Attitude toward the online advertising format

MISSING IMAGE

Material Information

Title:
Attitude toward the online advertising format a reexamination of the attitude toward the ad model in an online advertising context
Physical Description:
xvii, 264 leaves : ill. ; 29 cm.
Language:
English
Creator:
Burns, Kelli Suzanne
Publication Date:

Subjects

Subjects / Keywords:
Journalism and Communications thesis, Ph. D   ( lcsh )
Dissertations, Academic -- Journalism and Communications -- UF   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2003.
Bibliography:
Includes bibliographical references.
Statement of Responsibility:
by Kelli Suzanne Burns.
General Note:
Printout.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 030435051
oclc - 80381838
System ID:
AA00013638:00001


This item is only available as the following downloads:


Full Text









ATTITUDE TOWARD THE ONLINE ADVERTISING FORMAT:
A REEXAMINATION OF THE ATTITUDE TOWARD THE AD MODEL IN AN
ONLINE ADVERTISING CONTEXT

















By

KELLI SUZANNE BURNS


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
































Copyright 2003

by

Kelli Suzanne Bums































Dedicated in loving memory to my grandmother














ACKNOWLEDGMENTS

I would like to acknowledge the dedicated professors at the University of Florida

who guided me throughout this dissertation. I would like to thank Dr. Richard Lutz, who

went above and beyond his duties as a professor in the Warrington College of Business to

supervise and cochair a dissertation for a communications student. His intellectual

contributions and financial resources were critical to the successful completion of this

dissertation and are greatly appreciated. I would like to thank my dissertation chair, Dr.

John Sutherland, who contributed to this study through his incredible facility for data

analysis and knowledge of research methods. Committee members Dr. Marilyn Roberts,

Dr. Joseph Pisani, and Dr. Bart Weitz also deserve commendation for their feedback,

support, and participation.

My appreciation extends to Dr. David Eason and Dr. Bob Wyatt for inspiring me as

a graduate student at Middle Tennessee State University and for later hiring me.

I would like to thank my husband, Corey, who believed in me and supported my

dream. I am also grateful to John and Anne Berg, who generously and warmly welcomed

me into their home and lives for two years. I would like to thank my parents for instilling

in me all the values that made this possible and for giving me so many opportunities to

lead a fulfilling life. Finally, I would like to acknowledge my son, Griffin, who arrived

during this journey and added to the adventure.















TABLE OF CONTENTS
Page

ACKNOWLEDGMENTS ............................................................................ ..................... iv

L IST O F TA B LE S ...................... ......................... ...................................................... x

L IST O F F IG U R E S ................................................................................ ....................... xv

A B STR A C T ........................... ......................................................... ..................... xvi

CHAPTER

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

The Origins of Online Advertising................................................... ......................2...
Online Advertising Today ................. ...........................................................6......
Current Online Advertising Formats ...............................................................6...
Demand for New Online Advertising Formats ................................................. 11
Online Advertising Mix...................... ... .........................12
Online Advertising Spending ............................ ...................................................13
Online A advertising Effectiveness.................................................. ........................ 15
Statement of Purpose ............... ........................................... 18
Im portance of the Study................................................................ ......................... 19
O outline ..................................................................................... ........... ............ 20

2 LITERATURE REVIEW........... ................................................................................21

Attitude Toward the Ad........................... .........................23
Attitude Toward the Ad as a Mediator...........................................................24
L ikability Studies ..................................................................... ....................... 26
Attitude Tow ard the Ad M odel ............................................... ....................... 27
Attitudes Toward Advertising in General..............................................................31
Beliefs About Advertising in General ...................................................................34
The Relationship Between Beliefs and Attitudes..............................................35
Categorizing Beliefs About Advertising ...........................................................36
Belief and Attribute Dimensions Included in Previous Studies.......................37
Attitudes Toward Advertising in a Specific Media Vehicle.................................... 48
Online Advertising Effectiveness....................................................................51
Effectiveness of Executional Elements ................................ .................... 51
Online Consumer Behavior as a Measure of Effectiveness .............................52
Attitudes Toward Online Advertising ...................................................................53








Proposed M odel. ............................... ....................................... ........................ 59
Conclusion .......................................................... .................................................. 61

3 STUD Y 1 ........................................................... ................................................... 63

Purpose ............................................................................................ ..................63
Research Questions........................................................................ ........................ 63
Critique of M ethodology in Previous Research................................. .................... 63
M ethod ............................................ ......................................... ...................................64
Depth Interview s W ith Industry Experts................................. ....................... 64
Depth Interview s W ith Experienced Online U sers .......................................... 66
Procedure for Selecting Online A advertising Form ats............................................... 68
Form at Selection Results................................................................ ....................... 69
Perceptual D im tensions of Online Advertising Form ats.............................................74
Irritation ..................................... ............ ...... ..................... 74
Entertainm ent ................................................... ............................................. 77
Inform action ..................................................... ............................................... 78
Novelty .............................................................. ............................................ 79
Interactivity............................................... 79
Com position ............................................................................... ...................... 80
D iscussion............................................................................. ................................ 82
Online Advertising Form ats ................ ................................................. 82
Perceptual Dim tensions ...................................................... ....................... 83

4 STUD Y 2 ........................................................................................ 86

Purpose ........................................................................................... ...................86
Hypotheses.......................................................... .................................................. 89
M ethod ........................................................................................ .......................90
Sam ple ......................................................... .................................................. 90
Procedure............................................................................ ........................... 90
Stim ulus A ds ................................................... .............................................. 91
M easures....................................................... ................................................. 91
Results.................... .......................... .................................................... 94
Sam ple Description ............................ ........................................................ 94
Overall D ata Structure................................ ..................................................... 97
V erification of M measures ............... ..................................................... 98
Test-Retest Reliability............................................................ ........................ 99
Factor Analysis of Perceptual Items and Attitude Measures...........................100
Correlations Am ong Variables for Each Form at............................................. 107
Predictors of Attitude Toward the Online Ad Format....................................... 113
Predictors of Attitude Toward the Ad .......................... ............................. 120
Behavioral M measures ........................... ...................................................... 123
D iscussion......................................................... .................................................. 127
Overview ..................................................... ................................................ 127
Interpretation of Results ........................... .................................................... 128
Study Lim stations ............... .................................... 131


vi








Im plications for Online Advertising Theory................................................ 135
Im plications for Online Advertising Industry................................................ 135
Future Research .................................................. ........ .......................... 135
Conclusion ......................... ........................................................................... 136

5 STUDY 3 ....................................................................................... .....................137

Purpose .............................. ................................................... ........................... 137
Hypotheses........................................................ .................................................. 137
M ethod ............................... ................................................... ........................... 138
Sam ple ........................................................ ................................................. 138
Procedure ...................................................... ............................................... 139
Stimulus Ads .................................................. ............................................. 140
M easures...................................................... ................................................ 140
Results............................... ............................................................................... 142
Sam ple ................................................................................... ...................... 142
Attitude Toward Online Advertising............................................................... 150
Perceptions of and Attitude Toward Online Advertising Format ...................150
Relationships am ong Variables ............................................. ....................... 153
Behavioral M oderators .................. ....................................................... 156
Discussion......................................................... .................................................. 162
Overview ..................................................... ................................................ 162
Interpretation of Results ................................................ 162
Study Lim stations ................................................................. ....................... 163
Im plications for Online Advertising Theory...................... .......................... 166
Im plications for Online Advertising Industry ................................................. 167
Future Research ................... ............ ........... ....................... 168

6 IM PLICATION S ................................................... ............................................. 170

Introduction........................................................ ................................................. 170
Results Overview ................................................... ............................................. 172
Future Research .................................................... .............................................. 174
Conclusion ........................................................ .................................................. 176

APPENDIX

A INTERVIEW GUIDE FOR ONLINE ADVERTISING EXPERTS........................177

Screener ..............................................................................................................1...... 77
Inform ed Consent .................................................. ............................................. 177
Interview Guide ................................................................. ................................. 179

B INTERVIEW GUIDE FOR EXPERIENCED WEB SURFERS .............................181

Screener ................................................................................................................... 181
Inform ed Consent .................................................. ........................................ .....182


vii








Interview G uide .................................................... .............................................. 183

C ONLINE SURVEYS FOR STUDY 2 .................................... .......................... 189

V version A ......................................................... ................................................... 189
Introduction ................ .........................................189
Informed Consent .....................................................................................189
E xtra C credit ................................................... .............................................. 19 1
Surfing the Web......................................................................................... 191
Advertising on the Web................ ...................................................... 191
First Online Ad ................. ..................................... 192
Second Online Ad................. ......................................................... 195
Third Online Ad .......................................................................................198
Demographics........................................................................ ....................201
C conclusion ..................................................... .............................................. 202
V version B .......................................................... .................................................. 202
Introduction ................................................... .............................................. 202
Informed Consent ....................................................................................202
E xtra C credit ................................................... .............................................. 204
Surfing the W eb ................................................ ........................................... 204
Advertising on the Web.............. ....................................................... 204
First Online Ad ................. ....................................205
Second Online Ad...................................................................................... 208
Third Online Ad ................................... ........................ .........................211
D em ographics.................................................. ............................................ 215
C conclusion ..................................................... .............................................. 2 16

D ONLINE SURVEY FOR STUDY 3 ..................................... ........................... 217

Invitation ................................. ............................................................................ 2 17
Survey .............................................................................................. ............... .2 17
Introduction ......................................................... ........................................ 2 17
Informed Consent ...................................................................................... 218
E search.com ID ...................................................... ..................................... 219
Advertising on the Web................ ...................................................... 219
Format Introduction................. .......................................................... 220
First Online Ad Format ................ ...................................................... 221
Second Online Ad Format................. ................ ...... ................... 223
Third Online Ad Format.................. ..................................................... 225
Progress Report ......................................................................................... 227
Fourth Online Ad Format................... ..................................................... 227
Fifth Online Ad Format................ ........................................................229
Sixth Online Ad Format ............... ..................................................... 231
D em ographics.................................................. ............................................ 233
C conclusion ..................................................... .............................................. 235




viii









E FACTOR ANALYSES FOR PERCEPTUAL AND ATTITUDINAL MEASURES
BY ONLINE AD FORMAT .......................... .....................................................236

Analyses for Banner Ads............................. ..................................................... 236
Analyses for Pop-up Ads ...................... ...................................... ................... 238
Analyses for Skyscraper Ads................ ...................................................... 241
Analyses for Large Rectangle Ads ....................................... ............................243
Analyses for Floating Ads ................. ......................................................... 246
Analyses for Interstitial Ads ................ ........................................................ 248

LIST O F R EFEREN C ES ................................................ ........................................... 251

BIOGRAPHICAL SKETCH ................ ..............................................................264







































ix














LIST OF TABLES


Table page

3-1 Advertisers Represented in Stimulus Ads.................................... ....................... 67

3-2 Unaided Recall and Recognition of Online Ad Formats by Experienced Web
Surfers (N = 10)..................... .... .. ... ....................... 70

3-3 Summary of Performance of Chosen Formats across Selection Criteria..............73

3-4 Descriptors of Online Ad Formats Used by Online Advertising Experts................75

3-5 Descriptors of Online Ad Formats Used by Experienced Web Surfers................76

3-6 Summary of Dimensions and Corresponding Descriptors....................................81

4-1 Demographic Characteristics of Respondents.....................................................95

4-2 Internet Usage Characteristics of Respondents...................................................96

4-3 Online Purchase Behaviors of Respondents........................................................97

4-4 Verification of Common Attitude Measures.......................................................99

4-5 Mean Ratings for Attitude Indices for Respondents who Completed Both Versions
(n = 69 ) ............................................................ ................................................. 10 0

4-6 Summary of Factor Loadings for the Rotated Three-Factor Solution for Perceptual
Item s ...................................... .......................................................... ........... 102

4-7 Means, Standard Deviations, and Alphas for Three-Factor Solution...................103

4-8 Mean Scores for Attitude Indices (with Coefficient Alpha) for Each Online Ad
F o rm at .................................................................................................................... 104

4-9 Mean Scores for Perceptual Factor Indices (with Coefficient Alpha) for Each
O line A d Form at........................ ............................................................ 105

4-10 One-Way Analyses of Variance for the Effects of Format on Attitude and
Perceptual Factor Indices ............................. ..................................................... 106








4-11 Post-Hoc Comparisons of Attitude and Perceptual Factor Means Across Format
U sing L SD ........................... ............................................................................. 107

4-12 Intercorrelations for Attitude and Perceptual Factor Indices for Banner Ads........108

4-13 Intercorrelations for Attitude and Perceptual Factor Indices for Pop-Up Ads.......109

4-14 Intercorrelations for Attitude and Perceptual Factor Indices for Skyscraper Ads .110

4-15 Intercorrelations for Attitude and Perceptual Factor Indices for Large Rectangle
A d s .................................................... ...... .................... ....... ...... 11

4-16 Intercorrelations for Attitude and Perceptual Factor Indices for Floating Ads......112

4-17 Intercorrelations for Attitude and Perceptual Factor Indices for Interstitial Ads... 113

4-18 Regression Analysis Summary for Variables Predicting Attitude Toward Banner
A d s ...................................................................................... ...................... ... .. 1 15

4-19 Regression Analysis Summary for Variables Predicting Attitude Toward Pop-Up
A ds ....................... .......................................................... ........................ 115

4-20 Regression Analysis Summary for Variables Predicting Attitude Toward
Skyscraper A ds............................................................................... ..................... 116

4-21 Regression Analysis Summary for Variables Predicting Attitude Toward Large
R ectangle A d s ....................................................................................................... 1 17

4-22 Regression Analysis Summary for Variables Predicting Attitude Toward Floating
A d s ......................................................................................................................... 1 1 7

4-23 Regression Analysis Summary for Variables Predicting Attitude Toward Interstitial
A ds .......................... ......................................................... ................... 118

4-24 Summary of Significance of Predictor Variables Across All Formats ................118

4-25 Regression Analysis Summary for Variables Predicting Attitude Toward Banner
A d s ....................................................................... .......... ............ ............................ 12 1

4-26 Regression Analysis Summary for Variables Predicting Attitude Toward Pop-Up
A d s ...................................................................... ............ ........... ........................... 12 1

4-27 Regression Analysis Summary for Variables Predicting Attitude Toward
Skyscraper A ds.......................................................................... ....................... 122

4-28 Regression Analysis Summary for Variables Predicting Attitude Toward Large
R ectangle A ds ........................................................................... ....................... 122








4-29 Regression Analysis Summary for Variables Predicting Attitude Toward Floating
A d s .................................................................... ... .......... ........... ............................. 12 2

4-30 Regression Analysis Summary for Variables Predicting Attitude Toward Interstitial
A d s ......................................................................................................................... 12 3

4-31 One-Way Analyses of Variance for Effects of Format Familiarity on Aforma....... 124

4-32 Post-Hoc Comparisons of Aiormat Means Across Format Familiarity Using LSD .125

4-33 Group Differences for Attitude Toward Online Ad Format Between Respondents
Who Had Clicked Through on Certain Online Ad Formats and Respondents Who
Had Not Clicked Through..................................................... .......................... 126

4-34 Post-Hoc Comparisons of Aforat Means Across Clickthrough Frequency Using
L SD ........................ ............................................................................ .......... 126

4-35 Group Differences for Aforat Between Respondents Who Had Later Visited Sites
Advertised Using Certain Online Ad Formats and Respondents Who Had Not Later
V isited Sites........................................................................... 127

4-36 Regression Analysis Summary for Attitude Toward Large Rectangle Ads for First-
T im e R espondents................................................................. ............................ 133

4-37 Regression Analysis Summary for Attitude Toward Large Rectangle Ads for
Second-Tim e R espondents.................................................... ............................ 133

4-38 Post-Hoc Comparisons of Mean Familiarity Scores Across Formats Using LSD.134

5-1 Gender, Age, Race, and Marital Status of Respondents ......................................143

5-2 Education, Income, and Employment of Respondents ....................................... 146

5-3 Geographic Region of Residence of Respondents............................................ 148

5-4 Internet Usage Characteristics of Respondents.................................................... 149

5-5 Online Purchase Behaviors of Respondents........................ ........................... 150

5-6 Coefficient Alphas for Measures Across Formats ............................................151

5-7 ANOVA for Formats on Attitude and Perceptual Item Factor Indices................152

5-8 Post-Hoc Comparisons of Attitude and Perceptual Format Means Across Formats
U sing LSD ................................................................................................... 153

5-9 Regression Analysis Summary for Perceptual Factors Predicting Attitude Toward
B manner A ds ................................ ..........................................154








5-10 Regression Analysis Summary for Perceptual Factors Predicting Attitude Toward
P op-U p A ds ............................... ....................................................................... 154

5-11 Regression Analysis Summary for Perceptual Factors Predicting Attitude Toward
Skyscraper A ds.............................................................................. ...................... 155

5-12 Regression Analysis Summary for Perceptual Factors Predicting Attitude Toward
Large R ectangle A ds ......................... ............................................................... 155

5-13 Regression Analysis Summary for Perceptual Factors Predicting Attitude Toward
F loating A ds ............................................................................. ......................... 155

5-14 Regression Analysis Summary for Perceptual Factors Predicting Attitude Toward
Interstitial A ds ......................................................................... .......................... 156

5-15 Group Differences for Aformat Between Early and Late Internet Adopters.............157

5-16 Group Differences for Aformat Between High-Speed Internet Access and Low-Speed
Internet A access Respondents................................................... .......................... 158

5-17 Group Differences for Aformat Between Respondents Who Had Clicked Through on
Certain Online Ad Formats and Respondents Who Had Not.................159

5-18 One-Way Analyses of Variance for Effects Format Familiarity on Aformat............160

5-19 Post-Hoc Comparisons of Aformat Means Across Format Familiarity Using LSD .161

5-20 Group Differences for Afonmt Between Respondents Who Had Made an Online
Purchase and Those Who Had Not .................................... ............................161

E-I Rotated Component Matrix of Perceptual Items for Banner Ads........................236

E-2 Means, Standard Deviations, and Alphas for Banner Ad Factors........................237

E-3 Factor Analysis of Aad for Banner Ads ................................. .........................237

E-4 Factor Analysis of Afomat for Banner Ads ............................ ............................ 237

E-5 Factor Analysis of Asnie for Banner Ads ................................ ......................... 238

E-6 Rotated Component Matrix of Perceptual Items for Pop-up Ads ........................238

E-7 Means, Standard Deviations, and Alphas for Pop-up Ad Factors........................239

E-8 Factor Analysis of Aad for Pop-Up Ads .............................. ...........................239

E-9 Factor Analysis of Afom t for Pop-up Ads............................................................ 240

E-10 Factor Analysis of Aste for Pop-up Ads............................... .......................... 240








E-ll Rotated Component Matrix of Perceptual Items for Skyscraper Ads..................241

E-12 Means, Standard Deviations, and Alphas for Skyscraper Ad Factors..................241

E-13 Factor Analysis of Aad for Skyscraper Ads......................... ...........................242

E-14 Factor Analysis of Afoma, for Skyscraper Ads ...................... ..........................242

E-15 Factor Analysis of Asi for Skyscraper Ads......................... ...........................243

E-16 Rotated Component Matrix of Perceptual Items for Large Rec Ads ...................243

E-17 Means, Standard Deviations, and Alphas for Large Rec Factors.........................244

E-18 Factor Analysis of Aad for Large Rec Ads .......................... ...........................244

E-19 Factor Analysis of Aformat for Large Rec Ads....................... ...........................245

E-20 Factor Analysis of Asi, for Large Rec Ads .......................... ..........................245

E-21 Rotated Component Matrix of Perceptual Items for Floating Ads ......................246

E-22 Means, Standard Deviations, and Alphas for Floating Ad Factors......................247

E-23 Factor Analysis of Aad for Floating Ads.............................. ........................... 247

E-24 Factor Analysis of Aformt for Floating Ads........................... .........................247

E-25 Factor Analysis of Asite for Floating Ads ............................ ........................... 248

E-26 Rotated Component Matrix of Perceptual Items for Interstitial Ads ...................248

E-27 Means, Standard Deviations, and Alphas for Interstitial Ad Factors...................249

E-28 Factor Analysis of Aad for Interstitial Ads .......................... ...........................249

E-29 Factor Analysis of Afomat for Interstitial Ads....................................................... 250















LIST OF FIGURES


Figure page

2-1 Modified Structural Model of Aad Formation .....................................................28

2-2 Proposed Structural Model of Aad Formation (showing two antecedents) in an
O nline A advertising Context ......................................................... ....................... 60

4-1 Modified Attitude-Toward-the-Ad Model for Testing Relationships in the
O nline C ontext .......................................................................... .......................... 89

4-2 Modified Attitude-Toward-the-Ad Model for the Online Context Using Shading to
Indicate the First Set of Relationships to be Tested............................................1...14

4-3 Modified Attitude-Toward-the-Ad Model for the Online Context Using Shading to
Indicate the Second Set of Relationships to be Tested.........................................120

4-4 Significant Relationships in the Modified Attitude-Toward-the-Ad Model for the
O line C context .......................................................................... ....................... 13 1

5-1 Comparison of Formats Across Perceptual Descriptors ......................................164














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

ATTITUDE TOWARD THE ONLINE ADVERTISING FORMAT:
A REEXAMINATION OF THE ATTITUDE TOWARD THE AD MODEL IN AN
ONLINE ADVERTISING CONTEXT

By

Kelli Suzanne Bums

August 2003

Chair: John C. Sutherland
Cochair: Richard J. Lutz
Major Department: Journalism and Communications

This dissertation introduced a new construct-attitude toward the online

advertising format (Aformat)-and demonstrated its relevance in the attitude-toward-the-ad

(Aad) model.

As the online advertising environment becomes more cluttered, technological

possibilities expand, and expenditures show improvement, an understanding of

consumers' attitudes toward the various online advertising formats is critical.

In Study 1, depth interviews with Web surfers and online advertising experts

identified 15 perceptions of online advertising formats. The interviews also determined

six online advertising formats for future study including banners, pop-ups, skyscrapers,

large rectangles, floating ads, and interstitials.








In Study 2, a survey with a student sample was used to determine the influence of

Aformat on Aad and the ability of the perceptions identified in the first study and other

variables to predict Aformat.

The regression equation revealed that while Af&onat was a significant predictor of

Aad, attitude toward online advertising was not. The influence of Aformat on Aad is

particularly important given that Aad is a documented precursor of brand attitude, brand

choice, and purchase intentions.

Online ad perceptions were found to be related to Aformat for all six online ad

formats tested. The formats differed in terms of the specific perceptions that were

significantly correlated with attitude toward each format. The other hypothesized

predictors of Armat (i.e., attitude toward online advertising, attitude toward the Web site,

and attitude toward the Internet) were found to be either significantly correlated with only

certain formats or not significantly correlated at all.

Study 3 produced descriptive data on Afonna, using a national survey of 1,075 adults.

This study also determined how each format was rated on the perceptual dimensions and

tested the ability of perceptions to predict A&nrmat.

The data supported the three hypotheses of this study. Web users possess

significantly different attitudes across formats. Users also hold a varied combination of

perceptions about each format. Furthermore, the three perceptions of entertainment,

annoyance, and information have a significant impact on Aformat.













CHAPTER 1
INTRODUCTION

If you don't get noticed, you don't have anything. You just have to be noticed, but
the art is in getting noticed naturally, without screaming or without tricks.
-Leo Burnett, The Art of Writing Advertising

In 2001, the notoriously ubiquitous online ads for X10 wireless cameras garnered

much attention in trade publications and caused frustration and annoyance for many Web

users. Using an ad contained within its own window that "popped under" the user's

browser window, X10 Wireless Technologies was successful in achieving mass reach

online (32.8% of the Web's entire audience between January 2001 and May 2001) and a

total of 28 million unique visitors by the end of May 2001 (Jupiter Media Metrix, 2001b).

According to Nielsen//NetRatings, 388,000 unique visitors entered the X10 Web site the

month prior to the campaign launch (Mediapost, 2001). By May 2001, the number of

unique visitors for the month had climbed to 3.5 million (Mediapost, 2001). However, the

company experienced a steep decline in traffic with 73% of unique visitors leaving the

site or closing the window within 20 seconds (Jupiter Media Metrix, 2001b). Only 4.2%

of visitors spent more than three minutes on the site (Jupiter Media Metrix, 2001b).

While this campaign was successful in increasing the number of unique visitors to

the site, several indicators suggest it failed to enhance the company's image. These ads

generated negative publicity for the company and were met with disapproving reactions

by consumers (Olsen, 2001). In addition, chat rooms discussions focused on how to

disengage these ads and many articles were written describing how to restrain X10's

efforts (Mediapost, 2001). Downloads of the ad-blocking software AdSubtract have been








on the rise since the launch of this campaign (Taylor, 2001) and were predicted to reach 2

million users by the end of 2001 (Lefton, 2001a).

Yet, X10 Wireless Technologies remained undeterred in its advertising approach.

One year later, the company was the leading pop-up advertiser with over one billion pop-

up impressions in the first seven months of the year (Martin & Ryan, 2002).

The X10 example suggests the possibility of a relationship between the online

advertising format and consumer attitude toward that ad itself. However, before this

relationship can be explored, it is important to consider the specific dimensions or

perceptions of online ads that contribute to the attitude a consumer has about an ad

format. The intrusiveness of the pop-under ad, for example, may constitute a dimension

that shapes consumer attitudes toward pop-under ads.

Pop-under ads are just one of many online advertising formats. A review of the

origins of online advertising will illustrate how the efforts of a few advertisers, content

providers, and programmers set into motion what is today a multi-billion dollar industry

teeming with a variety of advertising formats.

The Origins of Online Advertising

The Internet is a product of ARPANET, which was launched by the Advanced

Research Projects Agency of the U.S. government in 1969 to connect research computers

at universities (Public Broadcasting Service [PBS], 1997). By 1971, ARPANET grew to

23 hosts, up from the original group of four host universities (PBS, 1997). In the 1980s,

Vint Cerf, known as the "Father of the Internet," and Bob Kahn created TCP/IP, which is

the language shared by Internet computers, and soon after, the ARPANET computers

began to be referred to as the Internet (PBS, 1997). The explosion of the personal

computer industry in the 1980s helped spur the use of the Internet in corporate America








(PBS, 1997). Today, the Internet consists of a worldwide system of networked computers

accessible to hundreds of millions of people.

The World Wide Web' was invented in 1991 by Tim Bemers-Lee while he was

working for CERN, the European Particle Physics Laboratory, in Switzerland (PBS,

1997). The code he posted in a newsgroup allowed users to combine text, images, and

sound on Web pages and easily publish information on the Internet (PBS, 1997).

The development of Mosaic, a user-friendly, graphical interface browser, by Marc

Andreessen and others at the University of Illinois in 1993 helped move a mass audience

online. Mosaic incorporated the new programming language Hypertext Markup

Language (HTML). When Mosaic was released, Internet traffic increased at an annual

growth rate of 341,634% (PBS, 1997). The emergence of several national commercial

Internet service providers in the early 1990s (i.e., America Online, CompuServe, and

Prodigy) also contributed to the astronomical growth rate. As more and more consumers

went to the Web, this medium became even more enticing to advertisers.

Some early browsers only supported a text-based interface, containing no graphics,

while browsers with graphical interfaces were not always used with a high-speed modem.

Although advertisers dreamed of sending television commercials over the Internet, the

necessary bandwidth was not available at the time (Koprowski, 1999), and graphical

interfaces were not widely used by consumers. Under these constraints, online

advertising naturally evolved into a form of advertising similar to that of print media.




I While often used interchangeably, the Internet and World Wide Web are not the same. The World Wide
Web is the term used to describe the collection of documents that are linked to one another through the use
of Hypertext Markup Language (HTML) and are hosted by computers connected to the Internet.








Online advertising emerged during the summer of 1994 when Coors Brewing

Company and Modem Media launched the first national consumer brand Web site for

Zima beer (Koprowski, 1999). Around the same time, McDonald's sponsored an online

chat on America Online (Koprowski, 1999).

Wired magazine's online magazine Hot Wired drew serious attention from the

advertising community (Koprowski, 1999). As the first online magazine supported by

advertising, Hot Wired was introduced in 1994 with 12 advertisers paying $30,000 for 12

weeks (Koprowski, 1999). Advertisers included AT&T, Sprint, MCI, and Volvo

(DiBlasi, 1997). While HotWired has often been credited for offering the first banner

advertisement, the first online banner ad may have been hosted by Prodigy and its origin

may be further traced back to previous videotext and teletext services (Dolley, 1998;

Elwell, 1998; Williamson, 1998).

Also launched in 1994 was Time Wamer's Pathfinder site, which included access

to magazines and featured test ads from AT&T (DiBlasi, 1997). Ziff Davis launched

ZDNet the same year. Both sites acquired their first advertisers a year later.

The spring of 1995 saw the explosion of Web sites for major national brands from

Maytag to United Airlines to Ragu. These advertisers used banner ads to lure customers

to their sites (Koprowski, 1999). Also in 1995, the Procter & Gamble online advertising

account was awarded to Grey Interactive and Modem Media acquired the AT&T account

(DiBlasi, 1997).

About the same time, Oldsmobile developed commercial chat rooms to provide a

showcase for the Oldsmobile brand using celebrities to attract consumer attention. Again,

banner ads were used to publicize the chat rooms (Koprowski, 1999). Also in 1995,








America Online changed its anti-advertising policy and became more receptive to online

advertising. The company began to accept ads from advertisers such as DLJdirect,

General Motors, and 1-800-FLOWERS on "sponsored sites" (Koprowski, 1999).

Introduced in 1995, the Java computer language provided advertisers with the

technology for creating more elaborate graphics, audio, and animation (Koprowski,

1999). AT&T used this technology in 1996 to create animated banner ads (DiBlasi,

1997).

The dominance of banner advertising and lack of consistency in banner ad sizes led

the Interactive Advertising Bureau (IAB), formerly the Intemet Advertising Bureau, a

nonprofit trade association founded in 1996, to adopt standards for online banner sizes

(IAB, 1996). This development ensured advertisers would not have to redesign their

banner ads to meet the specifications of individual sites.

While banners were receiving attention from advertisers and the LAB, consumer

interest in clicking through banner ads dropped significantly, from a 40% clickthrough

rate in 1995 and 1996 to less than 1% in 1997 (Koprowski, 1999). In 1996, Yahoo!

agreed to allow Procter & Gamble to pay for advertising space on the basis of

clickthrough rates rather than ad impressions (Williamson, 1996). The decline in

clickthrough rates on banner ads certainly challenged advertising agencies to rethink the

purpose and format of online advertising.

In recent years, faster modem speeds have enabled the application of new

technologies to online graphics, audio, and animation. The online interface can now have

the same look and feel as television. Subsequently, online advertising has evolved to

incorporate formats analogous to television commercials.








Online Advertising Today

Ads seem to be popping up and under everywhere online. Some of the more recent

and creative online advertising formats include floating ads (often referred to by the

trademarked name ShoshkeleTM ads), which resemble cartoons floating over text, and

Webmercials, which feature animation or streaming media to transmit an advertisement

that uses video and audio. Another recent online innovation is the use of large rectangle-

shaped advertisements wrapped by text. Sliding billboards are also being used by Web

sites such as USAToday.com (Elkin, 2002c). These ads appear as a large square when

first accessing a Web page and then slide up into a smaller banner ad. Another new

format is the full-screen ad introduced in 2003 by Unicast.

Current Online Advertising Formats

Below is an alphabetical list of current online advertising formats derived from

articles on online advertising, the Interactive Advertising Bureau's (2001c) Glossary of

Interactive Advertising Terms, and personal experience.

Banners are rectangular-shaped graphical elements often found at the top of Web

pages. The IAB guidelines are 468 x 60 pixels for a standard banner and 234 x 60 pixels

for a half banner (IAB, 2001a). Banners can be static or employ animated graphics to

capture the user's attention. Clicking on a banner directs the user to another Web page.

Buttons are clickable graphics that are similar to banners, but are smaller, often

shaped as a square, and placed anywhere on a page. The IAB guidelines are 125 x 125

pixels for a square button and 120 x 90 or 120 x 60 pixels for rectangular buttons (IAB,

2001a). At a size of 88 x 31 pixels, microbars can also be placed in this category (IAB,

n.d.)

Chat room advertising provides advertisers with access to chat room participants.








Classified ads are well-suited to the interactive environment. Users input their

requests and are presented with the appropriate listings to match their criteria.

Contest sponsorships are provided by content providers to allow advertisers to

sponsor and promote a contest.

Contextual advertising highlights words within the text of a Web page that are

hyperlinks to an advertiser. Toptext, developed by Ezula, is one of the technologies used

to highlight and provide the links. This program is unknowingly downloaded by the user

when downloading other software. Other programs for delivering contextual advertising

include SurfPlus and AdPointer.

Daughter window ads open in a separate browser window at the same time a

banner is displayed on the Web page of the primary browser window.

E-mail can be ad-supported by selling advertisements on the e-mail reader or home

page. The service then provides a free e-mail system to users. Two examples of free e-

mail services are Hotmail and Juno. Advertisers may also send direct e-mail to

consumers. Direct e-mail marketing is classified as either permission marketing, whereby

the recipient has provided permission to receive e-mail, or spam, which is unsolicited e-

mail. The e-mails are often formatted to resemble a letter, a newsletter, or a version of the

marketer's Web site. Rich media can also be incorporated in e-mails.

Expandable banners increase in size from 468 x 60 to as large as 468 x 200 when

the user clicks or moves the cursor over the banner. The user has the option to read the

ad, click through to the advertiser's Web site, or send the ad away by rolling the mouse to

another area on the page (Balian, 2001). One of the leading companies in this format is

PointRoll, which offers Point*Roll technology. Point*Roll produces additional messages








above, below, or to the side of the original ad when the user rolls over the ad. The

message disappears when the cursor is moved away to another part of the page.

Floating ads use a combination of Flash and Dynamic Hypertext Markup Language

(DHTML), which is an extended set of HTML commands. These ads create a translucent

or shaded layer over the content and then execute an animated ad within this layer using

Flash technology. The ads load after the parent page content has loaded and do not

require user initiation. United Virtualities and Eyeblaster are the two leading companies

using this technology to develop online ads. United Virtualities' patent-pending ad

technology creates ShoshkeleTM ads.

Full-screen overlays, introduced by Unicast in 2003 and also offered by Eyeblaster,

occupy the entire screen for 15 seconds (Elkin, 2003c).

Half-page units, used by the New York Times Digital, take up one side of the

screen and two of the site's four columns (Elkin. 2003a).

Interstitials are contained within the current browser window and are automatically

presented to a viewer when moving between two content pages. Once the requested page

loads, the interstitial fades away to the requested page. Interstitials are also referred to as

transition ads, intermercial ads, splash pages, and Flash pages (IAB, 2001b).

Large rectangle ads are large ads placed within the copy where an editorial photo or

graphic would normally go. The editorial copy either wraps around the side of the ad or

appears above and below the ad. Large rectangles provide advertisers with more space

than traditional banner ads. The space can feature rich media animation, interactive

product information, or e-commerce opportunities. Users can often click within the ad or

click through to the advertiser's site. The IAB guidelines for large rectangles are 336 x








280 pixels (IAB, 2001 a). Other rectangle sizes include the vertical rectangle (240 x 400

pixels), the medium rectangle (300 x 250 pixels), and the rectangle (180 x 150 pixels)

(IAB, 2001a).

Leaderboards are giant banners with the dimensions of 728 x 90 pixels (IAB, n.d.).

Nonlinking advertisements are name recognition builders. They are typically fixed

logos on a Web site that do not allow the user to click through to the advertiser's site.

Paid listings are available to feature advertisers' listings prominently on the results

pages of many search engines. Paid listings are also called search engine listings or key

word listings.

Pop-under ads open underneath the user's browser and usually are discovered when

the user has closed or minimized the browser. The user is forced to close the pop-under

ad separately.

Pop-up ads interrupt the user by opening another window over the user's browser.

The user must close or minimize the window to remove it from the screen. A pop-up ad

may appear while the user is on a Web page or while transitioning between two Web

pages. This ad format is similar to a daughter window but is not served with an associated

banner ad. The IAB guidelines for pop-up ads are 250 x 250 pixels (IAB, 2001a).

Portals reside in a tool bar. For example, CNET.com uses text portals for its

featured advertisers.

Sliding billboards appear as a large advertising unit when first accessing a Web

page and then slide up into a smaller banner unit.

Skyscrapers are similar to banners, but rather than being located at the top of a Web

page, these tall, thin ads are situated along the side of a Web page. The LAB guidelines








for skyscrapers are 160 x 600 and 120 x 600 pixels (IAB, 2001a). A tower ad, also called

a vertical banner, is a shorter version of a skyscraper.

Sponsored electronic mailing lists are e-mails distributed to subscribers of the

service. Many niche markets are available to advertisers through these lists. By

sponsoring an electronic mailing list, advertisers can distribute their content to a group of

people interested in the topic.

A sponsorship is an association with a Web site that provides more visibility for the

advertiser than run-of-site advertising (IAB, 2001b). With sponsorships, advertisers hope

users will favorably associate the content with the advertiser. ESPN Sportszone's Injury

Report is an example of sponsored content. Microsoft's $200,000 sponsorship of the

Superbowl Web site and Toyota and Chemical Bank's $120,000-per-year sponsorship of

the New York Times Digital were some of the earliest examples of online sponsorship

advertising (DiBlasi, 1997).

Superstitials, developed by Unicast, load into temporary memory while a user is

viewing a Web page and instantly appear when the user clicks to another page on the

same site. Superstitials range from the size of a postage stamp to 550 x 480 pixels

(about two thirds of the screen). Superstitials typically feature full animation, sound, and

graphics and can run as long as 30 seconds.

Surround sessions, first launched by the New York Times Digital, provide a user

with banners, large rectangles, skyscrapers, and buttons from a single advertiser during a

visit to the site (Saunders, 2001b).

Web sites are the most common format of advertising on the Internet. The site can

be structured as a destination site, whereby information and entertainment are used to








attract users to the site, or as a microsite (also called a jump page), which uses small

clusters of pages hosted by a content site or network. Although the most prevalent type of

online advertising, a Web site is similar in purpose to a physical store, or even a customer

service information hotline, and Web site hosting costs are not included in estimates of

online advertising expenditures.

Webmercials (also called intermercials or Webformercials) feature animation or

streaming video and audio and require a downloadable plug-in to be viewed or heard.

Webmercials are often launched through either the use of an interstitial or a hypertext

link. While a Webmercial may be the same commercial that runs on television, the image

is smaller and not as sharp. BMW and Jaguar are two companies that have used this type

of advertising.

Viral marketing is used to quickly spread information online and is typically

accomplished through the use of e-mail. For example, Hotmail tags a logo to all outgoing

e-mail messages to promote its free e-mail. Some sites allow users to e-mail content by

clicking on "e-mail this to a friend." Other sites encourage users to tell friends about the

site through e-mail. Viral marketing can also be used in newsgroups and chat rooms.

Advertisers can spread news about their company or products using this method.

Demand for New Online Advertising Formats

New online advertising formats have emerged as a result of the demands of

advertisers and their agencies and the economic situation of many online content

providers. For advertisers disappointed with the low clickthrough rates of banner ads,

more technologically innovative forms of online advertising evolved to supplement the

use of banner ads. Advertising agencies also recognize these more sophisticated formats

of ads as a way to increase their profit margins for online advertising (Black, 2001).








A sagging economy and the failure of many dot-corn companies have slowed

revenue growth for the online advertising industry (Heim, 2001). As a result, online

content providers have been desperate to sell advertising space and are willing to allow

their advertisers to try more daring advertising concepts to attract the attention of the user

(Mediapost, 2001).

At the same time, marketers and advertisers are also recognizing the value of

permission marketing, whereby the consumer grants permission to receive e-mail

solicitations. Research firm eMarketer estimated 226 billion permission-based e-mails

will be distributed by the end of 2003 (eMarketer, 2001b). Permission is typically

provided when a user registers at a Web site and checks (or unchecks) a box indicating

his or her willingness to accept communications from the Web site or one of its partners.

Permission marketing allows users to have more control over their online advertising

experience.

As demonstrated, the Internet provides the advertiser with a medium for

transmitting advertisements in a variety of formats. The variety of online advertising

formats has evolved greatly from the original static banner ads, and variations of current

formats seem to appear almost daily.

Online Advertising Mix

The most prevalent online advertising format is still the banner, which represented

33% of online advertising elements for the week of April 28, 2003 (AdRelevance, 2003).

Including half banners increases this percentage by 4% (AdRelevance, 2003).

Skyscrapers also represented a high percentage of online advertising elements at 9% for

standard skyscrapers, 4% for wide skyscrapers, and 4% for vertical banners








(AdRelevance, 2003). Buttons represented 14% of elements and squares and medium

rectangles totaled 10% (AdRelevance, 2003).

A report by Nielsen//NetRatings found similar results, citing the dominance of the

banner ad (Martin & Ryan, 2003). Of the impressions for the top 100 cross-media

advertisers in the fourth quarter of 2002, 29% were full banners and 10% were half

banners (Martin & Ryan, 2003). Rectangles (e.g., standard size, medium, large, and

vertical) totaled 24% of impressions and skyscrapers totaled 11% (Martin & Ryan, 2003).

Another Nielsen//NetRatings study also reported that pop-ups only represented

3.5% of all online advertising impressions for the fourth quarter of 2002 (Buchwalter &

Martin, 2003). Interestingly, pop-ups garnered over 11.3 billion ad impressions in the

first seven months of 2002 and 80% of the pop-up ads were used by only 63 advertisers

(Martin & Ryan, 2002). The remaining 20% of pop-up ads were distributed among 2,145

advertisers (Martin & Ryan, 2002). Over 9% of advertisers during these seven months

used a pop-up ad (Martin & Ryan, 2002).

In the two-year period before the fourth quarter of 2002, the average number of

online ad formats supported by Web sites more than doubled to 11 formats from 5.3

formats (Buchwalter & Martin, 2003). Almost all Web sites accepted the banner ad and

60% of advertisers in a Nielsen//NetRatings study were found to use banner ads

(Buchwalter & Martin, 2003). Only 10% of advertisers used the skyscraper format, while

70% of sites accepted the format, and less than 10% of advertisers used pop-ups, which

were also accepted by a high percentage of sites (Buchwalter & Martin, 2003).

Online Advertising Spending

Online advertising represented only a $0.2 billion industry in the U.S. in 1996

(Jackson, 2001 a). In 1998, online advertising revenue reached $1.92 billion, passing








outdoor advertising revenue for the first time (Koprowski, 1999). For 2000,

PricewaterhouseCoopers reported that online spending by U.S. advertisers totaled $8.2

billion (Black, 2001).

U.S. online advertising revenue was down 12% in 2001 from the

PricewaterhouseCoopers' estimate, resulting in expenditures of $7.2 billion for the year

(Jackson, 2001a). In 2002, online advertising spending totaled $5.95 billion (IAB, 2003).

The fourth quarter of 2002 showed the first consecutive quarterly increase in online

advertising revenue since the second quarter of 2000 (IAB, 2003).

Jupiter Research predicted a 10% growth rate for 2003 and $14 billion in

expenditures by 2007 (Jupiter Research, 2002). Worldwide online advertising

expenditures are expected to reach $42 billion in 2005 (Forrester Research, 2001).

Analysts at Jupiter Research anticipated that much of this growth will be fueled by

the rise in online classified ad spending (i.e., paid search engine listings) and the increase

in spending by traditional advertisers (Jupiter Research, 2002). Publishers are now

catering to traditional advertisers with better service, improved tools for measuring

results, and new technology for more creative ad format options (Green, 2003). In

addition, online ad prices have fallen dramatically and lowered the cost per thousand ad

impressions (Green, 2003).

In March 2001, a study by Nielsen//NetRatings reported that online advertising

spending by traditional advertisers surpassed spending by dot-coms for the first time

(Saunders, 2001a). In addition, of the top 100 online advertisers, more than half were

traditional advertisers. This trend has been attributed more to the increase in offline

advertisers moving online than to the recent failures in the dot-com industry (Saunders,








2001 la). A Nielsen//NetRatings report cited a number of the top ten cross-media

advertisers that increased their online presence in 2002, namely DaimlerChrysler with a

407% increase over 2001 online advertising expenditures, Verizon Communications with

an 88% increase, Johnson & Johnson with a 39% increase, Ford Motor Company with a

34% increase, and Walt Disney and AOL Time Warner, both with a 28% increase over

2001 expenditures (Martin & Ryan, 2003).

Online Advertising Effectiveness

Clickthrough rates, an early measure of advertising effectiveness, have been

dropping fast (Khermouch & Lowry, 2001). The increase in the sheer number of online

advertisements may provide some explanation for this decline (Song, 2001). Because

clickthrough rates are calculated by dividing clicks by impressions, it is possible that

people are not necessarily clicking less often, but that impressions have increased, driving

the clickthrough rate down.

These low clickthrough rates have made the medium less attractive to advertisers

(eMarketer, 2001a). In a 2001 study of marketers and ad agency executives by Myers

Mediaenomics, 85% of respondents cited "driving traffic to the Web site" as one of the

top five reasons for using online advertising (receiving a higher percentage of responses

than any other alternative), and 49% of marketers and 57% of ad agency executives

indicated that the clickthrough rates were not high enough to motivate them to increase

their online ad spending for 2001 (second only to budget limitations) (eMarketer, 2001 a).

The focus on clickthrough as a measure of online advertising effectiveness has been

downplayed since the initial online advertising boom. An alternative and more popular

view is that the value of online advertising cannot be solely measured by clickthrough

rates (Briggs & Hollis, 1997).








One problem with the use of clickthough rates as a measure of effectiveness is they

do not fully represent the totality of banner ad conversions. Data from Engage (2001)

suggested only 25% of sale or lead conversions by consumers who were exposed to an

online advertising campaign result from clickthroughs on an ad itself and slightly less

than half of all conversions occur one or more days after being exposed to a banner ad for

the site.

A study by ad agency Avenue A found that 20% of consumers who made a

purchase on a travel site clicked through on a banner ad, while 80% had previously seen

the ad and later went directly to the site to make a purchase (Gilliam, 2000). In addition,

consumers who saw ads accounted for 10% more sales and traffic than those who did not

(Gilliam, 2000).

Research has supported the idea that mere exposure to the banner ad itself (even

without a clickthrough) can have a positive effect on the brand. The 1996 HotWired Ad

Effectiveness Study (Briggs & Hollis, 1997) found banners have a brand building effect.

The 1997 IAB Online Advertising Effectiveness Study conducted by Millward Brown

(IAB, 1997) found a single exposure to a banner ad was enough to generate lifts in ad

awareness, brand awareness, purchase intent, and product attribute communication.

Using a survey of 18,000 respondents covering multiple product categories, Dynamic

Logic reported the average brand awareness lift for banner advertising to be 6%

(Dynamic Logic, 2000a). The same organization conducted a study for Travelocity and

saw a 16% lift in aided brand awareness, a percentage that increased to 44% for

respondents who saw the ad four or more times (Dynamic Logic, 2000b). A report by

Morgan Stanley Dean Witter (2001) concluded that consumers are 27% more likely to








recall a brand after seeing an Internet ad, representing a recall level higher than that for

magazines (26%), newspapers (23%), and television (17%). A study by Dynamic Logic

of advertisers in a program by Forbes.com that guaranteed brand improvement results

found the online campaigns of 12 participating marketers increased message association

by 28%, purchase consideration by 14%, aided awareness by 11%, and brand favorability

by 6% (Elkin, 2003b).

Research by Millward Brown Interactive has also confirmed that various online

advertising formats, including interstitials, Superstitial ads, rich media, adaptable

cursors, and streaming media, have a positive effect on brand awareness, brand

perceptions, and intent to purchase (Briggs & Stipp, 1999). A study by Morgan Stanley

found banner ads using streaming media were five times as effective in generating brand

recall than traditional banner ads (Morgan Stanley, 2001).

As these studies have shown, online advertising is capable of impacting brand

image, and therefore, clickthrough rates are certainly not the only measure of

effectiveness. Brand awareness, image, and intent-to-purchase measures may be better

indicators of long-term advertising effectiveness. Advertisers and marketers are

recognizing the implications of these studies and have been changing the way they

measure advertising effects. A report by Jupiter Media Metrix (2001a) found only 15% of

marketers measured online branding effects, while many chose to use direct response

metrics such as clickthrough rate (60%) and cost per conversion (75%). By 2002, the

percentage of marketers who measured long-term metrics, such as branding, had risen to

35% (Jupiter Research, 2002).








Marketers may also want to consider measuring offline sales resulting from online

advertisements. A joint study by Procter & Gamble and Information Resources, Inc.

found offline sales for an impulse food product to be 19% higher for the test group with

three online ad exposures than the control group with none (Welch & Krishnamoorthy,

2000).

As the advertising industry moves away from measuring the effectiveness of online

advertising through direct response metrics, such as clickthrough rates, traditional

measures of effectiveness, including brand awareness and intent-to-purchase, are being

embraced. A natural extension of these measures is attitude toward a specific

advertisement, which may be predicted, in part, by attitude toward the online advertising

format.

Statement of Purpose

The cluttered online advertising environment, the expanding options for online

advertising, and the estimates for future growth in online advertising expenditures all

suggest the need for the advertising industry to be concerned about consumers' attitudes

toward online advertising and attitudes toward individual online ad formats.

The current study hypothesizes that attitude toward the online ad format plays a

critical role in determining attitude toward the ad. As Dynamic Logic director of client

services Jeffrey Graham wrote in his company newsletter column, "You can't expect

people to separate the medium (pop-ups) from the message (bad advertising)" (Graham,

2001). The oft-quoted "medium is the message" pronouncement by Marshall McLuhan

(1964, p. 7) further suggests a way to think about online advertising formats as these

formats themselves communicate a message.








Identifying the specific perceptions of online advertising that may raise or lower

attitudes toward an online advertising format was the first purpose of this study. This

study also collected descriptive data on attitudes toward different formats of online

advertisements and developed and tested a model of online advertising attitudes that

specified the antecedents of attitudes toward these new advertising formats and the effect

of attitudes toward online ad formats on attitude toward the ad (Aad).

Importance of the Study

This study has a number of potential implications for advertisers and advertising

agencies. First, a greater awareness of attitudes toward online advertising formats should

influence the use of online advertising in general and choice of online advertising by

advertisers and their agencies. Second, the findings from this study identified the specific

perceptions that raise or lower attitudes toward a particular online ad format. These

results will be useful during the creation of an individual ad.

This research also makes an original contribution to the flourishing body of

literature in the area of attitudes toward advertising in general, attitudes toward

advertising in a specific media vehicle, and attitude toward the ad. This research will also

be directly useful in future studies of online advertising effectiveness and attitude toward

the online ad.

While deriving and testing dimensions of attitudes toward current online

advertising formats has both practical and theoretical significance, these findings also

have implications for emerging online ad formats, further strengthening the importance of

this study. This research can help guide the development of new online advertising

formats.








Outline

Chapter 2 reviews the research on the attitude-toward-the-ad model, which

incorporates, as an antecedent of Aad, the concept of attitude toward advertising in

general. The research streams on attitude toward advertising in general and attitude

toward advertising in a specific media vehicle are then reviewed, with particular

emphasis on attitude toward online advertising. The literature review concludes with a

discussion of the hypothesized model guiding the present study, illustrating the proposed

role of attitude toward the online advertising format in a modified attitude-toward-the-ad

model.

This research utilized a multi-method approach, using qualitative methods in the

first study and surveys in two additional studies. Chapter 3 describes the first study,

which used a qualitative approach. Chapter 4 discusses the second study, which tested the

hypothesized model using a student sample. Chapter 5 describes the third study, which

used an online survey to gather descriptive data from a nationwide sample of adults. In

Chapter 6, the implications of the findings of these studies are addressed in relation to the

future of the online advertising industry and theory.













CHAPTER 2
LITERATURE REVIEW

This study introduces a new construct-attitude toward the online advertising

format (Aformat)-and proposes to demonstrate both its determinants and its relevance in

the attitude-toward-the-ad model. According to Rodgers and Thorson (2000, para. 85),

"Attitude toward the ad.. .is a response easily applied to interactive advertising."

Paralleling the definition of attitude toward the ad (Lutz, 1985, p. 46; MacKenzie & Lutz,

1989, p. 49), attitude toward the online advertising format is defined as a predisposition

to respond in a consistently favorable or unfavorable manner toward an online

advertising format.

Ad format has been simply defined as "the manner in which [an ad] appears"

(Rodgers & Thorson, 2000, para. 47). For example, television advertising can be

categorized in terms of the length of the commercial (e.g., 30 seconds), while magazine

advertising can be classified according to size (e.g., full page) (Rodgers & Thorson,

2000). The need to consider the ad format variable in a study of online advertising stems

from the proliferation of various online advertising formats, from banner ads to the more

television-like Webmercials. While research on attitudes toward advertising in general

has been extended to analyses of attitudes towarJ advertising in specific media vehicles,

it has barely addressed the existence of multiple formats of advertising within one

medium.

Research has demonstrated that consumers possess different beliefs about

advertising in various media. When comparing beliefs about advertising, several studies








have found that consumers perceive newspaper and magazine advertisements to be the

most informative (Mittal, 1994), with consumers significantly more satisfied with the

informational value of magazine advertising than with television advertising (Soley &

Reid, 1983). Bauer and Greyser (1968) found television advertising to contain the highest

proportion of ads classified as annoying, while print advertising was more likely to be

categorized as informative and enjoyable. Similarly, in another study, newspaper and

magazine advertisements have been classified as less irritating and annoying than

television advertising (Mittal, 1994). These studies demonstrate that consumers have

different beliefs about advertising in various media and suggest the possibility that

consumers may also have unique beliefs about each online advertising format. These

belief sets are expected to lead to different attitudes toward each online ad format.

Academic studies in the area of attitudes toward online advertising are theoretically

and methodologically grounded in the tradition of research on attitudes toward

advertising in general, an area that has evolved to include a focus on attitudes toward

advertising in a specific media vehicle. The emphasis in both the trade and academic

literature on understanding attitudes toward advertising may be attributable to the

documented relationship between general attitudes toward advertising and attitude toward

a specific advertisement, i.e., the attitude-toward-the-ad construct (Bauer & Greyser,

1968; Lutz, 1985; MacKenzie & Lutz, 1989; Mehta, 2000). In turn, considerable research

has demonstrated a positive association between Aad and brand attitude, brand choices, or

purchase intention (Dr6ge, 1989; Gardner, 1985; Homer, 1990; MacKenzie & Lutz,

1989; MacKenzie, Lutz, & Belch, 1986; Miniard, Bhatla, & Rose, 1990; Mitchell &

Olson, 1981; see discussion in Shimp, 1981; see Brown & Stayman, 1992, for a








comprehensive review). In addition, the Advertising Research Foundation Copy Research

Validity Project identified likability of an advertisement as the single best discriminator

of advertising effectiveness (Haley, 1990).

The following literature review will first detail the Aad model, describing relevant

studies in this area and illustrating the antecedents of Aad. This review will then describe

the research on attitudes toward advertising in general in more detail. Recent trends in

research on attitudes toward advertising will also be addressed, including the emphasis on

understanding attitudes toward advertising in a specific media vehicle and attitudes

toward online advertising. Finally, this review will describe how the proposed concept-

attitude toward online advertising format-is hypothesized to fit into the existing model.

Attitude Toward the Ad

Attitude toward the ad is defined as a "predisposition to respond in a consistently

favorable or unfavorable manner to a particular advertising stimulus during a particular

exposure situation" (Lutz, 1985, p. 46). Early research focusing on the origins of Aad

incorporated a cognitive processing approach, in much the fashion of the central route to

persuasion in Petty & Cacioppo's elaboration likelihood model (ELM) (1981; see Lutz,

1985). The idea that cognitions about an ad (as opposed to cognitions about the

advertised brand) could have an influence on attitude toward the ad was extrapolated

from the findings of previous studies confirming the link between brand-related

cognitions and brand attitude (Mitchell & Olson, 1981) and the relationship between

cognitive responses to an advertising message and attitudinal message acceptance

(Wright, 1973). Subsequently, the relationship between ad-related cognitions and attitude

toward the ad has been documented (Lutz, MacKenzie, & Belch, 1983).








Aad represents an affective response to an advertisement and the validity of such a

response is further bolstered by persuasion theories from social cognition, which

recognize the influence of not only cognitive, but also affective responses on message

effectiveness. Petty and Cacioppo's (1981) ELM, which describes two routes to

persuasion, posits a "central route" to persuasion occurring through diligent processing of

message content and a "peripheral route" to persuasion resulting from more casual

processing of the message source or other contextual factors. The level of involvement

determines which path to persuasion dominates.

Attitude Toward the Ad as a Mediator

Introduced by Mitchell and Olson (1981) and Shimp (1981), Aad has been found to

be a mediator of brand attitude (Ab), brand choice, and purchase intentions (Gardner,

1985; Homer, 1990; MacKenzie & Lutz, 1989; MacKenzie, Lutz, & Belch, 1986;

Miniard, Bhatla, & Rose, 1990; Mitchell & Olson, 1981). A meta-analysis by Brown and

Stayman (1992) of 47 samples confirmed a significant relationship between Aad and

brand attitudes, brand-related cognitions, and purchase intention. The significance of

brand attitude is its documented link to purchase intentions (Brown & Stayman, 1992).

MacKenzie, Lutz, and Belch (1986) applied the concepts of ELM to two of their

four structural specifications of the mediating role of Aad. In the "affect transfer

hypothesis," the central route to persuasion explains the direct relationship between brand

cognitions and attitudes toward the brand, while the peripheral route serves to explain the

path from attitudes toward the ad to brand attitudes. In situations of high message

involvement (the cognitive effort directed toward processing message content) and low

ad execution involvement (the effort focused on processing non-content properties), the

central processing mechanism dominates persuasion and brand cognitions lead to brand








attitudes (MacKenzie & Lutz, 1989). As Droge noted, "Aad appears to be a peripheral cue

that has little or no impact when central processing predominates" (1989, p. 202). In

contrast, in situations of low ad message involvement, regardless of the level of ad

execution involvement, the peripheral route provides a framework for understanding how

attitudes toward the ad are directly related to brand attitude (MacKenzie & Lutz, 1989).

While Petty and Cacioppo's (1981) theory can be applied in situations of low

message involvement to explain how peripheral processes operate to allow a peripheral

cue such as attitude toward the ad to have persuasion abilities, it does not explain how

attitude toward the ad may serve as a peripheral cue to influence the central route to

persuasion (MacKenzie, Lutz, & Belch, 1986). Within this "dual-mediation" model,

attitude toward the ad is directly related to brand attitude and also indirectly related to

brand attitude by influencing the degree to which the audience incorporates message

content into its brand cognitions (MacKenzie & Lutz, 1989). Extending this line of

research, Miniard, Bhatla, and Rose (1990) determined that the Aad-brand attitude

relationship can be viewed as the result of not only peripheral processing, but that the two

constructs can be related even when persuasion follows the central route.

While MacKenzie, Lutz, and Belch (1986) found the relationship between Aad and

Ab to be stronger than any other relationship in the four models tested, they recognized

that shared method variance may have heightened this effect. MacKenzie and Lutz

(1989) also found Aad to have a strong effect on Ab, while cognitions about the brand did

not influence Ab as strongly as expected.

Brown and Stayman's (1992) meta-analysis demonstrated that while some path-

analytic studies did not find a significant relationship between brand cognitions and brand








attitude (e.g., MacKenzie & Lutz, 1989), others found a significant path (e.g., Homer,

1990). Based on aggregated study data, Brown and Stayman (1992) suggested that brand

cognitions do significantly affect brand attitudes, but that this relationship is the weakest

in the model. Furthermore, the meta-analysis supports the indirect effect of attitude

toward the ad on brand attitude through brand cognitions (Brown & Stayman, 1992). In

addition, while most studies found a substantial and significant direct relationship

between Aad and brand attitude (e.g., MacKenzie, Lutz, & Belch, 1986), Brown and

Stayman's (1992) meta-analysis found this relationship to be weaker than these studies

suggest.

Other studies have documented the circumstances under which Aad has strong

effects (Dr6ge, 1989; Gardner, 1985; Park & Young, 1986). Dr6ge (1989) found Aad to

be a significant predictor of Ab only for noncomparative, rather than comparative, ads.

Gardner (1985) found that a positive and significant relationship existed between Aad and

attitude toward the brand for both brand and nonbrand processing set conditions. Park

and Young (1986) differentiated cognitive, affective, or low involvement and found that

Aad influenced brand attitude only in situations of affective or low involvement.

Likability Studies

A number of studies have addressed liking of an advertisement, a concept virtually

identical to Aad (Haley, 1990; Walker & Dubitsky, 1994). Advertisement liking has been

linked to product liking, as positive feelings toward the ad are transferred to the brand

(see review by Thorson, 1991). Liking has also been found to increase the chance that a

viewer will pay attention to an advertisement and learn its message, thereby enhancing

the advertisement's effectiveness (Walker & Dubitsky, 1994). The ARF Copy Research








Validity Project (Haley, 1990) found liking of a commercial to be the strongest predictor

of the sales differences due to advertising for the cases evaluated.

Attitude Toward the Ad Model

While studies have demonstrated how Aad is predicted through cognitive responses

to the execution elements and the perception of advertiser credibility (Lutz, MacKenzie,

& Belch, 1983), Aad may also be the result of a peripheral processing mechanism (Lutz,

1985). Affective reactions to the advertiser and advertising in general, as well as the

mood of the consumer, may operate through peripheral processing to influence Aad (Lutz,

1985).

Lutz (1985) developed a structural model of five cognitive and affective

antecedents of Aad and MacKenzie and Lutz (1989) further refined Lutz's (1985) original

model. The five antecedents include ad credibility, ad perceptions, attitude toward

advertiser, attitude toward advertising in general, and mood. The model also incorporates

"second-order determinants," which directly influence the five antecedents of Aad and

indirectly impact Aad through the "first-order" antecedents. Figure 2-1 summarizes the

modified structural model (MacKenzie & Lutz, 1989). Descriptions of each antecedent as

defined by Lutz (1985) and MacKenzie and Lutz (1989) are also provided.

Ad credibility. The assessment of ad credibility, defined as "the extent to which the

audience perceives claims made about the brand in the ad to be truthful and believable"

(Lutz, 1985, p. 49), is a cognitive process requiring a central processing model. Ads

perceived to be credible receive more favorable responses by consumers.



























?J \Legend:
/ Perceptual Constructs
0 Affective Constructs
Brand A Exogenous Variables
Perceptions A Noncausal Identities

-- Hypothesized Causal
Relationships
Figure 2-1. Modified Structural Model of Aa Formation. From "An Empirical
Examination of the Structural Antecedents of Attitude Toward the Ad in an
Advertising Pretesting Context," by S. B. MacKenzie and R. J. Lutz, 1989,
Journal ofMarketing, 53, p. 53. Copyright 1989 by Scott B. MacKenzie and
Richard J. Lutz. Reprinted with permission.


Ad credibility results from three second-order determinants: perceived ad claim

discrepancy, advertiser credibility, and advertising credibility. Ad claim discrepancy is

the gap between the advertisement's claims about the brand and the consumer's

perceived performance of the brand, a perception influenced by past experience,

information about the advertised brand, and the content of the message. Advertiser

credibility reflects the consumer's perceived truthfulness of the ad's sponsor.

Furthermore, past experience and information about the advertiser directly influence








advertiser credibility. Advertiser credibility also serves as a component of the second-

order determinant of the multidimensional advertiser perceptions.

Advertising credibility, the perception of the believability of advertising in general,

also influences ad credibility. As with advertiser credibility, advertising credibility is one

component of advertising perceptions. Advertising credibility has also been modeled to

influence ad credibility indirectly through the more specific construct of advertiser

credibility (MacKenzie & Lutz, 1989).

While MacKenzie and Lutz (1989) established a significant relationship between ad

credibility and Aad, they could not find support for their hypothesis that advertising

credibility affects ad credibility. The findings did support a significant relationship

between ad credibility and another second-order determinant: advertiser credibility

(MacKenzie & Lutz, 1989).

Ad perceptions. As one of many perceptual responses to an ad, ad credibility is

actually a special case of ad perceptions, the second major antecedent to Aad. Because of

the amount of research in the area of ad credibility, a separate classification for this

variable was warranted (Lutz, 1985).

Like ad credibility, ad perceptions also entail some degree of central processing.

This construct incorporates only consumer perceptions of the advertising stimulus and

not perceptions of the advertised brand.

As a mediating variable, ad execution characteristics have been found to have a

strong positive relationship with Aad through ad perceptions (Lutz, MacKenzie, & Belch,

1983). Attitude toward advertising and attitude toward the advertiser are modeled to








impact ad perceptions, demonstrating the possible influence of affect on a perceptual

process (Fazio & Zanna, 1981).

Under ad pretest conditions, MacKenzie and Lutz (1989) found advertiser attitude

to have a strong positive relationship with ad perceptions, while the relationship between

attitude toward advertising and ad perceptions could not be cross-validated. Ad

perceptions were found to exhibit a strong positive correlation with Aad (MacKenzie &

Lutz, 1989).

Attitude toward the advertiser. Defined as "a learned predisposition to respond in a

consistently favorable or unfavorable manner to the sponsoring organization" (Lutz,

1985, p. 53), attitude toward the advertiser represents a more affective response to an

advertisement. Perceptions of the advertiser, including advertiser credibility, are expected

to influence attitude toward the advertiser. Perceptions emanate from consumers' past

experience and information about the company. Advertiser attitude was found to have a

strong positive correlation with Aad under ad pretest conditions (MacKenzie & Lutz,

1989).

Attitudes toward advertising. Attitude toward advertising represents a "learned

predisposition to respond in a consistently favorable or unfavorable manner toward

advertising in general" (Lutz, 1985, p. 53; MacKenzie & Lutz, 1989, p. 53-54). This

concept reflects consumers' general attitudes toward advertising, rather than attitudes

toward a specific advertisement or about advertising in a specific medium.

The study of the relationship between consumers' attitudes toward advertising in

general and ratings of specific ads dates back to Bauer and Greyser's (1968) classic study

described in Advertising in America: The Consumer View. Bauer and Greyser (1968, p.








121) suggested a relationship between overall attitudes toward advertising and the

proportion of ads classified as either favorable or unfavorable. In addition, the data

revealed a relationship between attitudes toward advertising and the perception of certain

ads as informative (Bauer & Greyser, 1968, p. 136).

The model of Aad formation proposes that attitudes toward advertising resulting

from perceptions of advertising have a direct impact on Aad. MacKenzie and Lutz (1989)

could not confirm a relationship between attitude toward advertising and Aad in a study

under ad pretest conditions; however, they suggested that in focusing attention on the

evaluation of a specific ad, subjects were less likely to base Aad assessments on general

constructs, such as attitude toward advertising, than on specific constructs. MacKenzie

and Lutz (1989) suggested that in a natural setting, as opposed to a forced exposure

situation, a stronger relationship between attitude toward advertising and Aad might exist.

Mood. As the most purely affective antecedent to Aad, mood is "the consumer's

affective state at the time of exposure to the ad stimulus" (Lutz, 1985, p. 54). Mood is

influenced by individual differences, which are the basic predispositional tendencies of

consumers; ad execution characteristics; and reception context, comprised of the nature

of the exposure, the amount of ad clutter, and the program or editorial context.

Attitudes Toward Advertising in General

A more thorough discussion of attitudes toward advertising is provided below and

will be followed by a discussion of attitudes toward advertising in a media vehicle and

attitudes toward online advertising. This review is provided to demonstrate how the

construct of attitude toward the online ad format emerges as a natural extension of this

body of research.








Researchers and industry practitioners have long been interested in attitudes toward

advertising (see Mittal, 1994; O'Donohoe, 1995; Pollay & Mittal, 1993; Zanot, 1981,

1984 for reviews), a construct found to influence attitudes toward specific advertisements

(Bauer & Greyser, 1968). The earliest studies in this area date back to 1939 and are

characterized by consumers' generally favorable attitudes toward advertising (Bauer &

Greyser, 1968). A 1942 survey by the Association of National Advertisers found more

than 80% of respondents to be supportive of advertising during the war (as cited in Bauer

& Greyser, 1968).

In the 1950s, attitudes toward advertising remained favorable as indicated by a

1951 survey by Mcfadden Publications (as cited in Bauer & Greyser, 1968) in which

90% of respondents agreed that advertising has played an important role in raising the

standard of living in the U.S. In the late 1950s, a Redbook magazine survey conducted by

the Gallup Organization, Inc. (1959, as cited in Bauer & Greyser, 1968) determined that

more than 80% of the over 1,600 respondents believed advertising helped raise

nationwide prosperity. In addition, 75% reported that they liked advertising and the most

frequently cited reason for liking advertising was its informational value.

General attitudes toward advertising have been on the decline since these early

studies. The percentage of Americans holding a generally favorable view of advertising

dropped to 54% in 1960 (Universal Marketing Research, 1961, as cited in Bauer &

Greyser, 1968). Bauer and Greyser (1968) found the percentage of respondents with a

favorable attitude toward advertising to be 41% by 1964. Although a majority of

respondents in this study believed advertising to be misleading and capable of persuading








people to buy products they should not buy, they still considered advertising to be

essential.

Zanot's (1981) review of 38 public opinion polls from the early 1930s to the 1970s

revealed that attitudes toward advertising became increasingly more unfavorable during

that time. According to Zanot's analysis:

The number of surveys conducted...during the 1970s increased dramatically; 20
are presented here...they reflect a decidedly negative public opinion toward
advertising. In almost every instance where a study was replicated, the later one
shows more negative attitudes. (1981, p. 146)

Research in recent years has focused more on attitudes toward advertising in a

specific medium than attitudes toward advertising in general (Alwitt & Prabhaker, 1992;

Mittal, 1994). One exception is a 1998 study of 1,000 adult consumers' current attitudes

toward and confidence in advertising (Shavitt, Lowrey, & Haefner, 1998). This study

revealed more favorable public attitudes than suggested by previous studies. Only one

fourth of respondents in this study indicated that they disliked advertising.

Another focus in recent years has been an attempt to understand the structure of

advertising attitudes. Because these studies have tended to use smaller and less nationally

representative samples, results are not generalizable to the American public (e.g., Alwitt

& Prabhaker, 1992; Andrews, 1989; Mittal, 1994; Muchling, 1987; Pollay & Mittal,

1993; Reid & Soley, 1982; Sandage & Leckenby, 1980).

Other studies have tested the relationship between general attitudes toward

advertising and advertising effectiveness. Research has demonstrated that attitudes

toward advertising in general are related to ad recall and buying interest (Donthu,

Cherian, & Bhargava, 1993; Mehta, 2000).








A study of outdoor advertising found that consumers with positive attitudes toward

advertising in general exhibited greater recall of outdoor advertisements than those with

negative attitudes (Donthu, Cherian, & Bhargava, 1993). Mehta (2000) found that

respondents who reported that they like advertising, feel that it provides useful

information, and view it as not being manipulative were more likely than those who did

not feel this way to notice and recall advertisements. In addition, buying interest was

found to be positively related to almost all of the advertising belief statements tested in

the study (Mehta, 2000).

Another study examined the influence of attitudes toward advertising in general on

involvement with specific advertisements, operationalized as the amount of time spent

looking at the advertisement (James & Kover, 1992). The group of subjects that believed

advertising to be manipulative and the group that found advertising to be irritating were

both more involved in the advertisements.

Beliefs About Advertising in General

While early studies often measured favorability or unfavorability toward

advertising, later studies focused on beliefs about certain aspects of advertising (Mittal,

1994). Referred to as "consequences" of advertising in some studies (Mittal, 1994) and

"functions" in others (Alwitt & Prabhaker, 1992), these statements generally reflect

beliefs about advertising.

Beliefs represent descriptive statements about the attributes an object possesses,

creating a link between an object and an attribute. Beliefs are generally considered to

contribute to the formation of attitude (Ajzen & Fishbein, 1980). As noted by Fishbein

and Ajzen, "a person's attitude is a function of his salient beliefs at a given point in time"

(1975, p. 222). Attitudes are summary evaluations of the perception that an object








possesses certain attributes and the desirability of those attributes (Ajzen & Fishbein,

1980). While an attitude is a "general and enduring positive and negative feeling about

some person, object, or issue," a belief can be described as "information that a person has

about other people, objects, and issues" and this "information may have positive,

negative, or no evaluative implications for the target of the information" (Petty &

Cacioppo, 1981, p. 7). Therefore, the feeling or attitude that people hold about online

advertising formats should be derived from what people think, know, or believe about

online advertising formats.

The Relationship Between Beliefs and Attitudes

Studies in the area of attitudes toward advertising often fail to draw a distinction

between attitudes and beliefs, often measuring beliefs in an attempt to measure attitudes

(Muehling, 1987). Other studies have examined perceptual dimensions without directly

relating them to advertising attitudes (Muehling, 1987).

Other studies have examined this correlation between beliefs and attitudes toward

advertising. According to Lutz (1985), attitude toward advertising in general is

determined in part by consumer beliefs about advertising in general. A number of studies

in the area of attitudes toward advertising have measured and then correlated attitudes

toward advertising and consumer perceptions of the evaluative attributes, or beliefs, that

form those attitudes (Aaker & Stayman, 1990; Alwitt & Prabhaker, 1992, 1994; Biel &

Bridgwater, 1990; Cho, 1999; Mittal, 1994; Schlosser, Shavitt, & Kanfer, 1999; Shavitt,

Lowrey, & Haeffer, 1998). Generally speaking, these studies have found a positive

correlation between attitudes and perceptions, but there are exceptions (see Mittal, 1994).

Mittal (1994) found ratings on ten evaluative items about product-specific

commercials to be congruent with overall likeability of the commercial, but this was not








always the case. More specifically, some commercials were rated as enjoyable but were

not liked. As Mittal concluded, "The merits and demerits of product specific commercials

do individually register on the consumer mind despite overall favorable or unfavorable

predispositions" (1994, p. 47). Applying this perspective to online advertising provides a

possible explanation for why a user may not like a pop-up ad, even though it is perceived

to be entertaining. Similarly, a user may rate a banner ad as informative, but because it is

associated with all online advertising, the user may not have a favorable attitude toward

this format.

Categorizing Beliefs About Advertising

The strong emphasis on belief dimensions is attributable to Bauer and Greyser's

(1968) influential study. In this study, Bauer & Greyser (1968, p. 124) demonstrated how

beliefs about advertising in general influence attitudes toward advertising in general.

Bauer and Greyser's (1968) study categorized eight beliefs about advertising as either

economic effects (e.g., "raises standard of living" or "results in better products") or social

effects (e.g., "persuades you to buy what you don't need" or "insults the intelligence of

an average consumer"). Others have adopted this approach to examining advertising

beliefs as classifiable under these two factors (Reid & Soley, 1982), and subsequent

factor analyses supported this distinction (Anderson, Engledow, & Becker, 1978;

Andrews, 1989).

While Bauer and Greyser (1968) found consumers to have a generally favorable

view of the economic role of advertising, they also found consumers to hold an

unfavorable view of the social role. Other studies have confirmed Bauer and Greyser's

(1968) finding that consumers feel more favorable toward the economic role of

advertising than the social role (Andrews, 1989; Greyser & Reece, 1971). A study by








Anderson et al. (1978) of Consumer Reports subscribers found that attitudes became less

favorable from 1970 to 1976 on both the economic and social factors of advertising.

Beliefs have been studied in terms of generalized and personalized levels (Reid &

Soley, 1982). Personalized belief items tap the influence of advertising on a person's own

behavior (e.g., "advertising misleads me" (Reid & Soley, 1982)), while generalized belief

items focus on how advertising affects the behavior of other people (e.g., "advertising

misleads people" (Reid & Soley, 1982)). Researchers have demonstrated a significant

difference in attitudes toward advertising's social and economic effects depending on

whether personalized or generalized beliefs are used (Reid & Soley, 1982).

Distinctions have also been made between the informational value of advertising

and its persuasive effects. Research has demonstrated that consumers tend to have

positive reactions toward advertising for its informational value and negative reactions

toward advertising as a result of any perceived manipulation, intrusion, or deceit (Mehta,

2000; Shavitt, Lowrey, & Haefner, 1998).

Sandage and Leckenby (1980) divided advertising attitudes into attitude toward the

institution and the instrument of advertising. While institution reflects the purpose and

effects of advertising, instrument refers to advertising's executional properties. Mittal

(1994) used this distinction in a study of attitudes toward television advertising.

Muehling (1987) examined the belief items that comprise these dimensions.

Belief and Attribute Dimensions Included in Previous Studies

Studies measuring attitudes toward the institution of advertising often incorporate

belief statements to tap underlying dimensions. These statements often reflect the effects

and consequences of advertising or the value of advertising. For example, Mittal (1994)

used belief statements to determine whether television advertising offers useful social-








image information, is a valuable source of information about local sales, and is

sometimes more entertaining than the programs.

Muehling (1987) measured the influence of 20 beliefs about advertising in general

on attitudes toward the institution and instrument of advertising and found five of these

beliefs to be significant and explain over 57% of the variance in attitudes toward

advertising in general. The significant beliefs included whether advertising insults the

intelligence of consumers, presents a true depiction of the advertised products, or wastes

natural resources by creating desires for necessary goods. Whether a limit should be

placed on the amount of money a company can spend on advertising and whether today's

standards for advertising are higher than 10 years ago were two additional significant

predictors of attitudes toward advertising in general. Muehling (1987) concluded that the

set of beliefs that influence attitudes toward advertising was smaller than expected and

that beliefs about both institutional and instrument aspects of advertising influenced

attitudes. However, attitudes toward the institution of advertising were higher than

attitudes toward the instrument, which was consistent with the findings of Sandage and

Leckenby (1980).

Measures of the advertising instrument often involve the use of attributes, as in the

earlier Reaction Profile studies (Wells, 1964). In these studies, respondents are asked to

indicate the extent to which attributes describe advertising or the percentage of

advertising that can be described by the attributes.

For example, Mittal (1994) used ten evaluative attributes (borrowed mainly from

Santos, 1976) and asked respondents to assess the proportion of television advertisements

that possesses each attribute. The list included such attributes as informative, honest,








enjoyable, boring, annoying, and silly. Mittal (1994) further categorized these attributes

under the headings of information/disinformation, enjoyment/annoyance, and silliness.

While information/disinformation was found to contribute the most to overall attitude

toward advertising, silliness was found to contribute the least (Mittal, 1994).

A number of belief dimensions have been identified by applying the Uses and

Gratifications approach to the understanding of advertising, particularly for

understanding the uses and gratifications of television commercials (Plummer, 1971;

Schlinger, 1979). Using a factor analysis of adjectives and descriptive statements,

Plummer (1971) found the "viewer rewards" of television to fall into seven categories

including entertainment or stimulation, irritation, familiarity, empathy or gratifying

involvement, confusion, informativeness or personal relevance, and brand reinforcement.

Schlinger (1979) had similar findings in a related study of 49 adjectives and descriptive

statements. The dimensions determined by this study included entertainment, confusion,

relevant news, brand reinforcement, empathy, familiarity, and alienation or irritation. In

both of these studies, confusion was described by items referring to the clarity of

expression and organization of commercials. Familiarity refers to the uniqueness or

novelty of an advertisement.

The Uses and Gratifications approach has been used to understand motivations for

and benefits of surfing the Web and the characteristics a Web site should exhibit to be

successful (Eighmey & McCord, 1997). Eighmey and McCord (1997) identified six

factors that discriminate between the best and worst Web sites and labeled these factors

marketing perceptions (referring to the business relationship between the site and users),

entertainment value, informational value, ease of use, credibility, and interactivity.








The purpose of the following review is to understand the dimensions that have been

established in previous studies. The current study will then derive its own dimensions for

quantitative research. The dimensions presented in this section are for later comparison

purposes and can be used as a benchmark for the dimensions derived in the current study.

These dimensions also provide a preview of the dimensions expected to be derived in this

study.

Studies on attitudes toward advertising, particularly attitudes toward online

advertising, also suggest dimensions relevant to this research. The three belief

dimensions that appear to be most relevant to the understanding of attitude toward online

advertising format date back to the 1968 study by Bauer and Greyser and were used in a

more recent study by Ducoffe (1996). Ducoffe (1996) found the informational,

entertainment, and irritation dimensions of advertising to be significant predictors of

attitudes toward advertising. Buying confidence will be addressed as a subcategory of the

informative dimension. Finally, offensiveness of advertising and other dimensions will be

discussed.

Information. Bauer and Greyser defined informative ads as follows:

These are ads that you learn something from that you are glad to know or know
about. They may tell you about a new product or service or they may tell you
something new about a product or service you were already familiar with. The main
thing is that they help you in one way or another because of the information they
provide. (1968, p. 182)

Bauer and Greyser's (1968) finding that attitude toward advertising is positively

related to information-related reasons for liking advertisements possibly motivated the

inclusion of the informational value of advertising in a number of subsequent advertising

belief taxonomies (Alwitt & Prabhaker, 1992, 1994; Ducoffe, 1996; Lee & Lumpkin,








1992; Mehta, 2000; Mittal, 1994; Muehling, 1987; Pollay & Mittal, 1993; Schlosser,

Shavitt, & Kanfer, 1999; Shavitt, Lowrey, & Haefner, 1998).

The idea that advertising provides information to consumers is grounded in

information theory (Gardner, 1983). The informational role of advertising has often been

regarded as its foremost legitimizing function (Rotzoll, Haefner, & Sandage, 1989) and

the ability of advertising to provide information was found to be the primary reason for

consumer approval (Bauer & Greyser, 1968). Stigler's (1961) classic study was the first

to demonstrate how advertising is an important source of product information. Product

information provided in advertisements is perceived to result in better decision-making

by consumers (Alwitt & Prabhaker, 1992). Furthermore, advertising has been found to

stimulate competition, promote new product or brand entry, and simplify consumer

shopping (Albion & Farris, 1981).

Resnik and Stem (1977) defined informative advertising as that which provides

relevant informational cues to help a consumer make an intelligent choice among

alternatives. Some of these cues include price, performance, quality, packaging, and

special offers (Resnik & Stem, 1977). Resnik and Stem (1977) found that over 49% of

the television ads sampled were informative. A replication of the Resnik and Stem study

found no significant differences in the overall proportion of informative ads in 1977 and

1991, but did find significant differences in the use of various types of informational cues

(Stem & Resnik, 1991).

In contrast, Aaker and Norris (1982) found just over 18% of a sample of 524

prime-time television ads to be perceived by respondents as informative. In a more recent








study, Mittal (1994) found that almost three fourths of respondents described only 25% or

less of television advertising as "informative and helpful."

Other studies of magazine advertising found 92% (Laczniak, 1979) and 86%

(Stem, Krugman, & Resnik, 1981) of the sampled consumer ads to be informative. Soley

and Reid (1983) found consumers to be more satisfied with the informational value of

magazine advertising than television advertising, although consumers were neither highly

satisfied nor dissatisfied with the informational value of the advertising in either medium.

More recent studies have examined the relationship between the informational

value of advertising and advertising attitudes. Mittal (1994) determined that of 10

perceptions considered, perceptions of the informational value of advertising explained

the most variance in overall attitude toward television advertising. Pollay and Mittal

(1993) found product information to be a significant predictor of attitudes toward

advertising. Lee and Lumpkin (1992) found that the informational dimension of attitudes

toward television advertising differentiates between those who rarely skip commercials

on pre-recorded television programs and those who skip commercials sometimes or

almost always.

Informativeness has been positively related to Internet advertising attitudes

(Schlosser, Shavitt, & Kanfer, 1999), overall advertising attitudes (Shavitt, Lowrey, &

Haefner, 1998), perceptions of advertising value (Ducoffe, 1996), and recall and buying

interest (Mehta, 2000). In contrast to studies that found a positive relationship between

perceptions of the informativeness of advertising and attitudes toward advertising, Alwitt

and Prabhaker (1992) found beliefs about the informational value of television

advertising to have little influence on attitudes toward television advertising. An








explanation provided by Alwitt and Prabhaker (1992) for this result is the high

intercorrelations among the four functions of knowledge, hedonic, social learning, and

affirmation of value with only the hedonic function emerging as a significant variable in

the multiple regression model.

Related to informativeness is the concept of buying confidence, which has been

addressed in a number of studies (see Mittal, 1994; Schlosser, Shavitt, & Kanfer, 1999).

Mittal (1994) found no significant relationship between perceptions of buying confidence

instilled by advertising and attitudes toward television advertising. In contrast, Schlosser,

Shavitt, and Kanfer (1999) found that the use of online advertising to make a purchase

decision contributes to Internet advertising attitudes.

Enjoyment/Entertainment. Bauer and Greyser defined enjoyable ads as follows:

These are ads that give you a pleasant feeling for any reason whatsoever. They may
be entertaining, amusing, especially attractive or well done. You might enjoy them
whether or not you are interested in what is advertised. The main thing is that you
like them and are pleased you saw or heard them. (1968, p. 182)

Perceptions of the entertainment value of advertising has been considered in a

number of previous studies (Alwitt & Prabhaker, 1992, 1994; Bauer & Greyser, 1968;

Ducoffe, 1996; Lee & Katz, 1993; Mehta, 2000; Mittal, 1994; Pollay & Mittal, 1993;

Schlosser, Shavitt, & Kanfer, 1999; Shavitt, Lowrey, & Haefner, 1998). Mayer (1991)

found that purchase behavior is based on not only the consumer's assessment of the

product itself, but also the entertainment value of its advertising.

Alwitt and Prabhaker (1992) found beliefs about advertising's hedonic function

contributed significantly to attitudes toward television advertising. Ducoffe (1996) found

entertainment to be significantly correlated with perceived advertising value. The

enjoyment of advertising has been shown to be the strongest predictor of attitudes toward








Internet advertising (Schlosser, Shavitt, & Kanfer, 1999) and attitudes toward advertising

(Shavitt, Lowrey, & Haefner, 1998). Pollay and Mittal (1993) found the dimension of

hedonic/pleasure to be significantly and positively related to advertising attitudes. Mehta

(2000) found that subjects who indicated that they enjoy looking at advertisements

exhibited higher recall and stronger buying interest than those who do not enjoy looking

at ads. Lee and Katz (1993) found over three fourths of their sample of video store

patrons disagreed that commercials on a videotape are fun to watch. In contrast, Mittal

(1994) did not find a relationship between entertainment value and attitudes toward

television advertising, but did find the group of evaluative dimensions labeled

enjoyment/annoyance to contribute significantly to variance in attitude toward

advertising.

Annoyance/Irritation. Bauer and Greyser defined annoying ads as follows:

These are ads that irritate you. They may be annoying because of what they say or
how they say it. They may annoy you because they are around so much, or because
of when and where they appear. There may be other reasons for ads to be
annoying-the main thing is that they bother or irritate you. (1968, p. 182)

The idea that advertising is defined by or can be described by its level of irritation

or annoyance is consistent with the foundations of Uses and Gratifications research

(Eighmey & McCord, 1998; Plummer, 1971; Schlinger 1979) and attitude toward

advertising studies (see Bauer & Greyser, 1968; Ducoffe, 1995, 1996; James & Kover,

1992; Mehta, 2000). Irritation is such a pervasive issue in advertising that this

characteristic has also merited a body of research about a common cause of irritation:

advertising clutter (Elliott & Speck, 1998; Ha, 1997).

Aaker and Bruzzone (1985) found irritation to be a reason for disliking advertising.

James and Kover's (1992) factor analysis of belief dimensions of attitudes toward








advertising resulted in just two factors, with one referring to the irritation experienced

from advertising. Ducoffe (1996) found irritation to be significantly related to perceptions

of advertising value. Mehta (2000) did not find a relationship between the belief that

advertising is annoying and either recall or buying interest.

As mentioned earlier, irritation may result from the advertising clutter (Elliott &

Speck, 1998; Ha, 1997). Ha (1997) defined perceived ad clutter as resulting from three

communication problems: hindered search, distraction, and disruption. Hindered search

hampers a person's ability to read or see the media content, while disruption is a

diversion from the media use experience and distraction is merely a lesser form of

disruption. Alwitt and Prabhaker (1994) found that respondents were more likely to

dislike television advertising when they believed that the same ads were shown too

frequently.

Other beliefs and attributes. A number of studies have examined the offensive

aspects of advertising (Alwitt & Prabhaker, 1992, 1994; Bauer & Greyser, 1968;

Schlosser, Shavitt, & Kanfer, 1999) or more specifically, poor taste and sex in advertising

(Larkin, 1977). While Bauer and Greyser (1968) contended that an offensive ad may be

considered annoying, they differentiated between these two dimensions by limiting

offensiveness to the moral aspects of the product or advertisement or the effect on

children.

Bauer and Greyser defined offensive ads as follows:

These are ads that are vulgar or morally bad in your opinion. They may be
dishonest, or untrue. They may be ads for something you don't think should be sold
or used. They may be offensive because of the way in which they were done, and
you may think that such ads should not be allowed. The main thing is that you feel
strongly that such ads are wrong. (1968, p. 182)








As defined by Bauer and Greyser (1968), advertising may be considered offensive

as a result of its use of deception. The deceptive nature of advertising has been examined

as a distinct belief dimension of attitudes toward advertising (Alwitt & Prabhaker, 1992;

Bauer & Greyser, 1968; Durand & Lambert, 1985; Larkin, 1977; Mehta, 2000; Muehling,

1987; Pollay & Mittal, 1993; Schlosser, Shavitt, & Kanfer, 1999; Shavitt, Lowrey, &

Haefner, 1998).

Alwitt and Prabhaker (1992) found that beliefs about the offensive aspects of

advertising have little influence on attitudes toward television advertising. Schlosser,

Shavitt, and Kanfer (1999) found that beliefs about the indignity of advertising (i.e.,

insulting intelligence and offensiveness) were not related to Internet advertising attitudes.

In contrast, Shavitt, Lowrey, and Haefner (1998) found perceptions of the indignity of

advertising (together with the entertainment value) to have the strongest effect on

predicting advertising attitudes. Alwitt and Prabhaker (1994) found a significant and

positive correlation between dislike of television advertising and the perception of the

offensiveness of the advertising.

While Pollay and Mittal (1993) found the perceived falsity of advertising to

influence attitudes, Alwitt and Prabhaker (1992) determined that the deceptive nature of

advertising has little influence on attitudes toward television advertising. Mehta (2000)

found perceptions of truth in advertising to influence buying interest. While Schlosser,

Shavitt, and Kanfer (1999) found that trust of online advertising did not contribute

significantly to Internet advertising attitudes, Shavitt, Lowrey, and Haefner (1998) found

perceptions of the trustworthiness of advertising to have a sizable effect on overall

advertising attitudes.








Other studies have focused on the belief that advertising promotes materialism

(Durand & Lambert, 1985; Larkin, 1977; Lee & Lumpkin, 1992; Mittal, 1994; Pollay &

Mittal, 1993). Both Mittal (1994) and Pollay and Mittal (1993) found perceptions of

materialism fostered by advertising to have a significant and negative relationship with

attitudes toward television advertising. Lee and Lumpkin (1992) did not find that

perceptions that advertising leads to wasteful buying discriminate between those who

rarely skip commercials on recorded programs and those who skip commercials

sometimes or always.

A significant and positive relationship has been established between the perception

that advertising is good for the economy and attitudes toward television advertising

(Mittal, 1994; Pollay & Mittal, 1993). The perception that advertising totally or partially

subsidizes the cost of media was also found to be a positive and significant contributor to

attitudes toward television advertising (Mittal, 1994).

Another belief perception in studies of attitudes toward advertising is social role

and image (Mittal, 1994; Pollay & Mittal, 1993). This belief reflects the idea that

advertising often attempts to sell the consumer an image or lifestyle, rather than a product

or service. Mittal (1994) found that social image information explained a significant

amount of variance in overall attitudes toward television advertising. This construct has

also been found to have a varied impact on attitudes toward advertising (Pollay & Mittal,

1993).

Other studies have addressed the need for government regulation of advertising

(Barksdale & Darden, 1972; Durand & Lambert, 1985; Larkin, 1977; Schlosser, Shavitt,

& Kanfer, 1999; Shavitt, Lowrey, & Haefner, 1998). Schlosser, Shavitt, and Kanfer








(1999) found beliefs about government regulation of advertising to be unrelated to

Internet advertising attitudes. Similarly, Shavitt, Lowrey, and Haefner (1998) found

perceptions of advertising regulation and effects of advertising on prices to account for an

insignificant amount of variance in overall advertising attitudes.

Other beliefs examined in previous studies that were found to be unrelated to

advertising attitudes include manipulation (Mittal, 1994), social learning (i.e., using

advertising to learn how to behave in social situations) (Alwitt & Prabhaker, 1992), and

value affirmation and corruption (Alwitt & Prabhaker, 1992; Pollay & Mittal, 1993).

Any inconsistencies in the findings of the studies reported above may be

attributable to differences in the operationalizations of the dimensions, the sample, or the

focus of the study (whether examining attitudes toward advertising in general or in a

specific medium).

Attitudes Toward Advertising in a Specific Media Vehicle

Since Bauer and Greyser (1968) noted the moderating effects of the advertising

medium on attitudes toward advertising in general, research has focused on attitudes

toward advertising in a specific media vehicle. In these studies, researchers have used

belief dimensions from studies of attitudes toward advertising in general to understand

attitudes toward advertising in specific media vehicles. These studies, however, rarely

distinguish between different formats of advertisements within the same medium.

While studies in this area have traditionally focused on television advertising

(Alwitt & Prabhaker, 1992; Mittal, 1994), advertising in other media vehicles has also

been studied. For example, Korgaonkar, Karson, and Akaah (1997) found that general

advertising attitude scales are adaptable to direct mail advertising and that beliefs toward

advertising in general are similar to beliefs toward direct mail advertising (Korgaonkar,








Karson, & Akaah, 1997). This study found that while respondents had negative beliefs

toward certain aspects of direct mail advertising, overall, beliefs were generally positive.

The study also concluded that respondents who spent more money as a result of direct

mail advertising and ordered more frequently had significantly more favorable beliefs

toward direct mail advertising (Korgaonkar, Karson, & Akaah, 1997).

Donthu, Cherian, and Bhargava (1993) examined the relationship between attitudes

toward advertising and ad recall in an outdoor advertising context and found that

consumers with a positive attitude toward advertising in general were more likely to

recall outdoor advertisements than those with a negative attitude. Using a sample of video

store patrons, Lee and Katz (1993) concluded that respondents had generally negative

feelings toward advertising on videocassettes.

Historically, overall attitudes toward television advertising have been negative

(Alwitt & Prabhaker, 1992; Bauer & Greyser, 1968; Bartos & Dunn, 1974; Mittal, 1994).

Bauer and Greyser (1968) found that consumers perceive television advertising to be

more annoying than advertising in other media. Alwitt and Prabhaker (1992) found

perceptions of television advertising to be more negative than perceptions of advertising

in general from two earlier Ogilvy and Mather studies (1974, 1985, as cited in Alwitt &

Prabhaker, 1992). Mittal (1994) found nearly half of his respondents reported that they do

not like television advertising and only one fourth reported liking it somewhat. More

recent studies have suggested that attitudes about television advertising may be becoming

more favorable (Shavitt, Lowrey, & Haefner, 1999).

Specific issues with regard to television advertising include a general mistrust of

advertising and feelings of insult. For example, Alwitt and Prabhaker (1992) found that








about 16% of respondents believed advertising presents advertised products accurately

and 66% felt that advertised products do not perform as claimed. Mittal (1994) found the

majority of respondents to consider less than one fourth of television commercials to be

honest and believable.

In contrast, research points to some favorable attitudes toward certain aspects of

television advertising. For example, Alwitt and Prabhaker (1992) found that slightly more

than half of their respondents considered television advertising to be funny or clever.

Similarly, Mittal (1994) found that almost half of his respondents believed that television

commercials are sometimes more enjoyable than the programs.

Recently, researchers have begun to explore attitudes toward Web sites, which

serve as both advertising and a vehicle for advertising (Bruner & Kumar, 2000; Chen &

Wells, 1999; Stevenson, Bruner, & Kumar, 2000). Chen and Wells (1999) developed a

scale to measure attitude toward the Web site, a construct that may be antecedent to the

effectiveness of online advertising on that site. A study by Stevenson, Bruner, and Kumar

(2000) found that as liking of a Web site increases, key variables in the advertising

hierarchy-of-effects, namely attitude toward the ad, attitude toward the brand, and

purchase intention, are improved.

An Online Publishers Association study of 5,000 Internet users classified

respondents into high-affinity and low-affinity users whereby affinity referred to the

user's connection to and engagement with a site (Elkin, 2002b). High-affinity users were

less likely than low-affinity users to feel that ads interfered with their surfing experience

and more likely than low-affinity users to believe the advertised brands are relevant, to








notice ads more, and to believe the sites carry advertising for high-quality products and

services (Elkin, 2002b).

Online Advertising Effectiveness

The popularity of the World Wide Web and the subsequent rise of online

advertising spending have led to studies of advertising in this medium.

Effectiveness of Executional Elements

A review of the literature reveals an emphasis on the impact of executional

elements of Internet advertising design (Bezjian-Avery, Calder, & lacobucci, 1998;

Bruner & Kumar, 2000; Chen & Wells, 1999; Dreze & Zufryden, 1997; Li & Bukovac,

1999; Stevenson, Bruner, & Kumar, 2000). Li and Bukovac (1999) found that animated

banner ads result in a faster response and higher recall than non-animated ads. In

addition, respondents were more likely to respond to and have higher comprehension of

larger, rather than smaller, banner ads.

Bruner and Kumar (2000) and Stevenson, Bruner, and Kumar (2000) examined the

influence of background complexity on the advertising hierarchy-of-effects. In addition,

Bruner and Kumar (2000) also considered the effects of dynamic content (e.g., animated

graphics and commercials) on attitudes. Bruner and Kumar (2000) found that dynamic

content had both a direct negative effect on attitudes toward the Web site and a positive

indirect effect. Dynamic content was found to result in less favorable attitudes toward the

site, which was attributed to the annoyance caused by this type of content. In contrast,

dynamic content makes the site more interesting and thus, it is positively related to

attitude toward the site. Stevenson, Bruner, and Kumar (2000) found that simpler

backgrounds on Web sites had a more positive influence on the advertising hierarchy-of-

effects, including attitude toward the ad, attitude toward the brand, purchase intentions,








and attitude toward the Web site. Bruner and Kumar (2000) could not confirm this

relationship is their study.

Dreze and Zufryden (1997) found a relationship between a number of executional

elements (i.e., background, image size, sound file display, celebrity endorsement, use of

java and frames, and operating system) and the dependent variables of number of pages

accessed and time spent on a Web site. Chen and Wells (1999) determined that the

informativeness, entertainment, and organization of a Web site influence consumer

response to the site.

Bezjian-Avery, Calder, and lacobucci (1998) found the interactivity of online

advertising may hamper persuasion under certain conditions as indicated by decreases in

purchase intention and time spent viewing advertisements when compared to the more

"linear" advertising format of traditional ads. The authors suggested that the best

combination appeared to be when the system properties (i.e., whether predominately

visual or verbal) matched the individual processing needs (i.e., preferring information

presented in a visual or verbal manner).

Online Consumer Behavior as a Measure of Effectiveness

Consumer behavior has been the focus of much of the online advertising

effectiveness literature (see Hoffman, Kalsbeek, & Novak, 1996, for a review). In

counting clicks and hits, researchers have attempted to quantify consumers' use of Web

sites and advertising (Berthon, Pitt, & Watson, 1996). While these techniques have

intuitive appeal and the data may appear more valid than that for other media,

measurement of consumer behavior produces an incomplete picture of the effectiveness

of Internet advertising.








Measures of online behavior have proven to be problematic, both overestimating

and underestimating actual effectiveness (Internet Advertising Bureau, 1997; Riphagen &

Kanfer, 1997). For example, the number of "hits" (i.e., a page view or impression) often

overestimates effectiveness because the user may not have attended to the message

content or the page may not have loaded properly. Clickthroughs on banner ads tend to

underestimate effectiveness, since exposure to the banner ad alone may impact consumer

attitude or future behavior (Briggs & Hollis, 1997). Another reason consumer behavior

should not be used exclusively as a measure of effectiveness is that simply observing

behavior (e.g., clickthroughs on online advertisements such as banner ads) does not

reveal the attitudes behind that behavior (Berthon, Pitt, & Watson, 1996).

Attitudes Toward Online Advertising

A review of the literature reveals a dearth of studies directly measuring attitudes

toward online advertising. The few published studies in this area of research have built a

solid foundation for continued study of attitudes toward online advertising by applying

methodological and theoretical approaches from studies of attitudes toward advertising in

general or attitudes toward advertising in a specific medium, such as television.

One focus of recent Internet advertising studies is attitudes toward Internet

advertising in general. Schlosser, Shavitt, and Kanfer (1999) found that overall attitudes

toward Internet advertising were mixed, with approximately one third of the sample

feeling positive, one third feeling negative, and the remaining one third feeling neutral.

When compared to a demographically-similar sample's perceptions about advertising in

general, fewer respondents felt positive toward Internet advertising than advertising in

general (Schlosser, Shavitt, & Kanfer, 1999). Previte (1998) found that 54% of








respondents agreed that online advertising was a good thing, 47% disagreed that their

opinion of online advertising was unfavorable, and 38% liked online advertising.

Some studies have incorporated belief dimensions from previous studies of

attitudes toward advertising in general. For example, Schlosser, Shavitt, and Kanfer

(1999) examined five dimensions of attitudes toward Internet advertising including

utility, indignity, trust, price perceptions, and regulation. They concluded that utility

(comprised of the traits of informative, entertaining, and useful for making decisions)

explained 43% of the variance in overall attitudes toward Internet advertising.

Ducoffe (1996) used a scale with the dimensions of informativeness, entertainment,

and irritation to determine perceived online advertising value. While the correlations

between these three dimensions and perceived value were all significant, the relationship

between informativeness and perceived value was the strongest. Furthermore, Ducoffe

(1996) found a positive and significant correlation between advertising value and attitude

toward online advertising.

Another focus of these studies is the relationship between attitudes toward online

advertising and attitude toward the ad. Cho (1999) found that subjects with more

favorable attitudes toward Web advertising overall had a more favorable attitude toward a

specific banner ad.

Advertising attitudes and clicking behavior have been examined as well (Brill,

1999; Cho, 1999). Cho (1999) studied the relationship between attitudes toward online

advertising in general and clicking behavior. He found that the three of the five belief

statements used to assess overall attitude toward Web advertising were capable of

discriminating between subjects with a high intention to click through a banner ad and








those with a low intention. The discriminating belief statements included the following:

Web advertising supplies valuable information, Web advertising is necessary, and Web

advertising is valuable. Brill (1999) found that consumers who had clicked on specific

banner ads had more favorable attitudes toward the banner ad and higher purchase

intentions toward the products or services advertised in the banner ad than for unclicked

banner ads.

An analysis by Briggs and Hollis (1997) focused on the influence of banner ads on

consumers' attitudes and behavior. They found that for five of the six banner ads that met

the threshold on brand perception items, consumers exhibited a significant positive

change in attitudes as a result of exposure to the ads.

The above discussion illustrates a recent focus of attitudes toward advertising

literature on attitudes toward online advertising. In examining belief dimensions of online

advertising attitudes and demonstrating the relationship between attitude toward online

advertising and attitude toward the ad, researchers have expanded the applicability of the

attitudes toward advertising construct.

While several studies have examined attitudes toward Internet advertising

(Ducoffe, 1996; Mehta & Sivadas, 1995; Schlosser, Shavitt, & Kanfer, 1999), fewer have

distinguished among the various online advertising formats. Previous studies on Internet

advertising often either considered attitudes toward online advertising in general

(Ducoffe, 1996; Schlosser, Shavitt, & Kanfer, 1999) or examined one format of online

advertising (e.g., newsgroup and e-mail advertising (Mehta & Sivadas, 1995)). Others

have compared one online ad format to advertising formats in traditional media (Dynamic

Logic, 2001b) or compared two or more online ad formats (Dynamic Logic, 2001b;








Harris Interactive, 2001; Statistical Research, 2001). No study has attempted to

understand the range of dimensions that influence attitudes toward various online

advertising formats or the impact of attitude toward online advertising format on other

variables.

Studies of attitudes toward Internet advertising often ask respondents to respond to

survey items with all online advertising formats in mind (e.g. Ducoffe, 1996; Schlosser,

Shavitt, & Kanfer, 1999). In the studies by Ducoffe (1996) and Schlosser, Shavitt, and

Kanfer (1999), the researchers did not define the range of online advertising formats for

the respondents. In the Schlosser et al. study, Internet advertising was defined as "any

form of commercial content available on the Internet that is designed by businesses to

inform consumers about a product or service" (1999, p. 36). As a result, respondents

almost certainly answered with unique representations of online advertising in mind. In

addition, although these studies collected data on respondents' definitions of online

advertising, no comparisons were made between these conceptualizations and overall

attitudes.

Other studies have focused on only a small subset of online ad formats. Mehta and

Sivadas (1995) concluded that consumers held unfavorable attitudes toward advertising

on newsgroups and via e-mail, regardless of the degree of relevance of the message to the

special interests of the group. Briggs and Hollis (1997) and Cho (1999) focused

specifically on banner advertising. While Briggs and Hollis (1997) considered attitude

toward a banner ad as an independent variable and studied its effect on brand attitude,

Cho (1999) studied the influence of attitude toward Web advertising in general on








attitude toward a banner ad and clicking intention. These findings of these studies may be

limited by the narrow focus.

A study by Statistical Research (2001; Jackson, 2001b) compared consumer

attitudes toward banner ads to attitudes toward pop-up ads. Respondents were more likely

to agree strongly or somewhat that they notice pop-up ads more than banner ads (76% vs.

69%) and that pop-up ads interfere with reading or using a Web page (84% vs. 54%).

Respondents were also more likely to disagree strongly or somewhat that companies that

use pop-up advertising are market leaders more so than companies that use banner

advertising (57% vs. 48%).

A study by Harris Interactive (2001) compared Superstitial" advertising to

television advertising in terms of some classic communication research measures,

including recall, communication, and persuasion. The study found that for the three ads

tested, Superstitials communicated the copy points as well or better than television ads.

Two of the three Superstitialso tested were as likable as the television commercials.

Intentions to use, buy, or consider the brand were comparable for both Superstitials and

television commercials in all three cases. Finally, brand recall for Superstitials was

slightly lower than that for television ads (81% vs. 93%).

The Interactive Advertising Bureau examined the use of the large rectangle ad

format in a study for Coca-Cola (Lefton, 2001b). The ad showed a lift in message

association by 56%, brand favorability by 7%, and purchase intent by 5% (Lefton,

2001b).

A study by Dynamic Logic (2001b) compared attitudes toward a number of online

ad formats. Over half of the respondents had a positive attitude toward banner advertising








(53%), followed by skyscraper ads (35%), large rectangles (17%), pop-ups (6%), and

interstitials (3%). This study also examined attitudes toward pop-up advertising and

traditional formats of advertising. Newspaper, magazine, radio, and billboard advertising

were found to be more desirable than pop-up advertising, while telemarketing, direct

mail, and television advertising were less desirable. While more comprehensive in terms

of formats than other studies, the Dynamic Logic study lacked a theoretical emphasis and

did not measure other important variables, such as Aad and online ad perceptions.

Studies measuring attitudes toward online advertising in general are often too broad

to provide practical value to the advertiser. Because each online ad format possesses

distinctive features, attitudes toward online advertising could differ depending on the

user's perception of what constitutes online advertising. Furthermore, the findings of

studies that focus on only one online ad format are not generalizable to other online ad

formats. As demonstrated, a review of the literature reveals no comprehensive study of

attitudes toward specific online advertising formats within a theoretical model of

advertising.

Although such a study has yet to be published, researchers are raising questions

about the relationship between attitude toward the online ad format and attitude toward

the ad. As previously mentioned, Rodgers and Thorson (2000) included ad formats and

attitude toward the ad in their Interactive Advertising Model, but did not test this

relationship. Schumann, Artis, and Rivera (2001) suggested a number of research

questions for future research including "What negative influences on consumer

perceptions are likely to arise from interactive advertising formats?" and "Which

interactive media formats will best facilitate relationship management?"








Advertising on the Internet is still evolving, manifesting itself in a variety of

formats, from banner ads to pop-ups to Webmer-ials. The Internet provides a more

versatile medium for advertising than traditional vehicles, and this versatility has led to

the development of the varied online advertising formats. While the Internet is an

appropriate context to study attitude toward ad format, it is also critical that attitudes

toward online formats are thoroughly examined. With the seemingly limitless

possibilities for online advertising formats, an understanding of consumer attitudes and

belief dimensions can certainly inform the future development of online advertising.

Proposed Model

The Interactive Advertising Model developed by Rodgers and Thorson (2000)

integrates the function of the Internet for consumers (i.e., consumer-controlled aspects)

and the structure of Internet ads (i.e., advertising-controlled aspects) to suggest consumer

responses, which include, among others, the formation of attitude toward the ad. One of

the advertiser-controlled Internet ad structures in the model is "ad formats," such as

banners, interstitials, and sponsorships (Rodgers & Thorson, 2000). Another is "ad

features," which are objective (i.e., advertiser-controlled) and subjective (i.e., consumer-

controlled) variables. The objective ad features for the Internet include color, animation,

and audio, while the subjective ad features include exciting, interesting, and boring

(Rodgers & Thorson, 2000).

This model has two implications for the current study. First, this model suggests a

relationship between ad format and attitude toward the ad. Second, this model

acknowledges the role of ad features or perceptions in attitude formation.

The foregoing discussions of the Interactive Advertising Model (Rodgers &

Thorson, 2000), the structural model of the antecedents to attitude toward the ad








(MacKenzie & Lutz, 1989; Lutz, 1985), and the literature of attitude toward advertising,

attitude toward advertising in a media vehicle, and attitude toward online advertising

provide a foundation for understanding how attitude toward online ad format may fit into

the existing attitude toward the ad model. Figure 2-2 illustrates the structure of a

subsection of the proposed model.


Figure 2-2. Proposed Structural Model of Aad Formation (showing two antecedents) in an
Online Advertising Context


Attitude toward online ad format is proposed as an antecedent to attitude toward the

ad. Figure 2-2 illustrates two antecedents for Aad: attitude toward online ad format and

attitude toward online advertising. The variable of attitude toward online advertising

serves to separate attitudes toward all advertising from those only related to online

advertising. As a relatively new form of advertising using a medium unlike other media,








online advertising may produce attitudes that are distinct from attitudes toward

advertising in general as found by Schlosser et al. (1999).

Underlying attitude toward online ad format are perceptions of online ad formats.

Just as attitude toward the ad is determined by ad perceptions (MacKenzie & Lutz, 1989),

attitude toward online ad format is predicted to be determined by online ad format

perceptions. Because these perceptions may parallel many of the perceptions of

advertising, a thorough discussion of these perceptions or beliefs was provided. This

research will identify other possible belief dimensions for the various online advertising

formats.

Furthermore, this model proposes that attitude toward online ad format is

influenced by attitudes toward online advertising. In addition, a user's attitude toward the

Internet may influence attitude toward online advertising formats.

For advertising hosted on a Web site, attitude toward a Web site may influence

attitude toward the ad format. This relationship is suggested by previous research that

found a strong and significant correlation between an attitude toward a television

program and attitude toward television advertising, even after controlling for

demographic variables (Alwitt & Prabhaker, 1992).

Because the current interest is focused on attitude toward the online advertising

format, only a subset of the entire Aad model was tested. The subsequent findings should

provide a framework for further research into other parts of the model.

Conclusion

This chapter presented the Aad model and then examined the research on attitudes

toward advertising in general in more detail. Recent trends in research on attitudes toward

advertising were also addressed, including the emphasis on understanding attitudes





62


toward advertising in a specific medium and attitudes toward online advertising. Finally,

this review described how the proposed concept-attitude toward online advertising

format-fits into the existing model and suggested hypotheses to be tested.

Chapter 3 discusses the first study in this research. This study used a qualitative

approach to derive perceptions of specific online advertising formats and determine

which formats should be included in a quantitative study that will test these relationships.













CHAPTER 3
STUDY 1

Purpose

The purpose of Study 1 was to use qualitative research to investigate the

determinants of attitude toward online advertising format, with special emphasis on

defining the dimensions of ad format perceptions, and to uncover online ad formats

appropriate for further study.

Research Questions

The research questions guiding this study were as follows:

1. What online advertising formats should be included in a study of attitudes toward
online ad formats?

2. What are the online ad format perceptions underlying online ad format attitudes?

Critique of Methodology ii Previous Research

The perceptual dimensions underlying advertising attitudes and the items used to

measure them have often been constructed through reviews of previous studies rather

than through exploratory methods (O'Donohoe, 1995). The influential study published by

Bauer and Greyser (1968) has historically been the basis for many perceptual inventories.

The use of items from previous studies offers the advantages of replication and

continuity.

While the precedent has been to adapt previous perceptual dimensions to the area

of study, the validity of the measures relevant to a particular area (e.g., online

advertising) may be improved by using exploratory research, such as focus groups or

interviews, to derive and define the appropriate dimensions (Churchill, 1979). For








example, Muehling (1987) and Pollay and Mittal (1993) incorporated thought-listing

techniques and open-ended questions to derive perceptual dimensions rather than relying

solely on previous research.

Perceptual dimensions from previous studies of attitudes toward advertising could

have been applied to a study of attitude toward online ad format. However, because

attitude toward the format is a newly-considered construct, it was important to enhance

the validity of the measures to be used by conducting preliminary qualitative research.

Method

Study 1 included depth interviews with industry experts and experienced online

users to explore online ad format perceptions. The literature review (Chapter 2) identified

perceptions of advertising in general or advertising in a specific medium. The purpose of

this review was to form a foundation of understanding advertising perceptions and for

comparison purposes following the depth interviews.

Depth Interviews With Industry Experts

Depth interviews with industry experts provided insight into the online advertising

formats that are most important, prevalent, distinct, and emerging, as well as the

perceptual dimensions on which the various online ad formats can be distinguished.

Sample. A total of 34 online advertising experts were identified representing

academe (6) and the advertising industry (26). Advertising academicians were selected

from the set of authors of papers or articles on the topic of online advertising published

during the past three years in the Proceedings of the American Academy ofAdvertising,

the Journal ofAdvertising, the Journal of Interactive Advertising, and the Journal of

Advertising Research. Practitioners were selected from those who either wrote or were

quoted in trade publication articles about online advertising. In addition, industry








members of the American Academy of Advertising were contacted for referrals and a

message was posted on the Online Advertising Discussion List to solicit additional

prospects.

Eleven members of the advertising community, including academicians and

practitioners, were interviewed. Of those in academe, five were contacted and three

completed the interview. Of those in industry, 11 were contacted and 8 completed the

interview. Sixty-nine percent of the contacts made resulted in a completed interview.

The academe participants represented the fields of marketing (2) and advertising

(1). The industry participants included three who work for research organizations, one

Web site advertising account director, two interactive creative directors, and two former

employees of online advertising networks.

The selected individuals were contacted by e-mail and invited to participate in this

research. The informed consent form was either faxed or sent via e-mail. The interviews

were conducted by phone and lasted approximately 20 minutes to one hour.

Measures. The depth interviews with industry experts served to narrow the online

ad formats that should be considered for further research to facilitate the later collection

of detailed data on online advertising formats and determine online ad format

perceptions.

Participants were asked to list prevalent online ad formats. They were also asked to

name formats that are and will be important in the future, should be included in an

attitude study, and are most similar or dissimilar, with special emphasis on the

dimensions that differentiate the formats. Participants were also asked to describe their








opinions of various online advertising formats. A discussion guide for the interviews with

industry experts is included in Appendix A.

Depth Interviews With Experienced Online Users

Depth interviews with experienced Web surfers were used to further develop a

typology of the dimensions of online ad formats and determine user familiarity with the

range of online advertising formats.

Sample. Experienced Web surfers were located through e-mail recruiting.

Seventeen students from a large southern university and six non-students from a southern

metropolitan area were screened for the interview. The screening process involved asking

questions about familiarity with online advertising formats and the amount of time spent

surfing the Web during the typical week. Prospective participants who exhibited the

highest levels of familiarity with multiple online advertising formats and spent the

greatest amount of time online were selected based on the assumption that they would be

able to discuss online advertising formats more knowledgeably, thereby producing more

valuable data. To somewhat disguise the purpose of the study prior to the actual

interview, prospective participants were also asked to name three Web sites they

regularly visit and up to three sites from which they have made an online purchase.

Of those recruited, six students and four non-students qualified to participate in the

interview. All students who were interviewed reported surfing a minimum of eight hours

per week (M = 10.8). All non-students except one estimated that they spend at least 20

hours per week surfing the Web, while one reported spending approximately 45 hours per

week surfing. All participants were familiar with at least three formats of advertising

prior to the interview.








Once an individual qualified to participate, an appointment for the hour-long

interview was arranged. Participants were paid $25 for their participation.

Stimuli. Stimulus ads were selected from the advertising displayed on several

popular Web sites, as well as various online galleries, collections, and portfolios. Table 3-

1 lists the advertisers used to illustrate each of the formats. The stimuli represent a broad

range of online advertising formats mentioned by Internet advertising experts including

banners, buttons, floating ads, pop-ups, interstitials, large rectangles, skyscrapers, and

Top Roll.

Table 3-1. Advertisers Represented in Stimulus Ads
Online Advertising Format Advertiser

Banners Apartmentguide.com, Casino on Net, UBid

Buttons Amazon, Staples, Wal-mart

Floating ads Circuit City, Emirates Airlines, Boston Red
Sox, ING

Pop-ups Air Force, Nikon, Ford Expedition

Interstitials Glaxo, Casino on Net

Large rectangles Absolut, Dell, IBM

Skyscrapers Best Buy, Classmates

Top Roll Ford Focus, Coca-Cola

During the interview, a laptop computer was used to demonstrate online ad

formats. Ads were shown in the context of a Web site to simulate an actual ad

impression.

Measures. Participants were first asked about their general impression of online

advertising. The interview also tapped specific online advertising formats to determine








familiarity with various online advertising formats, opinions of these formats, and the

differences among the formats.

Participants were then asked to focus specifically on one ad format. Three

examples of each format were presented for illustrative purposes. Thought-listing was

implemented at this point. This approach has been applied successfully in the study of

advertising (Batra & Ray, 1986; Lutz & MacKenzie, 1982) and more specifically, to the

study of attitudes toward advertising in general (Muehling, 1987).

Participants were invited to read the standard thought-listing instructions (Cacioppo

& Petty, 1981) and were given two minutes to complete each thought-listing exercise.

They were instructed to write their thoughts about each online ad format as it was

presented. The most prevalent thoughts were considered appropriate for further analysis.

Participants were also asked to describe what they like and dislike about the ad

format, as well as their opinion of Web sites that use the ad format. This process was

repeated for a total of five online advertising formats. Finally, participants were asked

describe the similarities or dissimilarities among various online advertising formats.

Questions used in the depth interviews with experienced Web users are included in

Appendix B.

Procedure for Selecting Online Advertising Formats

Four criteria were used to determine the inclusion of an online advertising format in

future studies. The ad formats chosen can be described as prevalent, important,

distinctive, or emerging, with many of the ad formats representing several of these

categories.

First, only the most prevalent online ad formats were considered. These formats

included those mentioned by Web surfers during unaided recall, those that Web surfers








were familiar with during aided recall, and formats cited as the most prevalent by

industry experts. These formats were also compared to those representing the highest

percentage of online creative elements as reported by AdRelevance (2003) and online

advertising revenue as reported by PricewaterhouseCoopers (2002) in the IAB Internet

Advertising Revenue Report.

Second, important ad formats were considered for inclusion. To determine the

important advertising formats, advertising experts were asked which formats should be

included in a consumer attitude study about online advertising formats.

Third, to further narrow the possibilities, only distinct advertising formats were

selected. Therefore, if two formats are virtually indistinguishable to Web surfers or the

experts, one was dropped from further study. Advertising experts were asked to describe

various categories of advertising formats. Experienced Web surfers were asked to

compare and contrast several formats.

Finally, emerging online ad formats cited by industry experts were considered. The

goal of this selection process was to determine five to eight prevalent, distinct, important,

and emerging online ad formats, as this number should be manageable for a later

quantitative study.

Format Selection Results

From the experienced Web surfer interviews, banners and pop-up ads emerged as

the two most frequently cited formats in terms of unaided recall. Almost all participants

mentioned pop-ups during unaided recall of online advertising formats and most

mentioned banner ads. Participants were familiar with most of the formats demonstrated.

Table 3-2 illustrates how the formats ranked in terms of unaided recall and recognition.








The ad formats with the most mentions during the interviews with Internet

advertising experts included banners, pop-ups, pop-unders, skyscrapers, large rectangles,

floating ads, sponsorships, and interstitials. Other ad formats mentioned included top

Rolls, jump pages, Superstitials, fixed logos, buttons, towers, Web sites, search engine

listings, electronic mailing lists, text links, streaming media, and e-mail.

Table 3-2. Unaided Recall and Recognition of Online Ad
Formats by Experienced Web Surfers (N= 10)
Unaided recall Recognition
Format N N

Pop-up 9 10

Banner 7 10

E-mail 4 *

Pop-under 2 10

Floating ads 2 8

Instant messaging 1 *

Tower I *

Large rectangle 1 10

Button 1 10

Contextual 1 6

Skyscraper I 10

Interstitial 0 9

Top Roll 0 8

Sponsorship 0 4
*Not demonstrated.

The IAB Internet Advertising Revenue Report conducted by

PricewaterhouseCoopers reported that the two formats garnering the highest percentage








of revenue for the first six months of 2002 were banners and sponsorships

(PricewaterhouseCoopers, 2002). Banners represented 33% of online banner revenue and

sponsorships represented 24% (PricewaterhouseCoopers, 2002). Interstitials and rich

media ads (e.g., floating ads) each represented 3% of revenue (PricewaterhouseCoopers,

2002).

In contrast to PricewaterhouseCoopers' method of calculating the proportion of

total online advertising revenues each format contributes, AdRelevance bases its data on

the total number of creative elements. According to AdRelevence data (2003) for the

week of April 28, 2003, banners were the most prevalent online advertising format,

representing 33% of online advertising elements. Including half banners increases this

percentage by 4% (AdRelevance, 2003). Skyscrapers also represented a high percentage

of online advertising elements at 17% for standard skyscrapers, wide skyscrapers, and

vertical banners combined (AdRelevance, 2003). Buttons represented 14% of elements,

and squares and medium rectangles totaled 10% (AdRelevance, 2003).

When asked what advertising formats should be included in a consumer attitude

study about online advertising formats, advertising experts were most likely to mention

banners, pop-ups, pop-unders, skyscrapers, floating ads, and sponsorships. Other formats

with fewer mentions included Point Roll, Superstitials, large rectangles, and text links.

Advertising experts suggested the following categorization schemas: flat, animated,

or interactive ads; small, larger, or floating ads; ads contained within Web page or ads

outside of Web page. Both experienced Web users and advertising experts often

perceived advertisements that were integrated into the content of the Web page to be one

distinct group and advertisements that appeared over or under the content to be another.








In further categorizing ads that are integrated into page content, skyscrapers and large

rectangles were often perceived to be distinct from banners and other smaller ad formats,

such as buttons. Another category identified were ads that appeared in the place of

content, such as interstitials and Superstitials.

Using this framework, banners, large rectangles, skyscrapers, towers, and buttons

would fall into one category while pop-ups, pop-unders, floating ads, and Top Rolls

would be a separate category. Interstitials and Superstitials would constitute a third

category. Ads integrated into page content might be further classified as large or small.

Finally, emerging online ad formats cited by industry experts were considered for

inclusion. From the interviews with the advertising experts, floating ads were often cited

as an emerging format that will become more popular in the future. Larger sizes, such as

large rectangles, were also mentioned as a trend in online advertising.

Analysis of formats using these criteria produced six advertising formats that were

used in subsequent studies: banners, pop-ups, floating ads, skyscrapers, large rectangles,

and interstitials. All of these formats were mentioned the most often by advertising

experts, and banners and pop-ups had the highest unaided recall by experienced Web

surfers. Furthermore, floating ads and large rectangles were often cited as emerging

formats. Banners, pop-ups, floating ads, and skyscrapers were all mentioned by

advertising experts to be important to include in an advertising attitudes study. While

interstitials were not cited as important or emerging, they were often cited as prevalent by

advertising experts. In addition, they represent a unique category of ads that appear

between content and are distinct from other formats. Table 3-3 illustrates how these six

formats rate on the four decision criteria.








Table 3-3. Summary of Performance of Chosen Formats across Selection Criteria
Criteria Banner Pop-up Skyscraper Floating Large Interstitial
Rec

Prevalent X X X X X X

Important X X X X

Distinct X X X X X X

Emerging X X

At this point, it is important to note that although a number of advertising experts

considered sponsorships to be an important online advertising format, and sponsorships

represented the second highest percentage of online advertising revenue for the first half

of 2002 (PricewaterhouseCoopers, 2002), this format was not considered in subsequent

studies for two reasons. First, sponsorships are often comprised of other online

advertising formats. Second, because sites offer a wide variety of sponsorship

opportunities, there is no single definition of a sponsorship.

Other ad formats were dropped from consideration because they were not

distinguishable enough from another format. For example, while pop-under ads were

often cited as prevalent and important, this format is quite similar to pop-up ads.

In contrast, although skyscrapers and banner ads are similar in shape, skyscrapers

were included in the list of formats for future study. Banners were one of the first formats

to emerge online, while skyscrapers are a more recent creation. For this reason, it is

possible that attitudes differ. In addition, some participants associated skyscrapers more

with large formats, such as large rectangles, and banners with smaller formats, such as

buttons. The placement of the ads also differs, with banners placed across the top of a

Web page and skyscrapers along the side.








Perceptual Dimensions of Online Advertising Formats

Online advertising perceptions emerged from an analysis of the responses to the

interview questions. These perceptions were then categorized into dimensions using

"conceptual factor analysis".

Internet advertising experts mentioned a total of 40 unique perceptions in

descriptions of online advertising, which were used to define perceptual dimensions. A

conceptual factor analysis of the perceptions was performed by grouping similar (or

opposite) perceptions into the same category. Within each category, the perceptions are

either synonyms (or antonyms) or were used by participants to describe the same concept.

This analysis resulted in the categories presented in Table 3-4. An indication of the

incidence of mention for each category is also provided.

Web surfers mentioned a total of 84 unique perceptions in descriptions of Internet

advertising, which were also used create dimensions. Once again, a conceptual factor

analysis of the adjectives was performed in the same manner described previously. This

analysis resulted in categories presented in Table 3-5. Because of the vast number of

adjectives mentioned by Web surfers, an attempt was made to further refine the

categories developed from the data for the advertising expert sample. An indication of the

incidence of mention for each category is also provided.

Irritation

"Annoying" and "irritating" clearly emerged as the most common descriptors of

online advertising. This dimension was more generally described by advertising experts

with the following adjectives: annoying, interrupts, intrusive, abrupt, distracting, and

interferes.








Table 3-4. Descriptors of Online Ad Formats Used by Online Advertising Experts
Descriptors N

Annoying, Interrupts, Intrusive, Abrupt, Distracting, Interferes 7

Informational value, Content, Compelling message, Room for content 7

% of screen occupied, Position, Size of ad, Clutter, Obtrusive 6

Relevance, Targeted 5

Interactive, Involvement 5

Control, Forced vs. voluntary, Choice 4

Cutting edge, Different, Sophisticated, Innovative, Clever, Creative 4

Animation, Flashing, Static (ant.) 3

Enjoyable, Entertaining 2

Classic, Tasteful, Cordial 1

Copy heavy 1

Provides reward 1

Ubiquitous 1

Visual 1
Note. N= number of participants who mentioned at least one of the descriptors.

Other categories which may contribute to the annoyance of online advertising

include one referring to the clutter caused by online advertising (percentage of screen

occupied, position of ad, size of ad, clutter, obtrusive), one referring to the activity of

online ads (flashing, animated), one referring to the ubiquity of ads (ubiquitous), and one

referring to the ability of the user to control the surfing experience (control, forced,

voluntary, choice).








Table 3-5. Descriptors of Online Ad Formats Used by Experienced Web Surfers
Descriptors N

Animated, Flashy, Blinking, Movement, Hyperactive, Static (ant.) 10
Annoying, Bothersome, Frustrating 10
Disruptive, Distracting, Diverts attention, Distorts content, Gets in the 9
way, Takes over page, Takes up space, In your face/Out of the way,
Interferes with background, Intrusive
Inconvenient, Time-consuming, Quick (ant.) 9
Entertaining, Exciting, Fun, Appealing, Cool, Neat, Amusing 9
Catches attention, Holds attention, Eye-catching, Interesting, 9
Noticeable, Obtrusive, Blends with site (ant.), Easy to ignore (ant.),
Contrasts with background
Innovative, Inventive, Clever, Cutting-edge, Different, Creative 8
Forced exposure, Wanted/unwanted, Control, Removal requires action 7
Cluttered, Overbearing, Pervasive, Obtrusive 6
Cool graphics, Good pictures, Images of product, Interesting graphic 6
Bold, Bright colors, Colorful 5
Audience-driven, Relevant to content 5
Big, Small, Size, Space 4
Simple, Plain 4
Beneficial, Useful offerings, Pointless (ant.) 4
Cute, Eyesore (ant.) 3
Informative 3
Separate from page, On page 3
Easy to read, Too many words (ant.) 2
Respectful, Good/bad etiquette 2
Extravagant, Dynamic 2
Interactive, Involvement 2
Sound 2
Repetitious 1
Deceptive 1
Note. N= number of participants who mentioned at least one of the descriptors.








As mentioned previously, the volume of descriptors mentioned by Web surfers

warranted developing several categories which may by encapsulated into the broader

category of "annoying" from the general descriptions of annoying, bothersome, and

frustrating to more specific segmentations. These segmentations include one that refers to

the disruption of the surfing experience (disruptive, distracting, diverts attention, distorts

content, gets in the way, takes over page, takes up space, in your face/out of the way,

interferes with background, intrusive, cluttered, overbearing, pervasive).

Another segmentation of annoyance refers to the activity of the ad itself, which

includes such descriptors as animated, flashy, blinking, movement, and hyperactive. Yet

another segmentation refers to the time involved in dealing with online ads as expressed

by the descriptors of inconvenient and time-consuming. Finally, irritation may be caused

by the fact that online advertising is often forced onto the user, which was described as

forced exposure, wanted/unwanted, control, and requires action to remove. In fact, Li,

Edwards, and Lee (2002) found the measure of intrusiveness to be independent from that

of irritation, which provides some evidence for the separation of irritation and intrusion.

The finding that online advertising is defined by or can be described by its level of

irritation or annoyance is consistent with that of Uses and Gratifications research

(Eighmey & McCord, 1997; Plummer, 1971; Schlinger 1979) and attitude toward

advertising studies (Ducoffe, 1995, 1996). Irritation is such a pervasive issue in

advertising that this characteristic has also merited studies about it exclusively, as

demonstrated by the advertising clutter research (Elliott & Speck, 1998; Ha, 1997).

Entertainment

Online advertising was also described by participants in terms of its entertainment

value. These categories ranged from the general category of entertainment (entertaining,








enjoyable, exciting, fun, appealing, cool, neat, amusing) to more specific categories

referring to activity of the ad itself (animation, flashing, blinking, movement,

hyperactive), to the fact the ad is eye-catching (catches attention, holds attention, eye-

catching, interesting, noticeable, obtrusive, contrasts with background), or to the

entertainment value in the graphics of the ad (cool graphics, good pictures, images of

product, interesting graphic).

As noted in Chapter 2, perceptions of the entertainment dimension of advertising

have been considered in previous studies (Alwitt & Prabhaker, 1992, 1994; Bauer &

Greyser, 1968; Ducoffe, 1996; Lee & Katz, 1993; Mehta, 2000; Mittal, 1994; Pollay &

Mittal, 1993; Schlosser, Shavitt, & Kanfer, 1999; Shavitt, Lowrey, & Haefner, 1998).

Information

While the information dimension clearly emerged from the interviews with

advertising experts, Web surfers were less likely to mention it. While only three Web

surfer participants noted the informativeness of online advertising, advertising experts

described information in terms of informational value, content, compelling message, and

space in ad for content.

Because many of the advertising expert participants were responsible for creating

or selling online ads, they should be more attuned to the content possibilities of online

advertisements, providing a possible explanation their emphasis on information. The Web

surfer participants may have been more focused on the design or behavior of the online

ads during the interview, particularly if the advertised product was not of interest to them.

While this dimension was not described in great detail by Web surfer participants,

its mention by advertising experts and its dominance in both Uses and Gratifications

research (Eighmey & McCord, 1997; Plummer, 1971; Schlinger, 1979) and attitudes








toward advertising research (Alwitt & Prabhaker, 1992, 1994; Ducoffe, 1996; Lee &

Lumpkin, 1992; Mehta, 2000; Mittal, 1994; Muehling, 1987; Pollay & Mittal, 1993;

Schlosser, Shavitt, & Kanfer, 1999; Shavitt, Lowrey, & Haefner, 1998) warrants its

inclusion in further studies.

Informativeness is also linked to relevance, as determined by Plummer (1971)

who identified "informativeness and personal relevance" as a factor explaining attitude

toward television commercials. The relevance of an online ad was mentioned by

participants who used such adjectives as audience-driven, relevant to content, and

targeted to describe this dimension. Web surfers also referred to online ads as beneficial

or having useful offerings, which can also be incorporated under this dimension.

Novelty

The novelty dimension also emerged from the qualitative research. Advertising

was described as cutting-edge, different, sophisticated, innovative, clever, creative, and

inventive. Both Plummer (1971) and Schlinger (1979) identified this dimension in their

studies of attitudes toward television commercials.

Interactivity

The interactivity of or involvement with online ads also surfaced during the

interviews. Interactivity has been defined as "the extent to which users can participate in

modifying the form and content of a mediated environment in real time" (Steuer, 1992, p.

84). This characteristic was addressed more by advertising experts, which may be

explained by the experience of many of the experts with the sophisticated technologies

used to create interactive online ads. Again, this dimension has appeared in Uses and

Gratifications studies as either "involvement" (Plummer, 1971) or "interactivity"

(Eighmey & McCord, 1997).








Composition

The final dimension is the "look" or composition of the online advertisement.

Web surfers defined this characteristic with descriptors referring to the colors of the ad

(bright colors, colorful, bold), the size (big, small), the simplicity (simple, plain), the text

(easy to read, too many words), or the overall appearance of the ad (cute, eyesore).

Industry experts also addressed the look of the ad to a lesser extent using descriptors such

as classic, tasteful, cordial, copy heavy, and visual.

This dimension reflects the instrument of advertising, which Sandage and

Leckenby (1980) differentiated from the institution of advertising. While institution

refers to the purpose and effects of advertising, instrument refers to advertising's

executional properties.

Table 3-6 lists these dimensions and their subdimensions and corresponding

descriptors. Several descriptors were selected from each of these dimensions to represent

the dimension in the next stage of research. For the annoyance dimension, the descriptors

of annoying, intrusive, overbearing, and disruptive were selected. To represent the

entertainment dimension, the descriptors entertaining, amusing, and eye-catching were

selected. Information, useful, and beneficial were chosen to represent the information

dimension. Innovative, different, and sophisticated were selected for the novelty

dimension. Finally, attractive and elaborate (antonym of plain) were chosen to represent

the composition dimension. Because the focus of future studies will be collecting data on

subjective dimensions, the interactive dimension was disregarded as it tends to be more

objective.








Table 3-6. Summary of Dimensions and Corresponding Descriptors
Dimensions Subdimensions Descriptors

Annoyance General Annoying, bothersome, frustrating
Disruption of Interrupts, abrupt, disruptive, distracting, diverts
experience attention, intrusive, interferes, distorts content,
(Intrusion) gets in the way, takes over page, in your
face/out of the way, interferes with background
Clutter Percentage of screen occupied, position of ad,
size of ad, clutter, obtrusive, takes up space,
overbearing, pervasive
Activity Flashing, animated, flashy, blinking, movement,
hyperactive
Ubiquity Ubiquitous
Ability to control Control, forced, voluntary, choice, forced
exposure, wanted/unwanted, requires action to
remove
Time factor Inconvenient, time-consuming
Entertainment General Entertaining, enjoyable, exciting, fun,
appealing, cool, neat, amusing
Activity Flashing, animated, flashy, blinking, movement,
hyperactive
Eye-catching Catches attention, holds attention, eye-catching,
interesting, noticeable, obtrusive, contrasts with
background
Graphics Cool graphics, good pictures, images of
product, interesting graphic
Information General Informative, informational value, content,
compelling message, space in ad for content
Relevant Audience-driven, relevant to content, targeted,
beneficial, useful offerings
Novelty General Cutting-edge, different, sophisticated,
innovative, clever, creative, and inventive
Interactivity General Interactivity, involvement


Composition Colors
Size
Simplicity
Text
Overall appearance


Bright colors, colorful, bold
Big, small


Simple, plain
Easy to read, too many words
Cute, eyesore








Discussion

The depth interviews identified critical perceptual dimensions of online advertising

formats and informed the development of items to be used to measure each dimension.

The interviews also highlighted six online advertising formats worthy of future study.

Online Advertising Formats

The six formats selected were banners, pop-up ads, skyscrapers, large rectangles,

floating ads, and interstitials. These formats have often been included in research studies

either in combination, in isolation, or in comparison to a traditional medium, such as

television.

The Dynamic Logic (2001 a) Ad Unit Effectiveness Study measured the

effectiveness of banners, skyscrapers, and large rectangles. The Advertising Reaction

Study also by Dynamic Logic (2001b) measured attitudes toward banners, pop-up ads,

skyscrapers, large rectangles, and interstitials. Attitudes toward pop-ups and banners have

been compared (Statistical Research, 2001). Other studies have examined the

effectiveness of banner ads (Briggs & Hollis, 1997; Cho, 1999; Gilliam, 2000; Morgan

Stanley, 2001), large rectangles (Lefton, 2001), or Superstitials (Harris Interactive,

2001).

Two of these formats-the skyscraper and the large rectangle-are part of the

lAB's recommended universal ad package, which further validates their importance

(Elkin, 2002d). Furthermore, Nielsen//NetRatings reported that banners, rectangles, and

skyscrapers represented the highest percentages of online advertising impressions (Martin

& Ryan, 2003).

While floating ads have yet to be included in an attitude study, this format is

expected to attract the attention of consumers and researchers as it becomes more








prominent and widely used. A Nielsen//NetRatings study found that almost three times as

many advertisers in the fourth quarter of 2002 than in the same quarter of 2001 used the

format (Buchwalter & Martin, 2003).

In conclusion, these six formats are appropriate for future studies based on their

dominance in the depth interviews and confirmation provided by inclusion in other

research studies, recognition by the IAB, or data indicating the prevalence or predicted

growth of the use of the format.

Perceptual Dimensions

The five perceptual dimensions identified in this qualitative study were annoyance,

entertainment, information, novelty, and composition. Four of these five dimensions have

been identified in previous Uses and Gratifications studies (Plummer, 1971; Schlinger,

1979). In these studies, entertainment, familiarity (opposite of novelty), information, and

irritation (similar to annoyance) were recognized as categories of viewer rewards of

television.

These dimensions are consistent with those in studies of attitudes toward

advertising. Annoyance of advertising has been linked to attitudes toward advertising

(Aaker & Bruzzone, 1985; James & Kover, 1992). The study of entertainment of

advertising can be traced back to Bauer and Greyser's 1968 study, which inspired the

inclusion of this variable in many later studies, several of which found a relationship

between entertainment and advertising attitudes (Alwitt & Prabhaker, 1992, Pollay &

Mittal, 1993; Schlosser, Shavitt, & Kanfer, 1999; Shavitt, Lowrey, & Haefner, 1999).

Similarly, a number of studies have confirmed a relationship between the informational

value of advertising and advertising attitudes (Mittal, 1994; Poliay & Mittal, 1993;

Schlosser, Shavitt, & Kanfer, 1999; Shavitt, Lowrey, & Haefner, 1999). The composition




Full Text
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 EHRPD9WBU_3YNRFR INGEST_TIME 2013-04-08T23:14:03Z PACKAGE AA00013638_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


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 EG9JQX19L_2VUFQJ INGEST_TIME 2013-03-12T14:18:25Z PACKAGE AA00013638_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


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 EHRPD9WBU_3YNRFR INGEST_TIME 2013-04-08T23:14:03Z PACKAGE AA00013638_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES