ns ............................................................................ 7 8
C o n c lu s io n s .................................................................................... 8 0
Suggestions for Future Research .............................................. 82
Implications For the Field ............................................................. 84

APPENDIX: A Nielsen TV Metered Market Household Analysis
M a y 19 9 6 ....... .................................................................. 8 6

R E F E R E N C E S ............................................................................................ 8 8

BIOGRAPHICAL SKETCH ......................................................................... 95












iv














LIST OF TABLES

Table Page

1 Breakout of Nielsen Metered Overnight Database .............................. 55

2 Comparison of Sweep versus Nonsweep Cases: 1996
D escriptive H ousehold D ata .................................................................. 56

3 Comparison of Sweep versus Nonsweep Cases: Paired
Sam ple T-Test by Households ............................................................... 56

4 1 0-Year Monday through Friday Average Sweep Performance of
T hree N ew scasts ...................................................................................... 57

5 1996 Marshall Marketing and Communications Survey on
Exclusive (Loyal) Viewing of Newscasts ........... ................................... 59

6 Comparative Newscast Performance Based on Nielsen Metered
O ve rn ig ht R ating s ...................................................................................... 6 2

7 Differential Performance Between 10:45 PM Lead-in and
11:00 P M N ew scasts ................................................................................ 64

8 Share Index of Late Newscasts Compared to Lead-in
P ro g ra m m in g .............................................................................................. 6 5

9 Household Correlational Analysis Between 10:45 PM Lead-in
and 11 :00 P M N ew scasts ....................................................................... 66

10 Comparison of Household Correlations Derived From Single
Day versus Seven- Day Moving Average Data ...................................... 67

11 Regression Data Based on Overall Households ................................ 67

12 Nielsen Audience Flow Analysis: May Sweep 1996 ............................ 70





V














Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

CONCEPTUALIZING AND MEASURING THE EFFECTS OF BRAND EQUITY
ON TELEVISION PROGRAM RATINGS PERFORMANCE By

Walter S. McDowell

August 1998

Chairperson: Dr. John C. Sutherland Major Department: Mass Communication

The purpose of this study was to explore the plausibility of applying

brand equity theory to electronic media, particularly television programs. While the broadcasting and cable industries have embraced the jargon of brand management, there has been scant published research on the specifics of how conventional brand management theory can be applied to television program strategies and practices. Additionally, there has been little nonproprietary work on how to measure the influence of program brand equity on program ratings performance.

When examining program ratings performance, the powerful influence of lead-in programming (coined by media researchers as inheritance effects or tuning inertia) cannot be ignored. Synthesizing elements from several consumer-based brand equity theories, the investigator proposed that a


Vi








program brand equity model focuses on the differential ratings response of a program to its direct competitors and to its lead-in programming. After establishing this framework, a number of hypotheses were tested within a case study format. The daily ratings performance of three 11:00 PM local newscasts in a major southeast television market were analyzed using one station as an equity benchmark.

Acknowledging the lack of external validity and the need for further testing across many markets and program genres, the results of this exploratory study were encouraging. With all hypotheses supported to some degree, the study concludes with several recommendations for a program equity research agenda for future work in the field.


























Vii














CHAPTER
INTRODUCTION

In the spring of 1996, many media observers were startled to learn that the newly appointed general manager of the prestigious NBC owned and operated TV station in Los Angeles, KNBC, had never before worked at a television station, broadcast network, or production studio. Instead, Carole Lynn Black was hired from the management ranks of Proctor and Gamble. NBC discounted her lack of direct broadcasting experience by emphasizing her extraordinary talents in brand marketing. In an interview, Black (1996) stated, "My experience at P&G taught me that branding is about finding the uniqueness in two very similar products, presenting that positive uniqueness to the consumer and driving that message home" (p. 93). This unprecedented management decision to hire someone outside the core industry to direct the operations of a network "flagship" station serves as just one example of how the electronic media have become captivated with brand management theory. In an effort to cope with unprecedented competition, audience fragmentation, and declining market shares, broadcasters have looked to the retail consumer goods industry for inspiration. The result has been the eager adoption of the jargon, if not the substance, of brand management. The media trade press are filled with cursory references to "brand identity," "brand image," and "brand




1





2


extension." In particular, there has been a growing interest in that most muddled of brand management concepts, "brand equity."

Although there are many divergent conceptual izations of brand equity, there is universal agreement that equity enhances a product's performance in the consumer marketplace. That is, equity helps reinforce consumer loyalty, attract new customers, and insulate the product from competitive attack. Fierce competition is the catalyst for most businesses to look beyond short-term sales goals and focus on the more enduring advantages of consumer-based brand equity. Encountering dozens, if not hundreds, of cable and satellite program competitors, many television executives have embraced a management perspective where networks, stations, or programs are regarded as consumer brands. As with conventional consumer goods, the expectation is that brand equity will enhance the performance characteristics of a program in the audience marketplace. More specifically, brand equity theory is presumed to help broadcasters develop and schedule programs that will deliver dominant Nielsen ratings.

A goal of brand equity research has been to determine the proportion of a brand's market share that is most vulnerable to consumer brand switching. Recognizing the vulnerability of a brand allows brand managers to make informed strategic decisions. By ignoring the symptoms of poor equity, managers run the risk of unforeseen declines in market shares and disappointing bottom lines (Kapferer, 1992). Similarly, managers of the electronic media wish also to diagnose any vulnerabilities in a program's





3


audience performance. Although there is considerable professional and scholarly research on brand equity dealing with conventional consumer goods, there has been little work done on adapting these theoretical precepts to electronic media.

Furthermore, any meaningful study of program ratings performance cannot ignore the potent influence of lead-in programming or inheritance effects. It is no secret among programmers that the best predictor of a program's audience size is often the size of the audience leading into it. While there is considerable literature on this subject, no one has approached this program performance issue from the vantage point of brand equity.

The purpose of this study is twofold. The first is to increase the body of knowledge of brand equity theory by exploring its applicability to television program ratings performance. The second is to offer an exploratory case study that tests a proposed'theoretical approach that synthesizes aspects of traditional brand equity with Nielsen television ratings. Using diary-based "sweeps" data and electronically metered "overnight" ratings, this study attempts to measure the relative influence of program brand equity on the performance of three late evening newscasts in a designated southeast market. Particular attention is paid to the relationship between the newscasts to their lead-in programs as a .-means of evaluating program brand equity.














CHAPTER 2
REVIEW OF LITERATURE

Branding has become the buzzword of the day.... More is riding on the success or failure of broadcast branding
than at any time in the medium's history.
(Stay Tuned, 1995, p. 58)

ComRetition Changes Everything

Branding and many of the notions of brand management are not new to most American consumer goods. Some of today's most prominent brands such as Coke, Levi, Maxwell House, Budweizer, Campbell and Kellogg began their branding efforts in the 1880s (Aaker, 1991). What is new is the adoption of this paradigm by the electronic media.

As will be elaborated in later sections, the primary motivation for applying brand management to a consumer product or service is competition. As the number of similar products or services in the marketplace increases, the need for highly differentiated brands becomes more acute. Also, with a rise in competition, there is usually a similar rise in the speed and sophistication of measuring brand performance in the marketplace. The introduction of technological devices such as computerized UPC codes has enabled retailers to track almost instantly consumer purchases for thousands of brands. While attitudinal measures of equity may certainly be of value, measures of in-market consumer behavior are the preferred means of evaluating brand performance.



4





5


As a marketing vice president at Proctor and Gamble recently proclaimed, "In an information-rich environment, impatient retailers don't hesitate to relinquish brands that don't deliver consistent and profitable sales. In order to survive in an increasingly competitive marketplace, a high level of brand fitness is required" (Gleason, 1995, p. 37).

The same can be said for the brand fitness of a television program in an ever-increasing competitive environment. The business of commercial broadcasting is the selling of audiences to advertisers, and in an "information rich" world of overnight Nielsen ratings, impatient program executives do not hesitate to cancel programs that cannot deliver consistent and profitable audience ratings.

The late arrival of brand management to the business of American

television was due primarily to a lack of competition. For over three decades, the competitive arena for commercial television was restricted to three major players. Through the interplay of various legal, economic, and technological factors, a three-network program oligopoly dominated the television industry from the 1950s to the mid 1980s. ABC, CBS, and NBC provided over 75% of the daily programming for their local affiliated stations (Long, 1979). Syndicated programming filled the remaining hours, but here again the viewer choices were meager because of the few number of stations licensed to each community.

By in the mid 1980s the competitive picture began to change when Fox became a legitimate network player, siphoning off young adult viewers from the





6


"Big Three" networks. The 80s also witnessed the coming of age for cable television. Cable program networks such as CNN, ESPN, MTV, A&E, Nickelodeon, and Discovery began to encroach on the audience territory that was once the exclusive domain of the broadcast networks and their affiliated stations. As the number of unique cable programming options increased, so did the number of subscribers, and with more subscribers came more audience fragmentation (Lin, 1995). From 1977 to 1997, ABC-, CBS-, and NBCaffiliated stations combined lost over 40% of their viewing households to basic cable and other alternative media (Myers, 1997). In the near future, technological breakthroughs such as digital compression and fiber optics will enable cable and satellite companies to increase channel capacity from a few dozen channels to several hundred. With over 70% of the U.S. households already "wired," traditional over-the-air (terrestrial) broadcasters can no longer count on captive audiences and guaranteed profits. Former NBC programming executive and media entrepreneur, Brandon Tartikoff, claimed in a 1993 interview that with the advent of so many new program choices on the horizon, America soon will witness a paradigm shift in power from program providers to viewers. "Some say that television was once a medium controlled by tyranny. Now, it's going to be a medium controlled by democracy" (Barton, 1993, p. 16). A commentary by respected economist Adam Theirer (1995) of the Heritage Foundation claimed the introduction of so much choice is "a signal to the world that the old media empires are modern-day dinosaurs headed for extinction" (p. 5).





7


Complementing the significant changes in media technology have been equally important changes in government regulation of electronic media. The highly publicized Telecommunications Bill of 1996 was a mandate to increase market-driven competition by unshackling government restrictions on telecommunications. Broadcasting, cable, satellite, and telephone delivery systems have been given the legal green light to cross boundaries and compete for consumers. Congressional advocates for the bill claim it will soon unleash a torrent of competition heralding nothing less than "the dawn of a new information age" (Stern, 1996, p. 8).

Media Available versus Media Used

Adding fuel to the fire of this emerging competitive marketplace is the fact that while the amount of program choice continues to increase, there has been no corresponding increase in the amount of time people spend watching television. Nielsen Media Research reports that over the past 10 years (19861996) the number of minutes per household dedicated to television viewing has remained unchanged (Nielsen Audience, 1996). The disturbing implications for television programmers is that while the number of slices in the programming pie has increased, the overall size of the pie itself has not increased. So far in the 1990s, more choice has not translated into more viewing. A number of media observers have referred to this phenomenon as media "cannibalism" (Burgi & Katz, 1997, p. 3), where program providers are forced into "feeding" on the audiences of fellow program providers (Mandese, 1995, p. S2). Using brand marketing terminology, television viewing appears to





8


have evolved into a mature or zero sum market where the number of available customers for a product category is stagnant. Therefore, as more brands enter the marketplace, the only means of survival is to take customers (audiences) away from competing brands (Kapferer, 1992; Lehmann & Winer, 1994).

A further frustration for television programmers is that despite the

historic technological and regulatory breakthroughs that have freed audiences from the "tyranny" of limited choice, the typical viewer still prefers to deal with only a handful of options. Several studies indicate that for an average American household, there is a viewing threshold of approximately a dozen channels. For example, Nielsen Media Research conducts an annual national survey of television viewing characteristics of which "Channels Received vs. Channels Viewed" is a special section, The 1996 report revealed that while the number of channels available in the home continued to grow, the number of channels actually viewed had not grown appreciably. The study concluded that while the average cable household can now receive 41.1 channels, only 10.6 channels are actually viewed. When 70 or more channels are available, viewership increases only slightly to 13,6 channels (Nielsen Audience, 1996). Several researchers have adopted the concept of channel repertoire to describe this limited array of channels from which audiences select programming (Webster & Lichty, 1991). This channel repertoire is similar in definition to what advertising researchers call an evoked set, where a consumer considers only a portion of the brands available in a purchase situation. The mathematics of this new programming marketplace are not lost on NBC president Bob Wright





9


when he states, "My biggest worry is our ability to continue to attract a disproportionate-size audience for our broadcast stations versus other means of gathering audiences" (B&C Hall of Fame, 1996, p. 51). Rethinking Program Content

The rise in competition has led to a reexamination of program content theory. For example, prime time programs for years were produced according to the least objectionable program theory( LOP). The theory is based on the core premise of moderate liking. Audience researchers discovered early on that most people do not watch television alone. Therefore, individual preference is often mediated by group dynamics where extreme opinions, either positive or negative, are suppressed in favor of a group compromise. By definition, a compromise is a movement towards the common center or "lowest common denominator" (Eastman, 1997). This strategy might generate "repeat customers" for a program but not necessarily highly committed fans. The LOP theory works well when there are few competitors, but as viewer options increase dramatically, programs designed for moderate liking can become a liability. As brand managers of many consumer goods have been forced to market their products to select consumer segments, so media programmers are now obliged to target their program content to the specific needs of fairly narrow, specialized audience segments (Cooper, 1996). As will be demonstrated later, "moderate liking" is not the stuff of brand equity.





10


Birth of a Buzzword

One of the first occasions where the concept of media as brands was presented in a public forum was a keynote speech made in 1993 at the Advertising Research Foundation Annual Conference by David Bender, President and CEO of Mediamark Research. He professed that media (both print and electronic) are similar to consumer products and services and that traditional brand marketing theory and procedures can be applied successfully to media to generate more business. He stated that "media vehicles are one of the few examples of brands which don't, for the most part, conceive of themselves as brands; as a result, they often are not able to take advantage of the insight that accrues to those aware of thinking and research on brands" (Bender, 1993, p. 2). Less than three months after Bender's speech, a frontpage headline in Advertising Age declared, "Here Come The Newest Brands: NBC, CBS, ABC" (Mandese, 1993, p. 1).

Since then, media executives and media trade journals such as

Broadcasting and Cable, Electronic Media, Media Week, and Advertising Age have become ever more comfortable with the phraseology of branding. For example, organizations such as the National Association of Television Program Executives (NATPE) and Promotion and Marketing Executives In Electronic Media (Promax) have sponsored seminars addressing "The Branding Bug" (NATPE, 1996)," Our Brand New World" (NATPE, 1997), "Brand Building and Visual Design", "Welcome to The Brand Revolution" and "Branding Your Message To Win" (Promax & BDA, 1996). After winning the fall








premier season, NBC placed full-page ads in a number of publications boasting that it was "America's Leading Network Brand" (NBC, 1995). Media Week dedicated a major article to "Network Brand Marketing Fever" (Dupree, 1996, p. 27). A trade journal ad for the off-network syndicated version of "Due South" bragged that the program was "a proven brand with a loyal audience" (Due South, 1997, p. 35). A newly published NATPE brochure mentions how broadcasters must "tackle the challenges of branding your program or station" (NATPE, 1997). Even the mainstream press appears to have adopted branding jargon. A subheadline in a New York Times article states that "television stations fear for their channel brand as choices proliferate in the digital age" (Brinkley, 1997, p. C9).

Not to be outdone by their broadcast competitors, cable and satellite programmers have jumped on the brand wagon. The annual convention of Cable Television Administration and Marketing (CTAM) provided a panel discussion on "Cable Image Branding" (McConville, 1995a, p. 34). The Western Cable Show offered a seminar on "Building a Reputation: Creating Brand Loyalty" (McConville, 1995b, p. 70). A programming executive at A&E claims that high profile programs such "Biography" help "brand the Network identity" (Walley, 1995, p. 30). The headline for an article about the FX channel stated that "FX Will Build Its Brand In Originals" (Spring, 1997, p. 10). A classified employment ad for The Weather Channel claimed it was seeking a Creative Director who will "steward the brand" (Weather, 1997, p. 53). A review of MTV's "Real k/Vorld" program congratulated the producers for "establishing





12


its own brand of show" (Sharkey, 1997, p. 17). Rather than focusing on individual networks or programs, Jonathan Sims, vice president of research for the Cable Advertising Bureau (CAB), indicts the entire medium of broadcast television, proclaiming "a decline in broadcast equity" (Sims, 1997, p. 37).

In recent years, broadcasters have grappled with the question of

overlapping brand identities. Networks, stations, and individual programs create a sometimes awkward branding hierarchy that has been described as "two people in a donkey costume ... not always going in the same direction, but both trying to get to the same place" (Rathbun, 1995, p. 12). Some program producers, such as Barry Thurston of Columbia Tri-star, are critical of network intimidation that coerces stations into "sacrificing their own identities at the network's alter" (Freeman & Stanley, 1996, p. 8). Branding Local News

The notions of brand management and particularly brand equity are no more prevalent than in discussions of local newscasts. Providing as much as 45% of a station's overall sales revenue, the financial and public relations benefits of a successful news operation can be enormous (Eastman, 1997; Tobenkin, 1994), but there can also be a downside. News programming is expensive. A station that ranks third in the market may never see any return on its investment. Media observers such as Ron Alridge, Editorial Director of Electronic Media, claim that "some folks are already saying that the third-placed newscast is an endangered species" (Alridge, 1996, p. 6).





13


For many stations, local news has become the only enduring and familiar program asset. At a recent Radio and Television News Directors Association (RTNDA) convention, a journalist reported that "in an environment where decades-old network affiliations can be broken overnight, local newscasts remain the strongest and most stable form of brand equity' (Tobenkin 1994, p. 68).

Summary

From observing industry practices and reading media trade journals, it is obvious that branding has indeed become "the buzzword of the day." At the heart of all the brand discourse is the realization by broadcasters that it is no longer business as usual in the quest to win the hearts and minds of audiences. Compared to the early 1980s, the number of available program channels has tripled, but the amount of time spent in front of the television set by a typical household has not changed significantly in 10 years. Furthermore, as the amount of program choice has increased dramatically, the number of channels actually viewed by a typical American household remains at about a dozen. With the stakes so high, broadcasters have turned to the highly competitive consumer goods industry for inspiration. The result has been an almost evangelical acceptance of media as brands.

Today, there is no nonproprietary research available from industry or academia on applying brand equity theory specifically to broadcast programming. The upcoming sections of this literature review provide (a) an overview of brand equity research derived primarily from a conventional





14


consumer goods perspective, (b) a review of research on audience inheritance effects, and (c) a translation of brand equity theory to the needs of television programming.

Brands and Brand Equity

In a cluttered marketplace, where consumer products are often more similar than they are different, proper brand management increases the probability of consumer brand choice. Brand management is the reason why consumers buy Big Macs rather than hamburgers, Nikes rather than sneakers, Harley Davidsons rather than motorcycles.

A brand is a name, term, sign, design, or a unifying combination of them intended to identify and distinguish the product or service from its competitors. Brand names communicate attributes and meaning that are designed to enhance the value of a product beyond its functional value. The basic reason for branding is to provide a symbol that facilitates rapid identification of the product and its repurchase by customers (Belch & Belch, 1993; Cobb-Walgren, Ruble & Donthu, 1995; Srivastava & Shocker, 1991).

Because consumers often lack the motivation, capacity, or opportunity to process all product information to which they are exposed in a thoughtful or deliberative manner, they opt for quick resolution techniques stored in memory (Kardes, 1994). Strong brands assist in this heuristic process. Biel (1991) offers the following insight:

On a very practical level consumers like brands because they package
meaning. They form a kind of shorthand that makes choice easier. They
let one escape from a feature-by-feature analysis of category





15


alternatives, and so, in a world where time is an ever-diminishing
commodity, brands make it easier to store evaluations. (p. 6)

In social psychology this same process contributes to the formation of social stereotypes, where people use prior knowledge and feelings (attitudes) about groups to make swift inferences about individual members (Beike & Sherman, 1994). Stereotyping and branding are essentially two different manifestations of the same phenomenon. Each serves to streamline cognitive processing. In a highly competitive, time-sensitive environment, strong brands have a better chance of being purchased than weak brands.

Strong brands also cultivate habits. Rosenstein and Grant (1997)

maintain that in repetitive decision-making situations, habits save time and reduce the mental effort of decision making, thereby allowing us to maintain complex behavior patterns without becoming overwhelmed by a huge cognitive task load. Habitual purchasing also can be explained readily using behavior learning theory and particularly operant conditioning. This approach assumes that learning is an associative process where a specific response is associated repeatedly with a specific stimulus. Over time this reinforcement process strengthens the bond between stimulus and response. For example, if the repeated outcomes resulting from the use of a branded product are positive, the likelihood of that consumer buying that brand again is increased (Belch & Belch, 1993).





16


The Emergence of Brand Equity

While brand management had been on the American business scene for decades, the specific topic of brand equity did not become popular until the volatile 1980s when once proud brands such as Sears, IBM, and Cadillac began to lose ground to lower-priced generic competition. Market shares dropped as a serious nationwide recession further aggravated corporate profits. This historic economic downturn inspired a rash of company consolidations and "downsizing" actions (Morris, 1996). Prudent cost cutting is indeed one way to shore up sagging profit margins, but as Sears' CEO, Arthur Martinez, warns, "You can't shrink your way to greatness," brand building is essential for survival (Greenwald, 1996, p. 54). During many leveraged buyout negotiations, corporate executives and Wall Street investors had to come to terms with the portfolio value of brand names. The recognition of brands as valuable intangible assets led to an increased interest in brand equity as a topic for private and scholarly research (Aaker, 1991; Keller, 1993 ).

According to several business analysts, the initial success of less expensive generic brands during the 1980s implied that consumers were failing to acknowledge the supposed added value of a "name brand" commodity and were influenced more by structural factors such as pricing and shopping convenience (Morris, 1996; Reis & Trout, 1981). This added value that immunizes a brand from such competitive incursions is often referred to as its equity. Businesses coping with "mature" product categories that exhibit little or no growth are particularly interested in brand equity. In such cases,





17


competitive pressures are intense as volume gains (share increases) are derived from competitors rather than new category users. Under these "zero sum" circumstances, as the number of competitors increase, the battle for market share becomes more acute (Lehmann & Winer, 1994, p. 60). Product Brands and Product Categories

The challenge of quantifying the value of a brand name has resulted in a range of measures from consumer attitude surveys to more concrete measures of in-market consumer behavior. However, at the core of all brand equity theory is the distinction between a product category and a product brand. A category refers to the generic commodity with no differentiation among brands. Depending on a number of motivational factors, a consumer may choose a category of product to satisfy a need with no consideration of brands-any brand will do. In another purchase situation, the same consumer may carefully evaluate the merits of a number of alternative brands--which is the best brand for me? A third option is the consumer seeks a specific brand with no consideration of category alternatives--only this brand will do. An important assumption for any practical measure of brand equity is that the brand under scrutiny must be a direct competitor coming from the same product category (Kapferer, 1992; Keller, 1993). Mouthwashes should not be compared to deodorants, nor should TV sitcoms be compared to newscasts.

Conceptual Definitions of Brand Equity

It is difficult to find agreement on the proper conceptualization of equity, much less the relative weight of each element in driving brand choice. After a





18


particularly disappointing car selling season for General Motors, a frustrated GM market analyst admitted that "even though brand equity may sometimes be hard to define, it's pretty clear when you've lost it" (Wilkie, 1996, p. 267).

Moving from an understanding of brands to a similar understanding of brand equity can be perilous because so many other branding terms have been used as synonyms for equity. While academics have yet to reach perfect agreement on some of these items, media professionals and journalists have been the worst offenders. The following is a brief clarification of some common terms that, in many cases, serve as the conceptual building blocks of a larger construct known as brand equity. These definitions are a synthesis of definitions and conceptualizations found in several sources, including Hoyer and Brown( 1990), Belch and Belch (1993), and Keller (1993). Branding is the process of naming a product or service in order to

distinguish it from its category competitors,

" Brand extension is the process of branding a new product with an already

established brand name. (An example would be Ford creating a new type of

automobile or truck but still identifying it as a Ford product.)

" Brand awareness (or brand identity) is the first rung on the equity ladder. It

refers to the simple familiarity (recall or recognition) of a brand relative to its

product category.

" Brand image goes beyond mere awareness and deals with the thoughts

and feelings (meaning) of the brand to a consumer. It can be conceived as

a cluster of attributes and associations that consumers connect with a





19


specific branded product. Some equity researchers, such as Keller (1993),

compare brand associations in terms of favorability, strength, and

uniqueness.

" Brand positioning is the image of a brand defined in relationship to its

market competitors. The goal is to "position" a brand in a consumer's mind

so that it has a highly distinctive (differentiated) image in comparison to

brands offered by competitors.

" Brand trial (or brand sampling) is the initial purchase or experience with a

brand by a consumer.

" Brand attitude can be viewed as an extension of brand image in that the

term refers not only to thoughts and feelings about the brand but

evaluations and most importantly, predispositions to respond (purchase). A positive or negative attitude is reinforced by experiencing a brand first hand. Brand loyalty has often been viewed as another name for brand equity, but

for our purposes a much narrower definition has been adopted. Loyalty is

the degree to which a consumer purchases repeatedly a certain brand

without digressing to alternative brands. Within this context, brand loyalty is

a measure of consumer purchase behavior.

" Brand commitment addresses the underlying psychology of brand loyalty.

That is, commitment is the degree of fidelity to a brand that, regardless of

pricing differentials, promotions and other market factors, the consumer

remains loyal in a purchase situation.





20


Brand equity in a general sense is defined in terms of marketing effects

uniquely attributable to the brand, that is, when certain outcomes result from

the marketing of a brand name that would not occur if the same product or

service did not have that name. Depending on the selected theoretical approach, consumer-based brand equity combines some or all of the

above-mentioned brand concepts. A more detailed examination of brand

equity follows.

Before presenting a review of industry and academic conceptual definitions of brand equity, it is important to remember that all of these definitions can be divided roughly into either attitudinal or behavior categories. It is well documented that attitudes alone are generally poor predictors of marketplace behavior (Kraus, 1995). For years researchers have wrestled with the disparities between the psychology of consumer preference and the behavioral realities of consumer brand choice. Despite these incongruities, brand equity has been defined typically from a consumer attitude perspective because it is often less expensive than tracking and matching product sales with individual consumers (Aaker, 1991; Churchill, 1995). Consumer Attitudes versus Consumer Behavior

While few people would disagree that consumer-based brand equity is ultimately a state of mind, many equity researchers assume that most attitudinal or psychological components of equity (such as awareness, imagery, positioning, and commitment) manifest themselves in observable purchase behavior. This consumer behavior in turn translates into measures of





21


brand performance in the marketplace. In other words, certain indicators of brand performance can be accepted as circumstantial evidence of consumerbased brand equity.

Definitions From the Private Sector

The Marketing Science Institute (MSI) offers a kind of all-encompassing definition in that brand equity is said to be

a set of associations and behaviors on the part of a brand's consumers, channel members and parent corporation that permits the brand to earn
greater volume and greater margins than it could without the brand
name and that gives the brand a strong, sustainable and differentiated
competitive advantage. (Anantachart, 1995, p. 12)

The institute does not further define what the specific "associations and behaviors" are that constitute equity.

The private sector has been very active in promoting brand equity

research. While each company covets is own unique and patented formula, they do reveal through their sales brochures and other publications some of their conceptual definitions of consumer-based equity. The Interbrand company has developed one of the best known commercial methods of calculating brand equity from a purely behavioral perspective. The firm applies a multiplier to the revenue generated by the brand. This multiplier is a function of six brandrelated factors that tap into product market positions and customer behavior. These include leadership, stability, market trend, support, and protection (Interbrand, 1994).

Other companies such as the Lintas Advertising Agency combine

attitudinal and behavioral elements into their "equity audit." Measures such as





22


brand awareness, image attribute ratings, and assumed leadership are merged with market share data, sensitivity to pricing, and purchase consistency over time. The overall purpose of the audit is to determine the strong or weak bias consumers have relative to other brands (Lintas, 1993). The NPID Group "Brand Builder" model also combines consumer perceptions and actual in-market performance to arrive at an indicator called brand health. Factors such as consumer awareness, product heritage, or reputation and consumer loyalty contribute to the essential health of the branded product (NPD, 1995). The Simmons Market Research Bureau's EQX equity measure tracks brand usage and then integrates more abstract notions including attribute importance and desired improvements (EQX, 1995). Total Research Corporation's Equitrend attempts to penetrate deep into some common attitudinal measures. Rather than ascertain the mere familiarity of a brand, these researchers delve into the "depth of awareness" and the "levels of perceived quality" in association with key variables such as price elasticity and brand loyalty. A byproduct of these calculations is a basic measure of user satisfaction (Total Research Corporation, 1995).

Product Evaluations Inc. provides a Product Q measure of brand equity that is highly attitudinal and is designed to calculate consumer enthusiasm-"The absolute number of people who think your product is terrific!" Using a nationwide survey technique, factors such as familiarity, experience, overall appeal, and company image are used to determine the level of enthusiasm (Product Q, 1995).





23


An upshot to the growing interest in brand equity in the 1980s was the

introduction of positioning theory, made popular by Reis and Trout in their 1981 best seller Positioning: The Battle For Your Mind. The authors contended that in a overcommunicated society, a company must create a position in the consumer's mind that takes into consideration not only the company's own strength and weaknesses but also those of the competition as well (Reis & Trout, 1981). They insisted that in a crowded marketplace, a branded product must be differentiated easily from its competitors. Although this book was long on conjecture and rather short on science, it remains must reading for aspiring corporate executives. Although the term brand equity was not mentioned in the original book per se, much literature on brand equity from both the private and academic sectors has incorporated brand positioning. Definitions From Academia

On a more scholarly level, a number of marketing professionals and academics have published books or journal articles on brand equity. For example, Axelrod (1993) takes a decidedly behavioral approach, emphasizing pricing variables. He proposes a conceptual definition of brand equity as "the incremental amount that a customer will pay to obtain a brand rather than a functionally equivalent alternative with a different brand name. The more an individual values a brand, the more he will pay for it" (p. 91).

Srivastava and Shocker (1991) offer two related perspectives on equity by distinguishing brand value from brand strength. The first is a corporate perspective where brand value is a financial measure dependent upon a





24


brand's current monetary strengths and future prospects as well as its financial status within a company's total portfolio of products. On the other hand, brand strength comes from a consumer perspective where a brand is evaluated according to "competitive positioning and imagery relative to customer demands and desires" (p. 6). Using this conceptual ization, one could surmise that consumer brand strength drives corporate brand value.

Alexander Biel (1991) posits that brand equity deals with the value,

usually defined in economic terms, of a brand beyond the physical assets with its manufacture or provision with stronger brands having more equity than weaker competitors. This strength ingredient can also be evaluated from a consumer perspective where brand image factors such as salience, trust, and richness drive brand equity.

Another way to address equity is to focus on consumer satisfaction.

There are multiple definitions of satisfaction in brand literature, but one that is suitable for the purposes of this study is that satisfaction is an outcome of a consumer's direct experience (trial) with a product (Yi, 1990). A positive experience will induce positive evaluations of the product and cultivate a preference in a future purchase situation. More concisely, one can say that satisfaction drives preference.

Satisfaction itself is not always a good predictor of consumer behavior. A number of studies have shown that, within a product category, many brands can be perceived as equally satisfying. Wilkie (1996) asserts that the good news for American business is that hundreds of consumer surveys by





25


academics, governments agencies, and businesses reveal that the overall consumer satisfaction level in the U.S. is extremely high. The resulting bad news is that cultivating brand preference is more challenging than ever. Other studies have demonstrated through controlled experiments that a brand can have a positive satisfaction level with a consumer and still not be purchased because competing brands are more accessible in memory due to heightened exposure (Farquhar, 1989; Fazio, Powell, & Williams, 1989). This heightened exposure is often the result of extensive brand advertising (Nedungadi, 1990). Wilkie (1996) concludes that for many product categories "brands are likely losing sales, not because they are not liked by consumers, but because they never come to mind during the consumer's (often rapid) decision process: They are simply never considered!" (p. 515).

The above studies reinforce the proposition that simple brand

awareness can have a profound effect on brand choice without the addition of complex images or associations. This of top-of-mind recognition (accessibility to memory) is powerful particularly in low interest-low involvement purchase situations. For example, Hoyer and Brown, (1990) proposed that in the interest of conserving time and effort, consumers will often use simple choice heuristics when making repeat purchase decisions. An experimental study involving the purchase of ordinary, inexpensive consumer goods revealed that when an inexperienced decision-maker is faced with choosing among several competing brands, a known or recognized brand is more likely to be the





26


ultimate choice--regard less of the relative quality of the other competing brands.

Today, brand managers continue to face the challenge of preventing customers from defecting to competing brands. A recent study by Grey advertising concluded that "'brand promiscuity' is a global phenomenon. That is, consumers from the U.S., to the U.K., to Australia routinely makes choices from a group of brands they consider acceptable" (Shell, 1997, p. 40). If all brands within a product category are seen as equally satisfying, then consumers become vulnerable to competitive strategies such as pricing, promotions, and convenience, none of which is an ingredient of consumerbased brand equityTo many marketing researchers, the opposite of "brand promiscuity" is

brand loyalty. Christiani (1993) of General Foods USA confesses that one of the most "elusive and debated" subjects surrounding brand equity is the "intangible asset consisting of a brand's consumer loyalty base" (p. 123).

Aaker (1991) includes brand loyalty as the centerpiece of his definition of brand equity which consists of five broad categories: brand loyalty, brand awareness, perceived quality, brand associations, and other proprietary assets such as patents. All of these brand characteristics influence the consumer's purchase decision and usage satisfaction, but brand loyalty is the force that makes a brand truly powerful. He believes that customer brand loyalty is essential because it reflects the likelihood (or probability) that a customer will switch to another brand. This notion of loyalty surfaces often in research





27


literature. Maturo (1993), a marketing director for Nielsen Household Services (not to be confused with Nielsen Media Research), asserts that dozens of recently performed analyses of brand choice modeling revealed that brand loyalty was a consistent leading predictor of future brand purchases. Other respected private research organizations such as the NPID Group and Lintas advocate consumer loyalty as a core component of any brand equity model (Lintas, 1993; NPID, 1995).

Inertia

There is an obvious intuitive acceptance that loyalty must somehow be an integral part of equity, but several researchers have warned that loyalty should not be misconstrued with mere repeat purchasing. According to Donius (1994), information concerning simple habitual repeat purchasing may obscure a vulnerability to switch brands in the future. He states that "loyalty can also be inertia, not necessarily preference" (p. 56). The presumption is that in some purchase situations there are noncommitted consumers that can be motivated to choose a brand by simply controlling key structural factors that have little to do with a brand's perceived value. According to Ceurvorst (1994) and Beatty, Kahle, and Homer (1988), the underlying cause of brand equity is the core psychological relationship of commitment of which loyalty is the behavioral result. Ceurvorst (1994) states that commitment provides "the essential basis for distinguishing true loyalty from other forms of repeat purchasing" driven by factors such as inventory, promotions, and discounts (p. 2).





28


Although not a full-fledged marketing theory, the concept of inertia is

referred to in several marketing and mass communication publications. Long ago, broadcasters realized that the reason people watch a program is often dependent simply on its lead-in program. This inheritance effect is sometimes referred to as tuning inertia. Its definitional roots are in the natural sciences. A typical definition of inertia borrowed from the domain of physics is the following:

Inertia is the property of matter which manifests itself as a resistance to change in the motion of a body. Thus when no external force is acting, a body at rest remains at rest and a body in motion continues moving in a
straight line. (Encyclopedia of Science and Technology, 1997)

From a perceptive of consumer behavior, we can define our consumer as the "body" and the forces of inertia as those marketing factors that encourage "resistance to change" or habit. In order to have the body change direction there must be an adequate "external force" which can be defined as a competing brand. In a later section we elaborate more on this concept introducing momentum as a measure of inertia. Brand Equity According to Keller

Our final conceptual definition of brand equity, and the one chosen to be the theoretical underpinning for our exploration of television program equity, comes from Keller (1993) who conceptualizes brand equity according to two kinds of memory associations, brand awareness and brand image. The two in combination are called brand knowledge. Brand equity is the differential effect of brand knowledge on consumer response to the marketing of a brand. Awareness is the first step in the process where measures such as recall and





29


recognition are introduced. Image deals with the meaning of a brand to a consumer through various associations such as attributes, benefits, and attitudes. These associations can vary according to their favorability, strength, and uniqueness and can play critical roles in determining purchase outcomes. Keller states, "The presence of strongly held, favorably evaluated associations that are unique to the brand and imply superiority over other brands is critical to a brand's success" (p. 6). Keller also asserts that "fundamentally high levels of brand knowledge (awareness and image) increase the probability of choice, as well as produce greater consumer loyalty and decrease vulnerability to competitive marketing actions" (p. 3).

From a more behavioral vantage point, Keller describes how these brand associations can be reflected in the marketplace behavior.

A brand is said to have positive (negative) consumer-based brand equity
if consumers react more (or less) favorably to the marketing mix of the
brand than they do to the same marketing mix element when it is
attributed to a fictitiously named or unnamed version of the product or
service ... If a brand is seen by customers to be the same as a
prototypical version of the product or service in the category, their
response should not differ from their response to a hypothetical product
or service. If the brand has some salient, unique associations, these
responses should differ. (p. 4)

In other words, if the marketing mix is controlled by the researcher, the attitudinal component of brand equity, namely, brand knowledge, will manifest itself in consumer behavior.

Operational Definitions of Brand Equity

The specific measuring techniques for brand equity depend on the particular theoretical approach. One broad approach takes a corporate





30


perspective and dwells on the financial value of equity, examining items such as stock prices, momentum accounting, and cash expenditures needed to introduce brand extensions. Another broad approach and the direction of this study is consumer-based brand equity, where consumer attitudes (perceptions) and behaviors are the focus of study. Within this second approach there is little agreement among researchers as to the ideal operational izations of equity. Also, there are scant academic (nonproprietary) consumer-based equity studies that have actually attempted to measure the influence of equity using real world products and consumers.

Interviews or self-administered questionnaires are typically used to

obtain attitudinal information. For example, Cobb-Walgreen, Ruble and Donthu (1995) were interested in the power of brand equity to influence brand preference and purchase intent. Hotels and household cleansers were selected as brand categories for study. Using the perceptual components of Aaker's (1991) conceptualization, the researchers employed survey instruments that attempted to capture brand awareness and brand associations. Respondents were asked to list as many brands of a specific product as they could to determine top-of-mind awareness. For brand associations, the respondents were asked to list descriptive thoughts, words, characters, symbols, or images that came to mind for each brand. To assess preference and purchase intent, the respondents were presented with hypothetical decision-making situations in the form of a story. The importance of product attributes, favored brands, and intention to purchase were





31


operationalized through 7-point Likert scales. Another variable, advertising awareness, was obtained by asking respondents if they had ever seen advertising for a target brand and if could they recall any salient copy points. Results indicated that higher advertising budgets yielded higher equity scores, and, in turn, higher equity generated significantly greater preferences and purchase intentions.

Testing an equity model that incorporated components of consumer preference and consumer satisfaction, Anantachart (1995) looked at the relative consumer equity of various soft drinks among college students. A number of established 5-point scales were borrowed from other academic research and were modified to work with this model. Additionally, new scales were created to measure items such as product consumption. As a model for detecting differential response among brands, the results were highly predictive.

Keller (1993) operational ized equity from two vantage points. He sees attitudinal or perceptual measures of equity (brand knowledge) as "indirect" operationalizations. Conversely, measures of consumer behavior (or the impact of brand knowledge) are viewed as "direct" operation alIizations For example, indirect measures would include instruments that capture brand name recall and various image dimensions such as favorability, strength, and uniqueness of brand associations. Direct measures deal with actual consumer response (choice) to marketing activities. Keller recommends various experimental and q uasi-experi mental settings where the marketing mix





32


(product, price, distribution, and promotion) of brands can be controlled systematically to assess the relative impact of brands on market performance. The equity model proposed in this dissertation adopts this direct approach of analyzing consumer response. Although a pure experimental setting was not feasible, the methodology did impose appropriate controls on the analysis of secondary data.

We have seen that there is no universally accepted detailed

conceptual ization of brand equity, However, it seems safe to assume that most equity researchers would agree on the outcomes of having high consumerbased brand equity. The market performance of a brand with strong equity is characterized by the following:

1. Long-term market share stability and growth. While weekly or monthly fluctuations can occur, the performance of these brands over the long run, namely, years, is unwavering.

2. Dominant market rankings. Brands attributed with high equity not only demonstrate stability over time, but they tend to be long-term market leaders, achieving high market ranking among direct competitors. This status is reflected in increased revenue, lower costs, and greater profits.

Behind this exceptional brand performance, equity researchers would also agree that there is a consumer base that holds the brand in such high esteem that they buy the brand exclusively, regardless of competitive marketing actions.





33


Inheritance Effects

For decades, the single best predictor of a program's ratings

performance has been the ratings of the program scheduled immediately before it. That is, television programs tend to "inherit" sizable audiences from the program airing immediately prior to it on the same channel. This carryover phenomenon is known as inheritance effects or tuning inertia.

While it would be foolish for programmers not to take advantage of this scheduling tactic, Webster and Lichty (1991) warn that some people

assume the choice of a program centers upon the active expression of a
preference for a program or type of program. However, so called
structural factors have traditionally been considered important mediators
of the programs viewers choose and complicate the relationship
between viewing preference and viewing behavior. (p. 178) Overview of Research

While the overall impact of inheritance effects has been confirmed

myriad times by industry and academic researchers, the results have been puzzling because of the considerable variances among studies. Beginning in 1975, Goddhart, Ehrenberg, and Collins essentially coined the term when they worked on the broader issue of audience duplication. Based on television viewing in the United Kingdom, the researchers proposed a Duplication of Viewing Law which stated the proportion of the audience of any program who watch another program on another day of the week is directly proportional to the rating of the later program, times a constant. Furthermore, they discovered that when programs were adjacent to each other, an "inheritance effect" took over that exceeded the predictions derived from the original duplication law.





34


Headen, Klompmaker, and Rust (1979) proposed an improved model incorporating five independent variables--ratings, channel, program type, daypart, and repeat viewing. An examination of over 4,000 combinations of pairs of programming using Simmons Market Research data revealed that by far ratings were the single best predictor. A different model offered by Webster (1985) included the contributing factors of audience availability (which in most cases is a fixed quantity), lead-in program ratings, the number of program options, and program content. Using Arbitron ratings from one sweep period in Portland, Maine, Webster concluded that for adjacent program pairs, lead-in ratings and the number of program options in combination explained 80% of the variance.

Tiedge and Ksobiech (1986) conducted a massive 22-year study of network prime time programming from 1963-1 985 looking at the effects of lead-in program shares, available program options, and similarity of program types on inheritance effects. The conclusions were that programs with highranked lead-ins scored an average of 6.8 share points higher than those with low ranked lead-ins. Also, fewer program options produced higher lead-in correlations and visa versa. In 1988, the same research team using the same ratings data set investigated the program strategy of "sandwiching" (hammocking) where the influence of both lead-in and lead-out programming were analyzed. Their conclusions were that the effects of lead-out were minimal and that a strong lead-in was essential to have a sandwich strategy work properly (Tiedge & Ksobiech, 1988).





35


Walker (1988) looked at nine years of Nielsen ratings from 1976 to 1985. The correlational relationships among inheritance effects, lead-in, program type, and number of options followed identical patterns observed by Tiedge and Ksobiech (1986). Examining total audience exposure among syndicated programs rather than network shows, Cooper (1993) correlated the influence of several variables on program ratings. These included lead-in, lead-out, number of options, program type compatibility, network affiliation, and cable penetration. The results from a 50-market sample of May sweep books revealed that while some variables such as network affiliation had minor measurable influence, lead-in ratings were by far the strongest predictors of a syndicated program's ratings and appeared to be "strong surrogates for other variables in the model ... and completely overwhelmed any other factor in the model" (p. 409).

Viewer Motivation and Loyalty

The variances in results found among inheritance effects studies have not been explained adequately using structural factors such as program type, number of options, channel, and network affiliation. Part of the problem may involve the viewing motivation of the audience member. For example, Rubin (1984, 1994) divides television program viewing motivations into ritualized and instrumental categories. The first denotes a viewer that uses television out of habit and to pass the time (relieve boredom, entertainment, amusement, etc.). These "escapist" audiences place more emphasis on the medium itself rather than specific program content. The second type of motivation (instrumental), is





36


more goal-oriented where the viewer is seeking program content to gratify a specific need and is, therefore, more discriminating. From an inheritance effects perspective one could hypothesize that the ritualized viewer would tend to remain on the same channel while an instrumental viewer might hold certain program loyalties that would encourage more channel switching.

The notion of audience loyalty was broached by Boemer (1987) when he took a 2-year look at local late night newscasts in Dallas, Texas. Using Arbitron sweep ratings, he found high positive correlations between newscasts and prime time lead-in programming. However, he also found considerable variances among the three competing stations ranging from .30 to .69. Additionally, he found that some newscasts delivered better ratings than their lead-ins, providing "circumstantial evidence of a loyal viewership for local news in the market" (p. 93).

Inheritance effects (tuning inertia) can play a crucial role in determining the ratings performance of a television program. While the impact is noticeable, there is considerable variance among programs as to how effectively they "inherit" audiences. That is, some programs appear to capitalize on a lead-in audience better than other programs. Conversely, some programs appear more resilient than others to the effects of a poor lead-in. An explanation may come from the ranks of brand equity theory.





37


Applying Brand Equity to Television Program Performance Before the "Buzz" There Was Audience Loyalty

The desire to nurture favorable, strong, and unique program

associations has always been a concern of broadcasters. Before the buzz surrounding media as brands, television managers tried to gauge audience loyalty. For many years, major market television stations have enlisted the services of consumer research companies to conduct attitudinal and behavioral surveys. While Nielsen rating points remain the ultimate measure of program performance, broadcasters also scrutinize other research data in order to make better programming and promotion decisions.

An example of one such company that provides proprietary information is Marshall Marketing and Communications (MM&C), a nationally recognized broadcast marketing and communications consulting company based in Pittsburgh, Pennsylvania. Approximately 100 television stations throughout the country subscribe currently to the service. On a market-specific basis, MM&C provides annual surveys of consumer purchasing habits and attitudes for preselected categories in retail, service, and merchandise industries. Additionally, the company provides analysis of consumer communication habits encompassing, radio, television, newspaper, cable, and direct mail. Within the television viewing domain, MM&C targets specific programs, including local newscasts (Marshall Marketing, 1996).

Marshall Marketing and Communications conducts several hundred indepth telephone interviews over a three- to five- week time span. One specific





38


research area is consumer loyalty to specific brands. Loyalty is defined conceptually as recurring and exclusive purchasing. That is, people who are "heavy users" of a product category and purchase only one brand over a considerable time period are considered loyalists. The operational specifics vary depending on the type of product and typical buying cycles. For example, some products tend to be a daily or weekly purchase, while others tend to be more of a monthly purchase. Daily versus weekly scheduled television programs can be regarded in the same light. Although MM&C has yet to adopt branding jargon, this measure of consumer loyalty is consistent in many respects to Keller's (1993) conceptualizations and can serve as a comparative benchmark for our case study examining ratings performance.

In the final analysis, the media's infatuation with the buzz of brand management is spawned by the proposition that brand concepts such as brand equity will help protect and enhance a program's ratings performance. While the jargon of brand management is prevalent throughout the broadcasting and cable industries, there has yet to be a systematic effort to translate properly many of these precepts for use by the media. Borrowing from the work of Keller (1993) and proponents of consumer inertia, this section examines the plausibility of applying brand equity theory to television programming. Network versus Station versus Program Equity

While brand equity theory could be applied to several branding units, the latest academic literature indicates that, for the most part, people still watch programs before networks, stations, or channels. A study by Abelman, Atkin,





39


and Rand (1997) revealed that the majority of viewers "are relatively indiscriminate consumers with regard to the source of programming. They possess little network or station affinity" (p. 377). A small group within the study did exhibit station affinity, but an important caveat was that the station offered preferred programming content. For example, station affinity was attributed to a preference for local newscasts, not necessarily to the station itself. While the assessment of brand equity for stations and networks is a worthwhile topic for future research, this paper concentrates on brand equity issues involving the most basic equity unit, the individual program. Adapting Brand Terminology

By conceptualizing a program as a product or service and the program's audiences as consumers, most branding terminology can be adapted readily to broadcast programming. On a fairly cursory level, the following concepts have been redefined for broadcasting.

Brand awareness. The recall or recognition of a network, station, or program name,

Brand extension. The imposition of a brand name over several program it products" such as a network ID.

Brand trial, The first experience a viewer has with a program (sampling)

Brand image. The thoughts or feelings (meanings) that come to mind when the program name is mentioned.

Brand positioning. The image of the brand defined in relationship to its direct market competitors.





40


Brand attitude. The evaluation of a program (brand images) and predisposition to watch.

Brand loyalty. The degree of repeat viewing of a program at the exclusion of direct competitors.

Brand commitment. The psychology underlying program loyalty. The degree of fidelity to a program despite competitive program scheduling, promotions, lead-ins, and so forth.

There are other useful brand marketing parallels between conventional consumer goods and broadcasting. For example, there are obvious similarities between a viewer's limited channel repertoire and a consumer's evoked set of brand selection. Also, the precepts of consumer habit formation through simple operatant conditioning can be seen in how broadcasters reinforce positive viewer outcomes (or satisfaction) by manipulating program content and scheduling.

Controlling the Marketing Mix

Keller (1993) insists that brand equity cannot be assessed properly

unless the marketing mixes of the competing brand competitors are equivalent or at least controlled by the researcher. The marketing mix components of product, price, place (distribution), and promotion can be translated into a broadcasting domain in the following manner. The produce

The product is the specific brand of program content under scrutiny. We mentioned earlier that in order to understand brand equity, it is necessary to





41


deal with direct competitors, that is, products or services that reside in the same product category. Just as there can be no meaningful comparison of brand equity between packaged goods, such as toothpaste and pantyhose, so we should not attempt equity comparisons between disparate television genres, such as newscasts and soap operas. The RL 2t

Unlike most consumer goods, pricing is not a paramount concern for broadcasters. Even when broadcast stations are included on subscription cable services, there is a general perception that the broadcast channels are "free," However, basic and premium cable programmers do encounter pricing dilemmas with their paying consumers. For the purpose of this study, we are excluding the pricing component.

The place (distribution)

For electronic media, controlling the distribution component of the

marketing mix is challenging but essential to understanding true equity. In an ideal test environment, a program's direct competitors should be equally available to all consumers. From a retail perspective, this would seldom be a major obstacle, but for broadcasters there is an obvious problem, Unlike most consumer goods, broadcast "products" are time-bound and are not available simultaneously to audiences. (Imagine a retail outlet where the shelf displays changed every half hour to display different competing brands.) Attitudinal studies can ignore this dilemma, but a behavioral approach to assessing





42


consumer-based brand equity requires that not only program content be similar but also program scheduling as well.

Program brand distribution (availability) also includes technological

factors. Direct competitors may be scheduled at the same point in time but the electronic delivery systems may not be equivalent, thereby biasing a test situation. Variables, such as geographical signal coverage and cable retransmissions, can influence the distribution of a television program.

A final distribution factor is consumer traffic flow. Just as some

consumer brands in a retail setting have better store locations or better in-store shelf positioning, so certain television programs are exposed to better audience "traffic" than others. Audiences already present in a lead-in program on the same channel can be construed as more available than audiences found on other channels. From our earlier review of inheritance effects research, we know that the mere proximity of a large lead-in audience can offer a distinct advantage.

The promotion

Another component of the traditional marketing mix is promotion, which includes communication actions such as advertising, publicity, and, most importantly, sales promotions. Coupons, sales, contests, and other "tactical gimmickry" are intended to stimulate product trial (sampling) and boost temporarily market shares but seldom cultivate long-term brand loyalties (Keller, 1993, p. 6).





43


In the quest for short-term ratings gains during the infamous Nielsen "sweep weeks," television networks and stations resort to a variety of programming and promotional tactics that are designed often to distort typical program performance. These tactics range from altering program schedules and manipulating story lines to atypical advertising campaigns and on-air contests. Known by industry insiders as hyping, or hyping, these practices have been condemned by Nielsen and the advertising community (Gimein, 1996; McDowell, 1995; Miles, 1997). The best means for neutralizing the biasing effects of broadcast promotion is to look at long-term performance where transitory promotion tactics become random variables.

In summary, the marketing mix for electronic media must somehow be held in check in order to evaluate program equity. Therefore, in an ideal research setting, the following criteria should be applied: (a) identical or highly similar target program content (derived from same "product category"), (b) identical pricing considerations (same price or free), (c) identical distribution of target programs in terms of scheduling, signal coverage, and audience flow, and (d) identical or neutralized short-term promotional tactics of competing target programs.

Inertia, Momentum, and Equity

The well-doCumented influence of lead-in program ratings implies that there are significant numbers of passive or uncommitted viewers who are not motivated to change channels. The magnitude of this inertia or "resistance to change," however, has been difficult to predict in terms of rating points. Given





44


the same lead-in rating points, some programs appear to "inherit" more points than others. For example, a program with a big lead-in rating may be unable to capitalize fully on this advantage, while another program dealing with a much smaller lead-in can overcome this apparent disadvantage and exhibit impressive ratings performance.

Using Keller's (1993) conceptualizations of brand equity we can offer a partial explanation for these inconsistencies. A program that is unable to generate familiar, strong, and unique brand associations among its potential viewers is vulnerable to the program content on competing channels. Likewise, a program demonstrating strong equity characteristics will retain more of its lead-in audience and recruit audiences from other channels.

Earlier we demonstrated how research in both brand equity and inheritance effects have employed the concept of inertia to explain a consumer's resistance to change behavior patterns. For television, this reluctance to change channels has been "coined tuning inertia," However, inertia alone may not be as worthy a metaphor as another term borrowed from physics, momenhim. Intended as a measure of "how much" inertia is present in a body, momentum consists of two components, mass and velocity (Encyclopedia of Science and Technolog 1997). Based on this simple formula, we can understand better how two bodies with the identical mass can exhibit different degrees of momentum, depending on their relative speed. Similarly, measures of mass alone cannot predict how susceptible a body is to





45

"external forces." Only when the mass and the velocity are measured in tandem can we assess properly the power of inertia on a body.

From a consumer behavior perspective, we can surmise that inertia is not necessarily a bad thing. Rather, it is the underlying consumer momentum factors that must be evaluated: (a) the brand's ability to retain current customers in the face of external forces of direct competitors and (b) the ability of a brand to recruit customers from competing brands. The repeat customers of a brand exhibiting poor equity are vulnerable to a competitor's marketing mix and, therefore, have less momentum than a brand with stronger equity.

In a similar vein, one could propose that tuning inertia (inheritance

effects) is influenced by the two "momentum" factors of (a) retention of lead-in audiences and (b) recruitment of audiences from other sources. This approach explains how two programs with identical lead-in ratings can achieve different results. Equity, then, can be viewed as a program's ratings momentum. Interfacing Momentum with Keller's Conceptual izations

The above notions of inertia and momentum interface well with Keller's (1993) approach to brand equity. That is, if a consumer does not perceive any 4L salient and unique associations" from a brand name, there will be a resistance to change purchase behavior. This momentum can be disrupted by the external force of a competing brand whose marketing mix generates favorable, strong, and unique associations that "imply superiority over other brands." Similarly, a television audience member that is currently viewing one particular channel will remain on that channel unless the "external forces" from





46


a competing channel intervene sufficiently to motivate the viewer to change channels.

Equity and Performance

We will recall that for Keller (1993) brand equity is the differential effect of brand knowledge on consumer response in a competitive marketplace. This brand knowledge includes (a) awareness and (b) image associations. The more favorable, strong, and unique these associations, the more likely a particular brand will be chosen over its direct competitors. This consumer response (choice behavior) can be translated into measures of brand performance in the marketplace. An essential caveat to assessing this performance is the need to control the marketing mix components of direct competitors.

Given this theoretical framework, television programs with strong

consumer-based brand equity should perform better in the marketplace than programs with weak equity. That is, programs which elicit from viewers favorable, strong, and unique associations will have a greater probability of being chosen over direct competitors. Again, control of the marketing mix elements is crucial in evaluating performance. Operational izi ng program performance logically should involve measures of audience program choice, and for the past four decades within the broadcasting industry, the most recognized measure has been program "ratings" provided by Nielsen Media Research.





47


The Nielsens: Measuring Program Performance

By adopting a behavioral approach to understanding program brand equity and insisting that this work be relevant to the real-world business environment of the broadcasting industry, Nielsen ratings were selected as the most appropriate measure of program performance. The word rating is often misunderstood outside the broadcast industry because the more vernacular use of the term means a subjective judgment or opinion. That is, people are asked often to "rate" their favorite movies, restaurants, or rock groups. Broadcast and cable ratings, however, are simply estimates of audience tuning behavior, or media exposure. Ratings are not intended to be a valid attitudinal measure except in the sense that consumer behavior can be inferred to be what Webster and Lichty (1991) and other researchers call "revealed preference" (p. 27). The following is a brief synopsis of Nielsen research methods.

In addition to the well known national "People Meter" ratings, Nielsen also provides local program ratings to over 200 individual markets. Using a multistage clustering sampling technique, all markets experience the monthlong, diary-driven "sweeps" four times a year (November, February, May, and July). An alternative to the sweeps for many major market stations is metered it overnight" ratings. Presently, 42 local markets are measured daily, using a passive electronic metering system that continuously records household viewing. The viewing data are processed "overnight" and then faxed the following morning to subscribing stations. From a reliability perspective, one of





48


the major criticisms of the sweeps methodology is that it covers only 16 weeks of the year and that these few weeks often are dominated by programming and promotional hype that may distort results. Furthermore, program performance data by individual days are not provided in the standard reports. Instead, ratings are presented as four-week averages. Consequently, whenever possible, this study utilized the more reliable overnight ratings which are not averaged and offer continuous, long-run data.

The fundamental units of measure used in almost all ratings-based research are households (000), rating points, and shares points. While households are expressed as whole numbers, rating and share points are percentage points, A rating expresses a program's audience as a percentage of the total population. A share expresses the same audience but as a percentage of the households (or persons) using television (HUTs or PUTs). Age and gender demographics can also be translated into rating and share points (Nielsen Method, 1995).














CHAPTER 3
RESEARCH HYPOTHESES

From the basic research question, how does brand equity influence program ratings performance, a number of exploratory hypotheses can be offered that resonate with our proposed theoretical assumptions. The literature review contends that brands exhibiting strong equity tend to be market leaders and that, over time, short-term fluctuations in performance give way to a longterm durability in market rank. In other words, market leadership endures; therefore, the following hypotheses were proposed.

Hl: Programs with higher brand equity will exhibit higher broadcast ratings than programs with lower brand equity.

H2: Programs with higher brand equity will exhibit higher broadcast shares than programs with lower brand equity.

H3: Programs with higher brand equity will exhibit higher household audiences than programs with lower brand equity.

The literature review also focused on the relationship of a program to its lead-in (inheritance effects) and recognized the existence of a kind of tuning inertia where viewers resist changing channels. However, a closer examination brought to light the concept of momentum which suggests that, given the same size lead-in, programs with strong equity will perform differently than programs with weaker equity. That is, consumer-based brand equity can be found in the


49





50


differential ratings response between two adjacent programs. The following two hypotheses were offered.

H4: Programs with higher brand equity will reveal a significantly greater difference between lead-in program audience size and program audience size than programs with lower brand equity.

H5: Programs with higher brand equity will reveal a significantly greater differential share index than programs with lower brand equity.

Narrowing our focus to the dynamics underlying this differential response, the literature review proposed that program brand equity is a function of (a) the power to retain lead-in audiences from the same channel and (b) the power to recruit audiences from other sources. These two components address the momentum of a program as it passes through its competitive program environment. Accordingly, two additional hypotheses were offered.

H6: Programs with higher brand equity will retain a greater proportion of their lead-in audiences than programs with lower brand equity.

H7: Programs with higher brand equity will recruit a greater proportion of audiences from other sources than programs with lower brand equity.













CHAPTER 4
RESEARCH METHODOLOGY

Because of the exploratory nature of this study, a case study format is appropriate. While there is an extensive statistical analysis involved with this case study, the goal was not to generalize statistical results but rather explore theoretical possibilities within a fairly controlled environment. Respected case study researcher Robert Yin (1994) claims that

Case studies, like experiments, are generalizable to theoretical
propositions and not to populations or universes. In this sense, the case
study does not represent a 'sample' and the investigators goal is to
expand and generalize theories and not to enumerate frequencies (p.
10)

The hypotheses serve as initial "tests" of some theoretical offerings but do not exclude unexpected findings that may stimulate further exploration of program equity.

There were two primary sources of data for this study: (a) Nielsen
it overnight" metered ratings and (b) Nielsen "sweep" ratings. As presented leader in the literature review, these data ]represent two different methodologies for generating and reporting program ratings performance. Whenever possible, the more reliable continuous overnight ratings were used, but'there were several Situations where only sweeps data were suitable (the specific data sources are disclosed on all tables and within the appropriate text portions of the study). The basic units of analysis for program performance were 51






52

household ratings (it), household shares (sh), and total households (hh). Again, the prior literature review offers precise statistical definitions of these terms. Demographic information, such as gender and age, were not available from the overnight metered data.

With the written permission of Nielsen Media Research, ratings data for January through December 1996 for one major television market were turned over to the researcher for analysis. This included 365 days of overnights and four-month-long sweep periods (February, May, July, and November). Because these periodic reports are used primarily to estimate ongoing and future program performance, Nielsen averages the four-week ratings in order to increase reliability.

In order to test the hypotheses presented in this study, these basic units of analysis were sometimes manipulated statistically to reveal greater insight. For example, the influence of program brand equity on program ratings performance was evaluated from a number of perspectives. The first was a comparison of newscast performance based on conventional descriptive statistics of rating, share, and households. The second was an analysis of the differential ratings responses between newscasts and their respective lead-in programs (inheritance effects). This was accomplished through analysis of variance, indexing, and correlational analysis.

To assess a program's ability to retain lead-in audiences and recruit audiences from other sources, Nielsen provided a custom "Audience Flow Analysis" based on the May 1996 sweep period. This analysis tracks the







53
source and destination of the viewing audiences between an adjacent quarterhour. For this study, our focal point was the household "flow" from 10:45 PM to 11:00 PM. Regrettably, Nielsen does not provide such flow studies using overnight data, but the four-week sweep data still provided worthwhile insight.

This test market was chosen specifically for reasons which are

presented in subsequent portions of this methodology section. For reasons of confidentiality, the name of the market was not disclosed in the study, and the target stations were identified merely as stations A, B and C.

Controlling the Marketing Mix

Based on Keller's (1993) conceptualizations and discussions in the prior literature review, the differential effect of brand knowledge on consumer response must be measured under controlled circumstances. These controlled circumstances include marketing mix components where the target brands must be (a) direct competitors in terms of product category, (b) identical in price across all competing brands, (c) equally available in terms of distribution to consumers, and (d) unaffected by short-term promotion activities.

In a broadcast programming context, a direct competitor would be a

program of highly similar content, such as a local newscast. Equal availability would be defined as competing programs that are scheduled at the same day and time in the same market airing on comparable facilities. Typical transitory promotions would occur during the periodic "sweep" weeks.






54

To satisfy the above criteria, the 1996 Nielsen ratings performance of the late evening newscasts on three competing VHF stations in a top 20 southeast television market were selected for this study. These three stations are well established network-affiliated competitors that have been offering 11:00 PM newscasts for over 20 years. Their terrestrial signal coverage and cable penetration levels were highly similar. In fact, two stations share the same transmitting tower. Furthermore, from an audience perspective, the average weekly cumulative household delivery (Monday through Sunday, 6:00 AM to 1:00 AM) was almost identical for these target stations. That is, over the course of a typical week, each station reached the same number of individual (unduplicated) households within the market (Nielsen Cume, 1996).

In addition to obvious direct brand competition in terms of program

content, scheduling, and signal coverage, there was the added advantage of each newscast experiencing a different lead-in seven nights a week. While most stations in the country broadcast several news programs throughout the day and evening, the ratings performance of the 11:00 PM (10:00 PM central) newscast is considered often a primary indicator of a station's vulnerability to competitive attack (Eastman, 1997). With a different lead-in program every night, the late news is the most susceptible time slot for competitive program sampling by docile audiences that are not particularly committed to any one brand of newscast.

To assure genuine direct competition, the ratings database was further refined by extracting dates when the three newscasts did not compete head to







55

head at 11:00 PM. Throughout the year stations are often forced into "late starts" due to extended movies, specials events, and sports. For example, the ABC affiliate had to contend with late starts for a dozen weeks due to Monday Night Football, thus destabilizing the competitive marketing mix at 11:00 PM. Of the 365 available dates for study, 289 were found to offer the ideal competitive environment where at 11:00 PM news viewers had three legitimate news options.

To neutralize the possible bias of short-term promotion activities, two procedures were introduced. First, whenever possible, the continuous overnight ratings were used instead of the periodic sweep weeks, thus diluting the statistical influence of hyping. Table 1 provides a breakout of the overnight database.


Table 1

Breakout of Nielsen Metered Overnight Database Number of original cases 364
Number of cases of direct 11:00 PM News 289 competition among three stations
Number of direct competition cases during 86 sweep months
FN7mber of direct cases outside sweep mo-ths7 203


A second precautionary procedure involved a direct comparison between sweep weeks and nonsweep weeks data which revealed no significant differences for 1996, Therefore, the 16 sweep weeks were retained in the overall database. Tables 2 and 3 offer station -by-stati on comparisons.







56

Table 2

Comparison of Sweep versus Nonsweep Cases: 1996 Descriptive Household Data

Sweeps Nonsweeps ____Station Mean SID Mean SID%
_______ (000) ____ (000) _____difference
A 127 26 132 27 4.0
B 114 33 115 31 1.0
C 89 21 92 23 3.0

Note. Sweep cases, N =86; nonsweep cases, N = 203.
Mean scores based on Monday through Sunday newscasts.


Table 3

Comparison of Sweep versus Nonsweep Cases: Paired Sample T-Test by Households

Station t- value 2 tail sig.
A -1.19 .239
B j -.25 .806
C J -1.00 .321

Note. Sweep cases, N = 86; nonsweep cases, N = 203


Establishing a Comparative Equity Benchmark

In order to test the relative influence of brand equity on program

performance, it was first necessary to identify a benchmark newscast that demonstrated superior equity characteristics based on information garnered from sources independent of the one-year Nielsen overnight ratings used for this study. Based on the prior literature review, the following two criteria should suffice as the basis for such a comparative benchmark.







57
1.- Significant long-term market dominance. From an outcome

perspective, consistently high market rankings over several years is a valid indicator of strong equity.

2. Stated brand loyalty or commitment by heavy users of a product. Bypassing broadcast ratings data, a consumer survey that provides program viewing loyalty information can serve as an alternative measure of equity.

Benchmark #1: Nielsen 1 0-Year Audience Trend Analysis

As stated in the literature review, one of the primary advantages to high brand equity is long-term market share stability. Over time, strong brands exhibit a resiliency to market fluctuations and competitive attacks. Table 4 provides a 10-year sweep summary (1986-1996) of late evening newscasts in our test market by rating, share, and households.


Table 4

1 0-Year Monday through Friday Average Sweep Performance of Three Newscasts (February 1986-November 1996)

Measure Station A Station B Station C IF ratio Prob. Scheffee at .05
Rating & 11.9(1) 10.6(2) 7.8(3) 37.6 .000 A -B, B -C,
(ra nk) __ _ C-A
Share & 27.3 (1) 24.3 (2) 17.5 (3) 64.0 .000 A B, B C,
(ran k) I______ I______ I_ _I__I C-A
Household 116 (l) 99 (2) 71 (3) 28.2 1.000 IA -1B,B -C,
& (rank) I_____ I_____ I____ ___ __ C-A

Note. N = 39 sweeps or 116 cases, df = 115, p level = .05
The following sweeps data were not available: July 1986, November 1988 and
1991, and May 1986.
Households = (000)






58

Because overnight metered ratings were not available in this market until the mid 1990s, this trend analysis is derived from conventional sweeps data. Of the 40 consecutive sweep periods analyzed, station A seldom lost its number one ranking. Additionally, over the 1 0-year period, the overall differences among the three stations was found to be statistically significant in all cases.

The investigator realizes that the use of these 10-year Nielsen ratings to predict the outcome of current overnight Nielsen ratings could be construed as a tautology in that both equity and performance are evaluated using the same units of measure. Although these data do demonstrate Station A's long-term stability, the following benchmark offers better evidence of brand equity utilizing a completely different methodology.

Benchmark #2: Marshall Marketing Loyalty Survey

In addition to a 10-year trend analysis, high equity for station A was

reaffirmed using data collected from Marshall Marketing and Communications. As was discussed in the literature review, MM&C conducts annual surveys of consumer purchasing habits for select markets around the county. In 1996, the same year as the Nielsen ratings data, MM&C conducted such a survey in the test market. Over 1,000 telephone interviews were conducted over four weeks. Sample sizes were predetermined to yield a 2% to 4% margin of error within a 95% confidence level. Recurring and exclusive viewing were defined operationally as watching only one station's newscast at least three times over the past seven days. Table 5 presents the results of the 1996 survey. The first






59

two columns show that among subjects who watched at least one local newscast during a prior week, Station A dominated the category by a 2 to 1 margin over either competitor. Sixty-nine percent of the responses included a Station A newscast. Taking a narrower look at the exclusive viewing habits of heavy users, the remaining two columns of the table reveal Station A as the market leader, garnering over 31 % of the adults who watched only one station for local news. Here the margin for Station A was three to one over its competitors. Presuming a controlled marketing mix where all stations were equivalent in terms of time period, signal coverage, and market conditions, these data support our earlier position. Station A demonstrated the highest brand equity.


Table 5

1996 Marshall Marketing and Communications Survey on Exclusive (Loyal) Viewing of Newscasts

Stations Watched at % of once Watched % of Exclusive
least once per week exclusively viewing
during past viewing" 3 or more times
week during past week
Combined 562* 298 53.0
A 388 69.0 177 31.5
B 191 34.3 57 10.1
C 191 34.2 64 11.4

*Original sample frame, N = 1,200. **Total % of three stations exceeds 100% because of viewing of multiple stations during week.


Having established, at least tentatively, station A as the comparative equity benchmark for the market, we can approach the research question of







60

how does program brand equity influence program ratings performance on an everyday basis.














CHAPTER 5
RESULTS

Hypotheses One, Two, and Three

The initial three hypotheses predicted a program with higher brand

equity would outperform its direct competitors across a number of descriptive audience statistics including, ratings, shares, and households. Analyzing the adjusted 1996 overnight ratings data, the hypotheses were supported on all dimensions. Table 6 reveals that for the categories of (a) overall performance and (b) individual day performance, station A (our designated equity leader) achieved the majority of number one rankings in households, ratings, and shares. Furthermore, ANOVAs of these differences in program performance indicate they were significant. A significant f statistic, however, indicates only the station means are probably different. It does not pinpoint where the differences occur. Therefore, the table goes a step further by identifying the specific station combinations that are different, The Scheffee multiple comparison procedure is considered highly conservative because it requires larger differences between means for significance than most other methods (Norusis, 1996). At a significance level of .05 we see that comparisons of rating, share, and household data are nearly identical. Furthermore, on a dayby-day basis, the number of significant combinations of stations is highly similar except for the combination of station A and station B on several nights.

61










Table 6

Comparative Newscast Performance Based on Nielsen Metered Overnight Ratings

Daypart Measures Station A Station B Station C df F ratio Prob. Post Hoc Post hoc A compared
Scheffee** to average of B&C**
Overall rating 13.0 (1) 11.0 (2) 8.9 (3) 866 126.0 .000 A-B, B-C, C-A .000
share 24.0 (1) 20.8 (2) 17.1 (3) 866 172.0 .000 A-B, B-C, C-A .000
households 126.0 (1) 110.0 (2) 90.0 (3 866 126.0 .000 A-B, B-C, C-A .000
~Mon r rafing 12.4(l 1 12,0(2 9.9 (3) 86 9.5 .000 A-B, B-C .000
share 23,7 (1) 22.3 (2) 18.4 (3) 86 15.6 .000 A-B, B-C .000
households 124.0 1) 120.0(2) 99.0 (3) 86 9.5 .000 A-B B-C 000
Tues rating 12.5 (1) 11.5 (2) 8.2 (3) 134 60.0 .000 B-C, C-A .000
share 23.5 (1) 21.8 (2) 15.5 (3) 134 91.0 .000 A-B, B-C, C-A .000
households 126.0 1 116.0 2) 83.0(3) 134 60.8 .000 B-C, C-A .000
Wed rati6 14.2(1 9.7 2 8A 3 131 85.8 .000 A-B, B-C, C-A 000
share 27.2 4 .18582) 16,23.131 127,0 .000 A-B, B-C. C-A 000
households 143.0 1) 98.0 2) 85.0(3) 131 86.6 000 A-B B-C C-A 000
Thurs rating 11.9 (2) 1 4.4 (1) 8.0 (3) 131 107.0 .000 A-B, B-C, C-A .000
share 22.4 (2) 27.1 (1) 15.5(3) 131 184.0 .000 A-B, B-C, C-A .000
households 119.0(2 145.0(1) .0 1 0(3)1 131 109.0 .000 A-B,B-C,C-A .000



Sat rating 9.7 (1) 9.1 (3) 11.7 (1) 128 31.9 .000 B-C, C-A .000
share 18.6 (2) 17.3 (3) 22.5 (1) 128 59.7 .000 B-C, C-A .000

Sun ratin 12.5 1 11.0 2 9.0(3) 119 12.4 .000 B-C, C-A .000
share 23.8 1 20.6 2 17,03 119 23,1 .000 A-B, B-C, C-A .000
households 1 125 110 2 91.0 (3) 119 12.4 .000 B-C, C-A .000

*Measure rankings expressed within parentheses. ** Scheffe and post hoc analysis based on p = <.05.






63

Station C distinguishes itself for being significantly different from both competitors every day of the week but Friday. Referring back to earlier data, we see that these significant differences are not necessarily signs of outstanding ratings performance. On the contrary, station C languished in third place across all measures. A second post hoc analysis looks at the relative performance of station A against the averaged audience ratings of stations B and C in tandem. Again, station A's market leadership was significant.

Some of the above ANOVA results may violate certain statistical

assumptions or necessary conditions for a perfect analysis. In particular, the reader is cautioned not to make any strict interpretations of share figures, which by definition cannot be viewed as independent data units. Because the use of ratings, shares, and households is so pervasive in the broadcasting industry, these units of measure were included in the initial portions of this results chapter. However, in subsequent analyses when statistical independence is required, only household (000) data are analyzed.

Hypotheses Four and Five

Hypotheses four and five addressed the functional relationship of the newscasts to their respective lead-in programs. Programs with strong equity were expected to optimize lead-in ratings more efficiently than programs with weaker equity. This efficiency would be observed in the differential ratings response between the two programs. Depending on the type of analysis employed, these propositions were supported to varying degrees.







64
Hypothesis four stated that programs with higher brand equity will reveal a significantly greater difference between lead-in program audience size and program audience size than programs with lower brand equity. Table 7 looks at the overall household performance of the three stations. In terms of combined performance, we see that there was a 6% drop in households using television from 10:45 PM to 11:00 PM. However, the data also show that station A, as expected, witnessed a substantial gain in audience, surpassing its lead-in by 9%. Of the three stations, only station A achieved a positive differential response to its lead-in, thus supporting the hypothesis. Additionally, an ANOVA and Scheffee analysis of these differentials (news minus lead-in households) indicates that these differences were significant among all three stations.


Table 7

Differential Performance Between 10:45 PM Lead-in and 11:00 PM Newscasts (Overall Nielsen Metered Overnight Households [0001)

Station Station Station A&B&C F ratio Prob. Scheffee at .05
A B C (News
HUT)_
10:45 PM 115 127 99 341 33.4 .000 A-B3,A-C
11:00 PM 1127 110 91 326 126.0 .000 A-B3, B-C, C-A
Diq,~ rqpiia + ... 17 -'~ 8 -15 7S[,.2 A-B. B-C. A-C
% change + 9 -14 -9 6 __________Note. df = 866, p = .05.


Hypothesis five stated that programs with higher equity will reveal a significantly greater differential share index than programs with lower equity. Because the number of homes using television (HUT) can change dramatically







65

throughout the day, comparing household and ratings data from different time periods may be misleading. Share calculations, however, are based solely on the available audience at the time of the measurement. Therefore, Table 8 interprets the newscast/lead-in relationship as a share index. Index scores exceeding a magnitude of one indicate the degree to which a newscast outperformed its lead-in. Conversely, scores of less than one infer the degree to which the program was unable to hold its lead-in audiences. As expected, in almost all cases, station A (our equity leader) indexed higher than its two competitors.


Table 8

Share Index of Late Newscasts Compared to Lead-in Programming Station Overall Mon Tues Wed Thurs Fri Sat Sun
A 1.28 1.5 1.2 1.2 1.5 1.0 1.5 1.2
B 1.02 1.1 .97 .99 .83 1.1 1.1 1.1
C 1.08 1.0 .94 1.2 1.2 1.0 1.0 .96

*Indicates highest index for that evening.


Another way to evaluate the differential response of a newscast to its

lead-in is through correlational analysis. Nearly all work on inheritance effects has used this technique in describing the close association or dependence between adjacent programs. However program brand equity is presumed to be a function of independence. That is, a program with strong equity is not as dependent on its lead-in for audience ratings as a program exhibiting lesser equity. Therefore, one would expect a weaker association (less dependency)







66

for a program with relatively strong equity. It was presumed that evidence of this lack of dependency would provide indirect support of hypotheses four and five. Table 9 offers the results of a correlational analysis based on household data.

From an overall standpoint, the correlation data give partial support for the hypothesis in that station A's correlation is substantially lower than station B. However, Station C demonstrated an even lower figure. On a day-by-day basis, station A offered more convincing evidence by "winning" four out of the seven nights with the lowest correlation. Table 9

Household Correlational Analysis Between 10:45 PM Lead-in Programming and 11:00 PM Newscasts

Station Overall Mon Tues Wed Thurs Fri Sat Sun A .76 .53 .55 .75 .60 .61 .53 .89
B .85 .86 .65 .68 .86 .68 .60 .84
C .71 .78 .67 .83 .41 .65 .55 .59

Note. A = 4, B = 1, C = 2.

*Indicates lowest r among the three stations.


As a brief digression, the investigator explored a different method of

correlation of the overall data. Since most television programming recurs on a weekly basis, correlations were performed using seven-day rolling averages rather than on the more conventional single-day data. According to the data results presented in Table 10, station A revealed the greatest disparity between the two methods and also repositioned itself as the station exhibiting the lowest correlation.







67

Looking at the regression data for overall household performance

among the three competing stations, Table 11 reveals that each newscast had a strong positive linear relationship with its lead-in programming. Lead-in Table 10

Comparison of Household Correlations Derived from Single Day versus Seven-Day Moving Averaqe Data

Station r derived from r derived from
single day data seven-day moving average data .76 .59
B .85 .82
F-c .71 7 .72


Table 11

Regression Data Based on Overall Households (Independent Variable: Lead-in Program Dependent Variable: Newscasts)

Station Slope Intercept
A .76 53,337
B .85 49,951
C .71 44,478



households were strong predictors of newscast household performance. However, on face value the slope and intercepts do not indicate any remarkable differences among the three competitors. Based on these data, one could speculate that Station A's success cannot be attributed solely to its ability to convert lead-in audiences but rather to its power to recruit audiences from other sources. The final two hypotheses take a closer look at this possibility.







68

Hypotheses Six and Seven

Hypotheses six and seven take the above notions of differential

response (or momentum) a step further by dividing the phenomenon into two distinct segments: (a) the power to retain lean-in audiences and (b) the power to recruit audiences from other sources. Nielsen Media Research provided the investigator with a standard "Flow Analysis" study of the May 1996 diary sweep period. Using four-week averaged household ratings, Nielsen tracked weekday audience flow from 10:45 PM to 11:00 PM. While the complete study is shown in the appendix, the most salient data have been distilled into Table 12.

Table rows 1 through 4 (shaded area) provide data on the relative

audience composition of each newscast. That is, total news households (row 4) equals the sum of households retained from lead-ins (row 1), plus households recruited from other stations providing local news (row 2) plus households recruited from nonnews sources (row 3). The last two categories deal with audiences that "switched" channels in order to watch a specific newscast. The percentage columns within the shaded area give the proportion of the total news audience that each category holds. These four data rows, however, do not provide sufficient insight into the power of brand equity. Rows 5, 6, and 7 delve into dynamics of our final hypotheses.

Hypothesis six predicted that programs with higher brand equity would retain a greater proporfion of their lead-in audiences than programs with lower brand equity. As expected, row 5 of Table 12 shows station A to be the best performer, retaining 65% of its lead-in households.







69

Hypothesis seven stated that programs with higher brand equity would recruit a greater proportion of audiences from other sources than programs with lower brand equity. Rows 6 and 7 of Table 12 address this proposition. Perhaps the proportion of households recruited from other news stations (row 6) offers the most persuasive evidence. Here we have audiences that switched away from a channel that was about to present a local newscast. Of all the news channel "switchers," station A earned the largest proportion (44%). Similarly, station A recruited a greater proportion (51%) of the total nonnews switchers.






70


Table 12

Nielsen Audience Flow Analysis, May Sweep 1996 (10:45 to 11:00 PM, Monday Through Friday, Four-Week Average Households [0001)

Row Measure Station A Station B Station C
no.
0 000 000 000 %
1 HH retained from 805
lead-in
2 H recruited from 24
rom other news
stations
switchers)
S HH recruited from 21





tlnon newsore
switcher)
colapTotal News H= 16 100 121 100 94 1

5ttn% HH retained 65 = 5
from leads-in *___ 1__________ _6 % recruited from 4 12
total news
switchers (92) *
7 % recruited from 5 92
total non news
______ switchers (87) ___ _______ _____ ___Note. chi square p =.05:

*4.49 with stations B and C collapsed, df =1. ** 5.0 with stations B and C collapsed, df = 1. **12.84 among all three stations, df = 2 and 11.64 with stations B and C collapsed, df = 2.













CHAPTER 6
DISCUSSION

The purpose of this study was to explore the plausibility of applying brand equity theory to television program ratings performance. While the industry has embraced the "buzzwords" of brand management, there has been little scholarly work on making the transition from products as brands to media as brands. While specific hypothesis were presented, much of the effort invested in this study was a matter of curiosity and intellectual tinkering. As the title of this dissertation implies, the first challenge was conceptualization-How can conventional brand equity theory be applied in a meaningful way to television programming. The second challenge was to find a way to measure or operationalize these concepts using standard Nielsen ratings data.

In addition to an overview of brand equity theory, the conclusion of the literature review section provided some feasible conceptualizations for television programming. In particular, Keller (1993) and the notions of consumer inertia (momentum) were found to work well in explaining program performance in light of the powerful influence of inheritance effects. Presuming that essential marketing mix components are controlled or neutralized, the program equity model of audiences retained/audiences recruited appears to have merit.



71







72

Applying this model to inheritance effects offers some new theoretical insights. Instead of addressing structural factors such as program type or number of program options, this approach delved into the ability of a program to retain lead-in audiences and recruit audiences from other sources. In tandem, these two power dimensions define a program's momentum as it "travels" through time. A program exhibiting strong equity can overcome a poor lead-in and recruit loyal viewers from other sources. Additionally, this program will be less likely to lose portions of its lead-in audience to direct competitors.

The case study provided a more concrete way to "test" some of these

theoretical propositions. All seven hypotheses were supported to some degree and open the door for further theoretical work and more generalizable empirical research.

From a conceptual standpoint, the a major dilemma facing the

investigator was choosing a theoretical starting point. With so many divergent conceptualizations and theories available, brand equity can be almost anything a researcher wants it to be. From attitudinal notions of "heritage," "reputation," 99 perceived quality," "depth of awareness," "enthusiasm," and "commitment" to more behavioral concepts, such as "pricing elasticity," "repeat purchases," ii exclusive heavy user," and "consumer inertia," marketing professionals and scholars have yet to agree on a precise definition of brand equity. Compounding the problem is that many in the field use marketing terms such as brand image, brand loyalty, and brand equity interchangeably. After much searching and reflection, Keller (1993) emerged from this theoretical muddle







73

with an approach that can be translated readily into broadcast programming terms. Keller's theoretical underpinnings combined with a more rigorously defined branding vocabulary can form the bedrock of a brand equity research agenda.

A far more perplexing issue was discovered on the operational side of program brand equity. Keller's (1993) overall conceptual framework requires a controlled marketing mix which enables a researcher to capture the relative power of brand equity. That is, strong, favorable, and unique brand associations (brand knowledge) cannot be measured properly until the possible confounding elements of the brand's marketing mix are held in check. Marketing mix factors such as pricing, distribution, and promotion can taint equity measures that rely on in-market brand performance data.

For our case study, controlling these marketing mix factors was an

unexpected challenge. The methodology section of this dissertation revealed a number of problems that had to be overcome. Among these were (a) identifying direct competitors by program type or genre and (b) creating the simultaneous availability of direct competitors. The selection of three 11:00 PIVI local newscasts, scheduled at the exact same time, made for an ideal test environment for our case study. However, most television programming does not offer this convenience.

Determining who is a direct competitor can be a frustrating task. While local newscasts do not pose a great problem, other program types can generate considerable disagreement among media researchers. For example,






74

should "Home Improvement" and "Married With Children?" be assigned to the same genre of sitcom? Within the category of adult drama, should "NYPD Blue" and "Dr. Quinn" be regarded as coming from the same product category? Webster (1985) addressed this issue of program typology and warned that the current industry categories would someday become "overly broad, and insensitive to subtle, though important, distinctions" (p. 125). In an environment of highly distinctive niche programming on dozens of channels, the definition of a direct competitor will become even more challenging.

The availability of direct competitors at the same point in time was

accomplished easily with our three 11:00 PM newscasts, but this situation is more the exception than the rule in television. In fact, in our test market there is currently a 10:00 PM local newscast on a small UHF station. Do people watch this program because it is the only 10:00 PM news in the market, or is there evidence of genuine brand equity based on content ? If this station scheduled its newscast at 11:00 PM, against the other entrenched news competitors, how well would it perform? Because of the time-bound nature of broadcasting, these types of questions are not answered easily. Perhaps a "what if" survey could be administered to an appropriate sample group. In this case, the controls would be far more contrived than a real-world setting where actual viewing behavior could be recorded.

Two related distribution factors that may be awkward to control are

signal coverage and available potential audiences. Just as retailers recognize the marketing advantages of a superior distribution system, so broadcast






75

researchers must be aware of certain structural factors, such as signal strength and the available audiences that can be reached by that signal. Programs that cannot be received cannot be watched. Unless there is a "level playing field" in terms of program distribution, consumer-based brand equity cannot be measured accurately. Fortunately, our test market exhibited equivalency in this area, but this may not be the case for some other markets. For example, the Gainesville, Florida, television market must "import" CBS and NBC signals via cable retransmission from other nearby markets.

The negative impact of certain promotion activities must also be

controlled. In recent months, there has been an outcry from the broadcast and advertising industries to control the marketing abuses unleashed during the infamous sweep weeks, and some progress has been made. In our case study, the sweeps did not appear to have any extraordinary influence on newscast ratings, but other program dayparts may have been more vulnerable to hyping. Of course, the best cure for sweeps distortion is to look beyond periodic measures and instead examine continuous overnight viewing data. Regrettably, only 42 out of the 212 Nielsen television markets currently have this research capability.

After determining a plausible theoretical approach and wrestling with the above-mentioned methodological challenges, the seven program performance hypotheses were then tested. A primary assumption of the study was that station A was a suitable equity benchmark. The two criteria of (a) long-term ratings leadership and (b) viewer loyalty data derived from the Marshall







76

Marketing survey served as yardsticks for isolating station A from its competitors.

While recognizing the lack of external validity, the hypotheses found

considerable support. As expected, the overall and daily performance of station A was exceptional when compared to its competitors. In fact, stations B and C often appeared quite similar in their audience performance, while station A stood apart from the crowd. When analyzing newscast performance in relationship to lead-in programming, the results were mixed. From simple differential and indexing perspectives, station A consistently outperformed its lead-in program. The correlational analyses were not as convincing but did show general support.. The correlations performed on the overall (seven-day) data revealed less equity value for station A than the individual day data. This may be the result of distortions that can occur sometimes when averaging data. For example, the .89 correlation on Sunday night for station A is far out of line compared with the other six days of the week and may have overstated the overall performance of the station. Similarly, the Thursday night correlation for station C (against "ER" on NBC) is extraordinarily low compared with other days. Another possibility may be a conceptual flaw. That is, the supposed independence of our high equity station may be revealed only when the lead-in is exceptionally poor. Another related avenue for future exploration is why the use of seven-day moving averages (Table 10) seemed to enhance the position of station A.







77

The regression analyses in this dissertation invite some theoretical

speculation on the interpretation of program equity. The two components of a single linear regression equation, its slope (coefficient) and its intercept (constant), could be construed as indicators of the two power dimensions of our equity model. Given the identical lead-in audiences (independent variable), it is possible mathematically for two programs to exhibit the same slope but different intercepts and visa versa. The slope of the equation may provide a rough measure of the efficiency of a program to retain its lead-in audience. Alternatively, the intercept, which would be defined mathematically as the size of a news audience (dependent variable) when the lead-in (independent variable) is zero, might be analogous to a measure of the program's ability to recruit audiences from other sources. In a typical regression analysis there are seldom any data points clustering at the actual intercept point. Therefore, this figure should not be interpreted in absolute terms. However, it may be of some value as a comparison tool. In the case study, station A revealed a modestly higher coefficient and intercept than its two competitors. Additional studies across several markets can determine if these results were merely coincidental.

The most illuminating data came from the Nielsen Audience flow study which actually tracked audience retention and outside recruitment for the May sweep period. Despite some methodological limitations, which are presented later, the data offered the most persuasive evidence for supporting hypotheses






78

six and seven. The ability of Station A to both retain and recruit audiences were unsurpassed.

Study Limitations

Aside from the above-mentioned difficulties in controlling the marketing mix, the researcher recognizes several limitations to this particular study. First, there is the lack of external validity. By design this was an exploratory effort under limited conditions that were familiar to the researcher, The three news programs in this particular market provided an ideal field setting where marketing mix factors could be controlled to a reasonable degree. By having different lead-in programs each night, the newscasts were placed in different 44 experimental" conditions. With the goal of taking the first steps in conceptualizing and measuring the effects of brand equity on television programs, the results of this case study are encouraging. The next logical step is to apply some or all of these notions across other markets and other types of programming.

A second limitation was the reliability of Nielsen ratings. Although

another measure of program performance would not be as relevant to the real world of commercial broadcasting, there is no denying the inadequacies of Nieslen ratings. The buyers and sellers of program audiences may look at these numbers as absolute measures, but media researchers (including Nielsen itself) recognize several statistical limitations. First, Nielsen does not guarantee a representative random sample due to a number of factors, the most important being dismal cooperation rates (averaging 45%) among







79

people asked to participate. Secondly, presuming the samples were perfectly random, there is considerable sampling error (standard error) that is not disclosed in the regular reports. At the beginning of every Nielsen market sweep report, the company warns subscribers against using certain ratings data to make management decisions because the relative standard error could exceed 50% (Nielsen Method, 1995). Therefore, television ratings data tend to fluctuate. Many media professionals want to believe that these changes are real, but savvy researchers know that much of this motion is due to sampling error or "bounce" rather than legitimate changes in audience tuning behavior.

While there are distinct advantages to using the metered overnight ratings compared to the four-week averaged sweeps data, a disappointing drawback was the lack of demographic information. Unlike the national Nielsen i1people meter" ratings, local metered markets have yet to adopt this recording technology. Currently, overnight ratings are restricted to household data. Based on the literature review, it seems possible that program loyalties could vary across age and gender demographic categories.

This same limitation arose when we examined Nielsen audience flow studies. To date, this detailed audience analysis is restricted to sweep weeks data. In our case, we had to look at Monday through Friday average newscast performance over one month. Despite these limitations, our theoretical precepts of (a) power to retain audiences and (b) the power to recruit audiences were substantiated, and Nielsen promises that overnight flow studies would be available soon to metered markets. Meanwhile, the sweep-







80

based flow studies (four per year) can still be applied to other markets in an attempt to replicate our case study findings. Of course, Nielsen charges broadcasters a substantial fee for these special studies, so an academic researcher may have a financial handicap.

The Marshal Marketing and Communications data were used primarily as a benchmark comparison for the Nielsen ratings. Within their statistical calculations was the determination of "heavy users." Offhand, this type of information seems most appropriate for equity research, but similar ratings data are not obtained easily from Nielsen. For an undisclosed fee, Nielsen will do a custom data analysis.

Conclusions

Returning to the essential purpose of this study, one can conclude that program brand equity is a plausible theoretical notion that can be applied to electronic media. However, broadcasters should be wary of all the brand management jargon that is so prevalent in the trade press and industry discussions. This study also showed that, given a highly controlled research setting, the impact of attitudinal components of program brand equity (Keller's brand knowledge) can be measured to some degree using the behavioral measures of standard Nielsen ratings. Additionally, this project offers new insight into the phenomenon of inheritance effects and tuning inertia.

After establishing conceptual and operational definitions of program

brand equity, this study examined how a program with substantial brand equity performs when compared to direct competitors with less brand equity. The







81

seven hypotheses approached this issue from several perspectives. Hypotheses one, two, and three looked at overall ratings performance and found that the program identified as having high brand equity had higher ratings, higher shares, and a larger household audience. Hypotheses four and five narrowed the investigation to the relationship of a program to its lead-in. A number of statistical indicators revealed that a program with strong equity processes or "inherits" audiences in ways that are different than its direct competitors. That is, while the high equity program consistently outperformed its 10:45 PM lead-ins to maintain 11:00 PM ratings leadership, the other stations usually lost ratings during this transition. Translating these rating point gains and losses into the components of where an audience has come from and where it has gone was the domain of the final two hypotheses. Using a "Flow Analysis" that differentiated between audiences retained from lead-ins and audiences recruited from other sources, the results indicated that the high equity program was superior to its low equity competitors not only in retaining lead-in households but was in drawing households from surrounding channels and the "off' position. Taking an even tighter focus, the results showed that the high equity program demonstrated exceptional power in recruiting audiences from channels offering similar local newscasts (direct competitors). Taken in combination, the hypotheses validate the importance of brand equity to the ratings success of a television program--at least for local news programming.






82

On a subordinate level, we can also conclude that this study explains some of the extreme variance found in prior inheritance effects research. Rather than relying solely on structural factors, such as program type and number of program options, we can now introduce the audiences retained/ audiences recruited equity model to help explain these disparities.

Suggestions for Future Research

This study fosters several research avenues for continued scholarly work. On a rudimentary level, there is a need to clarify exact research topics using proper terminology. For example, page 39 offered some standard brand concepts that could be applied to broadcast programming. Additionally, there is a need to segment further these branding domains into programs, stations, and networks. Below is a proposed research topic grid that may be useful for future investigations of media as brands.


Media As Brands Research Topic Grid Program Station Network
Brand Awareness
Brand Extensions
Brand Image
Brand Positioning
Brand Loyalty & commitment
Brand Equity


Taking the theoretical position that understanding audience flow is

crucial to understanding the influence of program brand equity, the following proposed audience member categories may assist future research.






83

1. Loyalists.

Definition: Viewers who hold strong consumer-based program brand equity and are found within the lead-in program's total audience.

Disposition: These viewers remain in place on the same channel because of brand commitment

2. Passives.

Definition: Viewers who do not hold any strong consumer-based

program brand equity towards any direct competitor and are found within the lead-in program's total audience.

Disposition: These viewers will remain in place on the same channel because of simple "inertia" rather than true commitment.

3. Converts.

Definition: Viewers who hold strong consumer-based program equity but were watching on a competing channel and, therefore, are motivated to switch channels.

Disposition: If necessary, these viewers will abandon a lead-in program channel and switch over to a more suitable program option (see number 5 defectors below).

4. Tune-ins.

Definition: "Appointment" viewers who hold very strong consumer-based program brand equity and make a deliberate effort to turn on the TV set in order to watch a specific program. It possible to have a subcategory of passive tune-







84

ins, where the desire is simply to have the set "on" with no regard for specific programming.

Disposition: These viewers are acquired audiences from what some researchers call the "off position."

5. Defectors.

Definition: "Converts" seen from an opposing perspective, that is,

viewers who are about to abandon a program channel in order to find a more suitable or preferred program. Defectors hold strong consumer-based program equity for a direct competitor. (One program's defector is another program's convert.)

Disposition: At the appropriate time, these viewers leave one program and go to another on a competing channel.

Using these categories, one could hypothesize that program equity (a) increases the number of converts, (b) reduces the number of defectors, (c) reinforces current loyalists, and (d) transforms passives into converts.

Implications For the Field

Well now, home entertainment was my baby's wish
so I hoped into town for a satellite dish
I tied it to the top of my Japanese car
I came home and pointed it out to the stars
A message came back from the great beyond
There's fifty-seven channels and nothing' on.
(Springsteen, 1989)

In an earlier decade, when cable and satellite delivery systems were just beginning to offer a broad range of programming options, singer/songwriter Bruce Springsteen composed a prophetic tune entitled "57 Channels And







85

Nothing' On." It recounts the tale of a frustrated young man who, after eagerly embracing these new technologies, finds that even with an array of 57 program channels, there was still nothing of value to watch on his TV. Finally, he retaliates by shooting his television set with his .44 magnum. Keller (1993), no doubt, would affirm that these programs failed to engender any favorable, strong, or unique brand associations and, indeed, from a scholarly viewpoint, Keller would be correct. While Springsteen was surely unfamiliar with the term brand equity, he did reveal delightful insights into consumer behavior and the challenge of creating successful media brands. Successfully applying the notions of consumer-based brand equity will be crucial for the economic survival of broadcast programs, stations, and networks, As we approach a new digital decade, where 57 choices will seem meager, the consumer anguish found in Springsteen's classic song will continue to ring true. Without proper media brand management, audiences in the 21st century will lament about 157 channels and nothing on.














APPENDIX
A NIELSEN TV METERED MARKET HOUSEHOLD ANALYSIS MAY 1996






NSI-PLUS.
NIELSEN TV METERED MARKET HOUSEHOLD ANALYSIS
ORLANDO MAY 1996
NO. 78931
FOR NIELSEN MEDIA RESEARCH

AUDIENCE SOURCE AND DESTINATION
DMA HOUSEHOLDS
SOURCE TIME: AVERAGE OF WKS 1-4 MON 10:45PM WKS 1-4 TUE 10:45PM WKS 1-4 WED-10:45PM WKS 1-4 THU 10:45PM WKS 1-4 FRI 10:45PM
DESTINATION: AVERAGE OF WKS 1-4 MON 11:00PM WKS 1-4 TUE 11:00PM WKS 1-4 WED 11:00PM WKS 1-4 THU 11:00PM WKS 1-4 FRI 11:00PM
AVG.SOURCE
AUDIENCE AVG. DESTINATION AUDIENCE (000)
STATION (000) WFTV WCPX WESH WKCF WOFL WRBW OTHR OFF
TOTAL 141 82 116 35 40 6
WFTV 124 80 10 15 4. 7 11 14
WCPX 103 18 53 14 2 6 8 14
WESH 142 22 13 77 4 7 1 14 21
WKCF 45 10 3 5 26 2 3 5
WOFL 35 8 2 4 1 18 4 5
WRBW 11 2 1 1 3 2 1
OTHER 19 9 11 2 6
TUNE-IN 5 3 4 3 2



AVG.SOURCE
AUDIENCE AVG. DESTINATION AUDIENCE (%)
STATION (%) WFTV WCPX WESH WKCF WOFL WRBW OTHR OFF
TOTAL 14.1 8.2 11.6 3.5 4.1 0.6
WFTV 12.4 8.0 1.0 1.5 0.4 0.7 1.1 1.4
WCPX 10.3 1.8 5.3 1.4 0.2 0.6 0.8 1. 4
WESH 14.2 2.2 1.3 7.7 0.4 0.7 0.1 1.4 2.1
WKCF 4.6 1.0 0.3 0.5 2.6 0.2 0.3 0.5
WOFL 3.6 0.8 0.2 0.4 0.1 1.8 0.4 0.5
WRBW 1.1 0.2 0.1 0.-I 0.3 0.2 0.1
OTHER 1.9 0.9 1.1 0.2 0.6
TUNE-IN 0.5 0.3 0.4 0.3 0.2




86






87





NSI-PLUS
NIELSEN TV METERED MARKET HOUSEHOLD ANALYSIS
ORLANDO MAY 1996
NO. 78931
FOR NIELSEN MEDIA RESEARCH

AUDIENCE SOURCE AND DESTINATION
DMA HOUSEHOLDS
SOURCE TIME: AVERAGE OF WKS 1-4 MON 10:45PM WKS 1-4 TUE 10:45PM WKS -1-4 WED 10:45PM WKS 1-4 THU 10:45PM WKS 1-4 FRI 10:45PM
DESTINATION: AVERAGE OF WKS 1-4 MON 11:00PM WKS 1-4 TUE 11:00PM WKS 1-4 WED 11:00PM WKS 1-4 THU 11:00PM WKS 1-4 FRI 11:00PM
DESTINATION OF AUDIENCE AVERAGE BASIS
STATION SHARE WFTV WCPX WESH WKCF WOFL WRBW- OTHR OFF

WFTV 100 64 8 12 3 5 9 11
WCPX 100 171 52 14- 2 6 8 14
WESH 100 15_ 9 55- 3 5 1 10 15
WKCF 100 23 7 12 56 4 7 11
WOFL 100 22 5 12 4 50 10 13
WRBW 100 20 9 11 31 19 11



SOURCE OF AUDIENCE AVERAGE BASIS
STATION WFTV WCPX WESH WKCF WOFL -WRBW
TOTAL SHARE 100 100 100 100 100 100
WFTV 57 12 13. 11 16
WCPX 13 65 12 7 16
WESH 16 16 67 13 18 17
WKCF 7 4 5 73 4
WOFL 6 2' 4 4 44
WRBW 2 1 1 56
OTHER 14 10 9 7 14
TUNE-IN 4 3 3 8 5



COPYRIGHT 1997 BY NIELSEN MEDIA RESEARCH INC. ALL AUDIENCE DATA CONTAINED IN THIS ANALYSIS WERE OBTAINED FROM THE SAME METER RECORDS USED TO PRODUCE THE REGULAR NIELSEN TV REPORT.














REFERENCES

Aaker, D. A. (1991). Managing brand equity. New York: The Free Press.

Abelman, R., Atkin, D., & Rand. M. (1997). What viewers watch when they watch
TV: Affiliation as case study. Journal of Broadcasting and Electronic
Media, 41, 360-379.

Airidge, R. (1996, July, 15). If you're trailing in news, try something different.
Electronic Media, p. 6.

Anantachart, S. (1995). Conceptualizations and a news measure of brand
eqiy Unpublished Masters Thesis, University of Florida College of
Journalism and Mass Communication.

Axelrod, J. (1993, October 27). The definition and measurement of brand equity.
Paper presented at ARF Brand Equity Research Day, New York.

Barton, L. (1993, June). Tartikoff: No more white noise! Promotion and
Marketing Executive In Electronic Media, p. 16.

B & C Hall of Fame. (1996, October, 30). Bob Wright profile. Broadcasting and
Cable, p. 51.

Beatty, S.E., Kahle, L.R., & Homner, P. (1988) The Involvement-Commitment
Model: Theory and implications. Journal of Business Research, 16,149168.

Beike, D.R, & Sherman, S.J. (1994). Social inference: Inductions, deductions,
and analogies. In Handbook of Social Cognition (2nd ed., pp. 281-326).
Palo Alto, CA: Annual Reviews.

Belch, G., & Belch, M. (1993). Introduction to advertising and promotion: An
integrated marketing communications perspective. Boston, MA: Irwin.

Bender, D. C. (1993, March 24). Media as brands. Speech presented at
Advertising Research Foundation 39th Annual Conference, New York.

Biel, A. L. (1991) How brand image drives brand equity. Journal of Advertising
Research, 6. RC6-RC 12.


88





89



Black (1996, June 3). Building a better brand. L.A. Broadcasting and Cable, p.
93.

Boemer. M. L. (1987). Correlation lead-in show ratings with local television
news ratings. Journal of Broadcasting and Electronic Media, 31, 89-94.

Brinkley, J, (1997, December 29). TV stations fear for their channel brand as
choices proliferate. New York Times, C9.

Burgii, M., & Katz, R. (1997, July 21). Big nets need Geritol. Media Week, p, 3.

Buzzard, K. (1992). Electronic media ratings; Turning audiences into dollars
and sense. Boston, MA: Focal Press.

Ceurvorist, R. W. (1994, February 15-16). A brand equity measure based on
consumer commitment to brands. Paper presented at ARF Brand Equity
Workshop, New York.

Churchill, G. A. (1995). Marketing research: Methodological foundations. Fort
Worth, TX: The Dryden Press.

Christiani, A. (1993, October 27). Measuring and monitoring brand loyalty and
its role in managing brand equity. Paper presented at ARF Brand Equity
Research Day, New York.

Cobb-Walgren, C.J., Ruble, C.A., & Donthu, N. (1995). Brand equity, brand
preference and purchase intent. Journal of Advertising, 24, 25-35.

Cooper, R. (1993). An expanded, integrated model for determining audience
exposure to television. Journal of Broadcasting and Electronic Media, 37,
404-410.

Cooper, R. (1996). The status and future of audience duplication research: an
assessment of ratings-based theories of audience behavior. Journal of
Broadcasting and Electronic Media, 40, 96 -111.

Donius, J. F. (1994, February 15-16). Brand equity: A holistic perspective. Paper
presented at ARF Equity Workshop, New York.

Due South. (1997, June 2). Advertisement for "Due South" in Broadcasting and
Cable, p. 35.

Dupree, S. (1996, February 12). Network TV: Building better brands. Media
Week, p. 27





90



Eastman, S. T. (1997). Broadcast/cable programming: Strategies and
practices. Belmont, CA: Wadsworth.

Encyclopedia of science and technology (p. 136, p. 427). (1997). New York:
McGraw-Hill.
EQX. (1995). Strengthening the power of the brand. Simmons Market Research
Bureau [Brochure]. New York.

Farquhar, P. H. (1989). Managing brand equity. Marketing Research, 1,24-33. Fazio, R. H., Powell, M. C., & Williams, C. J. (1989), The role of attitude
accessibility in the attitude-to-behavior process. Journal of Consumer
Research, 16, 280-288.

Freeman, M., & Stanley, T. L. (1996, June 24). Angling for promotion. Network's
and syndicators vie for valuable hype time on stations. Media Week, p. 8. Gimein, M. (1996, June 24). Sweeps contests up 55%. Media Week, p. 5. Gleason, M. (1995). Time to perform, or perish: Consumers gain more power
to determine survivors on shelf. Advertising Age, p. 37.

Goddhart, G. J., Ehrenberg, A. S. C, & Collins, M. A. (1975). The television
audience: Patterns of viewing. Westmead, UK: Saxon House. Greenwald, J. (1996, December 23). Reinventing Sears. Time, p. 54. Headen, R. S., Klompmaker, J. E.. & Rust, R. T. (1979). The duplication of
viewing law and television media schedule evaluation. Journal of
Marketing Research, 16, 333-340.

Hoyer, W. D, & Brown, S. P. (1990). Effects of brand awareness on choice for a
common, repeat-purchase product. Journal of Consumer Research, 17,
141-148.

Interbrand. (1994). Client sales [Brochure]. New York: Author. Kapferer, J. N. (1992) Strategic brand management. New York: The Free Press. Kardes, F. R. (1994). Consumer judgment and decision processes. In R.S.
Wyer & T.K. Srull, Handbook of social cognition (2nd ed., pp. 399-465).
Hillsdale, NJ: Erlbaum.





91


Keller, K. L. (1993). Conceptualizing, measuring, and managing customerbased brand equity. Journal of Marketing, 57, 1-22.

Kraus, S. J. (1995). Attitudes and the prediction of behavior: A meta-analysis of
the empirical literature. Personality and Social Psychological Bulletin, 21,
58-75.

Lehmann, D. R., & Winer, R. S. (1994). Analysis for marketing planning. Burr
Ridge, IL: Irwin Publishing.

Lin, C. A. (1995). Network prime-time programming strategies in the 1980s.
Journal of Broadcasting and Electronic Media, 39, 482-495.

Lintas. (1993). Lintas link brand equity audit and probe. In Russell & Lane,
Kleppner's advertising procedure (p. 210). Englewood, NJ: Prentice Hall.

Littleton, C. (1996, March 11). New day for local news. daytime newscasts are
boosting brand awareness and bottom lines. Broadcasting and Cable,,
p. 30.

Littleton, C. (1996, June 10). Murdoch bullish on over-the-air TV. Broadcasting
and Cable, p. 10.

Long, S. T. (1979). The development of the television network oligopoly. North
Stratford, NH: New Arno Press.

Mandese, J. (1993, May 17). Here come the newest brands: NBC, CBS, ABC.
Advertising Age, p. 1.

Mandese, J. (1995, March 25). In news growth phase, cable feeding on itself.
Advertising Age, p. S-1.

Marshall Marketing. (1996). Marshall Marketing and Communications client
sales brochure, personal letter to investigator and proprietary data from
1996 Orlando, Florida, market survey. Orlando, FL: Author.

Mature, R. (1993, February 1). Maintaining brand loyalty during times of
marketing chaos. Paper presented at ARF Brand Equity Workshop, New
York,

McClellan, S. (1995, April 17). Advertisers say stations need brand identity.
Broadcasting and Cable, p. 61.

McConville, J. (1995a, July 17). CTAM focuses on cable competition: Wake-up
call to include sessions on branding. Broadcasting and Cable, p. 15.





92



McConville, J. (1 995b, December 4). In branding, image is reality. Broadcasting
and Cable, p. 70.

McDowell, W.S. (1995, August). The ratings game: The ethics of hyping,
stunting, blunting, rigging, and other dirty tricks Paper presented at
AEJMC Convention, Advertising Division, Washington, DC.

Miles, L. (1997, February 10). Sweeps under suspicion. Media Week, p. 14. Morris, B. (1996, March 4). The brand's the thing. Fortune, p. 72, Myers. (1997, February 2). The Myers report: Media 2005. Electronic Media, pp.
28-29.

National Association of Television Program Executives. (1996). The branding
bug (Panel discussion). Annual Convention agenda. New York: Author. National Association of Television Program Executives. (1997, May 28). Our
brand new world (Televised seminar).

NBC, (1995, November). NBC, America's leading network television brand (Full
page ad). Advertisigg age. New York: NBC Marketing Department.

Nedungadi, P. (1990). Recall and consumer consideration sets: Influencing
choice without altering brand evaluations. Journal of Consumer
Research, IT 263-276.

Nielsen Audience. (1996). Television audience 1996 (Nielsen Media Research
Report). New York: Author.

Nielsen Cume. (1996). Nielsen station index for Orlando-Daytona BeachMelbourne, Florida. Average Weekly Cume Household Analysis for 1996.
New York: Author.

Nielsen Method. (1995). NSI methodology, techniques and data interpretation
(Nielsen station index supplement). New York: Author.

Norusis, M.M. (1996). SPSS guide to data analysis. Englewood Cliffs, NJPrentice Hall.

NPD. (1995). Brand builder: How can I measure, track and manage brand
equity? [Brochure]. New York: The NPD Group, Inc.

Product Q [Brochure]. (1995). New York: Product Evaluations, Inc.





93



Promax & BDA. (1996). Profiting from your powers of creativity [Convention
brochure]. Los Angeles: Promotion and Marketing Professionals In
Electronic Media and Broadcast Designers Association.

Rathbun, E. (1995, June 12). Marketers take promax spotlight. Broadcasting
and Cable, p. 12.

Ries, A., & Trout, J. (1981). Positioning: The battle for your mind. New York:
McGraw-Hill.

Rosenstein, A. W., & Grant, A. E. (1997). Reconcepualizing the role of habit: A
new model of television audience activity. Journal of Broadcasting and
Electronic Media, 41, 324-344.

Rubin, A. M. (1984). Ritualized and instrumental television viewing. Journal of
Communication, 34, 67-77.

Rubin, A. M. (1994). Media uses and effects: A uses-and-g ratifications
perspective. !n J. Bryant & D. Zillman (Eds.). Media effects: Advances In
theory and research (pp. 417-436). Hillsdale, NJ: Lawrence Erlbaum. Sharkey, B. (1997, July 21). Living in 'The Real World'. Media Week, p. 17. Shell, A. (1997, May 12). Brand loyalty? Ad Week, p. 40. Sims, J. (1997, April 14). How cable grabbed the prime time audience.
Electronic Media, p. 37

Spring, A. (1997, December 22). FX will brand it image through originals.
Electronic Media, p. 10.

Springsteen, B. (1989). 57 Channels and nothin' on. Lyrics from song.
Produced and recorded by B. Springsteen licensed through ASCAP. Srivastava, R. K., & Shocker, A. D. (1991). Brand equity: A perspective on its
meaning and measurement. Cambridge, MA: MSI.

Stay tuned. (1995, June 12). Broadcasting and Cable, p. 58. Stern, C. (1996, February 5). New law of the land. Broadcasting and Cable, p. 8. Thierer, A. D. (1995, January). Why mass media mergers are meaningless.
The FreeMan: Ideas on Liberty, p.5,




Full Text

PAGE 1

&21&(378$/,=,1* $1' 0($685,1* 7+( ())(&76 2) %5$1' (48,7< 21 7(/(9,6,21 352*5$0 5$7,1*6 3(5)250$1&( %\ :DOWHU V PFGRZHOO $ ',66(57$7,21 SUHVHQWHG WR WKH JUDGXDWH VFKRRO 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8/),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$

PAGE 2

$&.12:/('*0(176 7KH DXWKRU WKDQNV 0U 6NLS 6NLIILQJWRQ 3URJUDP 'LUHFWRU RI 6WDWLRQ $ DQG 0V -HQQ\ %XWOHU RI 1LHOVHQ 0HGLD 5HVHDUFK IRU SURYLGLQJ SURSULHWDU\ RYHUQLJKW 1LHOVHQ UDWLQJV GDWD

PAGE 3

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

PAGE 4

,QHUWLD 0RPHQWXP DQG (TXLW\ ,QWHUIDFLQJ 0RPHQWXP :LWK .HOOHUfV &RQFHSWXDOL]DWLRQV (TXLW\ DQG 3HUIRUPDQFH 7KH 1LHOVHQV 0HDVXULQJ 3URJUDP 3HUIRUPDQFH 5(6($5&+ +<327+(6(6 5(6($5&+ 0(7+2'2/2*< &RQWUROOLQJ 7KH 0DUNHWLQJ 0L[ (VWDEOLVKLQJ D &RPSDUDWLYH (TXLW\ %HQFKPDUN %HQFKPDUN 1LHOVHQ
PAGE 5

/,67 2) 7$%/(6 7DEOH 3DJH %UHDNRXW RI 1LHOVHQ 0HWHUHG 2YHUQLJKW 'DWDEDVH &RPSDULVRQ RI 6ZHHS YHUVXV 1RQVZHHS &DVHV 'HVFULSWLYH +RXVHKROG 'DWD &RPSDULVRQ RI 6ZHHS YHUVXV 1RQVZHHS &DVHV 3DLUHG 6DPSOH 77HVW E\ +RXVHKROGV
PAGE 6

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f FDQQRW EH LJQRUHG 6\QWKHVL]LQJ HOHPHQWV IURP VHYHUDO FRQVXPHUEDVHG EUDQG HTXLW\ WKHRULHV WKH LQYHVWLJDWRU SURSRVHG WKDW D YL

PAGE 7

SURJUDP EUDQG HTXLW\ PRGHO IRFXVHV RQ WKH GLIIHUHQWLDO UDWLQJV UHVSRQVH RI D SURJUDP WR LWV GLUHFW FRPSHWLWRUV DQG WR LWV OHDGLQ SURJUDPPLQJ $IWHU HVWDEOLVKLQJ WKLV IUDPHZRUN D QXPEHU RI K\SRWKHVHV ZHUH WHVWHG ZLWKLQ D FDVH VWXG\ IRUPDW 7KH GDLO\ UDWLQJV SHUIRUPDQFH RI WKUHH 30 ORFDO QHZVFDVWV LQ D PDMRU VRXWKHDVW WHOHYLVLRQ PDUNHW ZHUH DQDO\]HG XVLQJ RQH VWDWLRQ DV DQ HTXLW\ EHQFKPDUN $FNQRZOHGJLQJ WKH ODFN RI H[WHUQDO YDOLGLW\ DQG WKH QHHG IRU IXUWKHU WHVWLQJ DFURVV PDQ\ PDUNHWV DQG SURJUDP JHQUHV WKH UHVXOWV RI WKLV H[SORUDWRU\ VWXG\ ZHUH HQFRXUDJLQJ :LWK DLO K\SRWKHVHV VXSSRUWHG WR VRPH GHJUHH WKH VWXG\ FRQFOXGHV ZLWK VHYHUDO UHFRPPHQGDWLRQV IRU D SURJUDP HTXLW\ UHVHDUFK DJHQGD IRU IXWXUH ZRUN LQ WKH ILHOG

PAGE 8

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f VWDWHG f0\ H[SHULHQFH DW 3t* WDXJKW PH WKDW EUDQGLQJ LV DERXW ILQGLQJ WKH XQLTXHQHVV LQ WZR YHU\ VLPLODU SURGXFWV SUHVHQWLQJ WKDW SRVLWLYH XQLTXHQHVV WR WKH FRQVXPHU DQG GULYLQJ WKDW PHVVDJH KRPHf S f 7KLV XQSUHFHGHQWHG PDQDJHPHQW GHFLVLRQ WR KLUH VRPHRQH RXWVLGH WKH FRUH LQGXVWU\ WR GLUHFW WKH RSHUDWLRQV RI D QHWZRUN fIODJVKLSf VWDWLRQ VHUYHV DV MXVW RQH H[DPSOH RI KRZ WKH HOHFWURQLF PHGLD KDYH EHFRPH FDSWLYDWHG ZLWK EUDQG PDQDJHPHQW WKHRU\ ,Q DQ HIIRUW WR FRSH ZLWK XQSUHFHGHQWHG FRPSHWLWLRQ DXGLHQFH IUDJPHQWDWLRQ DQG GHFOLQLQJ PDUNHW VKDUHV EURDGFDVWHUV KDYH ORRNHG WR WKH UHWDLO FRQVXPHU JRRGV LQGXVWU\ IRU LQVSLUDWLRQ 7KH UHVXOW KDV EHHQ WKH HDJHU DGRSWLRQ RI WKH MDUJRQ LI QRW WKH VXEVWDQFH RI EUDQG PDQDJHPHQW 7KH PHGLD WUDGH SUHVV DUH ILOOHG ZLWK FXUVRU\ UHIHUHQFHV WR fEUDQG LGHQWLW\f fEUDQG LPDJHf DQG fEUDQG

PAGE 9

H[WHQVLRQf ,Q SDUWLFXODU WKHUH KDV EHHQ D JURZLQJ LQWHUHVW LQ WKDW PRVW PXGGOHG RI EUDQG PDQDJHPHQW FRQFHSWV fEUDQG HTXLW\f $OWKRXJK WKHUH DUH PDQ\ GLYHUJHQW FRQFHSWXDOL]DWLRQV RI EUDQG HTXLW\ WKHUH LV XQLYHUVDO DJUHHPHQW WKDW HTXLW\ HQKDQFHV D SURGXFWfV SHUIRUPDQFH LQ WKH FRQVXPHU PDUNHWSODFH 7KDW LV HTXLW\ KHOSV UHLQIRUFH FRQVXPHU OR\DOW\ DWWUDFW QHZ FXVWRPHUV DQG LQVXODWH WKH SURGXFW IURP FRPSHWLWLYH DWWDFN )LHUFH FRPSHWLWLRQ LV WKH FDWDO\VW IRU PRVW EXVLQHVVHV WR ORRN EH\RQG VKRUWWHUP VDOHV JRDOV DQG IRFXV RQ WKH PRUH HQGXULQJ DGYDQWDJHV RI FRQVXPHUEDVHG EUDQG HTXLW\ (QFRXQWHULQJ GR]HQV LI QRW KXQGUHGV RI FDEOH DQG VDWHOOLWH SURJUDP FRPSHWLWRUV PDQ\ WHOHYLVLRQ H[HFXWLYHV KDYH HPEUDFHG D PDQDJHPHQW SHUVSHFWLYH ZKHUH QHWZRUNV VWDWLRQV RU SURJUDPV DUH UHJDUGHG DV FRQVXPHU EUDQGV $V ZLWK FRQYHQWLRQDO FRQVXPHU JRRGV WKH H[SHFWDWLRQ LV WKDW EUDQG HTXLW\ ZLOO HQKDQFH WKH SHUIRUPDQFH FKDUDFWHULVWLFV RI D SURJUDP LQ WKH DXGLHQFH PDUNHWSODFH 0RUH VSHFLILFDOO\ EUDQG HTXLW\ WKHRU\ LV SUHVXPHG WR KHOS EURDGFDVWHUV GHYHORS DQG VFKHGXOH SURJUDPV WKDW ZLOO GHOLYHU GRPLQDQW 1LHOVHQ UDWLQJV $ JRDO RI EUDQG HTXLW\ UHVHDUFK KDV EHHQ WR GHWHUPLQH WKH SURSRUWLRQ RI D EUDQGfV PDUNHW VKDUH WKDW LV PRVW YXOQHUDEOH WR FRQVXPHU EUDQG VZLWFKLQJ 5HFRJQL]LQJ WKH YXOQHUDELOLW\ RI D EUDQG DOORZV EUDQG PDQDJHUV WR PDNH LQIRUPHG VWUDWHJLF GHFLVLRQV %\ LJQRULQJ WKH V\PSWRPV RI SRRU HTXLW\ PDQDJHUV UXQ WKH ULVN RI XQIRUHVHHQ GHFOLQHV LQ PDUNHW VKDUHV DQG GLVDSSRLQWLQJ ERWWRP OLQHV .DSIHUHU f 6LPLODUO\ PDQDJHUV RI WKH HOHFWURQLF PHGLD ZLVK DOVR WR GLDJQRVH DQ\ YXOQHUDELOLWLHV LQ D SURJUDPfV

PAGE 10

DXGLHQFH SHUIRUPDQFH $OWKRXJK WKHUH LV FRQVLGHUDEOH SURIHVVLRQDO DQG VFKRODUO\ UHVHDUFK RQ EUDQG HTXLW\ GHDOLQJ ZLWK FRQYHQWLRQDO FRQVXPHU JRRGV WKHUH KDV EHHQ OLWWOH ZRUN GRQH RQ DGDSWLQJ WKHVH WKHRUHWLFDO SUHFHSWV WR HOHFWURQLF PHGLD )XUWKHUPRUH DQ\ PHDQLQJIXO VWXG\ RI SURJUDP UDWLQJV SHUIRUPDQFH FDQQRW LJQRUH WKH SRWHQW LQIOXHQFH RI OHDGLQ SURJUDPPLQJ RU LQKHULWDQFH HIIHFWV ,W LV QR VHFUHW DPRQJ SURJUDPPHUV WKDW WKH EHVW SUHGLFWRU RI D SURJUDPfV DXGLHQFH VL]H LV RIWHQ WKH VL]H RI WKH DXGLHQFH OHDGLQJ LQWR LW :KLOH WKHUH LV FRQVLGHUDEOH OLWHUDWXUH RQ WKLV VXEMHFW QR RQH KDV DSSURDFKHG WKLV SURJUDP SHUIRUPDQFH LVVXH IURP WKH YDQWDJH SRLQW RI EUDQG HTXLW\ 7KH SXUSRVH RI WKLV VWXG\ LV WZRIROG 7KH ILUVW LV WR LQFUHDVH WKH ERG\ RI NQRZOHGJH RI EUDQG HTXLW\ WKHRU\ E\ H[SORULQJ LWV DSSOLFDELOLW\ WR WHOHYLVLRQ SURJUDP UDWLQJV SHUIRUPDQFH 7KH VHFRQG LV WR RIIHU DQ H[SORUDWRU\ FDVH VWXG\ WKDW WHVWV D SURSRVHG WKHRUHWLFDO DSSURDFK WKDW V\QWKHVL]HV DVSHFWV RI WUDGLWLRQDO EUDQG HTXLW\ ZLWK 1LHOVHQ WHOHYLVLRQ UDWLQJV 8VLQJ GLDU\EDVHG fVZHHSVf GDWD DQG HOHFWURQLFDOO\ PHWHUHG fRYHUQLJKWf UDWLQJV WKLV VWXG\ DWWHPSWV WR PHDVXUH WKH UHODWLYH LQIOXHQFH RI SURJUDP EUDQG HTXLW\ RQ WKH SHUIRUPDQFH RI WKUHH ODWH HYHQLQJ QHZVFDVWV LQ D GHVLJQDWHG VRXWKHDVW PDUNHW 3DUWLFXODU DWWHQWLRQ LV SDLG WR WKH UHODWLRQVKLS EHWZHHQ WKH QHZVFDVWV WR WKHLU OHDGLQ SURJUDPV DV D PHDQV RI HYDOXDWLQJ SURJUDP EUDQG HTXLW\

PAGE 11

&+$37(5 5(9,(: 2) /,7(5$785( %UDQGLQJ KDV EHFRPH WKH EX]]ZRUG RI WKH GD\ 0RUH LV ULGLQJ RQ WKH VXFFHVV RU IDLOXUH RI EURDGFDVW EUDQGLQJ WKDQ DW DQ\ WLPH LQ WKH PHGLXPfV KLVWRU\ 6WD\ 7XQHG S f &RPSHWLWLRQ &KDQJHV (YHU\WKLQJ %UDQGLQJ DQG PDQ\ RI WKH QRWLRQV RI EUDQG PDQDJHPHQW DUH QRW QHZ WR PRVW $PHULFDQ FRQVXPHU JRRGV 6RPH RI WRGD\fV PRVW SURPLQHQW EUDQGV VXFK DV &RNH /HYL 0D[ZHOO +RXVH %XGZHL]HU &DPSEHOO DQG .HOORJJ EHJDQ WKHLU EUDQGLQJ HIIRUWV LQ WKH V $DNHU f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

PAGE 12

$V D PDUNHWLQJ YLFH SUHVLGHQW DW 3URFWRU DQG *DPEOH UHFHQWO\ SURFODLPHG f,Q DQ LQIRUPDWLRQULFK HQYLURQPHQW LPSDWLHQW UHWDLOHUV GRQfW KHVLWDWH WR UHOLQTXLVK EUDQGV WKDW GRQfW GHOLYHU FRQVLVWHQW DQG SURILWDEOH VDOHV ,Q RUGHU WR VXUYLYH LQ DQ LQFUHDVLQJO\ FRPSHWLWLYH PDUNHWSODFH D KLJK OHYHO RI EUDQG ILWQHVV LV UHTXLUHG *OHDVRQ S f 7KH VDPH FDQ EH VDLG IRU WKH EUDQG ILWQHVV RI D WHOHYLVLRQ SURJUDP LQ DQ HYHULQFUHDVLQJ FRPSHWLWLYH HQYLURQPHQW 7KH EXVLQHVV RI FRPPHUFLDO EURDGFDVWLQJ LV WKH VHOOLQJ RI DXGLHQFHV WR DGYHUWLVHUV DQG LQ DQ fLQIRUPDWLRQ ULFKf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b RI WKH GDLO\ SURJUDPPLQJ IRU WKHLU ORFDO DIILOLDWHG VWDWLRQV /RQJ f 6\QGLFDWHG SURJUDPPLQJ ILOOHG WKH UHPDLQLQJ KRXUV EXW KHUH DJDLQ WKH YLHZHU FKRLFHV ZHUH PHDJHU EHFDXVH RI WKH IHZ QXPEHU RI VWDWLRQV OLFHQVHG WR HDFK FRPPXQLW\ %\ LQ WKH PLG V WKH FRPSHWLWLYH SLFWXUH EHJDQ WR FKDQJH ZKHQ )R[ EHFDPH D LHJLWLPDWH QHWZRUN SOD\HU VLSKRQLQJ RII \RXQJ DGXOW YLHZHUV IURP WKH

PAGE 13

f%LJ 7KUHHf QHWZRUNV 7KH V DOVR ZLWQHVVHG WKH FRPLQJ RI DJH IRU FDEOH WHOHYLVLRQ &DEOH SURJUDP QHWZRUNV VXFK DV &11 (631 079 $t( 1LFNHORGHRQ DQG 'LVFRYHU\ EHJDQ WR HQFURDFK RQ WKH DXGLHQFH WHUULWRU\ WKDW ZDV RQFH WKH H[FOXVLYH GRPDLQ RI WKH EURDGFDVW QHWZRUNV DQG WKHLU DIILOLDWHG VWDWLRQV $V WKH QXPEHU RI XQLTXH FDEOH SURJUDPPLQJ RSWLRQV LQFUHDVHG VR GLG WKH QXPEHU RI VXEVFULEHUV DQG ZLWK PRUH VXEVFULEHUV FDPH PRUH DXGLHQFH IUDJPHQWDWLRQ /LQ f )URP WR $%& &%6 DQG 1%& DIILOLDWHG VWDWLRQV FRPELQHG ORVW RYHU b RI WKHLU YLHZLQJ KRXVHKROGV WR EDVLF FDEOH DQG RWKHU DOWHUQDWLYH PHGLD 0\HUV f ,Q WKH QHDU IXWXUH WHFKQRORJLFDO EUHDNWKURXJKV VXFK DV GLJLWDO FRPSUHVVLRQ DQG ILEHU RSWLFV ZLOO HQDEOH FDEOH DQG VDWHOOLWH FRPSDQLHV WR LQFUHDVH FKDQQHO FDSDFLW\ IURP D IHZ GR]HQ FKDQQHOV WR VHYHUDO KXQGUHG :LWK RYHU b RI WKH 86 KRXVHKROGV DOUHDG\ fZLUHGf WUDGLWLRQDO RYHUWKHDLU WHUUHVWULDOf EURDGFDVWHUV FDQ QR ORQJHU FRXQW RQ FDSWLYH DXGLHQFHV DQG JXDUDQWHHG SURILWV )RUPHU 1%& SURJUDPPLQJ H[HFXWLYH DQG PHGLD HQWUHSUHQHXU %UDQGRQ 7DUWLNRII FODLPHG LQ D LQWHUYLHZ WKDW ZLWK WKH DGYHQW RI VR PDQ\ QHZ SURJUDP FKRLFHV RQ WKH KRUL]RQ $PHULFD VRRQ ZLOO ZLWQHVV D SDUDGLJP VKLIW LQ SRZHU IURP SURJUDP SURYLGHUV WR YLHZHUV f6RPH VD\ WKDW WHOHYLVLRQ ZDV RQFH D PHGLXP FRQWUROOHG E\ W\UDQQ\ 1RZ LWfV JRLQJ WR EH D PHGLXP FRQWUROOHG E\ GHPRFUDF\f %DUWRQ S f $ FRPPHQWDU\ E\ UHVSHFWHG HFRQRPLVW $GDP 7KHLUHU f RI WKH )OHULWDJH )RXQGDWLRQ FODLPHG WKH LQWURGXFWLRQ RI VR PXFK FKRLFH LV fD VLJQDO WR WKH ZRUOG WKDW WKH ROG PHGLD HPSLUHV DUH PRGHUQGD\ GLQRVDXUV KHDGHG IRU H[WLQFWLRQf S f

PAGE 14

&RPSOHPHQWLQJ WKH VLJQLILFDQW FKDQJHV LQ PHGLD WHFKQRORJ\ KDYH EHHQ HTXDOO\ LPSRUWDQW FKDQJHV LQ JRYHUQPHQW UHJXODWLRQ RI HOHFWURQLF PHGLD 7KH KLJKO\ SXEOLFL]HG 7HOHFRPPXQLFDWLRQV %LOO RI ZDV D PDQGDWH WR LQFUHDVH PDUNHWGULYHQ FRPSHWLWLRQ E\ XQVKDFNOLQJ JRYHUQPHQW UHVWULFWLRQV RQ WHOHFRPPXQLFDWLRQV %URDGFDVWLQJ FDEOH VDWHOOLWH DQG WHOHSKRQH GHOLYHU\ V\VWHPV KDYH EHHQ JLYHQ WKH OHJDO JUHHQ OLJKW WR FURVV ERXQGDULHV DQG FRPSHWH IRU FRQVXPHUV &RQJUHVVLRQDO DGYRFDWHV IRU WKH ELOO FODLP LW ZLOO VRRQ XQOHDVK D WRUUHQW RI FRPSHWLWLRQ KHUDOGLQJ QRWKLQJ OHVV WKDQ fWKH GDZQ RI D QHZ LQIRUPDWLRQ DJHf 6WHUQ S f 0HGLD $YDLODEOH YHUVXV 0HGLD 8VHG $GGLQJ IXHO WR WKH ILUH RI WKLV HPHUJLQJ FRPSHWLWLYH PDUNHWSODFH LV WKH IDFW WKDW ZKLOH WKH DPRXQW RI SURJUDP FKRLFH FRQWLQXHV WR LQFUHDVH WKHUH KDV EHHQ QR FRUUHVSRQGLQJ LQFUHDVH LQ WKH DPRXQW RI WLPH SHRSOH VSHQG ZDWFKLQJ WHOHYLVLRQ 1LHOVHQ 0HGLD 5HVHDUFK UHSRUWV WKDW RYHU WKH SDVW \HDUV f WKH QXPEHU RI PLQXWHV SHU KRXVHKROG GHGLFDWHG WR WHOHYLVLRQ YLHZLQJ KDV UHPDLQHG XQFKDQJHG 1LHOVHQ $XGLHQFH f 7KH GLVWXUELQJ LPSOLFDWLRQV IRU WHOHYLVLRQ SURJUDPPHUV LV WKDW ZKLOH WKH QXPEHU RI VOLFHV LQ WKH SURJUDPPLQJ SLH KDV LQFUHDVHG WKH RYHUDOO VL]H RI WKH SLH LWVHOI KDV QRW LQFUHDVHG 6R IDU LQ WKH V PRUH FKRLFH KDV QRW WUDQVODWHG LQWR PRUH YLHZLQJ $ QXPEHU RI PHGLD REVHUYHUV KDYH UHIHUUHG WR WKLV SKHQRPHQRQ DV PHGLD fFDQQLEDOLVPf %XUJL t .DW] S f ZKHUH SURJUDP SURYLGHUV DUH IRUFHG LQWR fIHHGLQJf RQ WKH DXGLHQFHV RI IHOORZ SURJUDP SURYLGHUV 0£QGHVH S 6f 8VLQJ EUDQG PDUNHWLQJ WHUPLQRORJ\ WHOHYLVLRQ YLHZLQJ DSSHDUV WR

PAGE 15

KDYH HYROYHG LQWR D PDWXUH RU ]HUR VXP PDUNHW ZKHUH WKH QXPEHU RI DYDLODEOH FXVWRPHUV IRU D SURGXFW FDWHJRU\ LV VWDJQDQW 7KHUHIRUH DV PRUH EUDQGV HQWHU WKH PDUNHWSODFH WKH RQO\ PHDQV RI VXUYLYDO LV WR WDNH FXVWRPHUV DXGLHQFHVf DZD\ IURP FRPSHWLQJ EUDQGV .DSIHUHU /HKPDQQ t :LQHU f $ IXUWKHU IUXVWUDWLRQ IRU WHOHYLVLRQ SURJUDPPHUV LV WKDW GHVSLWH WKH KLVWRULF WHFKQRORJLFDO DQG UHJXODWRU\ EUHDNWKURXJKV WKDW KDYH IUHHG DXGLHQFHV IURP WKH fW\UDQQ\f RI OLPLWHG FKRLFH WKH W\SLFDO YLHZHU VWLOO SUHIHUV WR GHDO ZLWK RQO\ D KDQGIXO RI RSWLRQV 6HYHUDO VWXGLHV LQGLFDWH WKDW IRU DQ DYHUDJH $PHULFDQ KRXVHKROG WKHUH LV D YLHZLQJ WKUHVKROG RI DSSUR[LPDWHO\ D GR]HQ FKDQQHOV )RU H[DPSOH 1LHOVHQ 0HGLD 5HVHDUFK FRQGXFWV DQ DQQXDO QDWLRQDO VXUYH\ RI WHOHYLVLRQ YLHZLQJ FKDUDFWHULVWLFV RI ZKLFK f&KDQQHOV 5HFHLYHG YV &KDQQHOV 9LHZHGf LV D VSHFLDO VHFWLRQ 7KH UHSRUW UHYHDOHG WKDW ZKLOH WKH QXPEHU RI FKDQQHOV DYDLODEOH LQ WKH KRPH FRQWLQXHG WR JURZ WKH QXPEHU RI FKDQQHOV DFWXDOO\ YLHZHG KDG QRW JURZQ DSSUHFLDEO\ 7KH VWXG\ FRQFOXGHG WKDW ZKLOH WKH DYHUDJH FDEOH KRXVHKROG FDQ QRZ UHFHLYH FKDQQHOV RQO\ FKDQQHOV DUH DFWXDOO\ YLHZHG :KHQ RU PRUH FKDQQHOV DUH DYDLODEOH YLHZHUVKLS LQFUHDVHV RQO\ VOLJKWO\ WR FKDQQHOV 1LHOVHQ $XGLHQFH f 6HYHUDO UHVHDUFKHUV KDYH DGRSWHG WKH FRQFHSW RI FKDQQHO UHSHUWRLUH WR GHVFULEH WKLV OLPLWHG DUUD\ RI FKDQQHOV IURP ZKLFK DXGLHQFHV VHOHFW SURJUDPPLQJ :HEVWHU t /LFKW\ f 7KLV FKDQQHO UHSHUWRLUH LV VLPLODU LQ GHILQLWLRQ WR ZKDW DGYHUWLVLQJ UHVHDUFKHUV FDOO DQ HYRNHG VHW ZKHUH D FRQVXPHU FRQVLGHUV RQO\ D SRUWLRQ RI WKH EUDQGV DYDLODEOH LQ D SXUFKDVH VLWXDWLRQ 7KH PDWKHPDWLFV RI WKLV QHZ SURJUDPPLQJ PDUNHWSODFH DUH QRW ORVW RQ 1%& SUHVLGHQW %RE :ULJKW

PAGE 16

ZKHQ KH VWDWHV f0\ ELJJHVW ZRUU\ LV RXU DELOLW\ WR FRQWLQXH WR DWWUDFW D GLVSURSRUWLRQDWHVL]H DXGLHQFH IRU RXU EURDGFDVW VWDWLRQV YHUVXV RWKHU PHDQV RI JDWKHULQJ DXGLHQFHVf %t& +DOO RI )DPH S f 5HWKLQNLQJ 3URJUDP &RQWHQW 7KH ULVH LQ FRPSHWLWLRQ KDV OHG WR D UHH[DPLQDWLRQ RI SURJUDP FRQWHQW WKHRU\ )RU H[DPSOH SULPH WLPH SURJUDPV IRU \HDUV ZHUH SURGXFHG DFFRUGLQJ WR WKH OHDVW REMHFWLRQDEOH SURJUDP WKHRU\ /23f 7KH WKHRU\ LV EDVHG RQ WKH FRUH SUHPLVH RI PRGHUDWH OLNLQJ $XGLHQFH UHVHDUFKHUV GLVFRYHUHG HDUO\ RQ WKDW PRVW SHRSOH GR QRW ZDWFK WHOHYLVLRQ DORQH 7KHUHIRUH LQGLYLGXDO SUHIHUHQFH LV RIWHQ PHGLDWHG E\ JURXS G\QDPLFV ZKHUH H[WUHPH RSLQLRQV HLWKHU SRVLWLYH RU QHJDWLYH DUH VXSSUHVVHG LQ IDYRU RI D JURXS FRPSURPLVH %\ GHILQLWLRQ D FRPSURPLVH LV D PRYHPHQW WRZDUGV WKH FRPPRQ FHQWHU RU fORZHVW FRPPRQ GHQRPLQDWRUf (DVWPDQ f 7KLV VWUDWHJ\ PLJKW JHQHUDWH fUHSHDW FXVWRPHUVf IRU D SURJUDP EXW QRW QHFHVVDULO\ KLJKO\ FRPPLWWHG IDQV 7KH /23 WKHRU\ ZRUNV ZHOO ZKHQ WKHUH DUH IHZ FRPSHWLWRUV EXW DV YLHZHU RSWLRQV LQFUHDVH GUDPDWLFDOO\ SURJUDPV GHVLJQHG IRU PRGHUDWH OLNLQJ FDQ EHFRPH D OLDELOLW\ $V EUDQG PDQDJHUV RI PDQ\ FRQVXPHU JRRGV KDYH EHHQ IRUFHG WR PDUNHW WKHLU SURGXFWV WR VHOHFW FRQVXPHU VHJPHQWV VR PHGLD SURJUDPPHUV DUH QRZ REOLJHG WR WDUJHW WKHLU SURJUDP FRQWHQW WR WKH VSHFLILF QHHGV RI IDLUO\ QDUURZ VSHFLDOL]HG DXGLHQFH VHJPHQWV &RRSHU f $V ZLOO EH GHPRQVWUDWHG ODWHU fPRGHUDWH OLNLQJf LV QRW WKH VWXII RI EUDQG HTXLW\

PAGE 17

%LUWK RI D %X]]ZRUG 2QH RI WKH ILUVW RFFDVLRQV ZKHUH WKH FRQFHSW RI PHGLD DV EUDQGV ZDV SUHVHQWHG LQ D SXEOLF IRUXP ZDV D NH\QRWH VSHHFK PDGH LQ DW WKH $GYHUWLVLQJ 5HVHDUFK )RXQGDWLRQ $QQXDO &RQIHUHQFH E\ 'DYLG %HQGHU 3UHVLGHQW DQG &(2 RI 0HGLDPDUN 5HVHDUFK +H SURIHVVHG WKDW PHGLD ERWK SULQW DQG HOHFWURQLFf DUH VLPLODU WR FRQVXPHU SURGXFWV DQG VHUYLFHV DQG WKDW WUDGLWLRQDO EUDQG PDUNHWLQJ WKHRU\ DQG SURFHGXUHV FDQ EH DSSOLHG VXFFHVVIXOO\ WR PHGLD WR JHQHUDWH PRUH EXVLQHVV +H VWDWHG WKDW fPHGLD YHKLFOHV DUH RQH RI WKH IHZ H[DPSOHV RI EUDQGV ZKLFK GRQfW IRU WKH PRVW SDUW FRQFHLYH RI WKHPVHOYHV DV EUDQGV DV D UHVXOW WKH\ RIWHQ DUH QRW DEOH WR WDNH DGYDQWDJH RI WKH LQVLJKW WKDW DFFUXHV WR WKRVH DZDUH RI WKLQNLQJ DQG UHVHDUFK RQ EUDQGVf %HQGHU S f /HVV WKDQ WKUHH PRQWKV DIWHU %HQGHUfV VSHHFK D IURQWn SDJH KHDGOLQH LQ $GYHUWLVLQJ $JH GHFODUHG f+HUH &RPH 7KH 1HZHVW %UDQGV 1%& &%6 $%&f 0£QGHVH S f 6LQFH WKHQ PHGLD H[HFXWLYHV DQG PHGLD WUDGH MRXUQDOV VXFK DV %URDGFDVWLQJ DQG &DEOH (OHFWURQLF 0HGLD 0HGLD :HHN DQG $GYHUWLVLQJ $JH KDYH EHFRPH HYHU PRUH FRPIRUWDEOH ZLWK WKH SKUDVHRORJ\ RI EUDQGLQJ )RU H[DPSOH RUJDQL]DWLRQV VXFK DV WKH 1DWLRQDO $VVRFLDWLRQ RI 7HOHYLVLRQ 3URJUDP ([HFXWLYHV 1$73(f DQG 3URPRWLRQ DQG 0DUNHWLQJ ([HFXWLYHV ,Q (OHFWURQLF 0HGLD 3URPD[f KDYH VSRQVRUHG VHPLQDUV DGGUHVVLQJ f7KH %UDQGLQJ %XJf 1$73( ff 2XU %UDQG 1HZ :RUOGf 1$73( f f%UDQG %XLOGLQJ DQG 9LVXDO 'HVLJQf f:HOFRPH WR 7KH %UDQG 5HYROXWLRQf DQG f%UDQGLQJ
PAGE 18

SUHPLHU VHDVRQ 1%& SODFHG IXOOSDJH DGV LQ D QXPEHU RI SXEOLFDWLRQV ERDVWLQJ WKDW LW ZDV f$PHULFDfV /HDGLQJ 1HWZRUN %UDQGf 1%& f 0HGLD :HHN GHGLFDWHG D PDMRU DUWLFOH WR f1HWZRUN %UDQG 0DUNHWLQJ )HYHUf 'XSUHH S f $ WUDGH MRXUQDO DG IRU WKH RIIQHWZRUN V\QGLFDWHG YHUVLRQ RI f'XH 6RXWKf EUDJJHG WKDW WKH SURJUDP ZDV fD SURYHQ EUDQG ZLWK D OR\DO DXGLHQFHf 'XH 6RXWK S f $ QHZO\ SXEOLVKHG 1$73( EURFKXUH PHQWLRQV KRZ EURDGFDVWHUV PXVW fWDFNOH WKH FKDOOHQJHV RI EUDQGLQJ \RXU SURJUDP RU VWDWLRQf 1$73( f (YHQ WKH PDLQVWUHDP SUHVV DSSHDUV WR KDYH DGRSWHG EUDQGLQJ MDUJRQ $ VXEKHDGOLQH LQ D 1HZ
PAGE 19

LWV RZQ EUDQG RI VKRZf 6KDUNH\ S f 5DWKHU WKDQ IRFXVLQJ RQ LQGLYLGXDO QHWZRUNV RU SURJUDPV -RQDWKDQ 6LPV YLFH SUHVLGHQW RI UHVHDUFK IRU WKH &DEOH $GYHUWLVLQJ %XUHDX &$%f LQGLFWV WKH HQWLUH PHGLXP RI EURDGFDVW WHOHYLVLRQ SURFODLPLQJ fD GHFOLQH LQ EURDGFDVW HTXLW\f 6LPV S f ,Q UHFHQW \HDUV EURDGFDVWHUV KDYH JUDSSOHG ZLWK WKH TXHVWLRQ RI RYHUODSSLQJ EUDQG LGHQWLWLHV 1HWZRUNV VWDWLRQV DQG LQGLYLGXDO SURJUDPV FUHDWH D VRPHWLPHV DZNZDUG EUDQGLQJ KLHUDUFK\ WKDW KDV EHHQ GHVFULEHG DV fWZR SHRSOH LQ D GRQNH\ FRVWXPH QRW DOZD\V JRLQJ LQ WKH VDPH GLUHFWLRQ EXW ERWK WU\LQJ WR JHW WR WKH VDPH SODFHf 5DWKEXQ S f 6RPH SURJUDP SURGXFHUV VXFK DV %DUU\ 7KXUVWRQ RI &ROXPELD 7ULVWDU DUH FULWLFDO RI QHWZRUN LQWLPLGDWLRQ WKDW FRHUFHV VWDWLRQV LQWR fVDFULILFLQJ WKHLU RZQ LGHQWLWLHV DW WKH QHWZRUNfV DOWHUf )UHHPDQ t 6WDQOH\ S f %UDQGLQJ /RFDO 1HZV 7KH QRWLRQV RI EUDQG PDQDJHPHQW DQG SDUWLFXODUO\ EUDQG HTXLW\ DUH QR PRUH SUHYDOHQW WKDQ LQ GLVFXVVLRQV RI ORFDO QHZVFDVWV 3URYLGLQJ DV PXFK DV b RI D VWDWLRQfV RYHUDOO VDOHV UHYHQXH WKH ILQDQFLDO DQG SXEOLF UHODWLRQV EHQHILWV RI D VXFFHVVIXO QHZV RSHUDWLRQ FDQ EH HQRUPRXV (DVWPDQ 7REHQNLQ f EXW WKHUH FDQ DOVR EH D GRZQVLGH 1HZV SURJUDPPLQJ LV H[SHQVLYH $ VWDWLRQ WKDW UDQNV WKLUG LQ WKH PDUNHW PD\ QHYHU VHH DQ\ UHWXUQ RQ LWV LQYHVWPHQW 0HGLD REVHUYHUV VXFK DV 5RQ $OULGJH (GLWRULDO 'LUHFWRU RI (OHFWURQLF 0HGLD FODLP WKDW fVRPH IRONV DUH DOUHDG\ VD\LQJ WKDW WKH WKLUGSODFHG QHZVFDVW LV DQ HQGDQJHUHG VSHFLHVf $OULGJH S f

PAGE 20

)RU PDQ\ VWDWLRQV ORFDO QHZV KDV EHFRPH WKH RQO\ HQGXULQJ DQG IDPLOLDU SURJUDP DVVHW $W D UHFHQW 5DGLR DQG 7HOHYLVLRQ 1HZV 'LUHFWRUV $VVRFLDWLRQ 571'$f FRQYHQWLRQ D MRXUQDOLVW UHSRUWHG WKDW fLQ DQ HQYLURQPHQW ZKHUH GHFDGHVROG QHWZRUN DIILOLDWLRQV FDQ EH EURNHQ RYHUQLJKW ORFDO QHZVFDVWV UHPDLQ WKH VWURQJHVW DQG PRVW VWDEOH IRUP RI EUDQG HTXLW\f 7REHQNLQ S f 6XPPDU\ )URP REVHUYLQJ LQGXVWU\ SUDFWLFHV DQG UHDGLQJ PHGLD WUDGH MRXUQDOV LW LV REYLRXV WKDW EUDQGLQJ KDV LQGHHG EHFRPH fWKH EX]]ZRUG RI WKH GD\f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f DQ RYHUYLHZ RI EUDQG HTXLW\ UHVHDUFK GHULYHG SULPDULO\ IURP D FRQYHQWLRQDO

PAGE 21

FRQVXPHU JRRGV SHUVSHFWLYH Ef D UHYLHZ RI UHVHDUFK RQ DXGLHQFH LQKHULWDQFH HIIHFWV DQG Ff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t %HOFK &REE:DOJUHQ 5XEOH t 'RQWKX 6ULYDVWDYD t 6KRFNHU f %HFDXVH FRQVXPHUV RIWHQ ODFN WKH PRWLYDWLRQ FDSDFLW\ RU RSSRUWXQLW\ WR SURFHVV DOO SURGXFW LQIRUPDWLRQ WR ZKLFK WKH\ DUH H[SRVHG LQ D WKRXJKWIXO RU GHOLEHUDWLYH PDQQHU WKH\ RSW IRU TXLFN UHVROXWLRQ WHFKQLTXHV VWRUHG LQ PHPRU\ .DUGHV f 6WURQJ EUDQGV DVVLVW LQ WKLV KHXULVWLF SURFHVV %LHO f RIIHUV WKH IROORZLQJ LQVLJKW 2Q D YHU\ SUDFWLFDO OHYHO FRQVXPHUV OLNH EUDQGV EHFDXVH WKH\ SDFNDJH PHDQLQJ 7KH\ IRUP D NLQG RI VKRUWKDQG WKDW PDNHV FKRLFH HDVLHU 7KH\ OHW RQH HVFDSH IURP D IHDWXUHE\IHDWXUH DQDO\VLV RI FDWHJRU\

PAGE 22

DOWHUQDWLYHV DQG VR LQ D ZRUOG ZKHUH WLPH LV DQ HYHUGLPLQLVKLQJ FRPPRGLW\ EUDQGV PDNH LW HDVLHU WR VWRUH HYDOXDWLRQV S f ,Q VRFLDO SV\FKRORJ\ WKLV VDPH SURFHVV FRQWULEXWHV WR WKH IRUPDWLRQ RI VRFLDO VWHUHRW\SHV ZKHUH SHRSOH XVH SULRU NQRZOHGJH DQG IHHOLQJV DWWLWXGHVf DERXW JURXSV WR PDNH VZLIW LQIHUHQFHV DERXW LQGLYLGXDO PHPEHUV %HLNH t 6KHUPDQ f 6WHUHRW\SLQJ DQG EUDQGLQJ DUH HVVHQWLDOO\ WZR GLIIHUHQW PDQLIHVWDWLRQV RI WKH VDPH SKHQRPHQRQ (DFK VHUYHV WR VWUHDPOLQH FRJQLWLYH SURFHVVLQJ ,Q D KLJKO\ FRPSHWLWLYH WLPHVHQVLWLYH HQYLURQPHQW VWURQJ EUDQGV KDYH D EHWWHU FKDQFH RI EHLQJ SXUFKDVHG WKDQ ZHDN EUDQGV 6WURQJ EUDQGV DOVR FXOWLYDWH KDELWV 5RVHQVWHLQ DQG *UDQW f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t %HOFK f

PAGE 23

7KH (PHUJHQFH RI %UDQG (TXLW\ :KLOH EUDQG PDQDJHPHQW KDG EHHQ RQ WKH $PHULFDQ EXVLQHVV VFHQH IRU GHFDGHV WKH VSHFLILF WRSLF RI EUDQG HTXLW\ GLG QRW EHFRPH SRSXODU XQWLO WKH YRODWLOH V ZKHQ RQFH SURXG EUDQGV VXFK DV 6HDUV ,%0 DQG &DGLOODF EHJDQ WR ORVH JURXQG WR ORZHUSULFHG JHQHULF FRPSHWLWLRQ 0DUNHW VKDUHV GURSSHG DV D VHULRXV QDWLRQZLGH UHFHVVLRQ IXUWKHU DJJUDYDWHG FRUSRUDWH SURILWV 7KLV KLVWRULF HFRQRPLF GRZQWXUQ LQVSLUHG D UDVK RI FRPSDQ\ FRQVROLGDWLRQV DQG fGRZQVL]LQJf DFWLRQV 0RUULV f 3UXGHQW FRVW FXWWLQJ LV LQGHHG RQH ZD\ WR VKRUH XS VDJJLQJ SURILW PDUJLQV EXW DV 6HDUVn &(2 $UWKXU 0DUWLQH] ZDUQV f
PAGE 24

FRPSHWLWLYH SUHVVXUHV DUH LQWHQVH DV YROXPH JDLQV VKDUH LQFUHDVHVf DUH GHULYHG IURP FRPSHWLWRUV UDWKHU WKDQ QHZ FDWHJRU\ XVHUV 8QGHU WKHVH f]HUR VXPf FLUFXPVWDQFHV DV WKH QXPEHU RI FRPSHWLWRUV LQFUHDVH WKH EDWWOH IRU PDUNHW VKDUH EHFRPHV PRUH DFXWH /HKPDQQ t :LQHU S f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f 0RXWKZDVKHV VKRXOG QRW EH FRPSDUHG WR GHRGRUDQWV QRU VKRXOG 79 VLWFRPV EH FRPSDUHG WR QHZVFDVWV &RQFHSWXDO 'HILQLWLRQV RI %UDQG (TXLW\ ,W LV GLIILFXOW WR ILQG DJUHHPHQW RQ WKH SURSHU FRQFHSWXDOL]DWLRQ RI HTXLW\ PXFK OHVV WKH UHODWLYH ZHLJKW RI HDFK HOHPHQW LQ GULYLQJ EUDQG FKRLFH $IWHU D

PAGE 25

SDUWLFXODUO\ GLVDSSRLQWLQJ FDU VHOOLQJ VHDVRQ IRU *HQHUDO 0RWRUV D IUXVWUDWHG *0 PDUNHW DQDO\VW DGPLWWHG WKDW fHYHQ WKRXJK EUDQG HTXLW\ PD\ VRPHWLPHV EH KDUG WR GHILQH LWfV SUHWW\ FOHDU ZKHQ \RXnYH ORVW LWf :LONLH S f 0RYLQJ IURP DQ XQGHUVWDQGLQJ RI EUDQGV WR D VLPLODU XQGHUVWDQGLQJ RI EUDQG HTXLW\ FDQ EH SHULORXV EHFDXVH VR PDQ\ RWKHU EUDQGLQJ WHUPV KDYH EHHQ XVHG DV V\QRQ\PV IRU HTXLW\ :KLOH DFDGHPLFV KDYH \HW WR UHDFK SHUIHFW DJUHHPHQW RQ VRPH RI WKHVH LWHPV PHGLD SURIHVVLRQDOV DQG MRXUQDOLVWV KDYH EHHQ WKH ZRUVW RIIHQGHUV 7KH IROORZLQJ LV D EULHI FODULILFDWLRQ RI VRPH FRPPRQ WHUPV WKDW LQ PDQ\ FDVHV VHUYH DV WKH FRQFHSWXDO EXLOGLQJ EORFNV RI D ODUJHU FRQVWUXFW NQRZQ DV EUDQG HTXLW\ 7KHVH GHILQLWLRQV DUH D V\QWKHVLV RI GHILQLWLRQV DQG FRQFHSWXDOL]DWLRQV IRXQG LQ VHYHUDO VRXUFHV LQFOXGLQJ +R\HU DQG 5URZQ f %HOFK DQG %HOFK f DQG .HOOHU f f %UDQGLQJ LV WKH SURFHVV RI QDPLQJ D SURGXFW RU VHUYLFH LQ RUGHU WR GLVWLQJXLVK LW IURP LWV FDWHJRU\ FRPSHWLWRUV f %UDQG H[WHQVLRQ LV WKH SURFHVV RI EUDQGLQJ D QHZ SURGXFW ZLWK DQ DOUHDG\ HVWDEOLVKHG EUDQG QDPH $Q H[DPSOH ZRXOG EH )RUG FUHDWLQJ D QHZ W\SH RI DXWRPRELOH RU WUXFN EXW VWLOO LGHQWLI\LQJ LW DV D )RUG SURGXFWf f %UDQG DZDUHQHVV RU EUDQG LGHQWLW\f LV WKH ILUVW UXQJ RQ WKH HTXLW\ ODGGHU ,W UHIHUV WR WKH VLPSOH IDPLOLDULW\ UHFDOO RU UHFRJQLWLRQf RI D EUDQG UHODWLYH WR LWV SURGXFW FDWHJRU\ f %UDQG LPDJH JRHV EH\RQG PHUH DZDUHQHVV DQG GHDOV ZLWK WKH WKRXJKWV DQG IHHOLQJV PHDQLQJf RI WKH EUDQG WR D FRQVXPHU ,W FDQ EH FRQFHLYHG DV D FOXVWHU RI DWWULEXWHV DQG DVVRFLDWLRQV WKDW FRQVXPHUV FRQQHFW ZLWK D

PAGE 26

VSHFLILF EUDQGHG SURGXFW 6RPH HTXLW\ UHVHDUFKHUV VXFK DV .HOOHU f FRPSDUH EUDQG DVVRFLDWLRQV LQ WHUPV RI IDYRUDELOLW\ VWUHQJWK DQG XQLTXHQHVV f %UDQG SRVLWLRQLQJ LV WKH LPDJH RI D EUDQG GHILQHG LQ UHODWLRQVKLS WR LWV PDUNHW FRPSHWLWRUV 7KH JRDO LV WR fSRVLWLRQf D EUDQG LQ D FRQVXPHUfV PLQG VR WKDW LW KDV D KLJKO\ GLVWLQFWLYH GLIIHUHQWLDWHGf LPDJH LQ FRPSDULVRQ WR EUDQGV RIIHUHG E\ FRPSHWLWRUV f %UDQG WULDO RU EUDQG VDPSOLQJf LV WKH LQLWLDO SXUFKDVH RU H[SHULHQFH ZLWK D EUDQG E\ D FRQVXPHU f %UDQG DWWLWXGH FDQ EH YLHZHG DV DQ H[WHQVLRQ RI EUDQG LPDJH LQ WKDW WKH WHUP UHIHUV QRW RQO\ WR WKRXJKWV DQG IHHOLQJV DERXW WKH EUDQG EXW HYDOXDWLRQV DQG PRVW LPSRUWDQWO\ SUHGLVSRVLWLRQV WR UHVSRQG SXUFKDVHf $ SRVLWLYH RU QHJDWLYH DWWLWXGH LV UHLQIRUFHG E\ H[SHULHQFLQJ D EUDQG ILUVW KDQG f %UDQG OR\DOW\ KDV RIWHQ EHHQ YLHZHG DV DQRWKHU QDPH IRU EUDQG HTXLW\ EXW IRU RXU SXUSRVHV D PXFK QDUURZHU GHILQLWLRQ KDV EHHQ DGRSWHG /R\DOW\ LV WKH GHJUHH WR ZKLFK D FRQVXPHU SXUFKDVHV UHSHDWHGO\ D FHUWDLQ EUDQG ZLWKRXW GLJUHVVLQJ WR DOWHUQDWLYH EUDQGV :LWKLQ WKLV FRQWH[W EUDQG OR\DOW\ LV D PHDVXUH RI FRQVXPHU SXUFKDVH EHKDYLRU f %UDQG FRPPLWPHQW DGGUHVVHV WKH XQGHUO\LQJ SV\FKRORJ\ RI EUDQG OR\DOW\ 7KDW LV FRPPLWPHQW LV WKH GHJUHH RI ILGHOLW\ WR D EUDQG WKDW UHJDUGOHVV RI SULFLQJ GLIIHUHQWLDOV SURPRWLRQV DQG RWKHU PDUNHW IDFWRUV WKH FRQVXPHU UHPDLQV OR\DO LQ D SXUFKDVH VLWXDWLRQ

PAGE 27

f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f )RU \HDUV UHVHDUFKHUV KDYH ZUHVWOHG ZLWK WKH GLVSDULWLHV EHWZHHQ WKH SV\FKRORJ\ RI FRQVXPHU SUHIHUHQFH DQG WKH EHKDYLRUDO UHDOLWLHV RI FRQVXPHU EUDQG FKRLFH 'HVSLWH WKHVH LQFRQJUXLWLHV EUDQG HTXLW\ KDV EHHQ GHILQHG W\SLFDOO\ IURP D FRQVXPHU DWWLWXGH SHUVSHFWLYH EHFDXVH LW LV RIWHQ OHVV H[SHQVLYH WKDQ WUDFNLQJ DQG PDWFKLQJ SURGXFW VDOHV ZLWK LQGLYLGXDO FRQVXPHUV $DNHU &KXUFKLOO f &RQVXPHU $WWLWXGHV YHUVXV &RQVXPHU %HKDYLRU :KLOH IHZ SHRSOH ZRXOG GLVDJUHH WKDW FRQVXPHUEDVHG EUDQG HTXLW\ LV XOWLPDWHO\ D VWDWH RI PLQG PDQ\ HTXLW\ UHVHDUFKHUV DVVXPH WKDW PRVW DWWLWXGLQDO RU SV\FKRORJLFDO FRPSRQHQWV RI HTXLW\ VXFK DV DZDUHQHVV LPDJHU\ SRVLWLRQLQJ DQG FRPPLWPHQWf PDQLIHVW WKHPVHOYHV LQ REVHUYDEOH SXUFKDVH EHKDYLRU 7KLV FRQVXPHU EHKDYLRU LQ WXUQ WUDQVODWHV LQWR PHDVXUHV RI

PAGE 28

EUDQG SHUIRUPDQFH LQ WKH PDUNHWSODFH ,Q RWKHU ZRUGV FHUWDLQ LQGLFDWRUV RI EUDQG SHUIRUPDQFH FDQ EH DFFHSWHG DV FLUFXPVWDQWLDO HYLGHQFH RI FRQVXPHU EDVHG EUDQG HTXLW\ 'HILQLWLRQV )URP WKH 3ULYDWH 6HFWRU 7KH 0DUNHWLQJ 6FLHQFH ,QVWLWXWH 06,f RIIHUV D NLQG RI DOOHQFRPSDVVLQJ GHILQLWLRQ LQ WKDW EUDQG HTXLW\ LV VDLG WR EH D VHW RI DVVRFLDWLRQV DQG EHKDYLRUV RQ WKH SDUW RI D EUDQGfV FRQVXPHUV FKDQQHO PHPEHUV DQG SDUHQW FRUSRUDWLRQ WKDW SHUPLWV WKH EUDQG WR HDUQ JUHDWHU YROXPH DQG JUHDWHU PDUJLQV WKDQ LW FRXOG ZLWKRXW WKH EUDQG QDPH DQG WKDW JLYHV WKH EUDQG D VWURQJ VXVWDLQDEOH DQG GLIIHUHQWLDWHG FRPSHWLWLYH DGYDQWDJH $QDQWDFKDUW S f 7KH LQVWLWXWH GRHV QRW IXUWKHU GHILQH ZKDW WKH VSHFLILF fDVVRFLDWLRQV DQG EHKDYLRUVf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f 2WKHU FRPSDQLHV VXFK DV WKH 8QWDV $GYHUWLVLQJ $JHQF\ FRPELQH DWWLWXGLQDO DQG EHKDYLRUDO HOHPHQWV LQWR WKHLU fHTXLW\ DXGLWf 0HDVXUHV VXFK DV

PAGE 29

EUDQG DZDUHQHVV LPDJH DWWULEXWH UDWLQJV DQG DVVXPHG OHDGHUVKLS DUH PHUJHG ZLWK PDUNHW VKDUH GDWD VHQVLWLYLW\ WR SULFLQJ DQG SXUFKDVH FRQVLVWHQF\ RYHU WLPH 7KH RYHUDOO SXUSRVH RI WKH DXGLW LV WR GHWHUPLQH WKH VWURQJ RU ZHDN ELDV FRQVXPHUV KDYH UHODWLYH WR RWKHU EUDQGV /LQWDV f 7KH 13' *URXS f%UDQG %XLOGHUf PRGHO DOVR FRPELQHV FRQVXPHU SHUFHSWLRQV DQG DFWXDO LQPDUNHW SHUIRUPDQFH WR DUULYH DW DQ LQGLFDWRU FDOOHG EUDQG KHDOWK )DFWRUV VXFK DV FRQVXPHU DZDUHQHVV SURGXFW KHULWDJH RU UHSXWDWLRQ DQG FRQVXPHU OR\DOW\ FRQWULEXWH WR WKH HVVHQWLDO KHDOWK RI WKH EUDQGHG SURGXFW 13' f 7KH 6LPPRQV 0DUNHW 5HVHDUFK %XUHDXfV (4; HTXLW\ PHDVXUH WUDFNV EUDQG XVDJH DQG WKHQ LQWHJUDWHV PRUH DEVWUDFW QRWLRQV LQFOXGLQJ DWWULEXWH LPSRUWDQFH DQG GHVLUHG LPSURYHPHQWV (4; f 7RWDO 5HVHDUFK &RUSRUDWLRQnV (TXLWUHQG DWWHPSWV WR SHQHWUDWH GHHS LQWR VRPH FRPPRQ DWWLWXGLQDO PHDVXUHV 5DWKHU WKDQ DVFHUWDLQ WKH PHUH IDPLOLDULW\ RI D EUDQG WKHVH UHVHDUFKHUV GHOYH LQWR WKH fGHSWK RI DZDUHQHVVf DQG WKH fOHYHOV RI SHUFHLYHG TXDOLW\f LQ DVVRFLDWLRQ ZLWK NH\ YDULDEOHV VXFK DV SULFH HODVWLFLW\ DQG EUDQG OR\DOW\ $ E\SURGXFW RI WKHVH FDOFXODWLRQV LV D EDVLF PHDVXUH RI XVHU VDWLVIDFWLRQ 7RWDO 5HVHDUFK &RUSRUDWLRQ f 3URGXFW (YDOXDWLRQV ,QF SURYLGHV D 3URGXFW 4 PHDVXUH RI EUDQG HTXLW\ WKDW LV KLJKO\ DWWLWXGLQDO DQG LV GHVLJQHG WR FDOFXODWH FRQVXPHU HQWKXVLDVP f7KH DEVROXWH QXPEHU RI SHRSOH ZKR WKLQN \RXU SURGXFW LV WHUULILFf 8VLQJ D QDWLRQZLGH VXUYH\ WHFKQLTXH IDFWRUV VXFK DV IDPLOLDULW\ H[SHULHQFH RYHUDOO DSSHDO DQG FRPSDQ\ LPDJH DUH XVHG WR GHWHUPLQH WKH OHYHO RI HQWKXVLDVP 3URGXFW 4 f

PAGE 30

$Q XSVKRW WR WKH JURZLQJ LQWHUHVW LQ EUDQG HTXLW\ LQ WKH V ZDV WKH LQWURGXFWLRQ RI SRVLWLRQLQJ WKHRU\ PDGH SRSXODU E\ 5HLV DQG 7URXW LQ WKHLU EHVW VHOOHU 3RVLWLRQLQJ 7KH %DWWOH )RU
PAGE 31

EUDQGfV FXUUHQW PRQHWDU\ VWUHQJWKV DQG IXWXUH SURVSHFWV DV ZHOO DV LWV ILQDQFLDO VWDWXV ZLWKLQ D FRPSDQ\fV WRWDO SRUWIROLR RI SURGXFWV 2Q WKH RWKHU KDQG EUDQG VWUHQJWK FRPHV IURP D FRQVXPHU SHUVSHFWLYH ZKHUH D EUDQG LV HYDOXDWHG DFFRUGLQJ WR fFRPSHWLWLYH SRVLWLRQLQJ DQG LPDJHU\ UHODWLYH WR FXVWRPHU GHPDQGV DQG GHVLUHVf S f 8VLQJ WKLV FRQFHSWXDOL]DWLRQ RQH FRXOG VXUPLVH WKDW FRQVXPHU EUDQG VWUHQJWK GULYHV FRUSRUDWH EUDQG YDOXH $OH[DQGHU %LHO f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fV GLUHFW H[SHULHQFH WULDOf ZLWK D SURGXFW
PAGE 32

DFDGHPLFV JRYHUQPHQWV DJHQFLHV DQG EXVLQHVVHV UHYHDO WKDW WKH RYHUDOO FRQVXPHU VDWLVIDFWLRQ OHYHO LQ WKH 86 LV H[WUHPHO\ KLJK 7KH UHVXOWLQJ EDG QHZV LV WKDW FXOWLYDWLQJ EUDQG SUHIHUHQFH LV PRUH FKDOOHQJLQJ WKDQ HYHU 2WKHU VWXGLHV KDYH GHPRQVWUDWHG WKURXJK FRQWUROOHG H[SHULPHQWV WKDW D EUDQG FDQ KDYH D SRVLWLYH VDWLVIDFWLRQ OHYHO ZLWK D FRQVXPHU DQG VWLOO QRW EH SXUFKDVHG EHFDXVH FRPSHWLQJ EUDQGV DUH PRUH DFFHVVLEOH LQ PHPRU\ GXH WR KHLJKWHQHG H[SRVXUH )DUTXKDU )D]LR 3RZHOO t :LOOLDPV f 7KLV KHLJKWHQHG H[SRVXUH LV RIWHQ WKH UHVXOW RI H[WHQVLYH EUDQG DGYHUWLVLQJ 1HGXQJDGL f :LONLH f FRQFOXGHV WKDW IRU PDQ\ SURGXFW FDWHJRULHV fEUDQGV DUH OLNHO\ ORVLQJ VDOHV QRW EHFDXVH WKH\ DUH QRW OLNHG E\ FRQVXPHUV EXW EHFDXVH WKH\ QHYHU FRPH WR PLQG GXULQJ WKH FRQVXPHUfV RIWHQ UDSLGf GHFLVLRQ SURFHVV 7KH\ DUH VLPSO\ QHYHU FRQVLGHUHGf S f 7KH DERYH VWXGLHV UHLQIRUFH WKH SURSRVLWLRQ WKDW VLPSOH EUDQG DZDUHQHVV FDQ KDYH D SURIRXQG HIIHFW RQ EUDQG FKRLFH ZLWKRXW WKH DGGLWLRQ RI FRPSOH[ LPDJHV RU DVVRFLDWLRQV 7KLV RI WRSRIPLQG UHFRJQLWLRQ DFFHVVLELOLW\ WR PHPRU\f LV SRZHUIXO SDUWLFXODUO\ LQ ORZ LQWHUHVWORZ LQYROYHPHQW SXUFKDVH VLWXDWLRQV )RU H[DPSOH +R\HU DQG %URZQ f SURSRVHG WKDW LQ WKH LQWHUHVW RI FRQVHUYLQJ WLPH DQG HIIRUW FRQVXPHUV ZLOO RIWHQ XVH VLPSOH FKRLFH KHXULVWLFV ZKHQ PDNLQJ UHSHDW SXUFKDVH GHFLVLRQV $Q H[SHULPHQWDO VWXG\ LQYROYLQJ WKH SXUFKDVH RI RUGLQDU\ LQH[SHQVLYH FRQVXPHU JRRGV UHYHDOHG WKDW ZKHQ DQ LQH[SHULHQFHG GHFLVLRQPDNHU LV IDFHG ZLWK FKRRVLQJ DPRQJ VHYHUDO FRPSHWLQJ EUDQGV D NQRZQ RU UHFRJQL]HG EUDQG LV PRUH OLNHO\ WR EH WKH

PAGE 33

XOWLPDWH FKRLFHUHJDUGOHVV RI WKH UHODWLYH TXDOLW\ RI WKH RWKHU FRPSHWLQJ EUDQGV 7RGD\ EUDQG PDQDJHUV FRQWLQXH WR IDFH WKH FKDOOHQJH RI SUHYHQWLQJ FXVWRPHUV IURP GHIHFWLQJ WR FRPSHWLQJ EUDQGV $ UHFHQW VWXG\ E\ *UH\ DGYHUWLVLQJ FRQFOXGHG WKDW fEUDQG SURPLVFXLW\f LV D JOREDO SKHQRPHQRQ 7KDW LV FRQVXPHUV IURP WKH 86 WR WKH 8. WR $XVWUDOLD URXWLQHO\ PDNHV FKRLFHV IURP D JURXS RI EUDQGV WKH\ FRQVLGHU DFFHSWDEOH 6KHOO S f ,I DOO EUDQGV ZLWKLQ D SURGXFW FDWHJRU\ DUH VHHQ DV HTXDOO\ VDWLVI\LQJ WKHQ FRQVXPHUV EHFRPH YXOQHUDEOH WR FRPSHWLWLYH VWUDWHJLHV VXFK DV SULFLQJ SURPRWLRQV DQG FRQYHQLHQFH QRQH RI ZKLFK LV DQ LQJUHGLHQW RI FRQVXPHU EDVHG EUDQG HTXLW\ 7R PDQ\ PDUNHWLQJ UHVHDUFKHUV WKH RSSRVLWH RI EUDQG SURPLVFXLW\ LV EUDQG OR\DOW\ &KULVWLDQL f RI *HQHUDO )RRGV 86$ FRQIHVVHV WKDW RQH RI WKH PRVW fHOXVLYH DQG GHEDWHGf VXEMHFWV VXUURXQGLQJ EUDQG HTXLW\ LV WKH fLQWDQJLEOH DVVHW FRQVLVWLQJ RI D EUDQGfV FRQVXPHU OR\DOW\ EDVHf S f $DNHU f LQFOXGHV EUDQG OR\DOW\ DV WKH FHQWHUSLHFH RI KLV GHILQLWLRQ RI EUDQG HTXLW\ ZKLFK FRQVLVWV RI ILYH EURDG FDWHJRULHV EUDQG OR\DOW\ EUDQG DZDUHQHVV SHUFHLYHG TXDOLW\ EUDQG DVVRFLDWLRQV DQG RWKHU SURSULHWDU\ DVVHWV VXFK DV SDWHQWV $OO RI WKHVH EUDQG FKDUDFWHULVWLFV LQIOXHQFH WKH FRQVXPHUfV SXUFKDVH GHFLVLRQ DQG XVDJH VDWLVIDFWLRQ EXW EUDQG OR\DOW\ LV WKH IRUFH WKDW PDNHV D EUDQG WUXO\ SRZHUIXO +H EHOLHYHV WKDW FXVWRPHU EUDQG OR\DOW\ LV HVVHQWLDO EHFDXVH LW UHIOHFWV WKH OLNHOLKRRG RU SUREDELOLW\f WKDW D FXVWRPHU ZLOO VZLWFK WR DQRWKHU EUDQG 7KLV QRWLRQ RI OR\DOW\ VXUIDFHV RIWHQ LQ UHVHDUFK

PAGE 34

OLWHUDWXUH 0DWXUR f D PDUNHWLQJ GLUHFWRU IRU 1LHOVHQ +RXVHKROG 6HUYLFHV QRW WR EH FRQIXVHG ZLWK 1LHOVHQ 0HGLD 5HVHDUFKf DVVHUWV WKDW GR]HQV RI UHFHQWO\ SHUIRUPHG DQDO\VHV RI EUDQG FKRLFH PRGHOLQJ UHYHDOHG WKDW EUDQG OR\DOW\ ZDV D FRQVLVWHQW OHDGLQJ SUHGLFWRU RI IXWXUH EUDQG SXUFKDVHV 2WKHU UHVSHFWHG SULYDWH UHVHDUFK RUJDQL]DWLRQV VXFK DV WKH 13' *URXS DQG /LQWDV DGYRFDWH FRQVXPHU OR\DOW\ DV D FRUH FRPSRQHQW RI DQ\ EUDQG HTXLW\ PRGHO /LQWDV 13' f ,QHUWLD 7KHUH LV DQ REYLRXV LQWXLWLYH DFFHSWDQFH WKDW OR\DOW\ PXVW VRPHKRZ EH DQ LQWHJUDO SDUW RI HTXLW\ EXW VHYHUDO UHVHDUFKHUV KDYH ZDUQHG WKDW OR\DOW\ VKRXOG QRW EH PLVFRQVWUXHG ZLWK PHUH UHSHDW SXUFKDVLQJ $FFRUGLQJ WR 'RQLXV f LQIRUPDWLRQ FRQFHUQLQJ VLPSOH KDELWXDO UHSHDW SXUFKDVLQJ PD\ REVFXUH D YXOQHUDELOLW\ WR VZLWFK EUDQGV LQ WKH IXWXUH +H VWDWHV WKDW fOR\DOW\ FDQ DOVR EH LQHUWLD QRW QHFHVVDULO\ SUHIHUHQFHf S f 7KH SUHVXPSWLRQ LV WKDW LQ VRPH SXUFKDVH VLWXDWLRQV WKHUH DUH QRQFRPPLWWHG FRQVXPHUV WKDW FDQ EH PRWLYDWHG WR FKRRVH D EUDQG E\ VLPSO\ FRQWUROOLQJ NH\ VWUXFWXUDO IDFWRUV WKDW KDYH LLWWOH WR GR ZLWK D EUDQGfV SHUFHLYHG YDOXH $FFRUGLQJ WR &HXUYRUVW f DQG %HDWW\ .DKOH DQG +RPHU f WKH XQGHUO\LQJ FDXVH RI EUDQG HTXLW\ LV WKH FRUH SV\FKRORJLFDO UHODWLRQVKLS RI FRPPLWPHQW RI ZKLFK OR\DOW\ LV WKH EHKDYLRUDO UHVXOW &HXUYRUVW f VWDWHV WKDW FRPPLWPHQW SURYLGHV fWKH HVVHQWLDO EDVLV IRU GLVWLQJXLVKLQJ WUXH OR\DOW\ IURP RWKHU IRUPV RI UHSHDW SXUFKDVLQJf GULYHQ E\ IDFWRUV VXFK DV LQYHQWRU\ SURPRWLRQV DQG GLVFRXQWV S f

PAGE 35

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f )URP D SHUFHSWLYH RI FRQVXPHU EHKDYLRU ZH FDQ GHILQH RXU FRQVXPHU DV WKH fERG\f DQG WKH IRUFHV RI LQHUWLD DV WKRVH PDUNHWLQJ IDFWRUV WKDW HQFRXUDJH fUHVLVWDQFH WR FKDQJHf RU KDELW ,Q RUGHU WR KDYH WKH ERG\ FKDQJH GLUHFWLRQ WKHUH PXVW EH DQ DGHTXDWH fH[WHUQDO IRUFHf ZKLFK FDQ EH GHILQHG DV D FRPSHWLQJ EUDQG ,Q D ODWHU VHFWLRQ ZH HODERUDWH PRUH RQ WKLV FRQFHSW LQWURGXFLQJ PRPHQWXP DV D PHDVXUH RI LQHUWLD %UDQG (TXLW\ $FFRUGLQJ WR .HOOHU 2XU ILQDO FRQFHSWXDO GHILQLWLRQ RI EUDQG HTXLW\ DQG WKH RQH FKRVHQ WR EH WKH WKHRUHWLFDO XQGHUSLQQLQJ IRU RXU H[SORUDWLRQ RI WHOHYLVLRQ SURJUDP HTXLW\ FRPHV IURP .HOOHU f ZKR FRQFHSWXDOL]HV EUDQG HTXLW\ DFFRUGLQJ WR WZR NLQGV RI PHPRU\ DVVRFLDWLRQV EUDQG DZDUHQHVV DQG EUDQG LPDJH 7KH WZR LQ FRPELQDWLRQ DUH FDOOHG EUDQG NQRZOHGJH %UDQG HTXLW\ LV WKH GLIIHUHQWLDO HIIHFW RI EUDQG NQRZOHGJH RQ FRQVXPHU UHVSRQVH WR WKH PDUNHWLQJ RI D EUDQG $ZDUHQHVV LV WKH ILUVW VWHS LQ WKH SURFHVV ZKHUH PHDVXUHV VXFK DV UHFDOO DQG

PAGE 36

UHFRJQLWLRQ DUH LQWURGXFHG LPDJH GHDOV ZLWK WKH PHDQLQJ RI D EUDQG WR D FRQVXPHU WKURXJK YDULRXV DVVRFLDWLRQV VXFK DV DWWULEXWHV EHQHILWV DQG DWWLWXGHV 7KHVH DVVRFLDWLRQV FDQ YDU\ DFFRUGLQJ WR WKHLU IDYRUDELOLW\ VWUHQJWK DQG XQLTXHQHVV DQG FDQ SOD\ FULWLFDO UROHV LQ GHWHUPLQLQJ SXUFKDVH RXWFRPHV .HOOHU VWDWHV f7KH SUHVHQFH RI VWURQJO\ KHOG IDYRUDEO\ HYDOXDWHG DVVRFLDWLRQV WKDW DUH XQLTXH WR WKH EUDQG DQG LPSO\ VXSHULRULW\ RYHU RWKHU EUDQGV LV FULWLFDO WR D EUDQGfV VXFFHVVf S f .HOOHU DOVR DVVHUWV WKDW fIXQGDPHQWDOO\ KLJK OHYHOV RI EUDQG NQRZOHGJH DZDUHQHVV DQG LPDJHf LQFUHDVH WKH SUREDELOLW\ RI FKRLFH DV ZHOO DV SURGXFH JUHDWHU FRQVXPHU OR\DOW\ DQG GHFUHDVH YXOQHUDELOLW\ WR FRPSHWLWLYH PDUNHWLQJ DFWLRQVf S f )URP D PRUH EHKDYLRUDO YDQWDJH SRLQW .HOOHU GHVFULEHV KRZ WKHVH EUDQG DVVRFLDWLRQV FDQ EH UHIOHFWHG LQ WKH PDUNHWSODFH EHKDYLRU $ EUDQG LV VDLG WR KDYH SRVLWLYH QHJDWLYHf FRQVXPHUEDVHG EUDQG HTXLW\ LI FRQVXPHUV UHDFW PRUH RU OHVVf IDYRUDEO\ WR WKH PDUNHWLQJ PL[ RI WKH EUDQG WKDQ WKH\ GR WR WKH VDPH PDUNHWLQJ PL[ HOHPHQW ZKHQ LW LV DWWULEXWHG WR D ILFWLWLRXVO\ QDPHG RU XQQDPHG YHUVLRQ RI WKH SURGXFW RU VHUYLFH,I D EUDQG LV VHHQ E\ FXVWRPHUV WR EH WKH VDPH DV D SURWRW\SLFDO YHUVLRQ RI WKH SURGXFW RU VHUYLFH LQ WKH FDWHJRU\ WKHLU UHVSRQVH VKRXOG QRW GLIIHU IURP WKHLU UHVSRQVH WR D K\SRWKHWLFDO SURGXFW RU VHUYLFH ,I WKH EUDQG KDV VRPH VDOLHQW XQLTXH DVVRFLDWLRQV WKHVH UHVSRQVHV VKRXOG GLIIHU S f ,Q RWKHU ZRUGV LI WKH PDUNHWLQJ PL[ LV FRQWUROOHG E\ WKH UHVHDUFKHU WKH DWWLWXGLQDO FRPSRQHQW RI EUDQG HTXLW\ QDPHO\ EUDQG NQRZOHGJH ZLOO PDQLIHVW LWVHOI LQ FRQVXPHU EHKDYLRU 2SHUDWLRQDO 'HILQLWLRQV RI %UDQG (TXLW\ 7KH VSHFLILF PHDVXULQJ WHFKQLTXHV IRU EUDQG HTXLW\ GHSHQG RQ WKH SDUWLFXODU WKHRUHWLFDO DSSURDFK 2QH EURDG DSSURDFK WDNHV D FRUSRUDWH

PAGE 37

SHUVSHFWLYH DQG GZHOOV RQ WKH ILQDQFLDO YDOXH RI HTXLW\ H[DPLQLQJ LWHPV VXFK DV VWRFN SULFHV PRPHQWXP DFFRXQWLQJ DQG FDVK H[SHQGLWXUHV QHHGHG WR LQWURGXFH EUDQG H[WHQVLRQV $QRWKHU EURDG DSSURDFK DQG WKH GLUHFWLRQ RI WKLV VWXG\ LV FRQVXPHUEDVHG EUDQG HTXLW\ ZKHUH FRQVXPHU DWWLWXGHV SHUFHSWLRQVf DQG EHKDYLRUV DUH WKH IRFXV RI VWXG\ :LWKLQ WKLV VHFRQG DSSURDFK WKHUH LV OLWWOH DJUHHPHQW DPRQJ UHVHDUFKHUV DV WR WKH LGHDO RSHUDWLRQDOL]DWLRQV RI HTXLW\ $OVR WKHUH DUH VFDQW DFDGHPLF QRQSURSULHWDU\f FRQVXPHUEDVHG HTXLW\ VWXGLHV WKDW KDYH DFWXDOO\ DWWHPSWHG WR PHDVXUH WKH LQIOXHQFH RI HTXLW\ XVLQJ UHDO ZRUOG SURGXFWV DQG FRQVXPHUV ,QWHUYLHZV RU VHOIDGPLQLVWHUHG TXHVWLRQQDLUHV DUH W\SLFDOO\ XVHG WR REWDLQ DWWLWXGLQDO LQIRUPDWLRQ )RU H[DPSOH &REE:DOJUHHQ 5XEOH DQG 'RQWKX f ZHUH LQWHUHVWHG LQ WKH SRZHU RI EUDQG HTXLW\ WR LQIOXHQFH EUDQG SUHIHUHQFH DQG SXUFKDVH LQWHQW +RWHOV DQG KRXVHKROG FOHDQVHUV ZHUH VHOHFWHG DV EUDQG FDWHJRULHV IRU VWXG\ 8VLQJ WKH SHUFHSWXDO FRPSRQHQWV RI $DNHUfV f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

PAGE 38

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f ORRNHG DW WKH UHODWLYH FRQVXPHU HTXLW\ RI YDULRXV VRIW GULQNV DPRQJ FROOHJH VWXGHQWV $ QXPEHU RI HVWDEOLVKHG SRLQW VFDOHV ZHUH ERUURZHG IURP RWKHU DFDGHPLF UHVHDUFK DQG ZHUH PRGLILHG WR ZRUN ZLWK WKLV PRGHO $GGLWLRQDOO\ QHZ VFDOHV ZHUH FUHDWHG WR PHDVXUH LWHPV VXFK DV SURGXFW FRQVXPSWLRQ $V D PRGHO IRU GHWHFWLQJ GLIIHUHQWLDO UHVSRQVH DPRQJ EUDQGV WKH UHVXOWV ZHUH KLJKO\ SUHGLFWLYH .HOOHU f RSHUDWLRQDOL]HG HTXLW\ IURP WZR YDQWDJH SRLQWV +H VHHV DWWLWXGLQDO RU SHUFHSWXDO PHDVXUHV RI HTXLW\ EUDQG NQRZOHGJHf DV fLQGLUHFWf RSHUDWLRQDOL]DWLRQV &RQYHUVHO\ PHDVXUHV RI FRQVXPHU EHKDYLRU RU WKH LPSDFW RI EUDQG NQRZOHGJHf DUH YLHZHG DV fGLUHFWf RSHUDWLRQDOL]DWLRQV )RU H[DPSOH LQGLUHFW PHDVXUHV ZRXOG LQFOXGH LQVWUXPHQWV WKDW FDSWXUH EUDQG QDPH UHFDOO DQG YDULRXV LPDJH GLPHQVLRQV VXFK DV IDYRUDELOLW\ VWUHQJWK DQG XQLTXHQHVV RI EUDQG DVVRFLDWLRQV 'LUHFW PHDVXUHV GHDO ZLWK DFWXDO FRQVXPHU UHVSRQVH FKRLFHf WR PDUNHWLQJ DFWLYLWLHV .HOOHU UHFRPPHQGV YDULRXV H[SHULPHQWDO DQG TXDVLH[SHULPHQWDO VHWWLQJV ZKHUH WKH PDUNHWLQJ PL[

PAGE 39

SURGXFW SULFH GLVWULEXWLRQ DQG SURPRWLRQf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

PAGE 40

,QKHULWDQFH (IIHFWV )RU GHFDGHV WKH VLQJOH EHVW SUHGLFWRU RI D SURJUDPfV UDWLQJV SHUIRUPDQFH KDV EHHQ WKH UDWLQJV RI WKH SURJUDP VFKHGXOHG LPPHGLDWHO\ EHIRUH LW 7KDW LV WHOHYLVLRQ SURJUDPV WHQG WR fLQKHULWf VL]DEOH DXGLHQFHV IURP WKH SURJUDP DLULQJ LPPHGLDWHO\ SULRU WR LW RQ WKH VDPH FKDQQHO 7KLV FDUU\RYHU SKHQRPHQRQ LV NQRZQ DV LQKHULWDQFH HIIHFWV RU WXQLQJ LQHUWLD :KLOH LW ZRXOG EH IRROLVK IRU SURJUDPPHUV QRW WR WDNH DGYDQWDJH RI WKLV VFKHGXOLQJ WDFWLF :HEVWHU DQG /LFKW\ f ZDUQ WKDW VRPH SHRSOH DVVXPH WKH FKRLFH RI D SURJUDP FHQWHUV XSRQ WKH DFWLYH H[SUHVVLRQ RI D SUHIHUHQFH IRU D SURJUDP RU W\SH RI SURJUDP +RZHYHU VR FDOOHG VWUXFWXUDO IDFWRUV KDYH WUDGLWLRQDOO\ EHHQ FRQVLGHUHG LPSRUWDQW PHGLDWRUV RI WKH SURJUDPV YLHZHUV FKRRVH DQG FRPSOLFDWH WKH UHODWLRQVKLS EHWZHHQ YLHZLQJ SUHIHUHQFH DQG YLHZLQJ EHKDYLRU S f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fLQKHULWDQFH HIIHFWf WRRN RYHU WKDW H[FHHGHG WKH SUHGLFWLRQV GHULYHG IURP WKH RULJLQDO GXSOLFDWLRQ ODZ

PAGE 41

+HDGHQ .ORPSPDNHU DQG 5XVW f SURSRVHG DQ LPSURYHG PRGHO LQFRUSRUDWLQJ ILYH LQGHSHQGHQW YDULDEOHVUDWLQJV FKDQQHO SURJUDP W\SH GD\SDUW DQG UHSHDW YLHZLQJ $Q H[DPLQDWLRQ RI RYHU FRPELQDWLRQV RI SDLUV RI SURJUDPPLQJ XVLQJ 6LPPRQV 0DUNHW 5HVHDUFK GDWD UHYHDOHG WKDW E\ IDU UDWLQJV ZHUH WKH VLQJOH EHVW SUHGLFWRU $ GLIIHUHQW PRGHO RIIHUHG E\ :HEVWHU f LQFOXGHG WKH FRQWULEXWLQJ IDFWRUV RI DXGLHQFH DYDLODELOLW\ ZKLFK LQ PRVW FDVHV LV D IL[HG TXDQWLW\f OHDGLQ SURJUDP UDWLQJV WKH QXPEHU RI SURJUDP RSWLRQV DQG SURJUDP FRQWHQW 8VLQJ $UELWURQ UDWLQJV IURP RQH VZHHS SHULRG LQ 3RUWODQG 0DLQH :HEVWHU FRQFOXGHG WKDW IRU DGMDFHQW SURJUDP SDLUV OHDGLQ UDWLQJV DQG WKH QXPEHU RI SURJUDP RSWLRQV LQ FRPELQDWLRQ H[SODLQHG b RI WKH YDULDQFH 7LHGJH DQG .VRELHFK f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fVDQGZLFKLQJf KDPPRFNLQJf ZKHUH WKH LQIOXHQFH RI ERWK OHDGLQ DQG OHDGRXW SURJUDPPLQJ ZHUH DQDO\]HG 7KHLU FRQFOXVLRQV ZHUH WKDW WKH HIIHFWV RI OHDGRXW ZHUH PLQLPDO DQG WKDW D VWURQJ OHDGLQ ZDV HVVHQWLDO WR KDYH D VDQGZLFK VWUDWHJ\ ZRUN SURSHUO\ 7LHGJH t .VRELHFK f

PAGE 42

:DONHU f ORRNHG DW QLQH \HDUV RI 1LHOVHQ UDWLQJV IURP WR 7KH FRUUHODWLRQDO UHODWLRQVKLSV DPRQJ LQKHULWDQFH HIIHFWV OHDGLQ SURJUDP W\SH DQG QXPEHU RI RSWLRQV IROORZHG LGHQWLFDO SDWWHUQV REVHUYHG E\ 7LHGJH DQG .VRELHFK f ([DPLQLQJ WRWDO DXGLHQFH H[SRVXUH DPRQJ V\QGLFDWHG SURJUDPV UDWKHU WKDQ QHWZRUN VKRZV &RRSHU f FRUUHODWHG WKH LQIOXHQFH RI VHYHUDO YDULDEOHV RQ SURJUDP UDWLQJV 7KHVH LQFOXGHG OHDGLQ OHDGRXW QXPEHU RI RSWLRQV SURJUDP W\SH FRPSDWLELOLW\ QHWZRUN DIILOLDWLRQ DQG FDEOH SHQHWUDWLRQ 7KH UHVXOWV IURP D PDUNHW VDPSOH RI 0D\ VZHHS ERRNV UHYHDOHG WKDW ZKLOH VRPH YDULDEOHV VXFK DV QHWZRUN DIILOLDWLRQ KDG PLQRU PHDVXUDEOH LQIOXHQFH OHDGLQ UDWLQJV ZHUH E\ IDU WKH VWURQJHVW SUHGLFWRUV RI D V\QGLFDWHG SURJUDPfV UDWLQJV DQG DSSHDUHG WR EH fVWURQJ VXUURJDWHV IRU RWKHU YDULDEOHV LQ WKH PRGHO DQG FRPSOHWHO\ RYHUZKHOPHG DQ\ RWKHU IDFWRU LQ WKH PRGHOf S f 9LHZHU 0RWLYDWLRQ DQG /R\DOW\ 7KH YDULDQFHV LQ UHVXOWV IRXQG DPRQJ LQKHULWDQFH HIIHFWV VWXGLHV KDYH QRW EHHQ H[SODLQHG DGHTXDWHO\ XVLQJ VWUXFWXUDO IDFWRUV VXFK DV SURJUDP W\SH QXPEHU RI RSWLRQV FKDQQHO DQG QHWZRUN DIILOLDWLRQ 3DUW RI WKH SUREOHP PD\ LQYROYH WKH YLHZLQJ PRWLYDWLRQ RI WKH DXGLHQFH PHPEHU )RU H[DPSOH 5XELQ f GLYLGHV WHOHYLVLRQ SURJUDP YLHZLQJ PRWLYDWLRQV LQWR ULWXDOL]HG DQG LQVWUXPHQWDO FDWHJRULHV 7KH ILUVW GHQRWHV D YLHZHU WKDW XVHV WHOHYLVLRQ RXW RI KDELW DQG WR SDVV WKH WLPH UHOLHYH ERUHGRP HQWHUWDLQPHQW DPXVHPHQW HWFf 7KHVH fHVFDSLVWf DXGLHQFHV SODFH PRUH HPSKDVLV RQ WKH PHGLXP LWVHOI UDWKHU WKDQ VSHFLILF SURJUDP FRQWHQW 7KH VHFRQG W\SH RI PRWLYDWLRQ LQVWUXPHQWDOf LV

PAGE 43

PRUH JRDORULHQWHG ZKHUH WKH YLHZHU LV VHHNLQJ SURJUDP FRQWHQW WR JUDWLI\ D VSHFLILF QHHG DQG LV WKHUHIRUH PRUH GLVFULPLQDWLQJ )URP DQ LQKHULWDQFH HIIHFWV SHUVSHFWLYH RQH FRXOG K\SRWKHVL]H WKDW WKH ULWXDOL]HG YLHZHU ZRXOG WHQG WR UHPDLQ RQ WKH VDPH FKDQQHO ZKLOH DQ LQVWUXPHQWDO YLHZHU PLJKW KROG FHUWDLQ SURJUDP OR\DOWLHV WKDW ZRXOG HQFRXUDJH PRUH FKDQQHO VZLWFKLQJ 7KH QRWLRQ RI DXGLHQFH OR\DOW\ ZDV EURDFKHG E\ %RHPHU f ZKHQ KH WRRN D \HDU ORRN DW ORFDO ODWH QLJKW QHZVFDVWV LQ 'DOODV 7H[DV 8VLQJ $UELWURQ VZHHS UDWLQJV KH IRXQG KLJK SRVLWLYH FRUUHODWLRQV EHWZHHQ QHZVFDVWV DQG SULPH WLPH OHDGLQ SURJUDPPLQJ +RZHYHU KH DOVR IRXQG FRQVLGHUDEOH YDULDQFHV DPRQJ WKH WKUHH FRPSHWLQJ VWDWLRQV UDQJLQJ IURP WR $GGLWLRQDOO\ KH IRXQG WKDW VRPH QHZVFDVWV GHOLYHUHG EHWWHU UDWLQJV WKDQ WKHLU OHDGLQV SURYLGLQJ fFLUFXPVWDQWLDO HYLGHQFH RI D OR\DO YLHZHUVKLS IRU ORFDO QHZV LQ WKH PDUNHWf S f ,QKHULWDQFH HIIHFWV WXQLQJ LQHUWLDf FDQ SOD\ D FUXFLDO UROH LQ GHWHUPLQLQJ WKH UDWLQJV SHUIRUPDQFH RI D WHOHYLVLRQ SURJUDP :KLOH WKH LPSDFW LV QRWLFHDEOH WKHUH LV FRQVLGHUDEOH YDULDQFH DPRQJ SURJUDPV DV WR KRZ HIIHFWLYHO\ WKH\ fLQKHULWf DXGLHQFHV 7KDW LV VRPH SURJUDPV DSSHDU WR FDSLWDOL]H RQ D OHDGLQ DXGLHQFH EHWWHU WKDQ RWKHU SURJUDPV &RQYHUVHO\ VRPH SURJUDPV DSSHDU PRUH UHVLOLHQW WKDQ RWKHUV WR WKH HIIHFWV RI D SRRU OHDGLQ $Q H[SODQDWLRQ PD\ FRPH IURP WKH UDQNV RI EUDQG HTXLW\ WKHRU\

PAGE 44

$SSO\LQJ %UDQG (TXLW\ WR 7HOHYLVLRQ 3URJUDP 3HUIRUPDQFH %HIRUH WKH f%X]]f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t&f D QDWLRQDOO\ UHFRJQL]HG EURDGFDVW PDUNHWLQJ DQG FRPPXQLFDWLRQV FRQVXOWLQJ FRPSDQ\ EDVHG LQ 3LWWVEXUJK 3HQQV\OYDQLD $SSUR[LPDWHO\ WHOHYLVLRQ VWDWLRQV WKURXJKRXW WKH FRXQWU\ VXEVFULEH FXUUHQWO\ WR WKH VHUYLFH 2Q D PDUNHWVSHFLILF EDVLV 00t& SURYLGHV DQQXDO VXUYH\V RI FRQVXPHU SXUFKDVLQJ KDELWV DQG DWWLWXGHV IRU SUHVHOHFWHG FDWHJRULHV LQ UHWDLO VHUYLFH DQG PHUFKDQGLVH LQGXVWULHV $GGLWLRQDOO\ WKH FRPSDQ\ SURYLGHV DQDO\VLV RI FRQVXPHU FRPPXQLFDWLRQ KDELWV HQFRPSDVVLQJ UDGLR WHOHYLVLRQ QHZVSDSHU FDEOH DQG GLUHFW PDLO :LWKLQ WKH WHOHYLVLRQ YLHZLQJ GRPDLQ 00t& WDUJHWV VSHFLILF SURJUDPV LQFOXGLQJ ORFDO QHZVFDVWV 0DUVKDOO 0DUNHWLQJ f 0DUVKDOO 0DUNHWLQJ DQG &RPPXQLFDWLRQV FRQGXFWV VHYHUDO KXQGUHG LQ GHSWK WHOHSKRQH LQWHUYLHZV RYHU D WKUHH WR ILYH ZHHN WLPH VSDQ 2QH VSHFLILF

PAGE 45

UHVHDUFK DUHD LV FRQVXPHU OR\DOW\ WR VSHFLILF EUDQGV /R\DOW\ LV GHILQHG FRQFHSWXDOO\ DV UHFXUULQJ DQG H[FOXVLYH SXUFKDVLQJ 7KDW LV SHRSOH ZKR DUH fKHDY\ XVHUVf RI D SURGXFW FDWHJRU\ DQG SXUFKDVH RQO\ RQH EUDQG RYHU D FRQVLGHUDEOH WLPH SHULRG DUH FRQVLGHUHG OR\DOLVWV 7KH RSHUDWLRQDO VSHFLILFV YDU\ GHSHQGLQJ RQ WKH W\SH RI SURGXFW DQG W\SLFDO EX\LQJ F\FOHV )RU H[DPSOH VRPH SURGXFWV WHQG WR EH D GDLO\ RU ZHHNO\ SXUFKDVH ZKLOH RWKHUV WHQG WR EH PRUH RI D PRQWKO\ SXUFKDVH 'DLO\ YHUVXV ZHHNO\ VFKHGXOHG WHOHYLVLRQ SURJUDPV FDQ EH UHJDUGHG LQ WKH VDPH OLJKW $OWKRXJK 00t& KDV \HW WR DGRSW EUDQGLQJ MDUJRQ WKLV PHDVXUH RI FRQVXPHU OR\DOW\ LV FRQVLVWHQW LQ PDQ\ UHVSHFWV WR .HOOHUfV f FRQFHSWXDOL]DWLRQV DQG FDQ VHUYH DV D FRPSDUDWLYH EHQFKPDUN IRU RXU FDVH VWXG\ H[DPLQLQJ UDWLQJV SHUIRUPDQFH ,Q WKH ILQDO DQDO\VLV WKH PHGLDfV LQIDWXDWLRQ ZLWK WKH EX]] RI EUDQG PDQDJHPHQW LV VSDZQHG E\ WKH SURSRVLWLRQ WKDW EUDQG FRQFHSWV VXFK DV EUDQG HTXLW\ ZLOO KHOS SURWHFW DQG HQKDQFH D SURJUDPfV UDWLQJV SHUIRUPDQFH :KLOH WKH MDUJRQ RI EUDQG PDQDJHPHQW LV SUHYDOHQW WKRXJKRXW WKH EURDGn FDVWLQJ DQG FDEOH LQGXVWULHV WKHUH KDV \HW WR EH D V\VWHPDWLF HIIRUW WR WUDQVODWH SURSHUO\ PDQ\ RI WKHVH SUHFHSWV IRU XVH E\ WKH PHGLD %RUURZLQJ IURP WKH ZRUN RI .HOOHU f DQG SURSRQHQWV RI FRQVXPHU LQHUWLD WKLV VHFWLRQ H[DPLQHV WKH SODXVLELOLW\ RI DSSO\LQJ EUDQG HTXLW\ WKHRU\ WR WHOHYLVLRQ SURJUDPPLQJ 1HWZRUN YHUVXV 6WDWLRQ YHUVXV 3URJUDP (TXLW\ :KLOH EUDQG HTXLW\ WKHRU\ FRXOG EH DSSOLHG WR VHYHUDO EUDQGLQJ XQLWV WKH ODWHVW DFDGHPLF OLWHUDWXUH LQGLFDWHV WKDW IRU WKH PRVW SDUW SHRSOH VWLOO ZDWFK SURJUDPV EHIRUH QHWZRUNV VWDWLRQV RU FKDQQHOV $ VWXG\ E\ $EHOPDQ $WNLQ

PAGE 46

DQG 5DQG f UHYHDOHG WKDW WKH PDMRULW\ RI YLHZHUV fDUH UHODWLYHO\ LQGLVFULPLQDWH FRQVXPHUV ZLWK UHJDUG WR WKH VRXUFH RI SURJUDPPLQJ 7KH\ SRVVHVV OLWWOH QHWZRUN RU VWDWLRQ DIILQLW\f S f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fV DXGLHQFHV DV FRQVXPHUV PRVW EUDQGLQJ WHUPLQRORJ\ FDQ EH DGDSWHG UHDGLO\ WR EURDGFDVW SURJUDPPLQJ 2Q D IDLUO\ FXUVRU\ OHYHO WKH IROORZLQJ FRQFHSWV KDYH EHHQ UHGHILQHG IRU EURDGFDVWLQJ %UDQG DZDUHQHVV 7KH UHFDOO RU UHFRJQLWLRQ RI D QHWZRUN VWDWLRQ RU SURJUDP QDPH %UDQG H[WHQVLRQ 7KH LPSRVLWLRQ RI D EUDQG QDPH RYHU VHYHUDO SURJUDP fSURGXFWVf VXFK DV D QHWZRUN ,' %UDQG WULDO 7KH ILUVW H[SHULHQFH D YLHZHU KDV ZLWK D SURJUDP VDPSOLQJf %UDQG LPDJH 7KH WKRXJKWV RU IHHOLQJV PHDQLQJVf WKDW FRPH WR PLQG ZKHQ WKH SURJUDP QDPH LV PHQWLRQHG %UDQG SRVLWLRQLQJ 7KH LPDJH RI WKH EUDQG GHILQHG LQ UHODWLRQVKLS WR LWV GLUHFW PDUNHW FRPSHWLWRUV

PAGE 47

%UDQG DWWLWXGH 7KH HYDOXDWLRQ RI D SURJUDP EUDQG LPDJHVf DQG SUHGLVSRVLWLRQ WR ZDWFK %UDQG OR\DOW\ 7KH GHJUHH RI UHSHDW YLHZLQJ RI D SURJUDP DW WKH H[FOXVLRQ RI GLUHFW FRPSHWLWRUV %UDQG FRPPLWPHQW 7KH SV\FKRORJ\ XQGHUO\LQJ SURJUDP OR\DOW\ 7KH GHJUHH RI ILGHOLW\ WR D SURJUDP GHVSLWH FRPSHWLWLYH SURJUDP VFKHGXOLQJ SURPRWLRQV OHDGLQV DQG VR IRUWK 7KHUH DUH RWKHU XVHIXO EUDQG PDUNHWLQJ SDUDOOHOV EHWZHHQ FRQYHQWLRQDO FRQVXPHU JRRGV DQG EURDGFDVWLQJ )RU H[DPSOH WKHUH DUH REYLRXV VLPLODULWLHV EHWZHHQ D YLHZHUfV OLPLWHG FKDQQHO UHSHUWRLUH DQG D FRQVXPHUfV HYRNHG VHW RI EUDQG VHOHFWLRQ $OVR WKH SUHFHSWV RI FRQVXPHU KDELW IRUPDWLRQ WKURXJK VLPSOH RSHUDWDQW FRQGLWLRQLQJ FDQ EH VHHQ LQ KRZ EURDGFDVWHUV UHLQIRUFH SRVLWLYH YLHZHU RXWFRPHV RU VDWLVIDFWLRQf E\ PDQLSXODWLQJ SURJUDP FRQWHQW DQG VFKHGXOLQJ &RQWUROOLQJ WKH 0DUNHWLQJ 0L[ .HOOHU f LQVLVWV WKDW EUDQG HTXLW\ FDQQRW EH DVVHVVHG SURSHUO\ XQOHVV WKH PDUNHWLQJ PL[HV RI WKH FRPSHWLQJ EUDQG FRPSHWLWRUV DUH HTXLYDOHQW RU DW OHDVW FRQWUROOHG E\ WKH UHVHDUFKHU 7KH PDUNHWLQJ PL[ FRPSRQHQWV RI SURGXFW SULFH SODFH GLVWULEXWLRQf DQG SURPRWLRQ FDQ EH WUDQVODWHG LQWR D EURDGFDVWLQJ GRPDLQ LQ WKH IROORZLQJ PDQQHU 7KH SURGXFW 7KH SURGXFW LV WKH VSHFLILF EUDQG RI SURJUDP FRQWHQW XQGHU VFUXWLQ\ :H PHQWLRQHG HDUOLHU WKDW LQ RUGHU WR XQGHUVWDQG EUDQG HTXLW\ LW LV QHFHVVDU\ WR

PAGE 48

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fIUHH +RZHYHU EDVLF DQG SUHPLXP FDEOH SURJUDPPHUV GR HQFRXQWHU SULFLQJ GLOHPPDV ZLWK WKHLU SD\LQJ FRQVXPHUV )RU WKH SXUSRVH RI WKLV VWXG\ ZH DUH H[FOXGLQJ WKH SULFLQJ FRPSRQHQW 7KH SODFH GLVWULEXWLRQf )RU HOHFWURQLF PHGLD FRQWUROOLQJ WKH GLVWULEXWLRQ FRPSRQHQW RI WKH PDUNHWLQJ PL[ LV FKDOOHQJLQJ EXW HVVHQWLDO WR XQGHUVWDQGLQJ WUXH HTXLW\ ,Q DQ LGHDO WHVW HQYLURQPHQW D SURJUDPfV GLUHFW FRPSHWLWRUV VKRXOG EH HTXDOO\ DYDLODEOH WR DOO FRQVXPHUV )URP D UHWDLO SHUVSHFWLYH WKLV ZRXOG VHOGRP EH D PDMRU REVWDFOH EXW IRU EURDGFDVWHUV WKHUH LV DQ REYLRXV SUREOHP 8QOLNH PRVW FRQVXPHU JRRGV EURDGFDVW fSURGXFWVf DUH WLPHERXQG DQG DUH QRW DYDLODEOH VLPXOWDQHRXVO\ WR DXGLHQFHV ,PDJLQH D UHWDLO RXWOHW ZKHUH WKH VKHOI GLVSOD\V FKDQJHG HYHU\ KDOI KRXU WR GLVSOD\ GLIIHUHQW FRPSHWLQJ EUDQGVf $WWLWXGLQDO VWXGLHV FDQ LJQRUH WKLV GLOHPPD EXW D EHKDYLRUDO DSSURDFK WR DVVHVVLQJ

PAGE 49

FRQVXPHUEDVHG EUDQG HTXLW\ UHTXLUHV WKDW QRW RQO\ SURJUDP FRQWHQW EH VLPLODU EXW DOVR SURJUDP VFKHGXOLQJ DV ZHOO 3URJUDP EUDQG GLVWULEXWLRQ DYDLODELOLW\f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fWUDIILFf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fWDFWLFDO JLPPLFNU\f DUH LQWHQGHG WR VWLPXODWH SURGXFW WULDO VDPSOLQJf DQG ERRVW WHPSRUDULO\ PDUNHW VKDUHV EXW VHOGRP FXOWLYDWH ORQJWHUP EUDQG OR\DOWLHV .HOOHU S f

PAGE 50

,Q WKH TXHVW IRU VKRUWWHUP UDWLQJV JDLQV GXULQJ WKH LQIDPRXV 1LHOVHQ fVZHHS ZHHNVf WHOHYLVLRQ QHWZRUNV DQG VWDWLRQV UHVRUW WR D YDULHW\ RI SURJUDPPLQJ DQG SURPRWLRQDO WDFWLFV WKDW DUH GHVLJQHG RIWHQ WR GLVWRUW W\SLFDO SURJUDP SHUIRUPDQFH 7KHVH WDFWLFV UDQJH IURP DOWHULQJ SURJUDP VFKHGXOHV DQG PDQLSXODWLQJ VWRU\ OLQHV WR DW\SLFDO DGYHUWLVLQJ FDPSDLJQV DQG RQDLU FRQWHVWV .QRZQ E\ LQGXVWU\ LQVLGHUV DV K\SLQJ RU K\SRLQJ WKHVH SUDFWLFHV KDYH EHHQ FRQGHPQHG E\ 1LHOVHQ DQG WKH DGYHUWLVLQJ FRPPXQLW\ *LPHLQ 0F'RZHOO 0LOHV f 7KH EHVW PHDQV IRU QHXWUDOL]LQJ WKH ELDVLQJ HIIHFWV RI EURDGFDVW SURPRWLRQ LV WR ORRN DW ORQJWHUP SHUIRUPDQFH ZKHUH WUDQVLWRU\ SURPRWLRQ WDFWLFV EHFRPH UDQGRP YDULDEOHV ,Q VXPPDU\ WKH PDUNHWLQJ PL[ IRU HOHFWURQLF PHGLD PXVW VRPHKRZ EH KHOG LQ FKHFN LQ RUGHU WR HYDOXDWH SURJUDP HTXLW\ 7KHUHIRUH LQ DQ LGHDO UHVHDUFK VHWWLQJ WKH IROORZLQJ FULWHULD VKRXOG EH DSSOLHG Df LGHQWLFDO RU KLJKO\ VLPLODU WDUJHW SURJUDP FRQWHQW GHULYHG IURP VDPH fSURGXFW FDWHJRU\ff Ef LGHQWLFDO SULFLQJ FRQVLGHUDWLRQV VDPH SULFH RU IUHHf Ff LGHQWLFDO GLVWULEXWLRQ RI WDUJHW SURJUDPV LQ WHUPV RI VFKHGXOLQJ VLJQDO FRYHUDJH DQG DXGLHQFH IORZ DQG Gf LGHQWLFDO RU QHXWUDOL]HG VKRUWWHUP SURPRWLRQDO WDFWLFV RI FRPSHWLQJ WDUJHW SURJUDPV ,QHUWLD 0RPHQWXP DQG (TXLW\ 7KH ZHOOGRFXPHQWHG LQIOXHQFH RI OHDGLQ SURJUDP UDWLQJV LPSOLHV WKDW WKHUH DUH VLJQLILFDQW QXPEHUV RI SDVVLYH RU XQFRPPLWWHG YLHZHUV ZKR DUH QRW PRWLYDWHG WR FKDQJH FKDQQHOV 7KH PDJQLWXGH RI WKLV LQHUWLD RU fUHVLVWDQFH WR FKDQJHf KRZHYHU KDV EHHQ GLIILFXOW WR SUHGLFW LQ WHUPV RI UDWLQJ SRLQWV *LYHQ

PAGE 51

WKH VDPH OHDGLQ UDWLQJ SRLQWV VRPH SURJUDPV DSSHDU WR fLQKHULWf PRUH SRLQWV WKDQ RWKHUV )RU H[DPSOH D SURJUDP ZLWK D ELJ OHDGLQ UDWLQJ PD\ EH XQDEOH WR FDSLWDOL]H IXOO\ RQ WKLV DGYDQWDJH ZKLOH DQRWKHU SURJUDP GHDOLQJ ZLWK D PXFK VPDOOHU OHDGLQ FDQ RYHUFRPH WKLV DSSDUHQW GLVDGYDQWDJH DQG H[KLELW LPSUHVVLYH UDWLQJV SHUIRUPDQFH 8VLQJ .HOOHUfV f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fV UHVLVWDQFH WR FKDQJH EHKDYLRU SDWWHUQV )RU WHOHYLVLRQ WKLV UHOXFWDQFH WR FKDQJH FKDQQHOV KDV EHHQ FRLQHG WXQLQJ LQHUWLD +RZHYHU LQHUWLD DORQH PD\ QRW EH DV ZRUWK\ D PHWDSKRU DV DQRWKHU WHUP ERUURZHG IURP SK\VLFV PRPHQWXP ,QWHQGHG DV D PHDVXUH RI fKRZ PXFKf LQHUWLD LV SUHVHQW LQ D ERG\ PRPHQWXP FRQVLVWV RI WZR FRPSRQHQWV PDVV DQG YHORFLW\ (QF\FORSHGLD RI 6FLHQFH DQG 7HFKQRORJ\ f %DVHG RQ WKLV VLPSOH IRUPXOD ZH FDQ XQGHUVWDQG EHWWHU KRZ WZR ERGLHV ZLWK WKH LGHQWLFDO PDVV FDQ H[KLELW GLIIHUHQW GHJUHHV RI PRPHQWXP GHSHQGLQJ RQ WKHLU UHODWLYH VSHHG 6LPLODUO\ PHDVXUHV RI PDVV DORQH FDQQRW SUHGLFW KRZ VXVFHSWLEOH D ERG\ LV WR

PAGE 52

fH[WHUQDO IRUFHVf 2QO\ ZKHQ WKH PDVV DQG WKH YHORFLW\ DUH PHDVXUHG LQ WDQGHP FDQ ZH DVVHVV SURSHUO\ WKH SRZHU RI LQHUWLD RQ D ERG\ )URP D FRQVXPHU EHKDYLRU SHUVSHFWLYH ZH FDQ VXUPLVH WKDW LQHUWLD LV QRW QHFHVVDULO\ D EDG WKLQJ 5DWKHU LW LV WKH XQGHUO\LQJ FRQVXPHU PRPHQWXP IDFWRUV WKDW PXVW EH HYDOXDWHG Df WKH EUDQGfV DELOLW\ WR UHWDLQ FXUUHQW FXVWRPHUV LQ WKH IDFH RI H[WHUQDO IRUFHV RI GLUHFW FRPSHWLWRUV DQG Ef WKH DELOLW\ RI D EUDQG WR UHFUXLW FXVWRPHUV IURP FRPSHWLQJ EUDQGV 7KH UHSHDW FXVWRPHUV RI D EUDQG H[KLELWLQJ SRRU HTXLW\ DUH YXOQHUDEOH WR D FRPSHWLWRUfV PDUNHWLQJ PL[ DQG WKHUHIRUH KDYH OHVV PRPHQWXP WKDQ D EUDQG ZLWK VWURQJHU HTXLW\ ,Q D VLPLODU YHLQ RQH FRXOG SURSRVH WKDW WXQLQJ LQHUWLD LQKHULWDQFH HIIHFWVf LV LQIOXHQFHG E\ WKH WZR fPRPHQWXPf IDFWRUV RI Df UHWHQWLRQ RI OHDGLQ DXGLHQFHV DQG Ef UHFUXLWPHQW RI DXGLHQFHV IURP RWKHU VRXUFHV 7KLV DSSURDFK H[SODLQV KRZ WZR SURJUDPV ZLWK LGHQWLFDO OHDGLQ UDWLQJV FDQ DFKLHYH GLIIHUHQW UHVXOWV (TXLW\ WKHQ FDQ EH YLHZHG DV D SURJUDPfV UDWLQJV PRPHQWXP ,QWHUIDFLQJ 0RPHQWXP ZLWK .HOOHUfV &RQFHSWXDOL]DWLRQV 7KH DERYH QRWLRQV RI LQHUWLD DQG PRPHQWXP LQWHUIDFH ZHOO ZLWK .HOOHUfV f DSSURDFK WR EUDQG HTXLW\ 7KDW LV LI D FRQVXPHU GRHV QRW SHUFHLYH DQ\ fVDOLHQW DQG XQLTXH DVVRFLDWLRQVf IURP D EUDQG QDPH WKHUH ZLOO EH D UHVLVWDQFH WR FKDQJH SXUFKDVH EHKDYLRU 7KLV PRPHQWXP FDQ EH GLVUXSWHG E\ WKH H[WHUQDO IRUFH RI D FRPSHWLQJ EUDQG ZKRVH PDUNHWLQJ PL[ JHQHUDWHV IDYRUDEOH VWURQJ DQG XQLTXH DVVRFLDWLRQV WKDW fLPSO\ VXSHULRULW\ RYHU RWKHU EUDQGVf 6LPLODUO\ D WHOHYLVLRQ DXGLHQFH PHPEHU WKDW LV FXUUHQWO\ YLHZLQJ RQH SDUWLFXODU FKDQQHO ZLOO UHPDLQ RQ WKDW FKDQQHO XQOHVV WKH fH[WHUQDO IRUFHVf IURP

PAGE 53

D FRPSHWLQJ FKDQQHO LQWHUYHQH VXIILFLHQWO\ WR PRWLYDWH WKH YLHZHU WR FKDQJH FKDQQHOV (TXLW\ DQG 3HUIRUPDQFH :H ZLOO UHFDOO WKDW IRU .HOOHU f EUDQG HTXLW\ LV WKH GLIIHUHQWLDO HIIHFW RI EUDQG NQRZOHGJH RQ FRQVXPHU UHVSRQVH LQ D FRPSHWLWLYH PDUNHWSODFH 7KLV EUDQG NQRZOHGJH LQFOXGHV Df DZDUHQHVV DQG Ef LPDJH DVVRFLDWLRQV 7KH PRUH IDYRUDEOH VWURQJ DQG XQLTXH WKHVH DVVRFLDWLRQV WKH PRUH OLNHO\ D SDUWLFXODU EUDQG ZLOO EH FKRVHQ RYHU LWV GLUHFW FRPSHWLWRUV 7KLV FRQVXPHU UHVSRQVH FKRLFH EHKDYLRUf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fUDWLQJVf SURYLGHG E\ 1LHOVHQ 0HGLD 5HVHDUFK

PAGE 54

7KH 1LHOVHQV 0HDVXULQJ 3URJUDP 3HUIRUPDQFH %\ DGRSWLQJ D EHKDYLRUDO DSSURDFK WR XQGHUVWDQGLQJ SURJUDP EUDQG HTXLW\ DQG LQVLVWLQJ WKDW WKLV ZRUN EH UHOHYDQW WR WKH UHDOZRUOG EXVLQHVV HQYLURQPHQW RI WKH EURDGFDVWLQJ LQGXVWU\ 1LHOVHQ UDWLQJV ZHUH VHOHFWHG DV WKH PRVW DSSURSULDWH PHDVXUH RI SURJUDP SHUIRUPDQFH 7KH ZRUG UDWLQJ LV RIWHQ PLVXQGHUVWRRG RXWVLGH WKH EURDGFDVW LQGXVWU\ EHFDXVH WKH PRUH YHUQDFXODU XVH RI WKH WHUP PHDQV D VXEMHFWLYH MXGJPHQW RU RSLQLRQ 7KDW LV SHRSOH DUH DVNHG RIWHQ WR fUDWHf WKHLU IDYRULWH PRYLHV UHVWDXUDQWV RU URFN JURXSV %URDGFDVW DQG FDEOH UDWLQJV KRZHYHU DUH VLPSO\ HVWLPDWHV RI DXGLHQFH WXQLQJ EHKDYLRU RU PHGLD H[SRVXUH 5DWLQJV DUH QRW LQWHQGHG WR EH D YDOLG DWWLWXGLQDO PHDVXUH H[FHSW LQ WKH VHQVH WKDW FRQVXPHU EHKDYLRU FDQ EH LQIHUUHG WR EH ZKDW :HEVWHU DQG /LFKW\ f DQG RWKHU UHVHDUFKHUV FDOO fUHYHDOHG SUHIHUHQFHf S f 7KH IROORZLQJ LV D EULHI V\QRSVLV RI 1LHOVHQ UHVHDUFK PHWKRGV ,Q DGGLWLRQ WR WKH ZHOO NQRZQ QDWLRQDO f3HRSOH 0HWHUf UDWLQJV 1LHOVHQ DOVR SURYLGHV ORFDO SURJUDP UDWLQJV WR RYHU LQGLYLGXDO PDUNHWV 8VLQJ D PXOWLVWDJH FOXVWHULQJ VDPSOLQJ WHFKQLTXH DOO PDUNHWV H[SHULHQFH WKH PRQWKn ORQJ GLDU\GULYHQ fVZHHSVf IRXU WLPHV D \HDU 1RYHPEHU )HEUXDU\ 0D\ DQG -XO\f $Q DOWHUQDWLYH WR WKH VZHHSV IRU PDQ\ PDMRU PDUNHW VWDWLRQV LV PHWHUHG fRYHUQLJKWf UDWLQJV 3UHVHQWO\ ORFDO PDUNHWV DUH PHDVXUHG GDLO\ XVLQJ D SDVVLYH HOHFWURQLF PHWHULQJ V\VWHP WKDW FRQWLQXRXVO\ UHFRUGV KRXVHKROG YLHZLQJ 7KH YLHZLQJ GDWD DUH SURFHVVHG fRYHUQLJKWf DQG WKHQ ID[HG WKH IROORZLQJ PRUQLQJ WR VXEVFULELQJ VWDWLRQV )URP D UHOLDELOLW\ SHUVSHFWLYH RQH RI

PAGE 55

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f UDWLQJ SRLQWV DQG VKDUHV SRLQWV :KLOH KRXVHKROGV DUH H[SUHVVHG DV ZKROH QXPEHUV UDWLQJ DQG VKDUH SRLQWV DUH SHUFHQWDJH SRLQWV $ UDWLQJ H[SUHVVHV D SURJUDPfV DXGLHQFH DV D SHUFHQWDJH RI WKH WRWDO SRSXODWLRQ $ VKDUH H[SUHVVHV WKH VDPH DXGLHQFH EXW DV D SHUFHQWDJH RI WKH KRXVHKROGV RU SHUVRQVf XVLQJ WHOHYLVLRQ +87V RU 387Vf $JH DQG JHQGHU GHPRJUDSKLFV FDQ DOVR EH WUDQVODWHG LQWR UDWLQJ DQG VKDUH SRLQWV 1LHOVHQ 0HWKRG f

PAGE 56

&+$37(5 5(6($5&+ +<327+(6(6 )URP WKH EDVLF UHVHDUFK TXHVWLRQ KRZ GRHV EUDQG HTXLW\ LQIOXHQFH SURJUDP UDWLQJV SHUIRUPDQFH D QXPEHU RI H[SORUDWRU\ K\SRWKHVHV FDQ EH RIIHUHG WKDW UHVRQDWH ZLWK RXU SURSRVHG WKHRUHWLFDO DVVXPSWLRQV 7KH OLWHUDWXUH UHYLHZ FRQWHQGV WKDW EUDQGV H[KLELWLQJ VWURQJ HTXLW\ WHQG WR EH PDUNHW OHDGHUV DQG WKDW RYHU WLPH VKRUWWHUP IOXFWXDWLRQV LQ SHUIRUPDQFH JLYH ZD\ WR D ORQJn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f DQG UHFRJQL]HG WKH H[LVWHQFH RI D NLQG RI WXQLQJ LQHUWLD ZKHUH YLHZHUV UHVLVW FKDQJLQJ FKDQQHOV +RZHYHU D FORVHU H[DPLQDWLRQ EURXJKW WR OLJKW WKH FRQFHSW RI PRPHQWXP ZKLFK VXJJHVWV WKDW JLYHQ WKH VDPH VL]H OHDGLQ SURJUDPV ZLWK VWURQJ HTXLW\ ZLOO SHUIRUP GLIIHUHQWO\ WKDQ SURJUDPV ZLWK ZHDNHU HTXLW\ 7KDW LV FRQVXPHUEDVHG EUDQG HTXLW\ FDQ EH IRXQG LQ WKH

PAGE 57

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f WKH SRZHU WR UHWDLQ OHDGLQ DXGLHQFHV IURP WKH VDPH FKDQQHO DQG Ef WKH SRZHU WR UHFUXLW DXGLHQFHV IURP RWKHU VRXUFHV 7KHVH WZR FRPSRQHQWV DGGUHVV WKH PRPHQWXP RI D SURJUDP DV LW SDVVHV WKURXJK LWV FRPSHWLWLYH SURJUDP HQYLURQPHQW $FFRUGLQJO\ WZR DGGLWLRQDO K\SRWKHVHV ZHUH RIIHUHG + 3URJUDPV ZLWK KLJKHU EUDQG HTXLW\ ZLOO UHWDLQ D JUHDWHU SURSRUWLRQ RI WKHLU OHDGLQ DXGLHQFHV WKDQ SURJUDPV ZLWK ORZHU EUDQG HTXLW\ + 3URJUDPV ZLWK KLJKHU EUDQG HTXLW\ ZLOO UHFUXLW D JUHDWHU SURSRUWLRQ RI DXGLHQFHV IURP RWKHU VRXUFHV WKDQ SURJUDPV ZLWK ORZHU EUDQG HTXLW\

PAGE 58

&+$37(5 5(6($5&+ 0(7+2'2/2*< %HFDXVH RI WKH H[SORUDWRU\ QDWXUH RI WKLV VWXG\ D FDVH VWXG\ IRUPDW LV DSSURSULDWH :KLOH WKHUH LV DQ H[WHQVLYH VWDWLVWLFDO DQDO\VLV LQYROYHG ZLWK WKLV FDVH VWXG\ WKH JRDO ZDV QRW WR JHQHUDOL]H VWDWLVWLFDO UHVXOWV EXW UDWKHU H[SORUH WKHRUHWLFDO SRVVLELOLWLHV ZLWKLQ D IDLUO\ FRQWUROOHG HQYLURQPHQW 5HVSHFWHG FDVH VWXG\ UHVHDUFKHU 5REHUW
PAGE 59

KRXVHKROG UDWLQJV UWf KRXVHKROG VKDUHV VKf DQG WRWDO KRXVHKROGV KKf $JDLQ WKH SULRU OLWHUDWXUH UHYLHZ RIIHUV SUHFLVH VWDWLVWLFDO GHILQLWLRQV RI WKHVH WHUPV 'HPRJUDSKLF LQIRUPDWLRQ VXFK DV JHQGHU DQG DJH ZHUH QRW DYDLODEOH IURP WKH RYHUQLJKW PHWHUHG GDWD :LWK WKH ZULWWHQ SHUPLVVLRQ RI 1LHOVHQ 0HGLD 5HVHDUFK UDWLQJV GDWD IRU -DQXDU\ WKURXJK 'HFHPEHU IRU RQH PDMRU WHOHYLVLRQ PDUNHW ZHUH WXUQHG RYHU WR WKH UHVHDUFKHU IRU DQDO\VLV 7KLV LQFOXGHG GD\V RI RYHUQLJKWV DQG IRXUPRQWKORQJ VZHHS SHULRGV )HEUXDU\ 0D\ -XO\ DQG 1RYHPEHUf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f 7KLV ZDV DFFRPSOLVKHG WKURXJK DQDO\VLV RI YDULDQFH LQGH[LQJ DQG FRUUHODWLRQDO DQDO\VLV 7R DVVHVV D SURJUDPfV DELOLW\ WR UHWDLQ OHDGLQ DXGLHQFHV DQG UHFUXLW DXGLHQFHV IURP RWKHU VRXUFHV 1LHOVHQ SURYLGHG D FXVWRP f$XGLHQFH )ORZ $QDO\VLVf EDVHG RQ WKH 0D\ VZHHS SHULRG 7KLV DQDO\VLV WUDFNV WKH

PAGE 60

VRXUFH DQG GHVWLQDWLRQ RI WKH YLHZLQJ DXGLHQFHV EHWZHHQ DQ DGMDFHQW TXDUWHU KRXU )RU WKLV VWXG\ RXU IRFDO SRLQW ZDV WKH KRXVHKROG fIORZf IURP 30 WR 30 5HJUHWWDEO\ 1LHOVHQ GRHV QRW SURYLGH VXFK IORZ VWXGLHV XVLQJ RYHUQLJKW GDWD EXW WKH IRXUZHHN VZHHS GDWD VWLOO SURYLGHG ZRUWKZKLOH LQVLJKW 7KLV WHVW PDUNHW ZDV FKRVHQ VSHFLILFDOO\ IRU UHDVRQV ZKLFK DUH SUHVHQWHG LQ VXEVHTXHQW SRUWLRQV RI WKLV PHWKRGRORJ\ VHFWLRQ )RU UHDVRQV RI FRQILGHQWLDOLW\ WKH QDPH RI WKH PDUNHW ZDV QRW GLVFORVHG LQ WKH VWXG\ DQG WKH WDUJHW VWDWLRQV ZHUH LGHQWLILHG PHUHO\ DV VWDWLRQV $ % DQG & &RQWUROOLQJ WKH 0DUNHWLQJ 0L[ %DVHG RQ .HOOHUfV f FRQFHSWXDOL]DWLRQV DQG GLVFXVVLRQV LQ WKH SULRU OLWHUDWXUH UHYLHZ WKH GLIIHUHQWLDO HIIHFW RI EUDQG NQRZOHGJH RQ FRQVXPHU UHVSRQVH PXVW EH PHDVXUHG XQGHU FRQWUROOHG FLUFXPVWDQFHV 7KHVH FRQWUROOHG FLUFXPVWDQFHV LQFOXGH PDUNHWLQJ PL[ FRPSRQHQWV ZKHUH WKH WDUJHW EUDQGV PXVW EH Df GLUHFW FRPSHWLWRUV LQ WHUPV RI SURGXFW FDWHJRU\ Ef LGHQWLFDO LQ SULFH DFURVV DLL FRPSHWLQJ EUDQGV Ff HTXDOO\ DYDLODEOH LQ WHUPV RI GLVWULEXWLRQ WR FRQVXPHUV DQG Gf XQDIIHFWHG E\ VKRUWWHUP SURPRWLRQ DFWLYLWLHV ,Q D EURDGFDVW SURJUDPPLQJ FRQWH[W D GLUHFW FRPSHWLWRU ZRXOG EH D SURJUDP RI KLJKO\ VLPLODU FRQWHQW VXFK DV D ORFDO QHZVFDVW (TXDO DYDLODELOLW\ ZRXOG EH GHILQHG DV FRPSHWLQJ SURJUDPV WKDW DUH VFKHGXOHG DW WKH VDPH GD\ DQG WLPH LQ WKH VDPH PDUNHW DLULQJ RQ FRPSDUDEOH IDFLOLWLHV 7\SLFDO WUDQVLWRU\ SURPRWLRQV ZRXOG RFFXU GXULQJ WKH SHULRGLF ffVZHHSf ZHHNV

PAGE 61

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f ZDV DOPRVW LGHQWLFDO IRU WKHVH WDUJHW VWDWLRQV 7KDW LV RYHU WKH FRXUVH RI D W\SLFDO ZHHN HDFK VWDWLRQ UHDFKHG WKH VDPH QXPEHU RI LQGLYLGXDO XQGXSOLFDWHGf KRXVHKROGV ZLWKLQ WKH PDUNHW 1LHOVHQ &XPH f ,Q DGGLWLRQ WR REYLRXV GLUHFW EUDQG FRPSHWLWLRQ LQ WHUPV RI SURJUDP FRQWHQW VFKHGXOLQJ DQG VLJQDO FRYHUDJH WKHUH ZDV WKH DGGHG DGYDQWDJH RI HDFK QHZVFDVW H[SHULHQFLQJ D GLIIHUHQW OHDGLQ VHYHQ QLJKWV D ZHHN :KLOH PRVW VWDWLRQV LQ WKH FRXQWU\ EURDGFDVW VHYHUDO QHZV SURJUDPV WKURXJKRXW WKH GD\ DQG HYHQLQJ WKH UDWLQJV SHUIRUPDQFH RI WKH 30 30 FHQWUDOf QHZVFDVW LV FRQVLGHUHG RIWHQ D SULPDU\ LQGLFDWRU RI D VWDWLRQfV YXOQHUDELOLW\ WR FRPSHWLWLYH DWWDFN (DVWPDQ f :LWK D GLIIHUHQW OHDGLQ SURJUDP HYHU\ QLJKW WKH ODWH QHZV LV WKH PRVW VXVFHSWLEOH WLPH VORW IRU FRPSHWLWLYH SURJUDP VDPSOLQJ E\ GRFLOH DXGLHQFHV WKDW DUH QRW SDUWLFXODUO\ FRPPLWWHG WR DQ\ RQH EUDQG RI QHZVFDVW 7R DVVXUH JHQXLQH GLUHFW FRPSHWLWLRQ WKH UDWLQJV GDWDEDVH ZDV IXUWKHU UHILQHG E\ H[WUDFWLQJ GDWHV ZKHQ WKH WKUHH QHZVFDVWV GLG QRW FRPSHWH KHDG WR

PAGE 62

KHDG DW 30 7KURXJKRXW WKH \HDU VWDWLRQV DUH RIWHQ IRUFHG LQWR fODWH VWDUWVf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

PAGE 63

7DEOH &RPSDULVRQ RI 6ZHHS YHUVXV 1RQVZHHS &DVHV 'HVFULSWLYH +RXVHKROG 'DWD 6ZHHSV 1RQVZHHSV 6WDWLRQ 0HDQ f 6' 0HDQ f 6' b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

PAGE 64

6LJQLILFDQW ORQJWHUP PDUNHW GRPLQDQFH )URP DQ RXWFRPH SHUVSHFWLYH FRQVLVWHQWO\ KLJK PDUNHW UDQNLQJV RYHU VHYHUDO \HDUV LV D YDOLG LQGLFDWRU RI VWURQJ HTXLW\ 6WDWHG EUDQG OR\DOW\ RU FRPPLWPHQW E\ KHDY\ XVHUV RI D SURGXFW %\SDVVLQJ EURDGFDVW UDWLQJV GDWD D FRQVXPHU VXUYH\ WKDW SURYLGHV SURJUDP YLHZLQJ OR\DOW\ LQIRUPDWLRQ FDQ VHUYH DV DQ DOWHUQDWLYH PHDVXUH RI HTXLW\ %HQFKPDUN 1LHOVHQ
PAGE 65

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fV ORQJWHUP VWDELOLW\ WKH IROORZLQJ EHQFKPDUN RIIHUV EHWWHU HYLGHQFH RI EUDQG HTXLW\ XWLOL]LQJ D FRPSOHWHO\ GLIIHUHQW PHWKRGRORJ\ %HQFKPDUN 0DUVKDOO 0DUNHWLQJ /R\DOW\ 6XUYH\ ,Q DGGLWLRQ WR D \HDU WUHQG DQDO\VLV KLJK HTXLW\ IRU VWDWLRQ $ ZDV UHDIILUPHG XVLQJ GDWD FROOHFWHG IURP 0DUVKDOO 0DUNHWLQJ DQG &RPPXQLFDWLRQV $V ZDV GLVFXVVHG LQ WKH OLWHUDWXUH UHYLHZ 00t& FRQGXFWV DQQXDO VXUYH\V RI FRQVXPHU SXUFKDVLQJ KDELWV IRU VHOHFW PDUNHWV DURXQG WKH FRXQW\ ,Q WKH VDPH \HDU DV WKH 1LHOVHQ UDWLQJV GDWD 00t& FRQGXFWHG VXFK D VXUYH\ LQ WKH WHVW PDUNHW 2YHU WHOHSKRQH LQWHUYLHZV ZHUH FRQGXFWHG RYHU IRXU ZHHNV 6DPSOH VL]HV ZHUH SUHGHWHUPLQHG WR \LHOG D b WR b PDUJLQ RI HUURU ZLWKLQ D b FRQILGHQFH OHYHO 5HFXUULQJ DQG H[FOXVLYH YLHZLQJ ZHUH GHILQHG RSHUDWLRQDOO\ DV ZDWFKLQJ RQO\ RQH VWDWLRQfV QHZVFDVW DW OHDVW WKUHH WLPHV RYHU WKH SDVW VHYHQ GD\V 7DEOH SUHVHQWV WKH UHVXOWV RI WKH VXUYH\ 7KH ILUVW

PAGE 66

WZR FROXPQV VKRZ WKDW DPRQJ VXEMHFWV ZKR ZDWFKHG DW OHDVW RQH ORFDO QHZVFDVW GXULQJ D SULRU ZHHN 6WDWLRQ $ GRPLQDWHG WKH FDWHJRU\ E\ D WR PDUJLQ RYHU HLWKHU FRPSHWLWRU 6L[W\QLQH SHUFHQW RI WKH UHVSRQVHV LQFOXGHG D 6WDWLRQ $ QHZVFDVW 7DNLQJ D QDUURZHU ORRN DW WKH H[FOXVLYH YLHZLQJ KDELWV RI KHDY\ XVHUV WKH UHPDLQLQJ WZR FROXPQV RI WKH WDEOH UHYHDO 6WDWLRQ $ DV WKH PDUNHW OHDGHU JDUQHULQJ RYHU b RI WKH DGXOWV ZKR ZDWFKHG RQO\ RQH VWDWLRQ IRU ORFDO QHZV +HUH WKH PDUJLQ IRU 6WDWLRQ $ ZDV WKUHH WR RQH RYHU LWV FRPSHWLWRUV 3UHVXPLQJ D FRQWUROOHG PDUNHWLQJ PL[ ZKHUH DOO VWDWLRQV ZHUH HTXLYDOHQW LQ WHUPV RI WLPH SHULRG VLJQDO FRYHUDJH DQG PDUNHW FRQGLWLRQV WKHVH GDWD VXSSRUW RXU HDUOLHU SRVLWLRQ 6WDWLRQ $ GHPRQVWUDWHG WKH KLJKHVW EUDQG HTXLW\ 7DEOH 0DUVKDOO 0DUNHWLQJ DQG &RPPXQLFDWLRQV 6XUYH\ RQ ([FOXVLYH /R\DOf 9LHZLQJ RI 1HZVFDVWV 6WDWLRQV :DWFKHG DW OHDVW RQFH GXULQJ SDVW ZHHN b RI RQFH SHU ZHHN YLHZLQJrr :DWFKHG H[FOXVLYHO\ RU PRUH WLPHV GXULQJ SDVW ZHHN b RI ([FOXVLYH YLHZLQJ &RPELQHG r $ % & r2ULJLQDO VDPSOH IUDPH 1 rr7RWDO b RI WKUHH VWDWLRQV H[FHHGV b EHFDXVH RI YLHZLQJ RI PXOWLSOH VWDWLRQV GXULQJ ZHHN +DYLQJ HVWDEOLVKHG DW OHDVW WHQWDWLYHO\ VWDWLRQ $ DV WKH FRPSDUDWLYH HTXLW\ EHQFKPDUN IRU WKH PDUNHW ZH FDQ DSSURDFK WKH UHVHDUFK TXHVWLRQ RI

PAGE 67

KRZ GRHV SURJUDP EUDQG HTXLW\ LQIOXHQFH SURJUDP UDWLQJV SHUIRUPDQFH RQ DQ HYHU\GD\ EDVLV

PAGE 68

&+$37(5 5(68/76 +\SRWKHVHV 2QH 7ZR DQG 7KUHH 7KH LQLWLDO WKUHH K\SRWKHVHV SUHGLFWHG D SURJUDP ZLWK KLJKHU EUDQG HTXLW\ ZRXOG RXWSHUIRUP LWV GLUHFW FRPSHWLWRUV DFURVV D QXPEHU RI GHVFULSWLYH DXGLHQFH VWDWLVWLFV LQFOXGLQJ UDWLQJV VKDUHV DQG KRXVHKROGV $QDO\]LQJ WKH DGMXVWHG RYHUQLJKW UDWLQJV GDWD WKH K\SRWKHVHV ZHUH VXSSRUWHG RQ DOO GLPHQVLRQV 7DEOH UHYHDOV WKDW IRU WKH FDWHJRULHV RI Df RYHUDOO SHUIRUPDQFH DQG Ef LQGLYLGXDO GD\ SHUIRUPDQFH VWDWLRQ $ RXU GHVLJQDWHG HTXLW\ OHDGHUf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f $W D VLJQLILFDQFH OHYHO RI ZH VHH WKDW FRPSDULVRQV RI UDWLQJ VKDUH DQG KRXVHKROG GDWD DUH QHDUO\ LGHQWLFDO )XUWKHUPRUH RQ D GD\ E\GD\ EDVLV WKH QXPEHU RI VLJQLILFDQW FRPELQDWLRQV RI VWDWLRQV LV KLJKO\ VLPLODU H[FHSW IRU WKH FRPELQDWLRQ RI VWDWLRQ $ DQG VWDWLRQ % RQ VHYHUDO QLJKWV

PAGE 69

7DEOH &RPSDUDWLYH 1HZVFDVW 3HUIRUPDQFH %DVHG RQ 1LHOVHQ 0HWHUHG 2YHUQLJKW 5DWLQJV 0HDVXUH UDQNLQJV H[SUHVVHG ZLWKLQ SDUHQWKHVHV rr 6FKHIIH DQG SRVW KRF DQDO\VLV EDVHG RQ S &' K2

PAGE 70

6WDWLRQ & GLVWLQJXLVKHV LWVHOI IRU EHLQJ VLJQLILFDQWO\ GLIIHUHQW IURP ERWK FRPSHWLWRUV HYHU\ GD\ RI WKH ZHHN EXW )ULGD\ 5HIHUULQJ EDFN WR HDUOLHU GDWD ZH VHH WKDW WKHVH VLJQLILFDQW GLIIHUHQFHV DUH QRW QHFHVVDULO\ VLJQV RI RXWVWDQGLQJ UDWLQJV SHUIRUPDQFH 2Q WKH FRQWUDU\ VWDWLRQ & ODQJXLVKHG LQ WKLUG SODFH DFURVV DOO PHDVXUHV $ VHFRQG SRVW KRF DQDO\VLV ORRNV DW WKH UHODWLYH SHUIRUPDQFH RI VWDWLRQ $ DJDLQVW WKH DYHUDJHG DXGLHQFH UDWLQJV RI VWDWLRQV % DQG & LQ WDQGHP $JDLQ VWDWLRQ $n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f GDWD DUH DQDO\]HG +\SRWKHVHV )RXU DQG )LYH +\SRWKHVHV IRXU DQG ILYH DGGUHVVHG WKH IXQFWLRQDO UHODWLRQVKLS RI WKH QHZVFDVWV WR WKHLU UHVSHFWLYH OHDGLQ SURJUDPV 3URJUDPV ZLWK VWURQJ HTXLW\ ZHUH H[SHFWHG WR RSWLPL]H OHDGLQ UDWLQJV PRUH HIILFLHQWO\ WKDQ SURJUDPV ZLWK ZHDNHU HTXLW\ 7KLV HIILFLHQF\ ZRXOG EH REVHUYHG LQ WKH GLIIHUHQWLDO UDWLQJV UHVSRQVH EHWZHHQ WKH WZR SURJUDPV 'HSHQGLQJ RQ WKH W\SH RI DQDO\VLV HPSOR\HG WKHVH SURSRVLWLRQV ZHUH VXSSRUWHG WR YDU\LQJ GHJUHHV

PAGE 71

+\SRWKHVLV IRXU VWDWHG WKDW SURJUDPV ZLWK KLJKHU EUDQG HTXLW\ ZLOO UHYHDO D VLJQLILFDQWO\ JUHDWHU GLIIHUHQFH EHWZHHQ OHDGLQ SURJUDP DXGLHQFH VL]H DQG SURJUDP DXGLHQFH VL]H WKDQ SURJUDPV ZLWK ORZHU EUDQG HTXLW\ 7DEOH ORRNV DW WKH RYHUDOO KRXVHKROG SHUIRUPDQFH RI WKH WKUHH VWDWLRQV ,Q WHUPV RI FRPELQHG SHUIRUPDQFH ZH VHH WKDW WKHUH ZDV D b GURS LQ KRXVHKROGV XVLQJ WHOHYLVLRQ IURP 30 WR 30 +RZHYHU WKH GDWD DOVR VKRZ WKDW VWDWLRQ $ DV H[SHFWHG ZLWQHVVHG D VXEVWDQWLDO JDLQ LQ DXGLHQFH VXUSDVVLQJ LWV OHDGLQ E\ b 2I WKH WKUHH VWDWLRQV RQO\ VWDWLRQ $ DFKLHYHG D SRVLWLYH GLIIHUHQWLDO UHVSRQVH WR LWV OHDGLQ WKXV VXSSRUWLQJ WKH K\SRWKHVLV $GGLWLRQDOO\ DQ $129$ DQG 6FKHIIHH DQDO\VLV RI WKHVH GLIIHUHQWLDOV QHZV PLQXV OHDGLQ KRXVHKROGVf LQGLFDWHV WKDW WKHVH GLIIHUHQFHV ZHUH VLJQLILFDQW DPRQJ DOO WKUHH VWDWLRQV 7DEOH 'LIIHUHQWLDO 3HUIRUPDQFH %HWZHHQ 30 /HDGLQ DQG 30 1HZVFDVWV 2YHUDOO 1LHOVHQ 0HWHUHG 2YHUQLJKW +RXVHKROGV >f 6WDWLRQ $ 6WDWLRQ % 6WDWLRQ & $t%t& 1HZV +87f ) UDWLR 3URE 6FKHIIHH DW 30 $% $& 30 $% %& &$ 'LIIHUHQWLDO $% %& $& b FKDQJH 1RWH GI S +\SRWKHVLV ILYH VWDWHG WKDW SURJUDPV ZLWK KLJKHU HTXLW\ ZLOO UHYHDO D VLJQLILFDQWO\ JUHDWHU GLIIHUHQWLD VKDUH LQGH[ WKDQ SURJUDPV ZLWK ORZHU HTXLW\ %HFDXVH WKH QXPEHU RI KRPHV XVLQJ WHOHYLVLRQ +87f FDQ FKDQJH GUDPDWLFDOO\

PAGE 72

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f LQGH[HG KLJKHU WKDQ LWV WZR FRPSHWLWRUV 7DEOH 6KDUH ,QGH[ RI /DWH 1HZVFDVWV &RPSDUHG WR /HDGLQ 3URJUDPPLQJ 6WDWLRQ 2YHUDOO 0RQ 7XHV :HG 7KXUV )UL 6DW 6XQ $ r r r r r r % r & r r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f

PAGE 73

IRU D SURJUDP ZLWK UHODWLYHO\ VWURQJ HTXLW\ ,W ZDV SUHVXPHG WKDW HYLGHQFH RI WKLV ODFN RI GHSHQGHQF\ ZRXOG SURYLGH LQGLUHFW VXSSRUW RI K\SRWKHVHV IRXU DQG ILYH 7DEOH RIIHUV WKH UHVXOWV RI D FRUUHODWLRQDO DQDO\VLV EDVHG RQ KRXVHKROG GDWD )URP DQ RYHUDOO VWDQGSRLQW WKH FRUUHODWLRQ GDWD JLYH SDUWLDO VXSSRUW IRU WKH K\SRWKHVLV LQ WKDW VWDWLRQ $fV FRUUHODWLRQ LV VXEVWDQWLDOO\ ORZHU WKDQ VWDWLRQ % +RZHYHU 6WDWLRQ & GHPRQVWUDWHG DQ HYHQ ORZHU ILJXUH 2Q D GD\E\GD\ EDVLV VWDWLRQ $ RIIHUHG PRUH FRQYLQFLQJ HYLGHQFH E\ fZLQQLQJf IRXU RXW RI WKH VHYHQ QLJKWV ZLWK WKH ORZHVW FRUUHODWLRQ 7DEOH +RXVHKROG &RUUHODWLRQDO $QDO\VLV %HWZHHQ 30 /HDGLQ 3URJUDPPLQJ DQG 30 1HZVFDVWV 6WDWLRQ 2YHUDOO 0RQ 7XHV :HG 7KXUV )UL 6DW 6XQ $ r r r r r % r & r 1RWH $ % & LQGLFDWHV ORZHVW U DPRQJ WKH WKUHH VWDWLRQV $V D EULHI GLJUHVVLRQ WKH LQYHVWLJDWRU H[SORUHG D GLIIHUHQW PHWKRG RI FRUUHODWLRQ RI WKH RYHUDOO GDWD 6LQFH PRVW WHOHYLVLRQ SURJUDPPLQJ UHFXUV RQ D ZHHNO\ EDVLV FRUUHODWLRQV ZHUH SHUIRUPHG XVLQJ VHYHQGD\ UROOLQJ DYHUDJHV UDWKHU WKDQ RQ WKH PRUH FRQYHQWLRQDO VLQJOHGD\ GDWD $FFRUGLQJ WR WKH GDWD UHVXOWV SUHVHQWHG LQ 7DEOH VWDWLRQ $ UHYHDOHG WKH JUHDWHVW GLVSDULW\ EHWZHHQ WKH WZR PHWKRGV DQG DOVR UHSRVLWLRQHG LWVHOI DV WKH VWDWLRQ H[KLELWLQJ WKH ORZHVW FRUUHODWLRQ

PAGE 74

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f 6WDWLRQ 6ORSH ,QWHUFHSW $ % & KRXVHKROGV ZHUH VWURQJ SUHGLFWRUV RI QHZVFDVW KRXVHKROG SHUIRUPDQFH +RZHYHU RQ IDFH YDOXH WKH VORSH DQG LQWHUFHSWV GR QRW LQGLFDWH DQ\ UHPDUNDEOH GLIIHUHQFHV DPRQJ WKH WKUHH FRPSHWLWRUV %DVHG RQ WKHVH GDWD RQH FRXOG VSHFXODWH WKDW 6WDWLRQ $fV VXFFHVV FDQQRW EH DWWULEXWHG VROHO\ WR LWV DELOLW\ WR FRQYHUW OHDGLQ DXGLHQFHV EXW UDWKHU WR LWV SRZHU WR UHFUXLW DXGLHQFHV IURP RWKHU VRXUFHV 7KH ILQDO WZR K\SRWKHVHV WDNH D FORVHU ORRN DW WKLV SRVVLELOLW\

PAGE 75

+\SRWKHVHV 6L[ DQG 6HYHQ +\SRWKHVHV VL[ DQG VHYHQ WDNH WKH DERYH QRWLRQV RI GLIIHUHQWLDO UHVSRQVH RU PRPHQWXPf D VWHS IXUWKHU E\ GLYLGLQJ WKH SKHQRPHQRQ LQWR WZR GLVWLQFW VHJPHQWV Df WKH SRZHU WR UHWDLQ OHDQLQ DXGLHQFHV DQG Ef WKH SRZHU WR UHFUXLW DXGLHQFHV IURP RWKHU VRXUFHV 1LHOVHQ 0HGLD 5HVHDUFK SURYLGHG WKH LQYHVWLJDWRU ZLWK D VWDQGDUG f)ORZ $QDO\VLV VWXG\ RI WKH 0D\ GLDU\ VZHHS SHULRG 8VLQJ IRXUZHHN DYHUDJHG KRXVHKROG UDWLQJV 1LHOVHQ WUDFNHG ZHHNGD\ DXGLHQFH IORZ IURP 30 WR 30 :KLOH WKH FRPSOHWH VWXG\ LV VKRZQ LQ WKH DSSHQGL[ WKH PRVW VDOLHQW GDWD KDYH EHHQ GLVWLOOHG LQWR 7DEOH 7DEOH URZV WKURXJK VKDGHG DUHDf SURYLGH GDWD RQ WKH UHODWLYH DXGLHQFH FRPSRVLWLRQ RI HDFK QHZVFDVW 7KDW LV WRWDO QHZV KRXVHKROGV URZ f HTXDOV WKH VXP RI KRXVHKROGV UHWDLQHG IURP OHDGLQV URZ f SOXV KRXVHKROGV UHFUXLWHG IURP RWKHU VWDWLRQV SURYLGLQJ ORFDO QHZV URZ f SOXV KRXVHKROGV UHFUXLWHG IURP QRQQHZV VRXUFHV URZ f 7KH ODVW WZR FDWHJRULHV GHDO ZLWK DXGLHQFHV WKDW fVZLWFKHGf FKDQQHOV LQ RUGHU WR ZDWFK D VSHFLILF QHZVFDVW 7KH SHUFHQWDJH FROXPQV ZLWKLQ WKH VKDGHG DUHD JLYH WKH SURSRUWLRQ RI WKH WRWDO QHZV DXGLHQFH WKDW HDFK FDWHJRU\n KROGV 7KHVH IRXU GDWD URZV KRZHYHU GR QRW SURYLGH VXIILFLHQW LQVLJKW LQWR WKH SRZHU RI EUDQG HTXLW\ 5RZV DQG GHOYH LQWR G\QDPLFV RI RXU ILQDO K\SRWKHVHV +\SRWKHVLV VL[ SUHGLFWHG WKDW SURJUDPV ZLWK KLJKHU EUDQG HTXLW\ ZRXOG UHWDLQ D JUHDWHU SURSRUWLRQ RI WKHLU OHDGLQ DXGLHQFHV WKDQ SURJUDPV ZLWK ORZHU EUDQG HTXLW\ $V H[SHFWHG URZ RI 7DEOH VKRZV VWDWLRQ $ WR EH WKH EHVW SHUIRUPHU UHWDLQLQJ b RI LWV OHDGLQ KRXVHKROGV

PAGE 76

+\SRWKHVLV VHYHQ VWDWHG WKDW SURJUDPV ZLWK KLJKHU EUDQG HTXLW\ ZRXOG UHFUXLW D JUHDWHU SURSRUWLRQ RI DXGLHQFHV IURP RWKHU VRXUFHV WKDQ SURJUDPV ZLWK ORZHU EUDQG HTXLW\ 5RZV DQG RI 7DEOH DGGUHVV WKLV SURSRVLWLRQ 3HUKDSV WKH SURSRUWLRQ RI KRXVHKROGV UHFUXLWHG IURP RWKHU QHZV VWDWLRQV URZ f RIIHUV WKH PRVW SHUVXDVLYH HYLGHQFH +HUH ZH KDYH DXGLHQFHV WKDW VZLWFKHG DZD\ IURP D FKDQQHO WKDW ZDV DERXW WR SUHVHQW D ORFDO QHZVFDVW 2I DOO WKH QHZV FKDQQHO fVZLWFKHUVf VWDWLRQ $ HDUQHG WKH ODUJHVW SURSRUWLRQ bf 6LPLODUO\ VWDWLRQ $ UHFUXLWHG D JUHDWHU SURSRUWLRQ bf RI WKH WRWDO QRQQHZV VZLWFKHUV

PAGE 77

7DEOH 1LHOVHQ $XGLHQFH )ORZ $QDO\VLV 0D\ 6ZHHS WR 30 0RQGD\ 7KURXJK )ULGD\ )RXU:HHN $YHUDJH +RXVHKROGV >f 5RZ QR 0HDVXUH 6WDWLRQ $ 6WDWLRQ % 6WDWLRQ & b b b ++ UHWDLQHG IURP OHDGLQ ++ UHFUXLWHG IURP IURP RWKHU QHZV VWDWLRQV VZLWFKHUVf ++ UHFUXLWHG IURP QRQ QHZV VRXUFHV VZLWFKHUf 7RWDO 1HZV ++ b ++ UHWDLQHG IURP OHDGVLQ r b UHFUXLWHG IURP WRWDO QHZV VZLWFKHUV f rr b UHFUXLWHG IURP WRWDO QRQ QHZV VZLWFKHUV f rrr 1RWH FKL VJXDUH S r ZLWK VWDWLRQV % DQG & FROODSVHG GI rr ZLWK VWDWLRQV % DQG & FROODSVHG GI rrr DPRQJ DOO WKUHH VWDWLRQV GI DQG ZLWK VWDWLRQV % DQG & FROODSVHG GI

PAGE 78

&+$37(5 ',6&866,21 7KH SXUSRVH RI WKLV VWXG\ ZDV WR H[SORUH WKH SODXVLELOLW\ RI DSSO\LQJ EUDQG HTXLW\ WKHRU\ WR WHOHYLVLRQ SURJUDP UDWLQJV SHUIRUPDQFH :KLOH WKH LQGXVWU\ KDV HPEUDFHG WKH fEX]]ZRUGVf RI EUDQG PDQDJHPHQW WKHUH KDV EHHQ OLWWOH VFKRODUO\ ZRUN RQ PDNLQJ WKH WUDQVLWLRQ IURP SURGXFWV DV EUDQGV WR PHGLD DV EUDQGV :KLOH VSHFLILF K\SRWKHVLV ZHUH SUHVHQWHG PXFK RI WKH HIIRUW LQYHVWHG LQ WKLV VWXG\ ZDV D PDWWHU RI FXULRVLW\ DQG LQWHOOHFWXDO WLQNHULQJ $V WKH WLWOH RI WKLV GLVVHUWDWLRQ LPSOLHV WKH ILUVW FKDOOHQJH ZDV FRQFHSWXDOL]DWLRQf§+RZ FDQ FRQYHQWLRQDO EUDQG HTXLW\ WKHRU\ EH DSSOLHG LQ D PHDQLQJIXO ZD\ WR WHOHYLVLRQ SURJUDPPLQJ 7KH VHFRQG FKDOOHQJH ZDV WR ILQG D ZD\ WR PHDVXUH RU RSHUDWLRQDOL]H WKHVH FRQFHSWV XVLQJ VWDQGDUG 1LHOVHQ UDWLQJV GDWD ,Q DGGLWLRQ WR DQ RYHUYLHZ RI EUDQG HTXLW\ WKHRU\ WKH FRQFOXVLRQ RI WKH OLWHUDWXUH UHYLHZ VHFWLRQ SURYLGHG VRPH IHDVLEOH FRQFHSWXDOL]DWLRQV IRU WHOHYLVLRQ SURJUDPPLQJ ,Q SDUWLFXODU .HOOHU f DQG WKH QRWLRQV RI FRQVXPHU LQHUWLD PRPHQWXPf ZHUH IRXQG WR ZRUN ZHOO LQ H[SODLQLQJ SURJUDP SHUIRUPDQFH LQ OLJKW RI WKH SRZHUIXO ,QIOXHQFH RI LQKHULWDQFH HIIHFWV 3UHVXPLQJ WKDW HVVHQWLD PDUNHWLQJ PL[ FRPSRQHQWV DUH FRQWUROOHG RU QHXWUDOL]HG WKH SURJUDP HTXLW\ PRGHO RI DXGLHQFHV UHWDLQHGDXGLHQFHV UHFUXLWHG DSSHDUV WR KDYH PHULW

PAGE 79

$SSO\LQJ WKLV PRGHO WR LQKHULWDQFH HIIHFWV RIIHUV VRPH QHZ WKHRUHWLFDO LQVLJKWV ,QVWHDG RI DGGUHVVLQJ VWUXFWXUDO IDFWRUV VXFK DV SURJUDP W\SH RU QXPEHU RI SURJUDP RSWLRQV WKLV DSSURDFK GHOYHG LQWR WKH DELOLW\ RI D SURJUDP WR UHWDLQ OHDGLQ DXGLHQFHV DQG UHFUXLW DXGLHQFHV IURP RWKHU VRXUFHV ,Q WDQGHP WKHVH WZR SRZHU GLPHQVLRQV GHILQH D SURJUDPnV PRPHQWXP DV LW fWUDYHOVf WKURXJK WLPH $ SURJUDP H[KLELWLQJ VWURQJ HTXLW\ FDQ RYHUFRPH D SRRU OHDGLQ DQG UHFUXLW OR\DO YLHZHUV IURP RWKHU VRXUFHV $GGLWLRQDOO\ WKLV SURJUDP ZLOO EH OHVV OLNHO\ WR ORVH SRUWLRQV RI LWV OHDGLQ DXGLHQFH WR GLUHFW FRPSHWLWRUV 7KH FDVH VWXG\ SURYLGHG D PRUH FRQFUHWH ZD\ WR fWHVWf VRPH RI WKHVH WKHRUHWLFDO SURSRVLWLRQV $OO VHYHQ K\SRWKHVHV ZHUH VXSSRUWHG WR VRPH GHJUHH DQG RSHQ WKH GRRU IRU IXUWKHU WKHRUHWLFDO ZRUN DQG PRUH JHQHUDOL]DUH HPSLULFDO UHVHDUFK )URP D FRQFHSWXDO VWDQGSRLQW WKH D PDMRU GLOHPPD IDFLQJ WKH LQYHVWLJDWRU ZDV FKRRVLQJ D WKHRUHWLFDO VWDUWLQJ SRLQW :LWK VR PDQ\ GLYHUJHQW FRQFHSWXDOL]DWLRQV DQG WKHRULHV DYDLODEOH EUDQG HTXLW\ FDQ EH DOPRVW DQ\WKLQJ D UHVHDUFKHU ZDQWV LW WR EH )URP DWWLWXGLQDO QRWLRQV RI fKHULWDJHf fUHSXWDWLRQf fSHUFHLYHG TXDOLW\f fGHSWK RI DZDUHQHVVf fHQWKXVLDVPf DQG fFRPPLWPHQWf WR PRUH EHKDYLRUDO FRQFHSWV VXFK DV fSULFLQJ HODVWLFLW\f fUHSHDW SXUFKDVHVf fH[FOXVLYH KHDY\ XVHUf DQG fFRQVXPHU LQHUWLDf PDUNHWLQJ SURIHVVLRQDOV DQG VFKRODUV KDYH \HW WR DJUHH RQ D SUHFLVH GHILQLWLRQ RI EUDQG HTXLW\ &RPSRXQGLQJ WKH SUREOHP LV WKDW PDQ\ LQ WKH ILHOG XVH PDUNHWLQJ WHUPV VXFK DV EUDQG LPDJH EUDQG OR\DOW\ DQG EUDQG HTXLW\ LQWHUFKDQJHDEO\ $IWHU PXFK VHDUFKLQJ DQG UHIOHFWLRQ .HOOHU f HPHUJHG IURP WKLV WKHRUHWLFDO PXGGOH

PAGE 80

ZLWK DQ DSSURDFK WKDW FDQ EH WUDQVODWHG UHDGLO\ LQWR EURDGFDVW SURJUDPPLQJ WHUPV .HOOHUfV WKHRUHWLFDO XQGHUSLQQLQJV FRPELQHG ZLWK D PRUH ULJRURXVO\ GHILQHG EUDQGLQJ YRFDEXODU\ FDQ IRUP WKH EHGURFN RI D EUDQG HTXLW\ UHVHDUFK DJHQGD $ IDU PRUH SHUSOH[LQJ LVVXH ZDV GLVFRYHUHG RQ WKH RSHUDWLRQDO VLGH RI SURJUDP EUDQG HTXLW\ .HOOHUfV f RYHUDOO FRQFHSWXDO IUDPHZRUN UHTXLUHV D FRQWUROOHG PDUNHWLQJ PL[ ZKLFK HQDEOHV D UHVHDUFKHU WR FDSWXUH WKH UHODWLYH SRZHU RI EUDQG HTXLW\ 7KDW LV VWURQJ IDYRUDEOH DQG XQLTXH EUDQG DVVRFLDWLRQV EUDQG NQRZOHGJHf FDQQRW EH PHDVXUHG SURSHUO\ XQWLO WKH SRVVLEOH FRQIRXQGLQJ HOHPHQWV RI WKH EUDQGfV PDUNHWLQJ PL[ DUH KHOG LQ FKHFN 0DUNHWLQJ PL[ IDFWRUV VXFK DV SULFLQJ GLVWULEXWLRQ DQG SURPRWLRQ FDQ WDLQW HTXLW\ PHDVXUHV WKDW UHO\ RQ LQPDUNHW EUDQG SHUIRUPDQFH GDWD )RU RXU FDVH VWXG\ FRQWUROOLQJ WKHVH PDUNHWLQJ PL[ IDFWRUV ZDV DQ XQH[SHFWHG FKDOOHQJH 7KH PHWKRGRORJ\ VHFWLRQ RI WKLV GLVVHUWDWLRQ UHYHDOHG D QXPEHU RI SUREOHPV WKDW KDG WR EH RYHUFRPH $PRQJ WKHVH ZHUH Df LGHQWLI\LQJ GLUHFW FRPSHWLWRUV E\ SURJUDP W\SH RU JHQUH DQG Ef FUHDWLQJ WKH VLPXOWDQHRXV DYDLODELOLW\ RI GLUHFW FRPSHWLWRUV 7KH VHOHFWLRQ RI WKUHH 30 ORFDO QHZVFDVWV VFKHGXOHG DW WKH H[DFW VDPH WLPH PDGH IRU DQ LGHDO WHVW HQYLURQPHQW IRU RXU FDVH VWXG\ +RZHYHU PRVW WHOHYLVLRQ SURJUDPPLQJ GRHV QRW RIIHU WKLV FRQYHQLHQFH 'HWHUPLQLQJ ZKR LV D GLUHFW FRPSHWLWRU FDQ EH D IUXVWUDWLQJ WDVN :KLOH ORFDO QHZVFDVWV GR QRW SRVH D JUHDW SUREOHP RWKHU SURJUDP W\SHV FDQ JHQHUDWH FRQVLGHUDEOH GLVDJUHHPHQW DPRQJ PHGLD UHVHDUFKHUV )RU H[DPSOH

PAGE 81

VKRXOG f+RPH ,PSURYHPHQWf DQG f0DUULHG :LWK &KLOGUHQ"f EH DVVLJQHG WR WKH VDPH JHQUH RI VLWFRP" :LWKLQ WKH FDWHJRU\ RI DGXOW GUDPD VKRXOG f1<3' %OXHf DQG f'U 4XLQQf EH UHJDUGHG DV FRPLQJ IURP WKH VDPH SURGXFW FDWHJRU\" :HEVWHU f DGGUHVVHG WKLV LVVXH RI SURJUDP W\SRORJ\ DQG ZDUQHG WKDW WKH FXUUHQW LQGXVWU\ FDWHJRULHV ZRXOG VRPHGD\ EHFRPH fRYHUO\ EURDG DQG LQVHQVLWLYH WR VXEWOH WKRXJK LPSRUWDQW GLVWLQFWLRQVf S f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fZKDW LIf VXUYH\ FRXOG EH DGPLQLVWHUHG WR DQ DSSURSULDWH VDPSOH JURXS ,Q WKLV FDVH WKH FRQWUROV ZRXOG EH IDU PRUH FRQWULYHG WKDQ D UHDOZRUOG VHWWLQJ ZKHUH DFWXDO YLHZLQJ EHKDYLRU FRXOG EH UHFRUGHG 7ZR UHODWHG GLVWULEXWLRQ IDFWRUV WKDW PD\ EH DZNZDUG WR FRQWURO DUH VLJQDO FRYHUDJH DQG DYDLODEOH SRWHQWLDO DXGLHQFHV -XVW DV UHWDLOHUV UHFRJQL]H WKH PDUNHWLQJ DGYDQWDJHV RI D VXSHULRU GLVWULEXWLRQ V\VWHP VR EURDGFDVW

PAGE 82

UHVHDUFKHUV PXVW EH DZDUH RI FHUWDLQ VWUXFWXUDO IDFWRUV VXFK DV VLJQDO VWUHQJWK DQG WKH DYDLODEOH DXGLHQFHV WKDW FDQ EH UHDFKHG E\ WKDW VLJQDO 3URJUDPV WKDW FDQQRW EH UHFHLYHG FDQQRW EH ZDWFKHG 8QOHVV WKHUH LV D fOHYHO SOD\LQJ ILHOGf LQ WHUPV RI SURJUDP GLVWULEXWLRQ FRQVXPHUEDVHG EUDQG HTXLW\ FDQQRW EH PHDVXUHG DFFXUDWHO\ )RUWXQDWHO\ RXU WHVW PDUNHW H[KLELWHG HTXLYDOHQF\ LQ WKLV DUHD EXW WKLV PD\ QRW EH WKH FDVH IRU VRPH RWKHU PDUNHWV )RU H[DPSOH WKH *DLQHVYLOOH )ORULGD WHOHYLVLRQ PDUNHW PXVW fLPSRUWf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f ORQJWHUP UDWLQJV OHDGHUVKLS DQG Ef YLHZHU OR\DOW\ GDWD GHULYHG IURP WKH 0DUVKDOO

PAGE 83

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f GDWD UHYHDOHG OHVV HTXLW\ YDOXH IRU VWDWLRQ $ WKDQ WKH LQGLYLGXDO GD\ GDWD 7KLV PD\ EH WKH UHVXOW RI GLVWRUWLRQV WKDW FDQ RFFXU VRPHWLPHV ZKHQ DYHUDJLQJ GDWD )RU H[DPSOH WKH FRUUHODWLRQ RQ 6XQGD\ QLJKW IRU VWDWLRQ $ LV IDU RXW RI OLQH FRPSDUHG ZLWK WKH RWKHU VL[ GD\V RI WKH ZHHN DQG PD\ KDYH RYHUVWDWHG WKH RYHUDOO SHUIRUPDQFH RI WKH VWDWLRQ 6LPLODUO\ WKH 7KXUVGD\ QLJKW FRUUHODWLRQ IRU VWDWLRQ & DJDLQVW f(5f RQ 1%&f LV H[WUDRUGLQDULO\ ORZ FRPSDUHG ZLWK RWKHU GD\V $QRWKHU SRVVLELOLW\ PD\ EH D FRQFHSWXDO IODZ 7KDW LV WKH VXSSRVHG LQGHSHQGHQFH RI RXU KLJK HTXLW\ VWDWLRQ PD\ EH UHYHDOHG RQO\ ZKHQ WKH OHDGLQ LV H[FHSWLRQDOO\ SRRU $QRWKHU UHODWHG DYHQXH IRU IXWXUH H[SORUDWLRQ LV ZK\ WKH XVH RI VHYHQGD\ PRYLQJ DYHUDJHV 7DEOH f VHHPHG WR HQKDQFH WKH SRVLWLRQ RI VWDWLRQ $

PAGE 84

7KH UHJUHVVLRQ DQDO\VHV LQ WKLV GLVVHUWDWLRQ LQYLWH VRPH WKHRUHWLFDO VSHFXODWLRQ RQ WKH LQWHUSUHWDWLRQ RI SURJUDP HTXLW\ 7KH WZR FRPSRQHQWV RI D VLQJOH OLQHDU UHJUHVVLRQ HTXDWLRQ LWV VORSH FRHIILFLHQWf DQG LWV LQWHUFHSW FRQVWDQWf FRXOG EH FRQVWUXHG DV LQGLFDWRUV RI WKH WZR SRZHU GLPHQVLRQV RI RXU HTXLW\ PRGHO *LYHQ WKH LGHQWLFDO OHDGLQ DXGLHQFHV LQGHSHQGHQW YDULDEOHf LW LV SRVVLEOH PDWKHPDWLFDOO\ IRU WZR SURJUDPV WR H[KLELW WKH VDPH VORSH EXW GLIIHUHQW LQWHUFHSWV DQG YLVD YHUVD 7KH VORSH RI WKH HTXDWLRQ PD\ SURYLGH D URXJK PHDVXUH RI WKH HIILFLHQF\ RI D SURJUDP WR UHWDLQ LWV OHDGLQ DXGLHQFH $OWHUQDWLYHO\ WKH LQWHUFHSW ZKLFK ZRXOG EH GHILQHG PDWKHPDWLFDOO\ DV WKH VL]H RI D QHZV DXGLHQFH GHSHQGHQW YDULDEOHf ZKHQ WKH OHDGLQ LQGHSHQGHQW YDULDEOHf LV ]HUR PLJKW EH DQDORJRXV WR D PHDVXUH RI WKH SURJUDPf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

PAGE 85

VL[ DQG VHYHQ 7KH DELOLW\ RI 6WDWLRQ $ WR ERWK UHWDLQ DQG UHFUXLW DXGLHQFHV ZHUH XQVXUSDVVHG 6WXG\ /LPLWDWLRQV $VLGH IURP WKH DERYHPHQWLRQHG GLIILFXOWLHV LQ FRQWUROOLQJ WKH PDUNHWLQJ PL[ WKH UHVHDUFKHU UHFRJQL]HV VHYHUDO OLPLWDWLRQV WR WKLV SDUWLFXODU VWXG\ )LUVW WKHUH LV WKH ODFN RI H[WHUQDO YDOLGLW\ %\ GHVLJQ WKLV ZDV DQ H[SORUDWRU\ HIIRUW XQGHU OLPLWHG FRQGLWLRQV WKDW ZHUH IDPLOLDU WR WKH UHVHDUFKHU 7KH WKUHH QHZV SURJUDPV LQ WKLV SDUWLFXODU PDUNHW SURYLGHG DQ LGHDO ILHOG VHWWLQJ ZKHUH PDUNHWLQJ PL[ IDFWRUV FRXOG EH FRQWUROOHG WR D UHDVRQDEOH GHJUHH %\ KDYLQJ GLIIHUHQW OHDGLQ SURJUDPV HDFK QLJKW WKH QHZVFDVWV ZHUH SODFHG LQ GLIIHUHQW fH[SHULPHQWDOf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f UHFRJQL]H VHYHUDO VWDWLVWLFDO OLPLWDWLRQV )LUVW 1LHOVHQ GRHV QRW JXDUDQWHH D UHSUHVHQWDWLYH UDQGRP VDPSOH GXH WR D QXPEHU RI IDFWRUV WKH PRVW LPSRUWDQW EHLQJ GLVPDO FRRSHUDWLRQ UDWHV DYHUDJLQJ bf DPRQJ

PAGE 86

SHRSOH DVNHG WR SDUWLFLSDWH 6HFRQGO\ SUHVXPLQJ WKH VDPSOHV ZHUH SHUIHFWO\ UDQGRP WKHUH LV FRQVLGHUDEOH VDPSOLQJ HUURU VWDQGDUG HUURUf WKDW LV QRW GLVFORVHG LQ WKH UHJXODU UHSRUWV $W WKH EHJLQQLQJ RI HYHU\ 1LHOVHQ PDUNHW VZHHS UHSRUW WKH FRPSDQ\ ZDUQV VXEVFULEHUV DJDLQVW XVLQJ FHUWDLQ UDWLQJV GDWD WR PDNH PDQDJHPHQW GHFLVLRQV EHFDXVH WKH UHODWLYH VWDQGDUG HUURU FRXOG H[FHHG b 1LHOVHQ 0HWKRG f 7KHUHIRUH WHOHYLVLRQ UDWLQJV GDWD WHQG WR IOXFWXDWH 0DQ\ PHGLD SURIHVVLRQDOV ZDQW WR EHOLHYH WKDW WKHVH FKDQJHV DUH UHDO EXW VDYY\ UHVHDUFKHUV NQRZ WKDW PXFK RI WKLV PRWLRQ LV GXH WR VDPSOLQJ HUURU RU fERXQFHf UDWKHU WKDQ OHJLWLPDWH FKDQJHV LQ DXGLHQFH WXQLQJ EHKDYLRU :KLOH WKHUH DUH GLVWLQFW DGYDQWDJHV WR XVLQJ WKH PHWHUHG RYHUQLJKW UDWLQJV FRPSDUHG WR WKH IRXUZHHN DYHUDJHG VZHHSV GDWD D GLVDSSRLQWLQJ GUDZEDFN ZDV WKH ODFN RI GHPRJUDSKLF LQIRUPDWLRQ 8QOLNH WKH QDWLRQDO 1LHOVHQ fSHRSOH PHWHUf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f SRZHU WR UHWDLQ DXGLHQFHV DQG Ef WKH SRZHU WR UHFUXLW DXGLHQFHV ZHUH VXEVWDQWLDWHG DQG 1LHOVHQ SURPLVHV WKDW RYHUQLJKW IORZ VWXGLHV ZRXOG EH DYDLODEOH VRRQ WR PHWHUHG PDUNHWV 0HDQZKLOH WKH VZHHS

PAGE 87

EDVHG IORZ VWXGLHV IRXU SHU \HDUf FDQ VWLOO EH DSSOLHG WR RWKHU PDUNHWV LQ DQ DWWHPSW WR UHSOLFDWH RXU FDVH VWXG\ ILQGLQJV 2I FRXUVH 1LHOVHQ FKDUJHV EURDGFDVWHUV D VXEVWDQWLDO IHH IRU WKHVH VSHFLDO VWXGLHV VR DQ DFDGHPLF UHVHDUFKHU PD\ KDYH D ILQDQFLDO KDQGLFDS 7KH 0DUVKDO 0DUNHWLQJ DQG &RPPXQLFDWLRQV GDWD ZHUH XVHG SULPDULO\ DV D EHQFKPDUN FRPSDULVRQ IRU WKH 1LHOVHQ UDWLQJV :LWKLQ WKHLU VWDWLVWLFDO FDOFXODWLRQV ZDV WKH GHWHUPLQDWLRQ RI f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fV EUDQG NQRZOHGJHf FDQ EH PHDVXUHG WR VRPH GHJUHH XVLQJ WKH EHKDYLRUDO PHDVXUHV RI VWDQGDUG 1LHOVHQ UDWLQJV $GGLWLRQDOO\ WKLV SURMHFW RIIHUV QHZ LQVLJKW LQWR WKH SKHQRPHQRQ RI LQKHULWDQFH HIIHFWV DQG WXQLQJ LQHUWLD $IWHU HVWDEOLVKLQJ FRQFHSWXDO DQG RSHUDWLRQDO GHILQLWLRQV RI SURJUDP EUDQG HTXLW\ WKLV VWXG\ H[DPLQHG KRZ D SURJUDP ZLWK VXEVWDQWLDO EUDQG HTXLW\ SHUIRUPV ZKHQ FRPSDUHG WR GLUHFW FRPSHWLWRUV ZLWK OHVV EUDQG HTXLW\ 7KH

PAGE 88

VHYHQ K\SRWKHVHV DSSURDFKHG WKLV LVVXH IURP VHYHUDO SHUVSHFWLYHV +\SRWKHVHV RQH WZR DQG WKUHH ORRNHG DW RYHUDOO UDWLQJV SHUIRUPDQFH DQG IRXQG WKDW WKH SURJUDP LGHQWLILHG DV KDYLQJ KLJK EUDQG HTXLW\ KDG KLJKHU UDWLQJV KLJKHU VKDUHV DQG D ODUJHU KRXVHKROG DXGLHQFH +\SRWKHVHV IRXU DQG ILYH QDUURZHG WKH LQYHVWLJDWLRQ WR WKH UHODWLRQVKLS RI D SURJUDP WR LWV OHDGLQ $ QXPEHU RI VWDWLVWLFDO LQGLFDWRUV UHYHDOHG WKDW D SURJUDP ZLWK VWURQJ HTXLW\ SURFHVVHV RU fLQKHULWVf DXGLHQFHV LQ ZD\V WKDW DUH GLIIHUHQW WKDQ LWV GLUHFW FRPSHWLWRUV 7KDW LV ZKLOH WKH KLJK HTXLW\ SURJUDP FRQVLVWHQWO\ RXWSHUIRUPHG LWV 30 OHDGLQV WR PDLQWDLQ 30 UDWLQJV OHDGHUVKLS WKH RWKHU VWDWLRQV XVXDOO\ ORVW UDWLQJV GXULQJ WKLV WUDQVLWLRQ 7UDQVODWLQJ WKHVH UDWLQJ SRLQW JDLQV DQG ORVVHV LQWR WKH FRPSRQHQWV RI ZKHUH DQ DXGLHQFH KDV FRPH IURP DQG ZKHUH LW KDV JRQH ZDV WKH GRPDLQ RI WKH ILQDO WZR K\SRWKHVHV 8VLQJ D f)ORZ $QDO\VLVf WKDW GLIIHUHQWLDWHG EHWZHHQ DXGLHQFHV UHWDLQHG IURP OHDGLQV DQG DXGLHQFHV UHFUXLWHG IURP RWKHU VRXUFHV WKH UHVXOWV LQGLFDWHG WKDW WKH KLJK HTXLW\ SURJUDP ZDV VXSHULRU WR LWV ORZ HTXLW\ FRPSHWLWRUV QRW RQO\ LQ UHWDLQLQJ OHDGLQ KRXVHKROGV EXW ZDV LQ GUDZLQJ KRXVHKROGV IURP VXUURXQGLQJ FKDQQHOV DQG WKH fRIIf SRVLWLRQ 7DNLQJ DQ HYHQ WLJKWHU IRFXV WKH UHVXOWV VKRZHG WKDW WKH KLJK HTXLW\ SURJUDP GHPRQVWUDWHG H[FHSWLRQDO SRZHU LQ UHFUXLWLQJ DXGLHQFHV IURP FKDQQHOV RIIHULQJ VLPLODU ORFDO QHZVFDVWV GLUHFW FRPSHWLWRUVf 7DNHQ LQ FRPELQDWLRQ WKH K\SRWKHVHV YDOLGDWH WKH LPSRUWDQFH RI EUDQG HTXLW\ WR WKH UDWLQJV VXFFHVV RI D WHOHYLVLRQ SURJUDPDW OHDVW IRU ORFDO QHZV SURJUDPPLQJ

PAGE 89

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t FRPPLWPHQW %UDQG (TXLW\ 7DNLQJ WKH WKHRUHWLFDO SRVLWLRQ WKDW XQGHUVWDQGLQJ DXGLHQFH IORZ LV FUXFLDO WR XQGHUVWDQGLQJ WKH LQIOXHQFH RI SURJUDP EUDQG HTXLW\ WKH IROORZLQJ SURSRVHG DXGLHQFH PHPEHU FDWHJRULHV PD\ DVVLVW IXWXUH UHVHDUFK

PAGE 90

/R\DOLVWV 'HILQLWLRQ 9LHZHUV ZKR KROG VWURQJ FRQVXPHUEDVHG SURJUDP EUDQG HTXLW\ DQG DUH IRXQG ZLWKLQ WKH OHDGLQ SURJUDPfV WRWDO DXGLHQFH 'LVSRVLWLRQ 7KHVH YLHZHUV UHPDLQ LQ SODFH RQ WKH VDPH FKDQQHO EHFDXVH RI EUDQG FRPPLWPHQW 3DVVLYHV 'HILQLWLRQ 9LHZHUV ZKR GR QRW KROG DQ\ VWURQJ FRQVXPHUEDVHG SURJUDP EUDQG HTXLW\ WRZDUGV DQ\ GLUHFW FRPSHWLWRU DQG DUH IRXQG ZLWKLQ WKH OHDGLQ SURJUDPfV WRWDO DXGLHQFH 'LVSRVLWLRQ 7KHVH YLHZHUV ZLOO UHPDLQ LQ SODFH RQ WKH VDPH FKDQQHO EHFDXVH RI VLPSOH fLQHUWLDf UDWKHU WKDQ WUXH FRPPLWPHQW &RQYHUWV 'HILQLWLRQ 9LHZHUV ZKR KROG VWURQJ FRQVXPHUEDVHG SURJUDP HTXLW\ EXW ZHUH ZDWFKLQJ RQ D FRPSHWLQJ FKDQQHO DQG WKHUHIRUH DUH PRWLYDWHG WR VZLWFK FKDQQHOV 'LVSRVLWLRQ ,I QHFHVVDU\ WKHVH YLHZHUV ZLOO DEDQGRQ D OHDGLQ SURJUDP FKDQQHO DQG VZLWFK RYHU WR D PRUH VXLWDEOH SURJUDP RSWLRQ VHH QXPEHU GHIHFWRUV EHORZf 7XQH LQV 'HILQLWLRQ f$SSRLQWPHQWf YLHZHUV ZKR KROG YHU\ VWURQJ FRQVXPHUEDVHG SURJUDP EUDQG HTXLW\ DQG PDNH D GHOLEHUDWH HIIRUW WR WXUQ RQ WKH 79 VHW LQ RUGHU WR ZDWFK D VSHFLILF SURJUDP ,W SRVVLEOH WR KDYH D VXEFDWHJRU\ RI SDVVLYH WXQH

PAGE 91

LQV ZKHUH WKH GHVLUH LV VLPSO\ WR KDYH WKH VHW fRQf ZLWK QR UHJDUG IRU VSHFLILF SURJUDPPLQJ 'LVSRVLWLRQ 7KHVH YLHZHUV DUH DFTXLUHG DXGLHQFHV IURP ZKDW VRPH UHVHDUFKHUV FDOO WKH fRII SRVLWLRQf 'HIHFWRUV 'HILQLWLRQn f&RQYHUWVf VHHQ IURP DQ RSSRVLQJ SHUVSHFWLYH WKDW LV YLHZHUV ZKR DUH DERXW WR DEDQGRQ D SURJUDP FKDQQHO LQ RUGHU WR ILQG D PRUH VXLWDEOH RU SUHIHUUHG SURJUDP 'HIHFWRUV KROG VWURQJ FRQVXPHUEDVHG SURJUDP HTXLW\ IRU D GLUHFW FRPSHWLWRU 2QH SURJUDPfV GHIHFWRU LV DQRWKHU SURJUDPfV FRQYHUWf 'LVSRVLWLRQ $W WKH DSSURSULDWH WLPH WKHVH YLHZHUV OHDYH RQH SURJUDP DQG JR WR DQRWKHU RQ D FRPSHWLQJ FKDQQHO 8VLQJ WKHVH FDWHJRULHV RQH FRXOG K\SRWKHVL]H WKDW SURJUDP HTXLW\ Df LQFUHDVHV WKH QXPEHU RI FRQYHUWV Ef UHGXFHV WKH QXPEHU RI GHIHFWRUV Ff UHLQIRUFHV FXUUHQW OR\DOLVWV DQG Gf WUDQVIRUPV SDVVLYHV LQWR FRQYHUWV ,PSOLFDWLRQV )RU WKH )LHOG :HOO QRZ KRPH HQWHUWDLQPHQW ZDV P\ EDE\fV ZLVK VR KRSHG LQWR WRZQ IRU D VDWHOOLWH GLVK WLHG LW WR WKH WRS RI P\ -DSDQHVH FDU FDPH KRPH DQG SRLQWHG LW RXW WR WKH VWDUV $ PHVVDJH FDPH EDFN IURP WKH JUHDW EH\RQG 7KHUHfV ILIW\VHYHQ FKDQQHOV DQG QRWKLQf RQ 6SULQJVWHHQ f LQ DQ HDUOLHU GHFDGH ZKHQ FDEOH DQG VDWHOOLWH GHOLYHU\ V\VWHPV ZHUH MXVW EHJLQQLQJ WR RIIHU D EURDG UDQJH RI SURJUDPPLQJ RSWLRQV VLQJHUVRQJZULWHU %UXFH 6SULQJVWHHQ FRPSRVHG D SURSKHWLF WXQH HQWLWOHG f &KDQQHOV $QG

PAGE 92

1RWKLQf 2Qf ,W UHFRXQWV WKH WDOH RI D IUXVWUDWHG \RXQJ PDQ ZKR DIWHU HDJHUO\ HPEUDFLQJ WKHVH QHZ WHFKQRORJLHV ILQGV WKDW HYHQ ZLWK DQ DUUD\ RI SURJUDP FKDQQHOV WKHUH ZDV VWLOO QRWKLQJ RI YDOXH WR ZDWFK RQ KLV 79 )LQDOO\ KH UHWDOLDWHV E\ VKRRWLQJ KLV WHOHYLVLRQ VHW ZLWK KLV PDJQXP .HOOHU f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fV FODVVLF VRQJ ZLOO FRQWLQXH WR ULQJ WUXH :LWKRXW SURSHU PHGLD EUDQG PDQDJHPHQW DXGLHQFHV LQ WKH VW FHQWXU\ ZLOO ODPHQW DERXW FKDQQHOV DQG QRWKLQJ RQ

PAGE 93

$33(1',; $ 1,(/6(1 79 0(7(5(' 0$5.(7 +286(+2/' $1$/<6,6 0$< 16,3/86f 1,(/6(1 79 0(7(5(' 0$5.(7 +286(+2/' $1$/<6,6 25/$1'2 0$< 12 )25 1,(/6(1 0(',$ 5(6($5&+ $8',(1&( 6285&( $1' '(67,1$7,21 '0$ +286(+2/'6 6285&( 7,0( $9(5$*( 2) :.6 021 30 :.6 :(' 30 :.6 )5, 30 '(67,1$7,21 $9(5$*( 2) :.6 021 30 :.6 :(' 30 :.6 )5, 30 :.6 78( 30 :.6 7+8 30 :.6 78( 30 :.6 7+8 30 $9*6285&( $8',(1&( $9* '(67,1$7,21 $8',(1&( f 67$7,21 f :)79 :&3; :(6+ :.&) :2)/ :5%: 27+5 2)) 727$/ :)79 B :&3; :(6+ n :.&) :2)/ n :5%: 27+(5 781(,1 $9* 6285&( $8',(1&( $9* '(67,1$7,21 $8',(1&( bf 67$7,21 bf :)79 :&3; :(6+ :.&) :2)/ :5%: 27+5 2)) 727$/ :)79 B :&3; :(6+ :.&) :2)/ :5%: 27+(5 781( ,1

PAGE 94

16,3/86 1,(/6(1 79 0(7(5(' 0$5.(7 +286(+2/' $1$/<6,6 25/$1'2 0$< 12 )25 1,(/6(1 0(',$ 5(6($5&+ $8',(1&( 6285&( $1' '(67,1$7,21 '0$ +286(+2/'6 6285&( 7,0( $9(5$*( 2) :.6 021 30 :.6 78( :.6 :(' 30 :.6 7+8 :.6 )5, 30 '(67,1$7,21 $9(5$*( 2) :.6 021 30 :.6 :.6 :(' 30 :.6 :.6 )5, 30 78( 7+8 30 30 2230 2230 '(67,1$7,21 2) $8',(1&( $9(5$*( %$6,6 67$7,21 6+$5( :)79 :&3; :(6+ :.&) :2)/ :5%: &7+5 2)) :)79 :&3; • ,$n :(6+ 66 :.&) :2)/ :5%: 6285&( 2) $8',(1&( $9(5$*( %$6,6 67$7,21 :)79 :&3; :(6+ :.&) :2)/ n :5%: 727$/ 6+$5( :)79 :&3; :(6+ :.&) :2)/ n :5%: 27+(5 781(,1 &23<5,*+7 %< 1,(/6(1 0(',$ 5(6($5&+ ,1& $// $8',(1&( '$7$ &217$,1(' ,1 7+,6 $1$/<6,6 :(5( 2%7$,1(' )520 7+( 6$0( 0(7(5 5(&25'6 86(' 72 352'8&( 7+( 5(*8/$5 1,(/6(1 79 5(3257

PAGE 95

5()(5(1&(6 $DNHU $ f 0DQDJLQJ EUDQG HTXLW\ 1HZ
PAGE 96

%ODFN -XQH f %XLOGLQJ D EHWWHU EUDQG /$ %URDGFDVWLQJ DQG &DEOH S %RHPHU 0 / f &RUUHODWLRQ OHDGLQ VKRZ UDWLQJV ZLWK ORFDO WHOHYLVLRQ QHZV UDWLQJV -RXUQDO RI %URDGFDVWLQJ DQG (OHFWURQLF 0HGLD %ULQNOH\ 'HFHPEHU f 79 VWDWLRQV IHDU IRU WKHLU FKDQQHO EUDQG DV FKRLFHV SUROLIHUDWH 1HZ
PAGE 97

(DVWPDQ 6 7 f %URDGFDVWFDEOH SURJUDPPLQJ 6WUDWHJLHV DQG SUDFWLFHV %HOPRQW &$ :DGVZRUWK (QF\FORSHGLD RI VFLHQFH DQG WHFKQRORJ\ S S f f 1HZ %URFKXUH@ 1HZ %URFKXUH@ 1HZ
PAGE 98

.HOOHU / f &RQFHSWXDOL]LQJ PHDVXULQJ DQG PDQDJLQJ FXVWRPHU EDVHG EUDQG HTXLW\ -RXUQDO RI 0DUNHWLQJ .UDXV 6 f $WWLWXGHV DQG WKH SUHGLFWLRQ RI EHKDYLRU $ PHWDDQDO\VLV RI WKH HPSLULFDO OLWHUDWXUH 3HUVRQDOLW\ DQG 6RFLDO 3V\FKRORJLFDO %XOOHWLQ /HKPDQQ 5 t :LQHU 5 6 f $QDO\VLV IRU PDUNHWLQJ SODQQLQJ %XUU 5LGJH ,/ ,UZLQ 3XEOLVKLQJ /LQ & $ f 1HWZRUN SULPHWLPH SURJUDPPLQJ VWUDWHJLHV LQ WKH V -RXUQDO RI %URDGFDVWLQJ DQG (OHFWURQLF 0HGLD /LQWDV f /LQWDV OLQN EUDQG HTXLW\ DXGLW DQG SUREH ,Q 5XVVHOO t /DQH .OHSSQHUfV DGYHUWLVLQJ SURFHGXUH S f (QJOHZRRG 13UHQWLFH +DOO /LWWOHWRQ & 0DUFK f 1HZ GD\ IRU ORFDO QHZV GD\WLPH QHZVFDVWV DUH ERRVWLQJ EUDQG DZDUHQHVV DQG ERWWRP OLQHV %URDGFDVWLQJ DQG &DEOH S /LWWOHWRQ & -XQH f 0XUGRFK EXOOLVK RQ RYHUWKHDLU 79 %URDGFDVWLQJ DQG &DEOH S /RQJ 6 7 f 7KH GHYHORSPHQW RI WKH WHOHYLVLRQ QHWZRUN ROLJRSRO\ 1RUWK 6WUDWIRUG 1+ 1HZ $UQR 3UHVV 0£QGHVH 0D\ f +HUH FRPH WKH QHZHVW EUDQGV 1%& &%6 $%& $GYHUWLVLQJ $JH S 0£QGHVH 0DUFK f ,Q QHZV JURZWK SKDVH FDEOH IHHGLQJ RQ LWVHOI $GYHUWLVLQJ $JH S 6 0DUVKDOO 0DUNHWLQJ f 0DUVKDOO 0DUNHWLQJ DQG &RPPXQLFDWLRQV FOLHQW VDOHV EURFKXUH SHUVRQDO OHWWHU WR LQYHVWLJDWRU DQG SURSULHWDU\ GDWD IURP 2UODQGR )ORULGD PDUNHW VXUYH\ 2UODQGR )/ $XWKRU 0DWXUR 5 )HEUXDU\ f 0DLQWDLQLQJ EUDQG OR\DOW\ GXULQJ WLPHV RI PDUNHWLQJ FKDRV 3DSHU SUHVHQWHG DW $5) %UDQG (TXLW\ :RUNVKRS 1HZ
PAGE 99

0F&RQYLOOH E 'HFHPEHU f ,Q EUDQGLQJ LPDJH LV UHDOLW\ %URDGFDVWLQJ DQG &DEOH S 0F'RZHOO :6 $XJXVWf 7KH UDWLQJV JDPH 7KH HWKLFV RI K\SLQJ VWXQWLQJ EOXQWLQJ ULJJLQJ DQG RWKHU GLUW\ WULFNV 3DSHU SUHVHQWHG DW $(-0& &RQYHQWLRQ $GYHUWLVLQJ 'LYLVLRQ :DVKLQJWRQ '& 0LOHV / )HEUXDU\ f 6ZHHSV XQGHU VXVSLFLRQ 0HGLD :HHN S 0RUULV % 0DUFK f 7KH EUDQGfV WKH WKLQJ )RUWXQH S 0\HUV )HEUXDU\ f 7KH 0\HUV UHSRUW 0HGLD (OHFWURQLF 0HGLD SS 1DWLRQDO $VVRFLDWLRQ RI 7HOHYLVLRQ 3URJUDP ([HFXWLYHV f 7KH EUDQGLQJ EXJ 3DQHO GLVFXVVLRQf $QQXDO &RQYHQWLRQ DJHQGD 1HZ %URFKXUH@ 1HZ %URFKXUH@ f 1HZ
PAGE 100

3URPD[ t %'$ f 3URILWLQJ IURP \RXU SRZHUV RI FUHDWLYLW\ >&RQYHQWLRQ EURFKXUH@ /RV $QJHOHV 3URPRWLRQ DQG 0DUNHWLQJ 3URIHVVLRQDOV ,Q (OHFWURQLF 0HGLD DQG %URDGFDVW 'HVLJQHUV $VVRFLDWLRQ 5DWKEXQ ( -XQH f 0DUNHWHUV WDNH SURPD[ VSRWOLJKW %URDGFDVWLQJ DQG &DEOH S 5LHV $ t 7URXW f 3RVLWLRQLQJ 7KH EDWWOH IRU \RXU PLQG 1HZ
PAGE 101

+HGJH 7 t .VRELHFK f 7KH fOHDGLQf VWUDWHJ\ IRU SULPHWLPH 79 'RHV LW ,QFUHDVH WKH DXGLHQFH" -RXUQDO RI &RPPXQLFDWLRQ f 7LHGJH 7 t .VRELHFK f 7KH VDQGZLFK SURJUDPPLQJ VWUDWHJ\ $ FDVH RI DXGLHQFH IORZ -RXUQDOLVP 4XDUWHUO\ 7REHQNLQ 2FWREHU f 1HZ OLIH IRU ORFDO QHZV 1HZV VWRFN ULVHV ZLWK DIILOLDWLRQ VZLWFK RSSRUWXQLWLHV DQG LQFUHDVHG HPSKDVLV RQ EUDQG HTXLW\ %URDGFDVWLQJ DQG &DEOH S 7RWDO 5HVHDUFK &RUSRUDWLRQ f (TXLWUHQG >%URFKXUH@ 1HZ
PAGE 102

%,2*5$3+,&$/ 6.(7&+ $IWHU HDUQLQJ D PDVWHUnV GHJUHH IURP 6\UDFXVH 8QLYHUVLW\ LQ :DOW 0F'RZHOO EHJDQ D SURIHVVLRQDO EURDGFDVWLQJ FDUHHU WKDW ZRXOG VSDQ DOPRVW \HDUV LQFOXGLQJ RYHU D GR]HQ \HDUV DV 'LUHFWRU RI 0DUNHWLQJ DQG &UHDWLYH 6HUYLFHV $%& DIILOLDWH :)79 LQ 2UODQGR )ORULGD +LV FROOHJH WHDFKLQJ H[SHULHQFH GDWHV EDFN WR ZKHQ KH MRLQHG WKH IDFXOW\ RI WKH 8QLYHUVLW\ RI &HQWUDO )ORULGD DV DQ DGMXQFW LQVWUXFWRU :KLOH ZRUNLQJ KLV fGD\ MREf DW :)79 KH WDXJKW HYHQLQJ FODVVHV IRU WKH 8&) 6FKRRO RI &RPPXQLFDWLRQ ,Q DIWHU HDUQLQJ D VHFRQG PDVWHUnV GHJUHH WKLV WLPH IURP 8&)f KH HQWHUHG WKH 3K SURJUDP DW WKH 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH ZKHUH KH WDXJKW VHYHUDO WHOHFRPPXQLFDWLRQV FRXUVHV ,Q WKH IDOO RI KH MRLQHG WKH IDFXOW\ RI 6RXWKHUQ ,OOLQRLV 8QLYHUVLW\ DW &DUERQGDOH DV DQ DVVLVWDQW SURIHVVRU LQ WKH &ROOHJH RI 0DVV &RPPXQLFDWLRQ DQG 0HGLD $UWV

PAGE 103

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ & 0KQ & 6XWKHUODQG &KDLU 3URIHVVRU RI -RXUQDOLVP DQG &RPPXQLFDWLRQV FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ n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

PAGE 104

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KMLRVRSK\ %DUWRQ $ :HLW] -& 3HQQH\ (PLQHQW 6FKRODU RI 0DUNHWLQJ 7KLV GLVVHUWDWLRQ ZDV VXEPLWWHG WR WKH *UDGXDWH )DFXOW\ RI WKH &ROOHJH RI -RXUQDOLVP DQG &RPPXQLFDWLRQV DQG WR WKH *UDGXDWH 6FKRRO DQG ZDV DFFHSWHG DV SDUWLDO IXOILOOPHQW RI WKH UHTXLUHPHQWV IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ $XJXVW E 'HDQ &ROOHJH RI -RXUQDOLVP DQG &RPPXQLFDWLRQV 'HDQ *UDGXDWH 6FKRRO


CONCEPTUALIZING AND MEASURING THE EFFECTS OF BRAND EQUITY ON
TELEVISION PROGRAM RATINGS PERFORMANCE
By
Walter s. mcdowell
A DISSERTATION presented to the graduate school
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1998

ACKNOWLEDGMENTS
The author thanks Mr. Skip Skiffington, Program Director of Station A, and
Ms. Jenny Butler of Nielsen Media Research for providing proprietary overnight
Nielsen ratings data.

TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ii
LIST OF TABLES v
ABSTRACT vi
CHAPTERS
1 INTRODUCTION 1
2 REVIEW OF LITERATURE 4
Competition Changes Everything 4
Media Available versus Media Used 7
Rethinking Program Content 9
Birth of a Buzzword 10
Branding Local News 12
Summary 13
Brands and Brand Equity 14
The Emergence of Brand Equity 16
Product Brand and Product Categories 17
Conceptual Definitions of Brand Equity 17
Consumer Attitudes versus Consumer Behavior 20
Definitions From the Private Sector 21
Definitions From Academia 23
Inertia 27
Brand Equity According To Keller 28
Operation Definitions of Brand Equity 29
Inheritance Effects 33
Overview of Research 33
Viewer Motivation and Loyalty 35
Applying Brand Equity to Television Program Performance.. 37
Before the Buzz, There Was Audience Loyalty 37
Network versus Station versus Program Equity 38
Adapting Brand Terminology 39
Controlling the Marketing Mix 40
iii

Inertia, Momentum, and Equity 43
Interfacing Momentum With Keller’s
Conceptualizations 45
Equity and Performance 46
The Nielsens: Measuring Program Performance 47
3 RESEARCH HYPOTHESES 49
4 RESEARCH METHODOLOGY 51
Controlling The Marketing Mix 53
Establishing a Comparative Equity Benchmark 56
Benchmark #1: Nielsen 10-Year Audience Trend Analysis... 57
Benchmark #2: Marshall Marketing Loyalty Survey 58
5 RESULTS 61
Hypotheses One, Two, and Three 61
Hypotheses Four and Five 63
Hypotheses Six and Seven 68
6 DISCUSSION 71
Study Limitations 78
Conclusions 80
Suggestions for Future Research 82
Implications For the Field 84
APPENDIX: A Nielsen TV Metered Market Household Analysis
May 1996 86
REFERENCES 88
BIOGRAPHICAL SKETCH 95
iv

LIST OF TABLES
Table Page
1 Breakout of Nielsen Metered Overnight Database 55
2 Comparison of Sweep versus Nonsweep Cases: 1996
Descriptive Household Data 56
3 Comparison of Sweep versus Nonsweep Cases: Paired
Sample T-Test by Households 56
4 10-Year Monday through Friday Average Sweep Performance of
Three Newscasts 57
5 1996 Marshall Marketing and Communications Survey on
Exclusive (Loyal) Viewing of Newscasts 59
6 Comparative Newscast Performance Based on Nielsen Metered
Overnight Ratings 62
7 Differential Performance Between 10:45 PM Lead-in and
11:00 PM Newscasts 64
8 Share Index of Late Newscasts Compared to Lead-in
Programming 65
9 Household Correlational Analysis Between 10:45 PM Lead-in
and 11 :00 PM Newscasts 66
10 Comparison of Household Correlations Derived From Single
Day versus Seven- Day Moving Average Data 67
11 Regression Data Based on Overall Households 67
12 Nielsen Audience Flow Analysis: May Sweep 1996 70
v

Abstract of Dissertation Presented to the Graduate School of the University of
Florida in Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy
CONCEPTUALIZING AND MEASURING THE EFFECTS OF BRAND EQUITY
ON TELEVISION PROGRAM RATINGS PERFORMANCE
By
Walter S. McDowell
August 1998
Chairperson: Dr. John C. Sutherland
Major Department: Mass Communication
The purpose of this study was to explore the plausibility of applying
brand equity theory to electronic media, particularly television programs. While
the broadcasting and cable industries have embraced the jargon of brand
management, there has been scant published research on the specifics of
how conventional brand management theory can be applied to television
program strategies and practices. Additionally, there has been little
nonproprietary work on how to measure the influence of program brand equity
on program ratings performance.
When examining program ratings performance, the powerful influence of
lead-in programming (coined by media researchers as inheritance effects or
tuning inertia) cannot be ignored. Synthesizing elements from several
consumer-based brand equity theories, the investigator proposed that a
vi

program brand equity model focuses on the differential ratings response of a
program to its direct competitors and to its lead-in programming. After
establishing this framework, a number of hypotheses were tested within a
case study format. The daily ratings performance of three 11:00 PM local
newscasts in a major southeast television market were analyzed using one
station as an equity benchmark.
Acknowledging the lack of external validity and the need for further testing
across many markets and program genres, the results of this exploratory study
were encouraging. With ail hypotheses supported to some degree, the study
concludes with several recommendations for a program equity research
agenda for future work in the field.

CHAPTER 1
INTRODUCTION
In the spring of 1996, many media observers were startled to learn that
the newly appointed general manager of the prestigious NBC owned and
operated TV station in Los Angeles, KNBC, had never before worked at a
television station, broadcast network, or production studio. Instead, Carole
Lynn Black was hired from the management ranks of Proctor and Gamble.
NBC discounted her lack of direct broadcasting experience by emphasizing her
extraordinary talents in brand marketing. In an interview, Black (1996) stated,
“My experience at P&G taught me that branding is about finding the uniqueness
in two very similar products, presenting that positive uniqueness to the
consumer and driving that message home” (p. 93). This unprecedented
management decision to hire someone outside the core industry to direct the
operations of a network “flagship” station serves as just one example of how
the electronic media have become captivated with brand management theory.
In an effort to cope with unprecedented competition, audience fragmentation,
and declining market shares, broadcasters have looked to the retail consumer
goods industry for inspiration. The result has been the eager adoption of the
jargon, if not the substance, of brand management. The media trade press are
filled with cursory references to “brand identity,” “brand image,” and “brand
1

extension.” In particular, there has been a growing interest in that most
muddled of brand management concepts, “brand equity.”
2
Although there are many divergent conceptualizations of brand equity,
there is universal agreement that equity enhances a product’s performance in
the consumer marketplace. That is, equity helps reinforce consumer loyalty,
attract new customers, and insulate the product from competitive attack. Fierce
competition is the catalyst for most businesses to look beyond short-term
sales goals and focus on the more enduring advantages of consumer-based
brand equity. Encountering dozens, if not hundreds, of cable and satellite
program competitors, many television executives have embraced a
management perspective where networks, stations, or programs are regarded
as consumer brands. As with conventional consumer goods, the expectation is
that brand equity will enhance the performance characteristics of a program in
the audience marketplace. More specifically, brand equity theory is presumed
to help broadcasters develop and schedule programs that will deliver dominant
Nielsen ratings.
A goal of brand equity research has been to determine the proportion of
a brand’s market share that is most vulnerable to consumer brand switching.
Recognizing the vulnerability of a brand allows brand managers to make
informed strategic decisions. By ignoring the symptoms of poor equity,
managers run the risk of unforeseen declines in market shares and
disappointing bottom lines (Kapferer, 1992). Similarly, managers of the
electronic media wish also to diagnose any vulnerabilities in a program’s

3
audience performance. Although there is considerable professional and
scholarly research on brand equity dealing with conventional consumer goods,
there has been little work done on adapting these theoretical precepts to
electronic media.
Furthermore, any meaningful study of program ratings performance
cannot ignore the potent influence of lead-in programming or inheritance
effects. It is no secret among programmers that the best predictor of a
program’s audience size is often the size of the audience leading into it. While
there is considerable literature on this subject, no one has approached this
program performance issue from the vantage point of brand equity.
The purpose of this study is twofold. The first is to increase the body of
knowledge of brand equity theory by exploring its applicability to television
program ratings performance. The second is to offer an exploratory case study
that tests a proposed theoretical approach that synthesizes aspects of
traditional brand equity with Nielsen television ratings. Using diary-based
“sweeps” data and electronically metered “overnight” ratings, this study
attempts to measure the relative influence of program brand equity on the
performance of three late evening newscasts in a designated southeast
market. Particular attention is paid to the relationship between the newscasts
to their lead-in programs as a means of evaluating program brand equity.

CHAPTER 2
REVIEW OF LITERATURE
Branding has become the buzzword of the day. . . . More
is riding on the success or failure of broadcast branding
than at any time in the medium’s history.
(Stay Tuned, 1995, p. 58)
Competition Changes Everything
Branding and many of the notions of brand management are not new to
most American consumer goods. Some of today’s most prominent brands
such as Coke, Levi, Maxwell House, Budweizer, Campbell and Kellogg began
their branding efforts in the 1880s (Aaker, 1991). What is new is the adoption of
this paradigm by the electronic media.
As will be elaborated in later sections, the primary motivation for applying
brand management to a consumer product or service is competition. As the
number of similar products or services in the marketplace increases, the need
for highly differentiated brands becomes more acute. Also, with a rise in
competition, there is usually a similar rise in the speed and sophistication of
measuring brand performance in the marketplace. The introduction of
technological devices such as computerized UPC codes has enabled retailers
to track almost instantly consumer purchases for thousands of brands. While
attitudinal measures of equity may certainly be of value, measures of in-market
consumer behavior are the preferred means of evaluating brand performance.
4

5
As a marketing vice president at Proctor and Gamble recently proclaimed, “In
an information-rich environment, impatient retailers don’t hesitate to relinquish
brands that don’t deliver consistent and profitable sales. In order to survive in
an increasingly competitive marketplace, a high level of brand fitness is
required11 (Gleason, 1995, p. 37).
The same can be said for the brand fitness of a television program in an
ever-increasing competitive environment. The business of commercial
broadcasting is the selling of audiences to advertisers, and in an “information
rich” world of overnight Nielsen ratings, impatient program executives do not
hesitate to cancel programs that cannot deliver consistent and profitable
audience ratings.
The late arrival of brand management to the business of American
television was due primarily to a lack of competition. For over three decades,
the competitive arena for commercial television was restricted to three major
players. Through the interplay of various legal, economic, and technological
factors, a three-network program oligopoly dominated the television industry
from the 1950s to the mid 1980s. ABC, CBS, and NBC provided over 75% of the
daily programming for their local affiliated stations (Long, 1979). Syndicated
programming filled the remaining hours, but here again the viewer choices
were meager because of the few number of stations licensed to each
community.
By in the mid 1980s the competitive picture began to change when Fox
became a iegitimate network player, siphoning off young adult viewers from the

6
“Big Three” networks. The 80s also witnessed the coming of age for cable
television. Cable program networks such as CNN, ESPN, MTV, A&E,
Nickelodeon, and Discovery began to encroach on the audience territory that
was once the exclusive domain of the broadcast networks and their affiliated
stations. As the number of unique cable programming options increased, so
did the number of subscribers, and with more subscribers came more
audience fragmentation (Lin, 1995). From 1977 to 1997, ABC-, CBS-, and NBC-
affiliated stations combined lost over 40% of their viewing households to basic
cable and other alternative media (Myers, 1997). In the near future,
technological breakthroughs such as digital compression and fiber optics will
enable cable and satellite companies to increase channel capacity from a few
dozen channels to several hundred. With over 70% of the U.S. households
already “wired,” traditional over-the-air (terrestrial) broadcasters can no longer
count on captive audiences and guaranteed profits. Former NBC programming
executive and media entrepreneur, Brandon Tartikoff, claimed in a 1993
interview that with the advent of so many new program choices on the horizon,
America soon will witness a paradigm shift in power from program providers to
viewers. “Some say that television was once a medium controlled by tyranny.
Now, it’s going to be a medium controlled by democracy” (Barton, 1993, p. 16).
A commentary by respected economist Adam Theirer (1995) of the Fleritage
Foundation claimed the introduction of so much choice is “a signal to the world
that the old media empires are modern-day dinosaurs headed for extinction”
(p. 5).

7
Complementing the significant changes in media technology have been
equally important changes in government regulation of electronic media. The
highly publicized Telecommunications Bill of 1996 was a mandate to increase
market-driven competition by unshackling government restrictions on
telecommunications. Broadcasting, cable, satellite, and telephone delivery
systems have been given the legal green light to cross boundaries and
compete for consumers. Congressional advocates for the bill claim it will soon
unleash a torrent of competition heralding nothing less than “the dawn of a new
information age” (Stern, 1996, p. 8).
Media Available versus Media Used
Adding fuel to the fire of this emerging competitive marketplace is the
fact that while the amount of program choice continues to increase, there has
been no corresponding increase in the amount of time people spend watching
television. Nielsen Media Research reports that over the past 10 years (1986-
1996) the number of minutes per household dedicated to television viewing
has remained unchanged (Nielsen Audience, 1996). The disturbing
implications for television programmers is that while the number of slices in
the programming pie has increased, the overall size of the pie itself has not
increased. So far in the 1990s, more choice has not translated into more
viewing. A number of media observers have referred to this phenomenon as
media “cannibalism” (Burgi & Katz, 1997, p. 3), where program providers are
forced into “feeding” on the audiences of fellow program providers (Mándese,
1995, p. S2). Using brand marketing terminology, television viewing appears to

8
have evolved into a mature or zero sum market where the number of available
customers for a product category is stagnant. Therefore, as more brands enter
the marketplace, the only means of survival is to take customers (audiences)
away from competing brands (Kapferer, 1992; Lehmann & Winer, 1994).
A further frustration for television programmers is that despite the
historic technological and regulatory breakthroughs that have freed audiences
from the “tyranny” of limited choice, the typical viewer still prefers to deal with
only a handful of options. Several studies indicate that for an average American
household, there is a viewing threshold of approximately a dozen channels. For
example, Nielsen Media Research conducts an annual national survey of
television viewing characteristics of which “Channels Received vs. Channels
Viewed” is a special section The 1996 report revealed that while the number of
channels available in the home continued to grow, the number of channels
actually viewed had not grown appreciably. The study concluded that while the
average cable household can now receive 41.1 channels, only 10.6 channels
are actually viewed. When 70 or more channels are available, viewership
increases only slightly to 13 6 channels (Nielsen Audience, 1996). Several
researchers have adopted the concept of channel repertoire to describe this
limited array of channels from which audiences select programming (Webster
& Lichty, 1991). This channel repertoire is similar in definition to what
advertising researchers call an evoked set, where a consumer considers only
a portion of the brands available in a purchase situation. The mathematics of
this new programming marketplace are not lost on NBC president Bob Wright

9
when he states, “My biggest worry is our ability to continue to attract a
disproportionate-size audience for our broadcast stations versus other means
of gathering audiences” (B&C Hall of Fame, 1996, p. 51).
Rethinking Program Content
The rise in competition has led to a reexamination of program content
theory. For example, prime time programs for years were produced according
to the least objectionable program theory( LOP). The theory is based on the
core premise of moderate liking. Audience researchers discovered early on
that most people do not watch television alone. Therefore, individual preference
is often mediated by group dynamics where extreme opinions, either positive or
negative, are suppressed in favor of a group compromise. By definition, a
compromise is a movement towards the common center or “lowest common
denominator” (Eastman, 1997). This strategy might generate “repeat
customers” for a program but not necessarily highly committed fans. The LOP
theory works well when there are few competitors, but as viewer options
increase dramatically, programs designed for moderate liking can become a
liability. As brand managers of many consumer goods have been forced to
market their products to select consumer segments, so media programmers
are now obliged to target their program content to the specific needs of fairly
narrow, specialized audience segments (Cooper, 1996). As will be
demonstrated later, “moderate liking” is not the stuff of brand equity.

10
Birth of a Buzzword
One of the first occasions where the concept of media as brands was
presented in a public forum was a keynote speech made in 1993 at the
Advertising Research Foundation Annual Conference by David Bender,
President and CEO of Mediamark Research. He professed that media (both
print and electronic) are similar to consumer products and services and that
traditional brand marketing theory and procedures can be applied successfully
to media to generate more business. He stated that “media vehicles are one of
the few examples of brands which don’t, for the most part, conceive of
themselves as brands; as a result, they often are not able to take advantage of
the insight that accrues to those aware of thinking and research on brands”
(Bender, 1993, p. 2). Less than three months after Bender’s speech, a front¬
page headline in Advertising Age declared, “Here Come The Newest Brands:
NBC, CBS, ABC” (Mándese, 1993, p. 1).
Since then, media executives and media trade journals such as
Broadcasting and Cable, Electronic Media, Media Week, and Advertising Age
have become ever more comfortable with the phraseology of branding. For
example, organizations such as the National Association of Television
Program Executives (NATPE) and Promotion and Marketing Executives In
Electronic Media (Promax) have sponsored seminars addressing “The
Branding Bug” (NATPE, 1996),” Our Brand New World” (NATPE, 1997), “Brand
Building and Visual Design”, “Welcome to The Brand Revolution” and
“Branding Your Message To Win” (Promax & BDA, 1996). After winning the fall

11
premier season, NBC placed full-page ads in a number of publications
boasting that it was “America’s Leading Network Brand” (NBC, 1995). Media
Week dedicated a major article to “Network Brand Marketing Fever” (Dupree,
1996, p. 27). A trade journal ad for the off-network syndicated version of “Due
South” bragged that the program was “a proven brand with a loyal audience”
(Due South, 1997, p. 35). A newly published NATPE brochure mentions how
broadcasters must “tackle the challenges of branding your program or station”
(NATPE, 1997). Even the mainstream press appears to have adopted branding
jargon. A subheadline in a New York Times article states that “television
stations fear for their channel brand as choices proliferate in the digital age”
(Brinkley, 1997, p. C9).
Not to be outdone by their broadcast competitors, cable and satellite
programmers have jumped on the brand wagon. The annual convention of
Cable Television Administration and Marketing (CTAM) provided a panel
discussion on “Cable Image Branding” (McConville, 1995a, p. 34). The
Western Cable Show offered a seminar on “Building a Reputation: Creating
Brand Loyalty” (McConville, 1995b, p. 70). A programming executive at A&E
claims that high profile programs such “Biography” help “brand the Network
identity” (Walley, 1995, p. 30). The headline for an article about the FX channel
stated that “FX Will Build Its Brand In Originals” (Spring, 1997, p. 10). A
classified employment ad for The Weather Channel claimed it was seeking a
Creative Director who will “steward the brand” (Weather, 1997, p. 53). A review
of MTV’s “Real World” program congratulated the producers for “establishing

12
its own brand of show” (Sharkey, 1997, p. 17). Rather than focusing on
individual networks or programs, Jonathan Sims, vice president of research for
the Cable Advertising Bureau (CAB), indicts the entire medium of broadcast
television, proclaiming “a decline in broadcast equity” (Sims, 1997, p. 37).
In recent years, broadcasters have grappled with the question of
overlapping brand identities. Networks, stations, and individual programs
create a sometimes awkward branding hierarchy that has been described as
“two people in a donkey costume ... not always going in the same direction,
but both trying to get to the same place” (Rathbun, 1995, p. 12). Some program
producers, such as Barry Thurston of Columbia Tri-star, are critical of network
intimidation that coerces stations into “sacrificing their own identities at the
network’s alter” (Freeman & Stanley, 1996, p. 8).
Branding Local News
The notions of brand management and particularly brand equity are no
more prevalent than in discussions of local newscasts. Providing as much as
45% of a station’s overall sales revenue, the financial and public relations
benefits of a successful news operation can be enormous (Eastman, 1997;
Tobenkin, 1994), but there can also be a downside. News programming is
expensive. A station that ranks third in the market may never see any return on
its investment. Media observers such as Ron Alridge, Editorial Director of
Electronic Media, claim that “some folks are already saying that the third-placed
newscast is an endangered species” (Alridge, 1996, p. 6).

13
For many stations, local news has become the only enduring and
familiar program asset. At a recent Radio and Television News Directors
Association (RTNDA) convention, a journalist reported that “in an environment
where decades-old network affiliations can be broken overnight, local
newscasts remain the strongest and most stable form of brand equity”
(Tobenkin ,1994, p. 68).
Summary
From observing industry practices and reading media trade journals, it is
obvious that branding has indeed become “the buzzword of the day.” At the
heart of all the brand discourse is the realization by broadcasters that it is no
longer business as usual in the quest to win the hearts and minds of
audiences. Compared to the early 1980s, the number of available program
channels has tripled, but the amount of time spent in front of the television set
by a typical household has not changed significantly in 10 years. Furthermore,
as the amount of program choice has increased dramatically, the number of
channels actually viewed by a typical American household remains at about a
dozen. With the stakes so high, broadcasters have turned to the highly
competitive consumer goods industry for inspiration. The result has been an
almost evangelical acceptance of media as brands.
Today, there is no nonproprietary research available from industry or
academia on applying brand equity theory specifically to broadcast
programming. The upcoming sections of this literature review provide (a) an
overview of brand equity research derived primarily from a conventional

14
consumer goods perspective, (b) a review of research on audience inheritance
effects, and (c) a translation of brand equity theory to the needs of television
programming.
Brands and Brand Equity
In a cluttered marketplace, where consumer products are often more
similar than they are different, proper brand management increases the
probability of consumer brand choice. Brand management is the reason why
consumers buy Big Macs rather than hamburgers, Nikes rather than sneakers,
Harley Davidsons rather than motorcycles.
A brand is a name, term, sign, design, or a unifying combination of them
intended to identify and distinguish the product or service from its competitors.
Brand names communicate attributes and meaning that are designed to
enhance the value of a product beyond its functional value. The basic reason
for branding is to provide a symbol that facilitates rapid identification of the
product and its repurchase by customers (Belch & Belch, 1993; Cobb-Walgren,
Ruble & Donthu, 1995; Srivastava & Shocker, 1991).
Because consumers often lack the motivation, capacity, or opportunity to
process all product information to which they are exposed in a thoughtful or
deliberative manner, they opt for quick resolution techniques stored in memory
(Kardes, 1994). Strong brands assist in this heuristic process. Biel (1991)
offers the following insight:
On a very practical level consumers like brands because they package
meaning. They form a kind of shorthand that makes choice easier. They
let one escape from a feature-by-feature analysis of category

15
alternatives, and so, in a world where time is an ever-diminishing
commodity, brands make it easier to store evaluations, (p. 6)
In social psychology this same process contributes to the formation of
social stereotypes, where people use prior knowledge and feelings (attitudes)
about groups to make swift inferences about individual members (Beike &
Sherman, 1994). Stereotyping and branding are essentially two different
manifestations of the same phenomenon. Each serves to streamline cognitive
processing. In a highly competitive, time-sensitive environment, strong brands
have a better chance of being purchased than weak brands.
Strong brands also cultivate habits. Rosenstein and Grant (1997)
maintain that in repetitive decision-making situations, habits save time and
reduce the mental effort of decision making, thereby allowing us to maintain
complex behavior patterns without becoming overwhelmed by a huge cognitive
task load. Habitual purchasing also can be explained readily using behavior
learning theory and particularly operant conditioning. This approach assumes
that learning is an associative process where a specific response is
associated repeatedly with a specific stimulus. Over time this reinforcement
process strengthens the bond between stimulus and response. For example, if
the repeated outcomes resulting from the use of a branded product are
positive, the likelihood of that consumer buying that brand again is increased
(Belch & Belch, 1993).

16
The Emergence of Brand Equity
While brand management had been on the American business scene
for decades, the specific topic of brand equity did not become popular until the
volatile 1980s when once proud brands such as Sears, IBM, and Cadillac
began to lose ground to lower-priced generic competition. Market shares
dropped as a serious nationwide recession further aggravated corporate
profits. This historic economic downturn inspired a rash of company
consolidations and “downsizing” actions (Morris, 1996). Prudent cost cutting is
indeed one way to shore up sagging profit margins, but as Sears' CEO, Arthur
Martinez, warns, “You can’t shrink your way to greatness,” brand building is
essential for survival (Greenwald, 1996, p. 54). During many leveraged buyout
negotiations, corporate executives and Wall Street investors had to come to
terms with the portfolio value of brand names. The recognition of brands as
valuable intangible assets led to an increased interest in brand equity as a
topic for private and scholarly research (Aaker, 1991; Keller, 1993 ).
According to several business analysts, the initial success of less
expensive generic brands during the 1980s implied that consumers were
failing to acknowledge the supposed added value of a “name brand”
commodity and were influenced more by structural factors such as pricing and
shopping convenience (Morris, 1996; Reis & Trout, 1981). This added value
that immunizes a brand from such competitive incursions is often referred to as
its equity. Businesses coping with “mature" product categories that exhibit little
or no growth are particularly interested in brand equity. In such cases,

17
competitive pressures are intense as volume gains (share increases) are
derived from competitors rather than new category users. Under these “zero
sum” circumstances, as the number of competitors increase, the battle for
market share becomes more acute (Lehmann & Winer, 1994, p. 60).
Product Brands and Product Categories
The challenge of quantifying the value of a brand name has resulted in a
range of measures from consumer attitude surveys to more concrete
measures of in-market consumer behavior. However, at the core of all brand
equity theory is the distinction between a product category and a product brand.
A category refers to the generic commodity with no differentiation among
brands. Depending on a number of motivational factors, a consumer may
choose a category of product to satisfy a need with no consideration of brands--
any brand will do. In another purchase situation, the same consumer may
carefully evaluate the merits of a number of alternative brands-which is the
best brand for me? A third option is the consumer seeks a specific brand with
no consideration of category alternatives-only this brand will do. An important
assumption for any practical measure of brand equity is that the brand under
scrutiny must be a direct competitor coming from the same product category
(Kapferer, 1992; Keller, 1993). Mouthwashes should not be compared to
deodorants, nor should TV sitcoms be compared to newscasts.
Conceptual Definitions of Brand Equity
It is difficult to find agreement on the proper conceptualization of equity,
much less the relative weight of each element in driving brand choice. After a

18
particularly disappointing car selling season for General Motors, a frustrated
GM market analyst admitted that “even though brand equity may sometimes be
hard to define, it’s pretty clear when you've lost it” (Wilkie, 1996, p. 267).
Moving from an understanding of brands to a similar understanding of
brand equity can be perilous because so many other branding terms have
been used as synonyms for equity. While academics have yet to reach perfect
agreement on some of these items, media professionals and journalists have
been the worst offenders. The following is a brief clarification of some common
terms that, in many cases, serve as the conceptual building blocks of a larger
construct known as brand equity. These definitions are a synthesis of
definitions and conceptualizations found in several sources, including Hoyer
and Rrown( 1990), Belch and Belch (1993), and Keller (1993).
• Branding is the process of naming a product or service in order to
distinguish it from its category competitors.
• Brand extension is the process of branding a new product with an already
established brand name. (An example would be Ford creating a new type of
automobile or truck but still identifying it as a Ford product.)
• Brand awareness (or brand identity) is the first rung on the equity ladder. It
refers to the simple familiarity (recall or recognition) of a brand relative to its
product category.
• Brand image goes beyond mere awareness and deals with the thoughts
and feelings (meaning) of the brand to a consumer. It can be conceived as
a cluster of attributes and associations that consumers connect with a

19
specific branded product. Some equity researchers, such as Keller (1993),
compare brand associations in terms of favorability, strength, and
uniqueness.
• Brand positioning is the image of a brand defined in relationship to its
market competitors. The goal is to “position” a brand in a consumer’s mind
so that it has a highly distinctive (differentiated) image in comparison to
brands offered by competitors.
• Brand trial (or brand sampling) is the initial purchase or experience with a
brand by a consumer.
• Brand attitude can be viewed as an extension of brand image in that the
term refers not only to thoughts and feelings about the brand but
evaluations and most importantly, predispositions to respond (purchase). A
positive or negative attitude is reinforced by experiencing a brand first hand.
• Brand loyalty has often been viewed as another name for brand equity, but
for our purposes a much narrower definition has been adopted. Loyalty is
the degree to which a consumer purchases repeatedly a certain brand
without digressing to alternative brands. Within this context, brand loyalty is
a measure of consumer purchase behavior.
• Brand commitment addresses the underlying psychology of brand loyalty.
That is, commitment is the degree of fidelity to a brand that, regardless of
pricing differentials, promotions and other market factors, the consumer
remains loyal in a purchase situation.

20
• Brand equity in a general sense is defined in terms of marketing effects
uniquely attributable to the brand, that is, when certain outcomes result from
the marketing of a brand name that would not occur if the same product or
service did not have that name. Depending on the selected theoretical
approach, consumer-based brand equity combines some or all of the
above-mentioned brand concepts. A more detailed examination of brand
equity follows.
Before presenting a review of industry and academic conceptual
definitions of brand equity, it is important to remember that all of these
definitions can be divided roughly into either attitudinal or behavior categories. It
is well documented that attitudes alone are generally poor predictors of
marketplace behavior (Kraus, 1995). For years researchers have wrestled with
the disparities between the psychology of consumer preference and the
behavioral realities of consumer brand choice. Despite these incongruities,
brand equity has been defined typically from a consumer attitude perspective
because it is often less expensive than tracking and matching product sales
with individual consumers (Aaker, 1991; Churchill, 1995).
Consumer Attitudes versus Consumer Behavior
While few people would disagree that consumer-based brand equity is
ultimately a state of mind, many equity researchers assume that most
attitudinal or psychological components of equity (such as awareness,
imagery, positioning, and commitment) manifest themselves in observable
purchase behavior. This consumer behavior in turn translates into measures of

21
brand performance in the marketplace. In other words, certain indicators of
brand performance can be accepted as circumstantial evidence of consumer-
based brand equity.
Definitions From the Private Sector
The Marketing Science Institute (MSI) offers a kind of all-encompassing
definition in that brand equity is said to be
a set of associations and behaviors on the part of a brand’s consumers,
channel members and parent corporation that permits the brand to earn
greater volume and greater margins than it could without the brand
name and that gives the brand a strong, sustainable and differentiated
competitive advantage. (Anantachart, 1995, p. 12)
The institute does not further define what the specific “associations and
behaviors” are that constitute equity.
The private sector has been very active in promoting brand equity
research. While each company covets is own unique and patented formula,
they do reveal through their sales brochures and other publications some of
their conceptual definitions of consumer-based equity. The Interbrand company
has developed one of the best known commercial methods of calculating
brand equity from a purely behavioral perspective. The firm applies a multiplier
to the revenue generated by the brand. This multiplier is a function of six brand-
related factors that tap into product market positions and customer behavior.
These include leadership, stability, market trend, support, and protection
(Interbrand, 1994).
Other companies such as the Untas Advertising Agency combine
attitudinal and behavioral elements into their “equity audit.” Measures such as

22
brand awareness, image attribute ratings, and assumed leadership are
merged with market share data, sensitivity to pricing, and purchase
consistency over time. The overall purpose of the audit is to determine the
strong or weak bias consumers have relative to other brands (Lintas, 1993).
The NPD Group “Brand Builder” model also combines consumer perceptions
and actual in-market performance to arrive at an indicator called brand health.
Factors such as consumer awareness, product heritage, or reputation and
consumer loyalty contribute to the essential health of the branded product
(NPD, 1995). The Simmons Market Research Bureau’s EQX equity measure
tracks brand usage and then integrates more abstract notions including
attribute importance and desired improvements (EQX, 1995). Total Research
Corporation's Equitrend attempts to penetrate deep into some common
attitudinal measures. Rather than ascertain the mere familiarity of a brand,
these researchers delve into the “depth of awareness” and the “levels of
perceived quality” in association with key variables such as price elasticity and
brand loyalty. A byproduct of these calculations is a basic measure of user
satisfaction (Total Research Corporation, 1995).
Product Evaluations Inc. provides a Product Q measure of brand equity
that is highly attitudinal and is designed to calculate consumer enthusiasm-
“The absolute number of people who think your product is terrific!” Using a
nationwide survey technique, factors such as familiarity, experience, overall
appeal, and company image are used to determine the level of enthusiasm
(Product Q, 1995).

23
An upshot to the growing interest in brand equity in the 1980s was the
introduction of positioning theory, made popular by Reis and Trout in their 1981
best seller Positioning: The Battle For Your Mind. The authors contended that in
a overcommunicated society, a company must create a position in the
consumer’s mind that takes into consideration not only the company’s own
strength and weaknesses but also those of the competition as well (Reis &
Trout, 1981). They insisted that in a crowded marketplace, a branded product
must be differentiated easily from its competitors. Although this book was long
on conjecture and rather short on science, it remains must reading for aspiring
corporate executives. Although the term brand equity was not mentioned in the
original book per se, much literature on brand equity from both the private and
academic sectors has incorporated brand positioning.
Definitions From Academia
On a more scholarly level, a number of marketing professionals and
academics have published books or journal articles on brand equity. For
example, Axelrod (1993) takes a decidedly behavioral approach, emphasizing
pricing variables. He proposes a conceptual definition of brand equity as “the
incremental amount that a customer will pay to obtain a brand rather than a
functionally equivalent alternative with a different brand name. The more an
individual values a brand, the more he will pay for it” (p. 91).
Srivastava and Shocker (1991) offer two related perspectives on equity
by distinguishing brand value from brand strength. The first is a corporate
perspective where brand value is a financial measure dependent upon a

24
brand’s current monetary strengths and future prospects as well as its financial
status within a company’s total portfolio of products. On the other hand, brand
strength comes from a consumer perspective where a brand is evaluated
according to “competitive positioning and imagery relative to customer
demands and desires” (p. 6). Using this conceptualization, one could surmise
that consumer brand strength drives corporate brand value.
Alexander Biel (1991) posits that brand equity deals with the value,
usually defined in economic terms, of a brand beyond the physical assets with
its manufacture or provision with stronger brands having more equity than
weaker competitors. This strength ingredient can also be evaluated from a
consumer perspective where brand image factors such as salience, trust, and
richness drive brand equity.
Another way to address equity is to focus on consumer satisfaction.
There are multiple definitions of satisfaction in brand literature, but one that is
suitable for the purposes of this study is that satisfaction is an outcome of a
consumer’s direct experience (trial) with a product (Yi, 1990). A positive
experience will induce positive evaluations of the product and cultivate a
preference in a future purchase situation. More concisely, one can say that
satisfaction drives preference.
Satisfaction itself is not always a good predictor of consumer behavior. A
number of studies have shown that, within a product category, many brands
can be perceived as equally satisfying. Wilkie (1996) asserts that the good
news for American business is that hundreds of consumer surveys by

25
academics, governments agencies, and businesses reveal that the overall
consumer satisfaction level in the U.S. is extremely high. The resulting bad
news is that cultivating brand preference is more challenging than ever. Other
studies have demonstrated through controlled experiments that a brand can
have a positive satisfaction level with a consumer and still not be purchased
because competing brands are more accessible in memory due to heightened
exposure (Farquhar, 1989; Fazio, Powell, & Williams, 1989). This heightened
exposure is often the result of extensive brand advertising (Nedungadi, 1990).
Wilkie (1996) concludes that for many product categories “brands are likely
losing sales, not because they are not liked by consumers, but because they
never come to mind during the consumer’s (often rapid) decision process:
They are simply never considered!” (p. 515).
The above studies reinforce the proposition that simple brand
awareness can have a profound effect on brand choice without the addition of
complex images or associations. This of top-of-mind recognition (accessibility
to memory) is powerful particularly in low interest-low involvement purchase
situations. For example, Hoyer and Brown, (1990) proposed that in the interest
of conserving time and effort, consumers will often use simple choice
heuristics when making repeat purchase decisions. An experimental study
involving the purchase of ordinary, inexpensive consumer goods revealed that
when an inexperienced decision-maker is faced with choosing among several
competing brands, a known or recognized brand is more likely to be the

26
ultimate choice-regardless of the relative quality of the other competing
brands.
Today, brand managers continue to face the challenge of preventing
customers from defecting to competing brands. A recent study by Grey
advertising concluded that "‘brand promiscuity’ is a global phenomenon. That
is, consumers from the U.S., to the U.K., to Australia routinely makes choices
from a group of brands they consider acceptable" (Shell, 1997, p. 40). If all
brands within a product category are seen as equally satisfying, then
consumers become vulnerable to competitive strategies such as pricing,
promotions, and convenience, none of which is an ingredient of consumer-
based brand equity.
To many marketing researchers, the opposite of "brand promiscuity" is
brand loyalty. Christiani (1993) of General Foods USA confesses that one of the
most “elusive and debated” subjects surrounding brand equity is the
“intangible asset consisting of a brand’s consumer loyalty base” (p. 123).
Aaker (1991) includes brand loyalty as the centerpiece of his definition of
brand equity which consists of five broad categories: brand loyalty, brand
awareness, perceived quality, brand associations, and other proprietary assets
such as patents. All of these brand characteristics influence the consumer’s
purchase decision and usage satisfaction, but brand loyalty is the force that
makes a brand truly powerful. He believes that customer brand loyalty is
essential because it reflects the likelihood (or probability) that a customer will
switch to another brand. This notion of loyalty surfaces often in research

27
literature. Maturo (1993), a marketing director for Nielsen Household Services
(not to be confused with Nielsen Media Research), asserts that dozens of
recently performed analyses of brand choice modeling revealed that brand
loyalty was a consistent leading predictor of future brand purchases. Other
respected private research organizations such as the NPD Group and Lintas
advocate consumer loyalty as a core component of any brand equity model
(Lintas, 1993; NPD, 1995).
Inertia
There is an obvious intuitive acceptance that loyalty must somehow be
an integral part of equity, but several researchers have warned that loyalty
should not be misconstrued with mere repeat purchasing. According to Donius
(1994), information concerning simple habitual repeat purchasing may
obscure a vulnerability to switch brands in the future. He states that “loyalty can
also be inertia, not necessarily preference” (p. 56). The presumption is that in
some purchase situations there are noncommitted consumers that can be
motivated to choose a brand by simply controlling key structural factors that
have iittle to do with a brand’s perceived value. According to Ceurvorst (1994)
and Beatty, Kahle, and Homer (1988), the underlying cause of brand equity is
the core psychological relationship of commitment of which loyalty is the
behavioral result. Ceurvorst (1994) states that commitment provides “the
essential basis for distinguishing true loyalty from other forms of repeat
purchasing” driven by factors such as inventory, promotions, and discounts (p.
2).

28
Although not a full-fledged marketing theory, the concept of inertia is
referred to in several marketing and mass communication publications. Long
ago, broadcasters realized that the reason people watch a program is often
dependent simply on its lead-in program. This inheritance effect is sometimes
referred to as tuning inertia. Its definitional roots are in the natural sciences. A
typical definition of inertia borrowed from the domain of physics is the following:
Inertia is the property of matter which manifests itself as a resistance to
change in the motion of a body. Thus when no external force is acting, a
body at rest, remains at rest and a body in motion continues moving in a
straight line. (Encyclopedia of Science and Technology, 1997)
From a perceptive of consumer behavior, we can define our consumer
as the “body” and the forces of inertia as those marketing factors that
encourage “resistance to change” or habit. In order to have the body change
direction there must be an adequate “external force” which can be defined as a
competing brand. In a later section we elaborate more on this concept
introducing momentum as a measure of inertia.
Brand Equity According to Keller
Our final conceptual definition of brand equity, and the one chosen to be
the theoretical underpinning for our exploration of television program equity,
comes from Keller (1993) who conceptualizes brand equity according to two
kinds of memory associations, brand awareness and brand image. The two in
combination are called brand knowledge. Brand equity is the differential effect
of brand knowledge on consumer response to the marketing of a brand.
Awareness is the first step in the process where measures such as recall and

29
recognition are introduced, image deals with the meaning of a brand to a
consumer through various associations such as attributes, benefits, and
attitudes. These associations can vary according to their favorability, strength,
and uniqueness and can play critical roles in determining purchase outcomes.
Keller states, “The presence of strongly held, favorably evaluated associations
that are unique to the brand and imply superiority over other brands is critical to
a brand’s success” (p. 6). Keller also asserts that “fundamentally high levels of
brand knowledge (awareness and image) increase the probability of choice, as
well as produce greater consumer loyalty and decrease vulnerability to
competitive marketing actions” (p. 3).
From a more behavioral vantage point, Keller describes how these
brand associations can be reflected in the marketplace behavior.
A brand is said to have positive (negative) consumer-based brand equity
if consumers react more (or less) favorably to the marketing mix of the
brand than they do to the same marketing mix element when it is
attributed to a fictitiously named or unnamed version of the product or
service...If a brand is seen by customers to be the same as a
prototypical version of the product or service in the category, their
response should not differ from their response to a hypothetical product
or service. If the brand has some salient, unique associations, these
responses should differ, (p. 4)
In other words, if the marketing mix is controlled by the researcher, the
attitudinal component of brand equity, namely, brand knowledge, will manifest
itself in consumer behavior.
Operational Definitions of Brand Equity
The specific measuring techniques for brand equity depend on the
particular theoretical approach. One broad approach takes a corporate

30
perspective and dwells on the financial value of equity, examining items such
as stock prices, momentum accounting, and cash expenditures needed to
introduce brand extensions. Another broad approach and the direction of this
study is consumer-based brand equity, where consumer attitudes
(perceptions) and behaviors are the focus of study. Within this second
approach there is little agreement among researchers as to the ideal
operationalizations of equity. Also, there are scant academic (nonproprietary)
consumer-based equity studies that have actually attempted to measure the
influence of equity using real world products and consumers.
Interviews or self-administered questionnaires are typically used to
obtain attitudinal information. For example, Cobb-Walgreen, Ruble and Donthu
(1995) were interested in the power of brand equity to influence brand
preference and purchase intent. Hotels and household cleansers were
selected as brand categories for study. Using the perceptual components of
Aaker’s (1991) conceptualization, the researchers employed survey
instruments that attempted to capture brand awareness and brand
associations. Respondents were asked to list as many brands of a specific
product as they could to determine top-of-mind awareness. For brand
associations, the respondents were asked to list descriptive thoughts, words,
characters, symbols, or images that came to mind for each brand. To assess
preference and purchase intent, the respondents were presented with
hypothetical decision-making situations in the form of a story. The importance
of product attributes, favored brands, and intention to purchase were

31
operationalized through 7-point Likert scales. Another variable, advertising
awareness, was obtained by asking respondents if they had ever seen
advertising for a target brand and if could they recall any salient copy points.
Results indicated that higher advertising budgets yielded higher equity scores,
and, in turn, higher equity generated significantly greater preferences and
purchase intentions.
Testing an equity model that incorporated components of consumer
preference and consumer satisfaction, Anantachart (1995) looked at the
relative consumer equity of various soft drinks among college students. A
number of established 5-point scales were borrowed from other academic
research and were modified to work with this model. Additionally, new scales
were created to measure items such as product consumption. As a model for
detecting differential response among brands, the results were highly
predictive.
Keller (1993) operationalized equity from two vantage points. He sees
attitudinal or perceptual measures of equity (brand knowledge) as “indirect”
operationalizations. Conversely, measures of consumer behavior (or the
impact of brand knowledge) are viewed as “direct” operationalizations. For
example, indirect measures would include instruments that capture brand
name recall and various image dimensions such as favorability, strength, and
uniqueness of brand, associations. Direct measures deal with actual
consumer response (choice) to marketing activities. Keller recommends
various experimental and quasi-experimental settings where the marketing mix

32
(product, price, distribution, and promotion) of brands can be controlled
systematically to assess the relative impact of brands on market performance.
The equity model proposed in this dissertation adopts this direct approach of
analyzing consumer response. Although a pure experimental setting was not
feasible, the methodology did impose appropriate controls on the analysis of
secondary data.
We have seen that there is no universally accepted detailed
conceptualization of brand equity, However, it seems safe to assume that most
equity researchers would agree on the outcomes of having high consumer-
based brand equity. The market performance of a brand with strong equity is
characterized by the following:
1. Long-term market share stability and growth. While weekly or monthly
fluctuations can occur, the performance of these brands over the long run,
namely, years, is unwavering.
2. Dominant market rankings. Brands attributed with high equity not only
demonstrate stability over time, but they tend to be long-term market leaders,
achieving high market ranking among direct competitors. This status is
reflected in increased revenue, lower costs, and greater profits.
Behind this exceptional brand performance, equity researchers would
also agree that there is a consumer base that holds the brand in such high
esteem that they buy the brand exclusively, regardless of competitive
marketing actions.

33
Inheritance Effects
For decades, the single best predictor of a program’s ratings
performance has been the ratings of the program scheduled immediately
before it. That is, television programs tend to “inherit” sizable audiences from
the program airing immediately prior to it on the same channel. This carryover
phenomenon is known as inheritance effects or tuning inertia.
While it would be foolish for programmers not to take advantage of this
scheduling tactic, Webster and Lichty (1991) warn that some people
assume the choice of a program centers upon the active expression of a
preference for a program or type of program. However, so called
structural factors have traditionally been considered important mediators
of the programs viewers choose and complicate the relationship
between viewing preference and viewing behavior, (p. 178)
Overview of Research
While the overall impact of inheritance effects has been confirmed
myriad times by industry and academic researchers, the results have been
puzzling because of the considerable variances among studies. Beginning in
1975, Goddhart, Ehrenberg, and Collins essentially coined the term when they
worked on the broader issue of audience duplication. Based on television
viewing in the United Kingdom, the researchers proposed a Duplication of
Viewing Law which stated the proportion of the audience of any program who
watch another program on another day of the week is directly proportional to the
rating of the later program, times a constant. Furthermore, they discovered that
when programs were adjacent to each other, an “inheritance effect” took over
that exceeded the predictions derived from the original duplication law.

34
Headen, Klompmaker, and Rust (1979) proposed an improved model
incorporating five independent variables-ratings, channel, program type,
daypart, and repeat viewing. An examination of over 4,000 combinations of
pairs of programming using Simmons Market Research data revealed that by
far ratings were the single best predictor. A different model offered by Webster
(1985) included the contributing factors of audience availability (which in most
cases is a fixed quantity), lead-in program ratings, the number of program
options, and program content. Using Arbitron ratings from one sweep period in
Portland, Maine, Webster concluded that for adjacent program pairs, lead-in
ratings and the number of program options in combination explained 80% of
the variance.
Tiedge and Ksobiech (1986) conducted a massive 22-year study of
network prime time programming from 1963-1985 looking at the effects of
lead-in program shares, available program options, and similarity of program
types on inheritance effects. The conclusions were that programs with high-
ranked lead-ins scored an average of 6.8 share points higher than those with
low ranked lead-ins. Also, fewer program options produced higher lead-in
correlations and visa versa. In 1988, the same research team using the same
ratings data set investigated the program strategy of “sandwiching”
(hammocking) where the influence of both lead-in and lead-out programming
were analyzed. Their conclusions were that the effects of lead-out were
minimal and that a strong lead-in was essential to have a sandwich strategy
work properly (Tiedge & Ksobiech, 1988).

35
Walker (1988) looked at nine years of Nielsen ratings from 1976 to
1985. The correlational relationships among inheritance effects, lead-in,
program type, and number of options followed identical patterns observed by
Tiedge and Ksobiech (1986). Examining total audience exposure among
syndicated programs rather than network shows, Cooper (1993) correlated the
influence of several variables on program ratings. These included lead-in,
lead-out, number of options, program type compatibility, network affiliation, and
cable penetration. The results from a 50-market sample of May sweep books
revealed that while some variables such as network affiliation had minor
measurable influence, lead-in ratings were by far the strongest predictors of a
syndicated program’s ratings and appeared to be “strong surrogates for other
variables in the model . . . and completely overwhelmed any other factor in the
model” (p. 409).
Viewer Motivation and Loyalty
The variances in results found among inheritance effects studies have
not been explained adequately using structural factors such as program type,
number of options, channel, and network affiliation. Part of the problem may
involve the viewing motivation of the audience member. For example, Rubin
(1984, 1994) divides television program viewing motivations into ritualized and
instrumental categories. The first denotes a viewer that uses television out of
habit and to pass the time (relieve boredom, entertainment, amusement, etc.).
These “escapist” audiences place more emphasis on the medium itself rather
than specific program content. The second type of motivation (instrumental), is

36
more goal-oriented where the viewer is seeking program content to gratify a
specific need and is, therefore, more discriminating. From an inheritance
effects perspective one could hypothesize that the ritualized viewer would tend
to remain on the same channel while an instrumental viewer might hold certain
program loyalties that would encourage more channel switching.
The notion of audience loyalty was broached by Boemer (1987) when he
took a 2-year look at local late night newscasts in Dallas, Texas. Using Arbitron
sweep ratings, he found high positive correlations between newscasts and
prime time lead-in programming. However, he also found considerable
variances among the three competing stations ranging from .30 to .69.
Additionally, he found that some newscasts delivered better ratings than their
lead-ins, providing “circumstantial evidence of a loyal viewership for local news
in the market” (p. 93).
Inheritance effects (tuning inertia) can play a crucial role in determining
the ratings performance of a television program. While the impact is noticeable,
there is considerable variance among programs as to how effectively they
“inherit” audiences. That is, some programs appear to capitalize on a lead-in
audience better than other programs. Conversely, some programs appear
more resilient than others to the effects of a poor lead-in. An explanation may
come from the ranks of brand equity theory.

37
Applying Brand Equity to Television Program Performance
Before the “Buzz” There Was Audience Loyalty
The desire to nurture favorable, strong, and unique program
associations has always been a concern of broadcasters. Before the buzz
surrounding media as brands, television managers tried to gauge audience
loyalty. For many years, major market television stations have enlisted the
services of consumer research companies to conduct attitudinal and
behavioral surveys. While Nielsen rating points remain the ultimate measure of
program performance, broadcasters also scrutinize other research data in
order to make better programming and promotion decisions.
An example of one such company that provides proprietary information is
Marshall Marketing and Communications (MM&C), a nationally recognized
broadcast marketing and communications consulting company based in
Pittsburgh, Pennsylvania. Approximately 100 television stations throughout the
country subscribe currently to the service. On a market-specific basis, MM&C
provides annual surveys of consumer purchasing habits and attitudes for
preselected categories in retail, service, and merchandise industries.
Additionaily, the company provides analysis of consumer communication
habits encompassing, radio, television, newspaper, cable, and direct mail.
Within the television viewing domain, MM&C targets specific programs,
including local newscasts (Marshall Marketing, 1996).
Marshall Marketing and Communications conducts several hundred in-
depth telephone interviews over a three- to five- week time span. One specific

38
research area is consumer loyalty to specific brands. Loyalty is defined
conceptually as recurring and exclusive purchasing. That is, people who are
“heavy users” of a product category and purchase only one brand over a
considerable time period are considered loyalists. The operational specifics
vary depending on the type of product and typical buying cycles. For example,
some products tend to be a daily or weekly purchase, while others tend to be
more of a monthly purchase. Daily versus weekly scheduled television
programs can be regarded in the same light. Although MM&C has yet to adopt
branding jargon, this measure of consumer loyalty is consistent in many
respects to Keller’s (1993) conceptualizations and can serve as a comparative
benchmark for our case study examining ratings performance.
In the final analysis, the media’s infatuation with the buzz of brand
management is spawned by the proposition that brand concepts such as
brand equity will help protect and enhance a program’s ratings performance.
While the jargon of brand management is prevalent thoughout the broad¬
casting and cable industries, there has yet to be a systematic effort to translate
properly many of these precepts for use by the media. Borrowing from the work
of Keller (1993) and proponents of consumer inertia, this section examines the
plausibility of applying brand equity theory to television programming.
Network versus Station versus Program Equity
While brand equity theory could be applied to several branding units, the
latest academic literature indicates that, for the most part, people still watch
programs before networks, stations, or channels. A study by Abelman, Atkin,

39
and Rand (1997) revealed that the majority of viewers “are relatively
indiscriminate consumers with regard to the source of programming. They
possess little network or station affinity” (p. 377). A small group within the study
did exhibit station affinity, but an important caveat was that the station offered
preferred programming content. For example, station affinity was attributed to a
preference for local newscasts, not necessarily to the station itself. While the
assessment of brand equity for stations and networks is a worthwhile topic for
future research, this paper concentrates on brand equity issues involving the
most basic equity unit, the individual program.
Adapting Brand Terminology
By conceptualizing a program as a product or service and the program’s
audiences as consumers, most branding terminology can be adapted readily to
broadcast programming. On a fairly cursory level, the following concepts have
been redefined for broadcasting.
Brand awareness. The recall or recognition of a network, station, or
program name
Brand extension. The imposition of a brand name over several program
“products” such as a network ID.
Brand trial. The first experience a viewer has with a program (sampling)
Brand image. The thoughts or feelings (meanings) that come to mind
when the program name is mentioned.
Brand positioning. The image of the brand defined in relationship to its
direct market competitors.

40
Brand attitude. The evaluation of a program (brand images) and
predisposition to watch.
Brand loyalty. The degree of repeat viewing of a program at the exclusion
of direct competitors.
Brand commitment. The psychology underlying program loyalty. The
degree of fidelity to a program despite competitive program scheduling,
promotions, lead-ins, and so forth.
There are other useful brand marketing parallels between conventional
consumer goods and broadcasting. For example, there are obvious similarities
between a viewer’s limited channel repertoire and a consumer’s evoked set of
brand selection. Also, the precepts of consumer habit formation through simple
operatant conditioning can be seen in how broadcasters reinforce positive
viewer outcomes (or satisfaction) by manipulating program content and
scheduling.
Controlling the Marketing Mix
Keller (1993) insists that brand equity cannot be assessed properly
unless the marketing mixes of the competing brand competitors are equivalent
or at least controlled by the researcher. The marketing mix components of
product, price, place (distribution), and promotion can be translated into a
broadcasting domain in the following manner.
The product
The product is the specific brand of program content under scrutiny. We
mentioned earlier that in order to understand brand equity, it is necessary to

41
deal with direct competitors, that is, products or services that reside in the
same product category. Just as there can be no meaningful comparison of
brand equity between packaged goods, such as toothpaste and pantyhose, so
we should not attempt equity comparisons between disparate television
genres, such as newscasts and soap operas.
The price
Unlike most consumer goods, pricing is not a paramount concern for
broadcasters. Even when broadcast stations are included on subscription
cable services, there is a general perception that the broadcast channels are
“free." However, basic and premium cable programmers do encounter pricing
dilemmas with their paying consumers. For the purpose of this study, we are
excluding the pricing component.
The place (distribution)
For electronic media, controlling the distribution component of the
marketing mix is challenging but essential to understanding true equity. In an
ideal test environment, a program’s direct competitors should be equally
available to all consumers. From a retail perspective, this would seldom be a
major obstacle, but for broadcasters there is an obvious problem. Unlike most
consumer goods, broadcast “products” are time-bound and are not available
simultaneously to audiences. (Imagine a retail outlet where the shelf displays
changed every half hour to display different competing brands.) Attitudinal
studies can ignore this dilemma, but a behavioral approach to assessing

42
consumer-based brand equity requires that not only program content be
similar but also program scheduling as well.
Program brand distribution (availability) also includes technological
factors. Direct competitors may be scheduled at the same point in time but the
electronic delivery systems may not be equivalent, thereby biasing a test
situation. Variables, such as geographical signal coverage and cable
retransmissions, can influence the distribution of a television program.
A final distribution factor is consumer traffic flow. Just as some
consumer brands in a retail setting have better store locations or better in-store
shelf positioning, so certain television programs are exposed to better
audience “traffic” than others. Audiences already present in a lead-in program
on the same channel can be construed as more available than audiences
found on other channels. From our earlier review of inheritance effects
research, we know that the mere proximity of a large lead-in audience can offer
a distinct advantage.
The promotion
Another component of the traditional marketing mix is promotion, which
includes communication actions such as advertising, publicity, and, most
importantly, sales promotions. Coupons, sales, contests, and other “tactical
gimmickry” are intended to stimulate product trial (sampling) and boost
temporarily market shares but seldom cultivate long-term brand loyalties
(Keller, 1993, p. 6).

43
In the quest for short-term ratings gains during the infamous Nielsen
“sweep weeks,” television networks and stations resort to a variety of
programming and promotional tactics that are designed often to distort typical
program performance. These tactics range from altering program schedules
and manipulating story lines to atypical advertising campaigns and on-air
contests. Known by industry insiders as hyping, or hypoing, these practices
have been condemned by Nielsen and the advertising community (Gimein,
1996; McDowell, 1995; Miles, 1997). The best means for neutralizing the
biasing effects of broadcast promotion is to look at long-term performance
where transitory promotion tactics become random variables.
In summary, the marketing mix for electronic media must somehow be
held in check in order to evaluate program equity. Therefore, in an ideal
research setting, the following criteria should be applied: (a) identical or highly
similar target program content (derived from same “product category”), (b)
identical pricing considerations (same price or free), (c) identical distribution of
target programs in terms of scheduling, signal coverage, and audience flow,
and (d) identical or neutralized short-term promotional tactics of competing
target programs.
Inertia, Momentum, and Equity
The well-documented influence of lead-in program ratings implies that
there are significant numbers of passive or uncommitted viewers who are not
motivated to change channels. The magnitude of this inertia or “resistance to
change,” however, has been difficult to predict in terms of rating points. Given

44
the same lead-in rating points, some programs appear to “inherit” more points
than others. For example, a program with a big lead-in rating may be unable to
capitalize fully on this advantage, while another program dealing with a much
smaller lead-in can overcome this apparent disadvantage and exhibit
impressive ratings performance.
Using Keller’s (1993) conceptualizations of brand equity we can offer a
partial explanation for these inconsistencies. A program that is unable to
generate familiar, strong, and unique brand associations among its potential
viewers is vulnerable to the program content on competing channels. Likewise,
a program demonstrating strong equity characteristics will retain more of its
lead-in audience and recruit audiences from other channels.
Earlier we demonstrated how research in both brand equity and
inheritance effects have employed the concept of inertia to explain a
consumer’s resistance to change behavior patterns. For television, this
reluctance to change channels has been "coined tuning inertia." However,
inertia alone may not be as worthy a metaphor as another term borrowed from
physics, momentum. Intended as a measure of “how much” inertia is present
in a body, momentum consists of two components, mass and velocity
(Encyclopedia of Science and Technology, 1997). Based on this simple
formula, we can understand better how two bodies with the identical mass can
exhibit different degrees of momentum, depending on their relative speed.
Similarly, measures of mass alone cannot predict how susceptible a body is to

45
“external forces.” Only when the mass and the velocity are measured in tandem
can we assess properly the power of inertia on a body.
From a consumer behavior perspective, we can surmise that inertia is
not necessarily a bad thing. Rather, it is the underlying consumer momentum
factors that must be evaluated: (a) the brand’s ability to retain current
customers in the face of external forces of direct competitors and (b) the ability
of a brand to recruit customers from competing brands. The repeat customers
of a brand exhibiting poor equity are vulnerable to a competitor’s marketing mix
and, therefore, have less momentum than a brand with stronger equity.
In a similar vein, one could propose that tuning inertia (inheritance
effects) is influenced by the two “momentum” factors of (a) retention of lead-in
audiences and (b) recruitment of audiences from other sources. This approach
explains how two programs with identical lead-in ratings can achieve different
results. Equity, then, can be viewed as a program’s ratings momentum.
Interfacing Momentum with Keller’s Conceptualizations
The above notions of inertia and momentum interface well with Keller’s
(1993) approach to brand equity. That is, if a consumer does not perceive any
“salient and unique associations” from a brand name, there will be a
resistance to change purchase behavior. This momentum can be disrupted by
the external force of a competing brand whose marketing mix generates
favorable, strong, and unique associations that “imply superiority over other
brands.” Similarly, a television audience member that is currently viewing one
particular channel will remain on that channel unless the “external forces” from

46
a competing channel intervene sufficiently to motivate the viewer to change
channels.
Equity and Performance
We will recall that for Keller (1993) brand equity is the differential effect of
brand knowledge on consumer response in a competitive marketplace. This
brand knowledge includes (a) awareness and (b) image associations. The
more favorable, strong, and unique these associations, the more likely a
particular brand will be chosen over its direct competitors. This consumer
response (choice behavior) can be translated into measures of brand
performance in the marketplace. An essential caveat to assessing this
performance is the need to control the marketing mix components of direct
competitors.
Given this theoretical framework, television programs with strong
consumer-based brand equity should perform better in the marketplace than
programs with weak equity. That is, programs which elicit from viewers
favorable, strong, and unique associations will have a greater probability of
being chosen over direct competitors. Again, control of the marketing mix
elements is crucial in evaluating performance. Operationalizing program
performance logically should involve measures of audience program choice,
and for the past four decades within the broadcasting industry, the most
recognized measure has been program “ratings” provided by Nielsen Media
Research.

47
The Nielsens: Measuring Program Performance
By adopting a behavioral approach to understanding program brand
equity and insisting that this work be relevant to the real-world business
environment of the broadcasting industry, Nielsen ratings were selected as the
most appropriate measure of program performance. The word rating is often
misunderstood outside the broadcast industry because the more vernacular
use of the term means a subjective judgment or opinion. That is, people are
asked often to “rate” their favorite movies, restaurants, or rock groups.
Broadcast and cable ratings, however, are simply estimates of audience tuning
behavior, or media exposure. Ratings are not intended to be a valid attitudinal
measure except in the sense that consumer behavior can be inferred to be
what Webster and Lichty (1991) and other researchers call “revealed
preference” (p. 27). The following is a brief synopsis of Nielsen research
methods.
In addition to the well known national “People Meter” ratings, Nielsen
also provides local program ratings to over 200 individual markets. Using a
multistage clustering sampling technique, all markets experience the month¬
long, diary-driven “sweeps” four times a year (November, February, May, and
July). An alternative to the sweeps for many major market stations is metered
“overnight” ratings. Presently, 42 local markets are measured daily, using a
passive electronic metering system that continuously records household
viewing. The viewing data are processed “overnight” and then faxed the
following morning to subscribing stations. From a reliability perspective, one of

48
the major criticisms of the sweeps methodology is that it covers only 16 weeks
of the year and that these few weeks often are dominated by programming and
promotional hype that may distort results. Furthermore, program performance
data by individual days are not provided in the standard reports. Instead, ratings
are presented as four-week averages. Consequently, whenever possible, this
study utilized the more reliable overnight ratings which are not averaged and
offer continuous, long-run data.
The fundamental units of measure used in almost all ratings-based
research are households (000), rating points, and shares points. While
households are expressed as whole numbers, rating and share points are
percentage points. A rating expresses a program’s audience as a percentage
of the total population. A share expresses the same audience but as a
percentage of the households (or persons) using television (HUTs or PUTs).
Age and gender demographics can also be translated into rating and share
points (Nielsen Method, 1995).

CHAPTER 3
RESEARCH HYPOTHESES
From the basic research question, how does brand equity influence
program ratings performance, a number of exploratory hypotheses can be
offered that resonate with our proposed theoretical assumptions. The literature
review contends that brands exhibiting strong equity tend to be market leaders
and that, over time, short-term fluctuations in performance give way to a long¬
term durability in market rank. In other words, market leadership endures;
therefore, the following hypotheses were proposed.
H1: Programs with higher brand equity will exhibit higher broadcast
ratings than programs with lower brand equity.
H2: Programs with higher brand equity will exhibit higher broadcast
shares than programs with lower brand equity.
H3: Programs with higher brand equity will exhibit higher household
audiences than programs with lower brand equity.
The literature review also focused on the relationship of a program to its
lead-in (inheritance effects) and recognized the existence of a kind of tuning
inertia where viewers resist changing channels. However, a closer examination
brought to light the concept of momentum which suggests that, given the same
size lead-in, programs with strong equity will perform differently than programs
with weaker equity. That is, consumer-based brand equity can be found in the
49

50
differential ratings response between two adjacent programs. The following
two hypotheses were offered.
H4: Programs with higher brand equity will reveal a significantly greater
difference between lead-in program audience size and program audience size
than programs with lower brand equity.
H5: Programs with higher brand equity will reveal a significantly greater
differential share index than programs with lower brand equity.
Narrowing our focus to the dynamics underlying this differential
response, the literature review proposed that program brand equity is a
function of (a) the power to retain lead-in audiences from the same channel
and (b) the power to recruit audiences from other sources. These two
components address the momentum of a program as it passes through its
competitive program environment. Accordingly, two additional hypotheses were
offered.
H6: Programs with higher brand equity will retain a greater proportion of
their lead-in audiences than programs with lower brand equity.
H7: Programs with higher brand equity will recruit a greater proportion of
audiences from other sources than programs with lower brand equity.

CHAPTER 4
RESEARCH METHODOLOGY
Because of the exploratory nature of this study, a case study format is
appropriate. While there is an extensive statistical analysis involved with this
case study, the goal was not to generalize statistical results but rather explore
theoretical possibilities within a fairly controlled environment. Respected case
study researcher Robert Yin (1994) claims that
Case studies, like experiments, are generalizadle to theoretical
propositions and not to populations or universes. In this sense, the case
study does not represent a ‘sample’ and the investigators goal is to
expand and generalize theories and not to enumerate frequencies (p.
10)
The hypotheses serve as initial “tests” of some theoretical offerings but do not
exclude unexpected findings that may stimulate further exploration of program
equity.
There were two primary sources of data for this study: (a) Nielsen
“overnight" metered ratings and (b) Nielsen “sweep” ratings. As presented
earlier in the literature review, these data represent two different methodologies
for generating and reporting program ratings performance. Whenever possible,
the more reliable continuous overnight ratings were used, but there were
several situations where only sweeps data were suitable (the specific data
sources are disclosed on all tables and within the appropriate text portions of
the study). The basic units of analysis for program performance were
51

52
household ratings (rt), household shares (sh), and total households (hh). Again,
the prior literature review offers precise statistical definitions of these terms.
Demographic information, such as gender and age, were not available from
the overnight metered data.
With the written permission of Nielsen Media Research, ratings data for
January through December 1996 for one major television market were turned
over to the researcher for analysis. This included 365 days of overnights and
four-month-long sweep periods (February, May, July, and November). Because
these periodic reports are used primarily to estimate ongoing and future
program performance, Nielsen averages the four-week ratings in order to
increase reliability.
In order to test the hypotheses presented in this study, these basic units
of analysis were sometimes manipulated statistically to reveal greater insight.
For example, the influence of program brand equity on program ratings
performance was evaluated from a number of perspectives. The first was a
comparison of newscast performance based on conventional descriptive
statistics of rating, share, and households. The second was an analysis of the
differential ratings responses between newscasts and their respective lead-in
programs (inheritance effects). This was accomplished through analysis of
variance, indexing, and correlational analysis.
To assess a program’s ability to retain lead-in audiences and recruit
audiences from other sources, Nielsen provided a custom “Audience Flow
Analysis” based on the May 1996 sweep period. This analysis tracks the

53
source and destination of the viewing audiences between an adjacent quarter-
hour. For this study, our focal point was the household “flow” from
10:45 PM to 11:00 PM. Regrettably, Nielsen does not provide such flow studies
using overnight data, but the four-week sweep data still provided worthwhile
insight.
This test market was chosen specifically for reasons which are
presented in subsequent portions of this methodology section. For reasons of
confidentiality, the name of the market was not disclosed in the study, and the
target stations were identified merely as stations A, B and C.
Controlling the Marketing Mix
Based on Keller’s (1993) conceptualizations and discussions in the
prior literature review, the differential effect of brand knowledge on consumer
response must be measured under controlled circumstances. These
controlled circumstances include marketing mix components where the target
brands must be (a) direct competitors in terms of product category, (b) identical
in price across aii competing brands, (c) equally available in terms of
distribution to consumers, and (d) unaffected by short-term promotion activities.
In a broadcast programming context, a direct competitor would be a
program of highly similar content, such as a local newscast. Equal availability
would be defined as competing programs that are scheduled at the same day
and time in the same market airing on comparable facilities. Typical transitory
promotions would occur during the periodic ’’sweep” weeks.

54
To satisfy the above criteria, the 1996 Nielsen ratings performance of the
late evening newscasts on three competing VHF stations in a top 20 southeast
television market were selected for this study. These three stations are well
established network-affiliated competitors that have been offering 11:00 PM
newscasts for over 20 years. Their terrestrial signal coverage and cable
penetration levels were highly similar. In fact, two stations share the same
transmitting tower. Furthermore, from an audience perspective, the average
weekly cumulative household delivery (Monday through Sunday, 6:00 AM to 1:00
AM) was almost identical for these target stations. That is, over the course of a
typical week, each station reached the same number of individual
(unduplicated) households within the market (Nielsen Cume, 1996).
In addition to obvious direct brand competition in terms of program
content, scheduling, and signal coverage, there was the added advantage of
each newscast experiencing a different lead-in seven nights a week. While
most stations in the country broadcast several news programs throughout the
day and evening, the ratings performance of the 11:00 PM (10:00 PM central)
newscast is considered often a primary indicator of a station’s vulnerability to
competitive attack (Eastman, 1997). With a different lead-in program every
night, the late news is the most susceptible time slot for competitive program
sampling by docile audiences that are not particularly committed to any one
brand of newscast.
To assure genuine direct competition, the ratings database was further
refined by extracting dates when the three newscasts did not compete head to

55
head at 11:00 PM. Throughout the year stations are often forced into “late
starts” due to extended movies, specials events, and sports. For example, the
ABC affiliate had to contend with late starts for a dozen weeks due to Monday
Night Football, thus destabilizing the competitive marketing mix at 11:00 PM. Of
the 365 available dates for study, 289 were found to offer the ideal competitive
environment where at 11:00 PM news viewers had three legitimate news
options.
To neutralize the possible bias of short-term promotion activities, two
procedures were introduced. First, whenever possible, the continuous
overnight ratings were used instead of the periodic sweep weeks, thus diluting
the statistical influence of hypoing. Table 1 provides a breakout of the overnight
database.
Table 1
Breakout of Nielsen Metered Overnight Database
Number of original cases
364
Number of cases of direct 11:00 PM News
competition among three stations
289
Number of direct competition cases during
sweep months
86
Number of direct cases outside sweep months
203
A second precautionary procedure involved a direct comparison between
sweep weeks and nonsweep weeks data which revealed no significant
differences for 1996. Therefore, the 16 sweep weeks were retained in the
overall database. Tables 2 and 3 offer station-by-station comparisons.

56
Table 2
Comparison of Sweep versus Nonsweep Cases: 1996 Descriptive
Household Data
Sweeps
Nonsweeps
Station
Mean
(000)
SD
Mean
(000)
SD
%
difference
A
127
26
132
27
4.0
B
114
33
115
31
1.0
C
89
21
92
23
3.0
Note. Sweep cases, N = 86; nonsweep cases, N = 203.
Mean scores based on Monday through Sunday newscasts.
Table 3
Comparison of Sweep versus Nonsweep Cases: Paired Sample
T-Test by Households
Station
t- value
2 tail sig.
A
-1.19
.239
B
-.25
.806
C
-1.00
.321
Note. Sweep cases, N = 86; nonsweep cases, N = 203
Establishing a Comparative Equity Benchmark
In order to test the relative influence of brand equity on program
performance, it was first necessary to identify a benchmark newscast that
demonstrated superior equity characteristics based on information garnered
from sources independent of the one-year Nielsen overnight ratings used for
this study. Based on the prior literature review, the following two criteria should
suffice as the basis for such a comparative benchmark.

57
1. Significant long-term market dominance. From an outcome
perspective, consistently high market rankings over several years is a valid
indicator of strong equity.
2. Stated brand loyalty or commitment by heavy users of a product.
Bypassing broadcast ratings data, a consumer survey that provides
program viewing loyalty information can serve as an alternative measure of
equity.
Benchmark #1: Nielsen 10-Year Audience Trend Analysis
As stated in the literature review, one of the primary advantages to high
brand equity is long-term market share stability. Over time, strong brands
exhibit a resiliency to market fluctuations and competitive attacks. Table 4
provides a 10-year sweep summary (1986-1996) of late evening newscasts in
our test market by rating, share, and households.
Table 4
10-Year Monday through Friday Average Sweep Performance of Three
Newscasts (February 1986-November 1996)
Measure
Station A
Station B
Station C
F ratio
Prob.
Scheffee at .05
Rating &
(rank)
11.9 (1)
10.6 (2)
7.8 (3)
37.6
.000
A-B, B-C,
C - A
Share &
(rank)
27.3 (1)
24.3 (2)
17.5 (3)
64.0
.000
A-B, B-C,
C-A
Household
& (rank)
116 (1)
99 (2)
71 (3)
28.2
.000
A-B, B-C,
C-A
Note. N = 39 sweeps or 116 cases, df = 115, p level = .05
The following sweeps data were not available: July 1986, November 1988 and
1991, and May 1986.
Households = (000)

58
Because overnight metered ratings were not available in this market
until the mid 1990s, this trend analysis is derived from conventional sweeps
data. Of the 40 consecutive sweep periods analyzed, station A seldom lost its
number one ranking. Additionally, over the 10-year period, the overall
differences among the three stations was found to be statistically significant in
all cases.
The investigator realizes that the use of these 10-year Nielsen ratings to
predict the outcome of current overnight Nielsen ratings could be construed as
a tautology in that both equity and performance are evaluated using the same
units of measure. Although these data do demonstrate Station A’s long-term
stability, the following benchmark offers better evidence of brand equity utilizing
a completely different methodology.
Benchmark #2: Marshall Marketing Loyalty Survey
In addition to a 10-year trend analysis, high equity for station A was
reaffirmed using data collected from Marshall Marketing and Communications.
As was discussed in the literature review, MM&C conducts annual surveys of
consumer purchasing habits for select markets around the county. In 1996, the
same year as the Nielsen ratings data, MM&C conducted such a survey in the
test market. Over 1,000 telephone interviews were conducted over four weeks.
Sample sizes were predetermined to yield a 2% to 4% margin of error within a
95% confidence level. Recurring and exclusive viewing were defined
operationally as watching only one station’s newscast at least three times over
the past seven days. Table 5 presents the results of the 1996 survey. The first

59
two columns show that among subjects who watched at least one local
newscast during a prior week, Station A dominated the category by a 2 to 1
margin over either competitor. Sixty-nine percent of the responses included a
Station A newscast. Taking a narrower look at the exclusive viewing habits of
heavy users, the remaining two columns of the table reveal Station A as the
market leader, garnering over 31% of the adults who watched only one station
for local news. Here the margin for Station A was three to one over its
competitors. Presuming a controlled marketing mix where all stations were
equivalent in terms of time period, signal coverage, and market conditions,
these data support our earlier position. Station A demonstrated the highest
brand equity.
Table 5
1996 Marshall Marketing and Communications Survey on Exclusive (Loyal)
Viewing of Newscasts
Stations
Watched at
least once
during past
week
% of once
per week
viewing**
Watched
exclusively
3 or more times
during past week
% of Exclusive
viewing
Combined
562*
-
298
53.0
A
388
69.0
177
31.5
B
191
34.3
57
10.1
C
191
34.2
64
11.4
*Original sample frame, N = 1,200. **Total % of three stations exceeds 100%
because of viewing of multiple stations during week.
Having established, at least tentatively, station A as the comparative
equity benchmark for the market, we can approach the research question of

60
how does program brand equity influence program ratings performance on an
everyday basis.

CHAPTER 5
RESULTS
Hypotheses One, Two, and Three
The initial three hypotheses predicted a program with higher brand
equity would outperform its direct competitors across a number of descriptive
audience statistics including, ratings, shares, and households. Analyzing the
adjusted 1996 overnight ratings data, the hypotheses were supported on all
dimensions. Table 6 reveals that for the categories of (a) overall performance
and (b) individual day performance, station A (our designated equity leader)
achieved the majority of number one rankings in households, ratings, and
shares. Furthermore, ANOVAs of these differences in program performance
indicate they were significant. A significant f statistic, however, indicates only
the station means are probably different. It does not pinpoint where the
differences occur. Therefore, the table goes a step further by identifying the
specific station combinations that are different. The Scheffee multiple
comparison procedure is considered highly conservative because it requires
larger differences between means for significance than most other methods
(Norusis, 1996). At a significance level of .05 we see that comparisons of
rating, share, and household data are nearly identical. Furthermore, on a day-
by-day basis, the number of significant combinations of stations is highly
similar except for the combination of station A and station B on several nights.
61

Table 6
Comparative Newscast Performance Based on Nielsen Metered Overnight Ratings
Measure rankings expressed within parentheses. ** Scheffe and post hoc analysis based on p = <.05.
CD
hO

63
Station C distinguishes itself for being significantly different from both
competitors every day of the week but Friday. Referring back to earlier data, we
see that these significant differences are not necessarily signs of outstanding
ratings performance. On the contrary, station C languished in third place
across all measures. A second post hoc analysis looks at the relative
performance of station A against the averaged audience ratings of stations B
and C in tandem. Again, station A's market leadership was significant.
Some of the above ANOVA results may violate certain statistical
assumptions or necessary conditions for a perfect analysis. In particular, the
reader is cautioned not to make any strict interpretations of share figures,
which by definition cannot be viewed as independent data units. Because the
use of ratings, shares, and households is so pervasive in the broadcasting
industry, these units of measure were included in the initial portions of this
results chapter. However, in subsequent analyses when statistical
independence is required, only household (000) data are analyzed.
Hypotheses Four and Five
Hypotheses four and five addressed the functional relationship of the
newscasts to their respective lead-in programs. Programs with strong equity
were expected to optimize lead-in ratings more efficiently than programs with
weaker equity. This efficiency would be observed in the differential ratings
response between the two programs. Depending on the type of analysis
employed, these propositions were supported to varying degrees.

64
Hypothesis four stated that programs with higher brand equity will reveal
a significantly greater difference between lead-in program audience size and
program audience size than programs with lower brand equity. Table 7 looks at
the overall household performance of the three stations. In terms of combined
performance, we see that there was a 6% drop in households using television
from 10:45 PM to 11:00 PM. However, the data also show that station A, as
expected, witnessed a substantial gain in audience, surpassing its lead-in by
9%. Of the three stations, only station A achieved a positive differential
response to its lead-in, thus supporting the hypothesis. Additionally, an
ANOVA and Scheffee analysis of these differentials (news minus lead-in
households) indicates that these differences were significant among all three
stations.
Table 7
Differential Performance Between 10:45 PM Lead-in and 11:00 PM Newscasts
(Overall Nielsen Metered Overnight Households [0001)
Station
A
Station
B
Station
C
A&B&C
(News
HUT)
F ratio
Prob.
Scheffee at .05
10:45 PM
115
127
99
341
33.4
.000
A-B, A-C
11:00 PM
127
110
91
326
126.0
.000
A-B, B-C, C-A
Differential
+11
-17
-8
-15
79.2
.000
A-B, B-C, A-C
% change
+ 9
-14
- 9
- 6
Note, df = 866, p = .05
Hypothesis five stated that programs with higher equity will reveal a
significantly greater differentia! share index than programs with lower equity.
Because the number of homes using television (HUT) can change dramatically

65
throughout the day, comparing household and ratings data from different time
periods may be misleading. Share calculations, however, are based solely on
the available audience at the time of the measurement. Therefore, Table 8
interprets the newscast/lead-in relationship as a share index. Index scores
exceeding a magnitude of one indicate the degree to which a newscast
outperformed its lead-in. Conversely, scores of less than one infer the degree
to which the program was unable to hold its lead-in audiences. As expected, in
almost all cases, station A (our equity leader) indexed higher than its two
competitors.
Table 8
Share Index of Late Newscasts Compared to Lead-in Programming
Station
Overall
Mon
Tues
Wed
Thurs
Fri
Sat
Sun
A
1.28
1.5*
1.2 *
1.2 *
1.5 *
1.0
1.5 *
1.2 *
B
1.02
1.1
.97
.99
.83
1.1 *
1.1
1.1
C
1.08
1.0
.94
1.2 *
1.2
1.0
1.0
.96
*lndicates highest index for that evening.
Another way to evaluate the differential response of a newscast to its
lead-in is through correlational analysis. Nearly all work on inheritance effects
has used this technique in describing the close association or dependence
between adjacent programs. However program brand equity is presumed to be
a function of independence. That is, a program with strong equity is not as
dependent on its lead-in for audience ratings as a program exhibiting lesser
equity. Therefore, one would expect a weaker association (less dependency)

66
for a program with relatively strong equity. It was presumed that evidence of this
lack of dependency would provide indirect support of hypotheses four and five.
Table 9 offers the results of a correlational analysis based on household data.
From an overall standpoint, the correlation data give partial support for
the hypothesis in that station A’s correlation is substantially lower than station
B. However, Station C demonstrated an even lower figure. On a day-by-day
basis, station A offered more convincing evidence by “winning” four out of the
seven nights with the lowest correlation.
Table 9
Household Correlational Analysis Between 10:45 PM Lead-in Programming
and 11:00 PM Newscasts
Station
Overall
Mon
Tues
Wed
Thurs
Fri
Sat
Sun
A
.76
.53*
.55 *
.75
.60*
.61 *
.53 *
.89
B
.85
.86
.65
.68 *
.86
.68
.60
.84
C
.71
.78
.67
.83
.41
.65
.55
.59 *
Note. A = 4, B= 1, C = 2.
indicates lowest r among the three stations.
As a brief digression, the investigator explored a different method of
correlation of the overall data. Since most television programming recurs on a
weekly basis, correlations were performed using seven-day rolling averages
rather than on the more conventional single-day data. According to the data
results presented in Table 10, station A revealed the greatest disparity between
the two methods and also repositioned itself as the station exhibiting the
lowest correlation.

67
Looking at the regression data for overall household performance
among the three competing stations, Table 11 reveals that each newscast had
a strong positive linear relationship with its lead-in programming. Lead-in
Table 10
Comparison of Household Correlations Derived from Single Day versus
Seven-Day Moving Average Data
Station
r derived from
single day data
r derived from
seven-day moving average data
A
.76
.59
B
.85
.82
C
.71
.72
Table 11
Regression Data Based on Overall Households (Independent Variable:
Lead-in Program Dependent Variable: Newscasts)
Station
Slope
Intercept
A
.76
53,337
B
.85
49,951
C
.71
44,478
households were strong predictors of newscast household performance.
However, on face value the slope and intercepts do not indicate any
remarkable differences among the three competitors. Based on these data,
one could speculate that Station A’s success cannot be attributed solely to its
ability to convert lead-in audiences but rather to its power to recruit audiences
from other sources. The final two hypotheses take a closer look at this
possibility.

68
Hypotheses Six and Seven
Hypotheses six and seven take the above notions of differential
response (or momentum) a step further by dividing the phenomenon into two
distinct segments: (a) the power to retain lean-in audiences and (b) the power
to recruit audiences from other sources. Nielsen Media Research provided the
investigator with a standard “Flow Analysis" study of the May 1996 diary sweep
period. Using four-week averaged household ratings, Nielsen tracked weekday
audience flow from 10:45 PM to 11:00 PM. While the complete study is shown
in the appendix, the most salient data have been distilled into Table 12.
Table rows 1 through 4 (shaded area) provide data on the relative
audience composition of each newscast. That is, total news households (row
4) equals the sum of households retained from lead-ins (row 1), plus
households recruited from other stations providing local news (row 2) plus
households recruited from nonnews sources (row 3). The last two categories
deal with audiences that “switched” channels in order to watch a specific
newscast. The percentage columns within the shaded area give the proportion
of the total news audience that each category' holds. These four data rows,
however, do not provide sufficient insight into the power of brand equity. Rows
5, 6, and 7 delve into dynamics of our final hypotheses.
Hypothesis six predicted that programs with higher brand equity would
retain a greater proportion of their lead-in audiences than programs with lower
brand equity. As expected, row 5 of Table 12 shows station A to be the best
performer, retaining 65% of its lead-in households.

69
Hypothesis seven stated that programs with higher brand equity would
recruit a greater proportion of audiences from other sources than programs
with lower brand equity. Rows 6 and 7 of Table 12 address this proposition.
Perhaps the proportion of households recruited from other news stations (row
6) offers the most persuasive evidence. Here we have audiences that switched
away from a channel that was about to present a local newscast. Of all the
news channel “switchers,” station A earned the largest proportion (44%).
Similarly, station A recruited a greater proportion (51%) of the total nonnews
switchers.

70
Table 12
Nielsen Audience Flow Analysis, May Sweep 1996 (10:45 to 11:00 PM, Monday
Through Friday, Four-Week Average Households [0001)
Row
no.
Measure
Station A
Station B
Station C
0
000
%
000
%
000
%
1
HH retained from
lead-in
80
49
77
59
53
56
2
HH recruited from
from other news
stations
(switchers)
40
25
29
22
23
24
3
HH recruited from
non news sources
(switcher)
44
26
25
19
18
20
4
Total News HH
164
100
131
100
94
100
5
% HH retained
from leads-in *
65
54
51
6
% recruited from
total news
switchers (92) **
44
31
25
7
% recruited from
total non news
switchers (87) ***
51
29
20
Note, chi sguare p = .05:
*4.49 with stations B and C collapsed, df = 1. ** 5.0 with stations B and C
collapsed, df = 1. *** 12.84 among all three stations, df = 2 and 11.64 with
stations B and C collapsed, df = 2.

CHAPTER 6
DISCUSSION
The purpose of this study was to explore the plausibility of applying
brand equity theory to television program ratings performance. While the
industry has embraced the “buzzwords” of brand management, there has been
little scholarly work on making the transition from products as brands to media
as brands. While specific hypothesis were presented, much of the effort
invested in this study was a matter of curiosity and intellectual tinkering. As the
title of this dissertation implies, the first challenge was conceptualization—How
can conventional brand equity theory be applied in a meaningful way to
television programming. The second challenge was to find a way to measure
or operationalize these concepts using standard Nielsen ratings data.
In addition to an overview of brand equity theory, the conclusion of the
literature review section provided some feasible conceptualizations for
television programming. In particular, Keller (1993) and the notions of
consumer inertia (momentum) were found to work well in explaining program
performance in light of the powerful influence of inheritance effects. Presuming
that essentia! marketing mix components are controlled or neutralized, the
program equity model of audiences retained/audiences recruited appears to
have merit.
71

72
Applying this model to inheritance effects offers some new theoretical
insights. Instead of addressing structural factors such as program type or
number of program options, this approach delved into the ability of a program
to retain lead-in audiences and recruit audiences from other sources. In
tandem, these two power dimensions define a program's momentum as it
“travels” through time. A program exhibiting strong equity can overcome a poor
lead-in and recruit loyal viewers from other sources. Additionally, this program
will be less likely to lose portions of its lead-in audience to direct competitors.
The case study provided a more concrete way to “test” some of these
theoretical propositions. All seven hypotheses were supported to some degree
and open the door for further theoretical work and more generalizare empirical
research.
From a conceptual standpoint, the a major dilemma facing the
investigator was choosing a theoretical starting point. With so many divergent
conceptualizations and theories available, brand equity can be almost anything
a researcher wants it to be. From attitudinal notions of “heritage,” “reputation,”
“perceived quality,” “depth of awareness,” “enthusiasm,” and “commitment” to
more behavioral concepts, such as “pricing elasticity,” “repeat purchases,”
“exclusive heavy user,” and “consumer inertia,” marketing professionals and
scholars have yet to agree on a precise definition of brand equity.
Compounding the problem is that many in the field use marketing terms such
as brand image, brand loyalty, and brand equity interchangeably. After much
searching and reflection, Keller (1993) emerged from this theoretical muddle

73
with an approach that can be translated readily into broadcast programming
terms. Keller’s theoretical underpinnings combined with a more rigorously
defined branding vocabulary can form the bedrock of a brand equity research
agenda.
A far more perplexing issue was discovered on the operational side of
program brand equity. Keller’s (1993) overall conceptual framework requires a
controlled marketing mix which enables a researcher to capture the relative
power of brand equity. That is, strong, favorable, and unique brand
associations (brand knowledge) cannot be measured properly until the
possible confounding elements of the brand’s marketing mix are held in check.
Marketing mix factors such as pricing, distribution, and promotion can taint
equity measures that rely on in-market brand performance data.
For our case study, controlling these marketing mix factors was an
unexpected challenge. The methodology section of this dissertation revealed a
number of problems that had to be overcome. Among these were (a) identifying
direct competitors by program type or genre and (b) creating the simultaneous
availability of direct competitors. The selection of three 11:00 PM local
newscasts, scheduled at the exact same time, made for an ideal test
environment for our case study. However, most television programming does
not offer this convenience.
Determining who is a direct competitor can be a frustrating task. While
local newscasts do not pose a great problem, other program types can
generate considerable disagreement among media researchers. For example,

74
should “Home Improvement” and “Married With Children?” be assigned to the
same genre of sitcom? Within the category of adult drama, should “NYPD Blue”
and “Dr. Quinn” be regarded as coming from the same product category?
Webster (1985) addressed this issue of program typology and warned that the
current industry categories would someday become “overly broad, and
insensitive to subtle, though important, distinctions” (p. 125). In an environment
of highly distinctive niche programming on dozens of channels, the definition of
a direct competitor will become even more challenging.
The availability of direct competitors at the same point in time was
accomplished easily with our three 11:00 PM newscasts, but this situation is
more the exception than the rule in television. In fact, in our test market there is
currently a 10:00 PM local newscast on a small UHF station. Do people watch
this program because it is the only 10:00 PM news in the market, or is there
evidence of genuine brand equity based on content ? If this station scheduled
its newscast at 11:00 PM, against the other entrenched news competitors, how
well would it perform? Because of the time-bound nature of broadcasting,
these types of questions are not answered easily. Perhaps a “what if” survey
could be administered to an appropriate sample group. In this case, the
controls would be far more contrived than a real-world setting where actual
viewing behavior could be recorded.
Two related distribution factors that may be awkward to control are
signal coverage and available potential audiences. Just as retailers recognize
the marketing advantages of a superior distribution system, so broadcast

75
researchers must be aware of certain structural factors, such as signal
strength and the available audiences that can be reached by that signal.
Programs that cannot be received cannot be watched. Unless there is a “level
playing field” in terms of program distribution, consumer-based brand equity
cannot be measured accurately. Fortunately, our test market exhibited
equivalency in this area, but this may not be the case for some other markets.
For example, the Gainesville, Florida, television market must “import” CBS and
NBC signals via cable retransmission from other nearby markets.
The negative impact of certain promotion activities must also be
controlled. In recent months, there has been an outcry from the broadcast and
advertising industries to control the marketing abuses unleashed during the
infamous sweep weeks, and some progress has been made. In our case
study, the sweeps did not appear to have any extraordinary influence on
newscast ratings, but other program dayparts may have been more vulnerable
to hypoing. Of course, the best cure for sweeps distortion is to look beyond
periodic measures and instead examine continuous overnight viewing data.
Regrettably, only 42 out of the 212 Nielsen television markets currently have
this research capability.
After determining a plausible theoretical approach and wrestling with the
above-mentioned methodological challenges, the seven program performance
hypotheses were then tested. A primary assumption of the study was that
station A was a suitable equity benchmark. The two criteria of (a) long-term
ratings leadership and (b) viewer loyalty data derived from the Marshall

76
Marketing survey served as yardsticks for isolating station A from its
competitors.
While recognizing the lack of external validity, the hypotheses found
considerable support. As expected, the overall and daily performance of station
A was exceptional when compared to its competitors. In fact, stations B and C
often appeared quite similar in their audience performance, while station A
stood apart from the crowd. When analyzing newscast performance in
relationship to lead-in programming, the results were mixed. From simple
differential and indexing perspectives, station A consistently outperformed its
lead-in program. The correlational analyses were not as convincing but did
show general support. The correlations performed on the overall (seven-day)
data revealed less equity value for station A than the individual day data. This
may be the result of distortions that can occur sometimes when averaging
data. For example, the .89 correlation on Sunday night for station A is far out of
line compared with the other six days of the week and may have overstated the
overall performance of the station. Similarly, the Thursday night correlation for
station C (against “ER" on NBC) is extraordinarily low compared with other
days. Another possibility may be a conceptual flaw. That is, the supposed
independence of our high equity station may be revealed only when the lead-in
is exceptionally poor. Another related avenue for future exploration is why the
use of seven-day moving averages (Table 10) seemed to enhance the position
of station A.

77
The regression analyses in this dissertation invite some theoretical
speculation on the interpretation of program equity. The two components of a
single linear regression equation, its slope (coefficient) and its intercept
(constant), could be construed as indicators of the two power dimensions of
our equity model. Given the identical lead-in audiences (independent variable),
it is possible mathematically for two programs to exhibit the same slope but
different intercepts and visa versa. The slope of the equation may provide a
rough measure of the efficiency of a program to retain its lead-in audience.
Alternatively, the intercept, which would be defined mathematically as the size
of a news audience (dependent variable) when the lead-in (independent
variable) is zero, might be analogous to a measure of the program’s ability to
recruit audiences from other sources. In a typical regression analysis there are
seldom any data points clustering at the actual intercept point. Therefore, this
figure should not be interpreted in absolute terms. However, it may be of some
value as a comparison tool. In the case study, station A revealed a modestly
higher coefficient and intercept than its two competitors. Additional studies
across several markets can determine if these results were merely
coincidental.
The most illuminating data came from the Nielsen Audience flow study
which actually tracked audience retention and outside recruitment for the May
sweep period. Despite some methodological limitations, which are presented
later, the data offered the most persuasive evidence for supporting hypotheses

78
six and seven. The ability of Station A to both retain and recruit audiences were
unsurpassed.
Study Limitations
Aside from the above-mentioned difficulties in controlling the marketing
mix, the researcher recognizes several limitations to this particular study. First,
there is the lack of external validity. By design this was an exploratory effort
under limited conditions that were familiar to the researcher. The three news
programs in this particular market provided an ideal field setting where
marketing mix factors could be controlled to a reasonable degree. By having
different lead-in programs each night, the newscasts were placed in different
“experimental” conditions. With the goal of taking the first steps in
conceptualizing and measuring the effects of brand equity on television
programs, the results of this case study are encouraging. The next logical step
is to apply some or all of these notions across other markets and other types of
programming.
A second limitation was the reliability of Nielsen ratings. Although
another measure of program performance would not be as relevant to the real
world of commercial broadcasting, there is no denying the inadequacies of
Nieslen ratings. The buyers and sellers of program audiences may look at
these numbers as absolute measures, but media researchers (including
Nielsen itself) recognize several statistical limitations. First, Nielsen does not
guarantee a representative random sample due to a number of factors, the
most important being dismal cooperation rates (averaging 45%) among

79
people asked to participate. Secondly, presuming the samples were perfectly
random, there is considerable sampling error (standard error) that is not
disclosed in the regular reports. At the beginning of every Nielsen market
sweep report, the company warns subscribers against using certain ratings
data to make management decisions because the relative standard error could
exceed 50% (Nielsen Method, 1995). Therefore, television ratings data tend to
fluctuate. Many media professionals want to believe that these changes are
real, but savvy researchers know that much of this motion is due to sampling
error or “bounce” rather than legitimate changes in audience tuning behavior.
While there are distinct advantages to using the metered overnight
ratings compared to the four-week averaged sweeps data, a disappointing
drawback was the lack of demographic information. Unlike the national Nielsen
“people meter” ratings, local metered markets have yet to adopt this recording
technology. Currently, overnight ratings are restricted to household data. Based
on the literature review, it seems possible that program loyalties could vary
across age and gender demographic categories.
This same limitation arose when we examined Nielsen audience flow
studies. To date, this detailed audience analysis is restricted to sweep weeks
data. In our case, we had to look at Monday through Friday average newscast
performance over one month. Despite these limitations, our theoretical
precepts of (a) power to retain audiences and (b) the power to recruit
audiences were substantiated, and Nielsen promises that overnight flow
studies would be available soon to metered markets. Meanwhile, the sweep-

80
based flow studies (four per year) can still be applied to other markets in an
attempt to replicate our case study findings. Of course, Nielsen charges
broadcasters a substantial fee for these special studies, so an academic
researcher may have a financial handicap.
The Marshal Marketing and Communications data were used primarily
as a benchmark comparison for the Nielsen ratings. Within their statistical
calculations was the determination of “heavy users." Offhand, this type of
information seems most appropriate for equity research, but similar ratings
data are not obtained easily from Nielsen. For an undisclosed fee, Nielsen will
do a custom data analysis.
Conclusions
Returning to the essential purpose of this study, one can conclude that
program brand equity is a plausible theoretical notion that can be applied to
electronic media. However, broadcasters should be wary of all the brand
management jargon that is so prevalent in the trade press and industry
discussions. This study also showed that, given a highly controlled research
setting, the impact of attitudinal components of program brand equity (Keller’s
brand knowledge) can be measured to some degree using the behavioral
measures of standard Nielsen ratings. Additionally, this project offers new
insight into the phenomenon of inheritance effects and tuning inertia.
After establishing conceptual and operational definitions of program
brand equity, this study examined how a program with substantial brand equity
performs when compared to direct competitors with less brand equity. The

81
seven hypotheses approached this issue from several perspectives.
Hypotheses one, two, and three looked at overall ratings performance and
found that the program identified as having high brand equity had higher
ratings, higher shares, and a larger household audience. Hypotheses four and
five narrowed the investigation to the relationship of a program to its lead-in. A
number of statistical indicators revealed that a program with strong equity
processes or “inherits” audiences in ways that are different than its direct
competitors. That is, while the high equity program consistently outperformed
its 10:45 PM lead-ins to maintain 11:00 PM ratings leadership, the other
stations usually lost ratings during this transition. Translating these rating point
gains and losses into the components of where an audience has come from
and where it has gone was the domain of the final two hypotheses. Using a
“Flow Analysis” that differentiated between audiences retained from lead-ins
and audiences recruited from other sources, the results indicated that the high
equity program was superior to its low equity competitors not only in retaining
lead-in households but was in drawing households from surrounding
channels and the “off’ position. Taking an even tighter focus, the results
showed that the high equity program demonstrated exceptional power in
recruiting audiences from channels offering similar local newscasts (direct
competitors). Taken in combination, the hypotheses validate the importance of
brand equity to the ratings success of a television program--at least for local
news programming.

82
On a subordinate level, we can also conclude that this study explains
some of the extreme variance found in prior inheritance effects research.
Rather than relying solely on structural factors, such as program type and
number of program options, we can now introduce the audiences retained/
audiences recruited equity model to help explain these disparities.
Suggestions for Future Research
This study fosters several research avenues for continued scholarly
work. On a rudimentary level, there is a need to clarify exact research topics
using proper terminology. For example, page 39 offered some standard brand
concepts that could be applied to broadcast programming. Additionally, there is
a need to segment further these branding domains into programs, stations,
and networks. Below is a proposed research topic grid that may be useful for
future investigations of media as brands.
Media As Brands Research Topic Grid
Program
Station
Network
Brand Awareness
Brand Extensions
Brand Image
Brand Positioning
Brand Loyalty & commitment
Brand Equity
Taking the theoretical position that understanding audience flow is
crucial to understanding the influence of program brand equity, the following
proposed audience member categories may assist future research.

83
1. Loyalists.
Definition: Viewers who hold strong consumer-based program brand
equity and are found within the lead-in program’s total audience.
Disposition: These viewers remain in place on the same channel
because of brand commitment
2. Passives.
Definition: Viewers who do not hold any strong consumer-based
program brand equity towards any direct competitor and are found within the
lead-in program’s total audience.
Disposition: These viewers will remain in place on the same channel
because of simple “inertia” rather than true commitment.
3. Converts.
Definition: Viewers who hold strong consumer-based program equity but
were watching on a competing channel and, therefore, are motivated to switch
channels.
Disposition: If necessary., these viewers will abandon a lead-in program
channel and switch over to a more suitable program option (see number 5
defectors below).
4. Tune- ins.
Definition: “Appointment” viewers who hold very strong consumer-based
program brand equity and make a deliberate effort to turn on the TV set in order
to watch a specific program. It possible to have a subcategory of passive tune-

84
ins, where the desire is simply to have the set “on” with no regard for specific
programming.
Disposition-. These viewers are acquired audiences from what some
researchers call the “off position.”
5. Defectors.
Definition'. “Converts” seen from an opposing perspective, that is,
viewers who are about to abandon a program channel in order to find a more
suitable or preferred program. Defectors hold strong consumer-based
program equity for a direct competitor. (One program’s defector is another
program’s convert.)
Disposition: At the appropriate time, these viewers leave one program
and go to another on a competing channel.
Using these categories, one could hypothesize that program equity (a)
increases the number of converts, (b) reduces the number of defectors, (c)
reinforces current loyalists, and (d) transforms passives into converts.
Implications For the Field
Well now, home entertainment was my baby’s wish
so I hoped into town for a satellite dish
I tied it to the top of my Japanese car
I came home and pointed it out to the stars
A message came back from the great beyond
There’s fifty-seven channels and nothin’ on.
(Springsteen, 1989)
in an earlier decade, when cable and satellite delivery systems were just
beginning to offer a broad range of programming options, singer/songwriter
Bruce Springsteen composed a prophetic tune entitled “57 Channels And

85
Nothin’ On.” It recounts the tale of a frustrated young man who, after eagerly
embracing these new technologies, finds that even with an array of 57 program
channels, there was still nothing of value to watch on his TV. Finally, he
retaliates by shooting his television set with his .44 magnum. Keller (1993), no
doubt, would affirm that these programs failed to engender any favorable,
strong, or unique brand associations and, indeed, from a scholarly viewpoint,
Keller would be correct. While Springsteen was surely unfamiliar with the term
brand equity, he did reveal delightful insights into consumer behavior and the
challenge of creating successful media brands. Successfully applying the
notions of consumer-based brand equity will be crucial for the economic
survival of broadcast programs, stations, and networks. As we approach a new
digital decade, where 57 choices will seem meager, the consumer anguish
found in Springsteen’s classic song will continue to ring true. Without proper
media brand management, audiences in the 21st century will lament about
157 channels and nothing on.

APPENDIX
A NIELSEN TV METERED MARKET HOUSEHOLD ANALYSIS MAY 1996
NSI-PLUS-
NIELSEN TV METERED MARKET HOUSEHOLD ANALYSIS
ORLANDO MAY 1996
NO. 78931
FOR NIELSEN MEDIA RESEARCH
AUDIENCE SOURCE AND DESTINATION
DMA HOUSEHOLDS
SOURCE TIME: AVERAGE OF WKS 1-4 MON 10:45PM
WKS 1-4 WED.10: 45PM
WKS 1-4 FRI 10:45PM
DESTINATION: AVERAGE OF WKS 1-4 MON 11: 00PM
WKS 1-4 WED 11:00PM
WKS 1-4 FRI 11:00PM
WKS 1-4 TUE 10:45PM
WKS 1-4 THU 10:45PM
WKS 1-4 TUE 11:00PM
WKS 1-4 THU 11:00PM
AVG.SOURCE
AUDIENCE AVG. DESTINATION AUDIENCE (000)
STATION
(000)
WFTV
WCPX
WESH
WKCF
WOFL
WRBW
OTHR
OFF
TOTAL
141
82
116
35
40
6
WFTV
124
80
10
15
4.
7
_
11
14
WCPX
103
18
53
14
2
6
-
8
14
WESH
142
22
13 '
77
4
7
1
14
21
WKCF
45
10
3
5
26
2
-
3
5
WOFL
' 35
8
2
4
1
18
-
4
5
WRBW
11
2
1
1
-
-
3
2
- 1
OTHER
19
9
11
2
6
-
TUNE-IN
5
3
4
3
2
-
AVG.SOURCE
AUDIENCE AVG. DESTINATION AUDIENCE (%)
STATION
(%)
WFTV
WCPX
WESH
WKCF
WOFL
WRBW
OTHR
OFF
TOTAL
14. 1
8.2
11. 6
3. 5
4. 1
0. 6
WFTV
12. 4
8. 0
1. 0
1. 5
0. 4
0. 7
_
1. 1
1.4
WCPX
10. 3
1. 8
5. 3
1. 4
0. 2
0. 6
-
0. 8
1. 4
WESH
14. 2
2. 2
1. 3
7. 7
0. 4
0. 7
0. 1
1. 4
2. 1
WKCF
4. 6
1. 0
0. 3
0. 5
2. 6
0. 2
-
0. 3
0. 5
WOFL
3. 6
0. 8
0. 2
0. 4
0.1
1. 8
-
0. 4
0. 5
WRBW
1. 1
0. 2
0. 1
0. 1
-
-
0. 3
0. 2
0. 1
OTHER
1. 9
0. 9
1. 1
0. 2
0. 6
-
TUNE-IN
0. 5
0. 3
0. 4
0. 3
0. 2
-
86

87
NSI-PLUS
NIELSEN TV METERED MARKET HOUSEHOLD ANALYSIS
ORLANDO MAY 1996
NO. 78931
FOR NIELSEN MEDIA RESEARCH
AUDIENCE SOURCE AND DESTINATION
DMA HOUSEHOLDS
SOURCE TIME: AVERAGE OF WKS 1-4 MON 10: 45PM WKS 1-4 TUE
WKS 1-4 WED 10: 45PM WKS 1-4 THU
WKS 1-4 FRI 10:45PM
DESTINATION: AVERAGE OF WKS 1-4 MON 11:00PM WKS 1-4
WKS 1-4 WED 11:00PM WKS 1-4
WKS 1-4 FRI 11:00PM
TUE
THU
10:45PM
10: 45PM
11: OOPM
11: OOPM
DESTINATION OF AUDIENCE - AVERAGE BASIS
STATION
SHARE
WFTV
WCPX
WESH
WKCF
WOFL
WRBW
CTHR
OFF
WFTV
100
64
8
12
3
5
9
. 11
WCPX
10Ó
17
52
IA¬
2
6
-
8
14
WESH
100
15 ,
9
SS
3
5
1
10
15
WKCF
100
23
. 7
12
56
4
-
7
11
WOFL
100
22
5
12
4
50
-
10
13
WRBW
100
20
9
11
31
19
11
SOURCE
OF AUDIENCE -
AVERAGE BASIS
STATION
WFTV
WCPX
WESH
WKCF
WOFL
- WRBW
TOTAL
SHARE
100
100
100
100
100
100
WFTV
57
12
13
11
16
-
WCPX
13
65
12
7
16
-
WESH
16
16
67
13
18
17
WKCF
7
4
5
73
4
-
WOFL
6
2 '
4
4
44
-
WRBW
2
1
1
-
-
56
OTHER
14
10
9
7
14
-
TUNE-IN
4
3
3
8
5
-
COPYRIGHT 1997 BY NIELSEN MEDIA RESEARCH, INC.
ALL AUDIENCE DATA CONTAINED IN THIS ANALYSIS
WERE OBTAINED FROM THE SAME METER RECORDS
USED TO PRODUCE THE REGULAR NIELSEN TV REPORT.

REFERENCES
Aaker, D. A. (1991). Managing brand equity. New York: The Free Press.
Abelman, R., Atkin, D., & Rand. M. (1997). What viewers watch when they watch
TV: Affiliation as case study. Journal of Broadcasting and Electronic
Media, 41, 360-379.
Alridge, R. (1996, July, 15). If you’re trailing in news, try something different.
Electronic Media, p. 6.
Anantachart, S. (1995). Conceptualizations and a news measure of brand
equity. Unpublished Masters Thesis, University of Florida College of
Journalism and Mass Communication.
Axelrod, J. (1993, October 27). The definition and measurement of brand equity.
Paper presented at ARF Brand Equity Research Day, New York.
Barton, L. (1993, June). Tartikoff: No more white noise! Promotion and
Marketing Executive In Electronic Media, p. 16.
B & C Hal! of Fame. (1996, October, 30). Bob Wright profile. Broadcasting and
Cable, p. 51.
Beatty, S.E., Kahle, L.R., & Homer, P. (1938) The Involvement-Commitment
Model: Theory and implications. Journal of Business Research, 16,149-
168.
Beike, D.R, & Sherman, S.J. (1994). Social inference: Inductions, deductions,
and analogies. In Handbook of Social Cognition (2nd ed., pp. 281-326).
Palo Alto, CA: Annual Reviews.
Belch, G., & Belch, M. (1993). Introduction to advertising and promotion: An
integrated marketing communications perspective. Boston, MA: Irwin.
Bender, D. C. (1993, March 24). Media as brands. Speech presented at
Advertising Research Foundation 39th Annual Conference, New York.
Biel, A. L. (1991) How brand image drives brand equity. Journal of Advertising
Research, 6, RC6-RC12.
88

89
Black (1996, June 3). Building a better brand. L.A. Broadcasting and Cable, p.
93.
Boemer. M. L. (1987). Correlation lead-in show ratings with local television
news ratings. Journal of Broadcasting and Electronic Media, 31, 89-94.
Brinkley, J. (1997, December 29). TV stations fear for their channel brand as
choices proliferate. New York Times, C9.
Burgii, M., & Katz, R. (1997, July 21). Big nets need Geritol. Media Week, p, 3.
Buzzard, K. (1992). Electronic media ratings: Turning audiences into dollars
and sense. Boston, MA: Focal Press.
Ceurvorist, R. W. (1994, February 15-16). A brand eguity measure based on
consumer commitment to brands. Paper presented at ARF Brand Equity
Workshop, New York.
Churchill, G. A. (1995). Marketing research: Methodological foundations. Fort
Worth, TX: The Dryden Press.
Christiani, A. (1993, October 27). Measuring and monitoring brand loyalty and
its role in managing brand equity. Paper presented at ARF Brand Equity
Research Day, New York.
Cobb-Walgren, C.J., Ruble, C.A., & Donthu, N. (1995). Brand equity, brand
preference and purchase intent. Journal of Advertising, 24, 25-35.
Cooper, R. (1993). An expanded, integrated model for determining audience
exposure to television. Journal of Broadcasting and Electronic Media, 37,
404-410.
Cooper, R. (1996). The status and future of audience duplication research: an
assessment of ratings-based theories of audience behavior. Journal of
Broadcasting and Electronic Media, 40, 96-111.
Donius, J. F. (1994, February 15-16). Brand eguity: A holistic perspective. Paper
presented at ARF Equity Workshop, New York.
Due South. (1997, June 2). Advertisement for “Due South” in Broadcasting and
Cable, p. 35.
Dupree, S. (1996, February 12). Network TV: Building better brands. Media
Week, p. 27

90
Eastman, S. T. (1997). Broadcast/cable programming: Strategies and
practices. Belmont, CA: Wadsworth.
Encyclopedia of science and technology (p. 136, p. 427). (1997). New York:
McGraw-Hill.
EQX. (1995). Strengthening the power of the brand. Simmons Market Research
Bureau [Brochure], New York.
Farquhar, P. H. (1989). Managing brand equity. Marketing Research, 1_, 24-33.
Fazio, R. H., Powell, M. C., & Williams, C. J. (1989), The role of attitude
accessibility in the attitude-to-behavior process. Journal of Consumer
Research, 16, 280-288.
Freeman, M., & Stanley, T. L. (1996, June 24). Angling for promotion. Network’s
and syndicators vie for valuable hype time on stations. Media Week, p. 8.
Gimein, M. (1996, June 24). Sweeps contests up 55%. Media Week, p. 5.
Gleason, M. (1995). Time to perform, or perish: Consumers gain more power
to determine survivors on shelf. Advertising Age, p. 37.
Goddhart, G. J., Ehrenberg, A. S. C, & Collins, M. A. (1975). The television
audience: Patterns of viewing. Westmead, UK: Saxon House.
Greenwald, J. (1996, December 23). Reinventing Sears. Time, p. 54.
Headen, R. S., Klompmaker, J. E., & Rust, R. T. (1979). The duplication of
viewing law and television media schedule evaluation. Journal of
Marketing Research, 16, 333-340.
Hoyer, W. D, & Brown, S. P. (1990). Effects of brand awareness on choice for a
common, repeat-purchase product. Journal of Consumer Research, 17,
141-148.
Interbrand. (1994). Client sales [Brochure], New York: Author.
Kapferer, J. N. (1992) Strategic brand management. New York: The Free Press.
Kardes, F. R. (1994). Consumer judgment and decision processes. In R.S.
Wyer & T.K. Srull, Handbook of social cognition (2nd ed., pp. 399-465).
Hillsdale, NJ: Erlbaum.

91
Keller, K. L. (1993). Conceptualizing, measuring, and managing customer-
based brand equity. Journal of Marketing, 57, 1-22.
Kraus, S. J. (1995). Attitudes and the prediction of behavior: A meta-analysis of
the empirical literature. Personality and Social Psychological Bulletin, 21,
58-75.
Lehmann, D. R., & Winer, R. S. (1994). Analysis for marketing planning. Burr
Ridge, IL: Irwin Publishing.
Lin, C. A. (1995). Network prime-time programming strategies in the 1980s.
Journal of Broadcasting and Electronic Media, 39, 482-495.
Lintas. (1993). Lintas link brand equity audit and probe. In Russell & Lane,
Kleppner’s advertising procedure (p. 210). Englewood, NJ: Prentice Hall.
Littleton, C. (1996, March 11). New day for local news, daytime newscasts are
boosting brand awareness and bottom lines. Broadcasting and Cable,
p. 30.
Littleton, C. (1996, June 10). Murdoch bullish on over-the-air TV. Broadcasting
and Cable, p. 10.
Long, S. T. (1979). The development of the television network oligopoly. North
Stratford, NH: New Arno Press.
Mándese, J. (1993, May 17). Here come the newest brands: NBC, CBS, ABC.
Advertising Age, p. 1.
Mándese, J. (1995, March 25). In news growth phase, cable feeding on itself.
Advertising Age, p. S-1.
Marshall Marketing. (1996). Marshall Marketing and Communications client
sales brochure, personal letter to investigator and proprietary data from
1996 Orlando, Florida, market survey. Orlando, FL: Author.
Maturo, R. (1993, February 1). Maintaining brand loyalty during times of
marketing chaos. Paper presented at ARF Brand Equity Workshop, New
York.
McClellan, S. (1995, April 17). Advertisers say stations need brand identity.
Broadcasting and Cable, p. 61.
McConville, J. (1995a, July 17). CTAM focuses on cable competition: Wake-up
call to include sessions on branding. Broadcasting and Cable, p. 15.

92
McConville, J. (1995b, December 4). In branding, image is reality. Broadcasting
and Cable, p. 70.
McDowell, W.S. (1995, August). The ratings game: The ethics of hyping,
stunting, blunting, rigging, and other dirty tricks. Paper presented at
AEJMC Convention, Advertising Division, Washington, DC.
Miles, L. (1997, February 10). Sweeps under suspicion. Media Week, p. 14.
Morris, B. (1996, March 4). The brand’s the thing. Fortune, p. 72.
Myers. (1997, February 2). The Myers report: Media 2005. Electronic Media, pp.
28-29.
National Association of Television Program Executives. (1996). The branding
bug (Panel discussion). Annual Convention agenda. New York: Author.
National Association of Television Program Executives. (1997, May 28). Our
brand new worid (Televised seminar).
NBC. (1995, November). NBC, America’s leading network television brand (Full
page ad). Advertising age. New York: NBC Marketing Department.
Nedungadi, P. (1990). Recall and consumer consideration sets: Influencing
choice without altering brand evaluations. Journal of Consumer
Research, 17, 263-276.
Nielsen Audience. (1996). Television audience 1996 (Nielsen Media Research
Report). New York: Author.
Nielsen Cume. (1996). Nielsen station index for Orlando-Daytona Beach-
Melbourne, Florida. Average Weekly Cume Household Analysis for 1996.
New York: Author.
Nielsen Method. (1995). NSI methodology, technigues and data interpretation
(Nielsen station index supplement). New York: Author.
Norusis, M.M. (1996). SPSS guide to data analysis. Englewood Cliffs, NJ:
Prentice Flail.
NPD. (1995). Brand builder: Flow can I measure, track and manage brand
eguity? [Brochure], New York: The NPD Group, Inc.
Product Q [Brochure], (1995). New York: Product Evaluations, Inc.

93
Promax & BDA. (1996). Profiting from your powers of creativity [Convention
brochure], Los Angeles: Promotion and Marketing Professionals In
Electronic Media and Broadcast Designers Association.
Rathbun, E. (1995, June 12). Marketers take promax spotlight. Broadcasting
and Cable, p. 12.
Ries, A., & Trout, J. (1981). Positioning: The battle for your mind. New York:
McGraw-Hill.
Rosenstein, A. W., & Grant, A. E. (1997). Reconcepualizing the role of habit: A
new model of television audience activity. Journal of Broadcasting and
Electronic Media, 41, 324-344.
Rubin, A. M. (1984). Ritualized and instrumental television viewing. Journal of
Communication, 34, 67-77.
Rubin, A. M. (1994). Media uses and effects: A uses-and-gratifications
perspective. In J. Bryant & D. Zillman (Eds.). Media effects: Advances In
theory and research (pp. 417-436). Hillsdale, NJ: Lawrence Erlbaum.
Sharkey, B. (1997, July 21). Living in The Real World’. Media Week, p. 17.
Shell, A. (1997, May 12). Brand loyalty? Ad Week, p. 40.
Sims, J. (1997, April 14). How cable grabbed the prime time audience.
Electronic Media, p. 37
Spring, A. (1997, December 22). FX will brand it image through originals.
Electronic Media, p. 10.
Springsteen, B (1989). 57 Channels and nothin' on. Lyrics from song.
Produced and recorded by B. Springsteen licensed through ASCAP.
Srivastava, R. K., & Shocker, A. D. (1991). Brand eguity: A perspective on its
meaning and measurement. Cambridge, MA: MSI.
Stay tuned. (1995, June 12). Broadcasting and Cable, p. 58.
Stern, C. (1996, February 5). New law of the land. Broadcasting and Cable, p. 8.
Thierer, A. D. (1995, January). Why mass media mergers are meaningless.
The FreeMan: Ideas on Liberty, p.5.

94
Hedge. J. T., & Ksobiech. K. J . (1986). The “lead-in” strategy for prime-time TV:
Does it Increase the audience? Journal of Communication, 36(2), 51-63.
Tiedge. J. T., & Ksobiech. K. J. (1988). The sandwich programming strategy: A
case of audience flow. Journalism Quarterly, 65, 376-383.
Tobenkin. D. (1994, October 10). New life for local news: News stock rises with
affiliation switch opportunities and increased emphasis on brand equity.
Broadcasting and Cable, p. 68.
Total Research Corporation (1995). Equitrend [Brochure]. New York: Author.
Walker, J. R. (1988). Inheritance effects in the new media environment. Journal
of Broadcasting and Electronic Media, 4, 391-401.
Walley, W. (1995, December 12). Getting along famously: How A&E's
“Biography” helps brand the network identity. Electronic Media, p. 30.
Weather. (1997, December 22). Classified ad placed by The Weather Channel.
Broadcasting and Cable, p. 54.
Webster, J. G. (1985). Program audience duplication: A study of television
inheritance effects. Journal of Broadcasting and Electronic Media, 29,
121-133.
Webster, J. G., & Lichty, L. W. (1991). Ratings analysis: Theory and practice.
Hillsdale, NJ. Lawrence Erlbaum.
Wilkie , W. L. (1996). Consumer behavior. New York: John Wiley & Sons.
Yi, Y. (1990). A critical review of consumer satisfaction. In V. A Zeithaml (Ed).
Review of marketing 1990 (pp. 68-123). Pittsburgh, PA: University of
Pittsburgh.
Yin, R. K. (1994). Case study research. Design and methods. New York: Sage.

BIOGRAPHICAL SKETCH
After earning a master's degree from Syracuse University in 1968, Walt
McDowell began a professional broadcasting career that would span almost
23 years, including over a dozen years as Director of Marketing and Creative
Services, ABC affiliate WFTV, in Orlando, Florida. His college teaching
experience dates back to 1989 when he joined the faculty of the University of
Central Florida as an adjunct instructor. While working his “day job” at WFTV,
he taught evening classes for the UCF School of Communication. In 1994, after
earning a second master's degree (this time from UCF), he entered the Ph D.
program at the University of Florida, Gainesville, where he taught several
telecommunications courses. In the fall of 1997 he joined the faculty of
Southern Illinois University at Carbondale as an assistant professor in the
College of Mass Communication and Media Arts.
95

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor of Philosophy.
C.
Mhn C. Sutherland, Chair
¿-Professor of Journalism and
Communications
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor of Philosophy.
'"Deborah Treise_, 7
Associate Professor of Journalism and
Communications
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degreerof Doctor of Philosophy.
David H. Ostroff /
Professor of Journalism and
Communications
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor of Philosophy.
John W. Wrigh
Professor of Journalism and
Communications

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope
and quality, as a dissertation for the degree of Doctor of Phjiosophy.
Barton A. Weitz
J.C. Penney Eminent Scholar of
Marketing
This dissertation was submitted to the Graduate Faculty of the College of
Journalism and Communications and to the Graduate School and was
accepted as partial fulfillment of the requirements for the degree of Doctor of
Philosophy.
August 1998
2b
Dean, College of Journalism and
Communications
Dean, Graduate School



xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EXMA9X4AA_9QO1CP INGEST_TIME 2014-08-27T22:27:24Z PACKAGE AA00024979_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES