 |
Internet Marketing · eMarketing · Internet Advertising · Online Branding |
|
Internet Marketing · eMarketing · Internet Advertising · Online Branding |
|
|
|
||
| ADVERTISING BY ICONOCAST |
||
Copytesting of Advertising on the WWW:
Clicking Motivation Profile
Copytesting of Advertising on the WWW:
Clicking Motivation Profile
bY
Chang-Hoan Cho
Doctoral Student
c.cho@mail.utexas.edu
http://uts.cc.utexas.edu/-echo
Department of Advertising
College of Communication
The University of Texas at Austin
Austin, Texas 787 12
And
John D. Leckenby
Everett D. Collier Centennial Chair
in Communication
john.leckenby@mail.utexas.edu
http://uts.cc.utexas.edu/-admedium/
Department of Advertising
College of Communication
The University of Texas at Austin
Austin, Texas 787 12
WORKING PAPER
(Do not cite without permission of authors.)
September, 1997
Copytesting of Advertising on the WWW:
Clicking Motivation Profile
Abstract
This paper develops a copytesting method for Web advertising named the
“Clicking Motivation Profile” (CMP). In the process of developing this copytesting
method, the concept of voluntary exposure is first explicated as it relates to the most di
stinguishing characteristic of Web advertising, i.e., interactivity. To check the viability of
the newly-developed copytesting method (CMP), the method is first empirically tested in
terms of its validity (construct validity and convergent validity). The CMP is then tested
for its functional aspects as a good copytesting method, namely its evaluative and
diagnostic functions. Results show that the CMP can work well as a copytesting method
for advertising on the WWW.
2
Copytesting of Advertising on the WWW:
Clicking Motivation Profile
Introduction
The growth of the Internet has been exponential. By the end of 1996, there were
10 million hosts on the Internet that connected 105,000 networks (Cyber Atlas, 1996 at
URL: http://www.cyberatlas.corn/news.html), supporting the total Internet users,
estimated between 32 million and 50 million worldwide (Forrester Research 1996, 1997
at URL: http://www.forrester.com). The numbers are doubling every year (Forrester
Research, 1996 at URL: http://www.forrester.com). As a result, marketers have already
become very active participants in this medium and are investing a lot of money in the
medium (McKenna, 1995).
The nature of the Internet has been dramatically changed from the original
military-industrial research network for noncommercial purposes (Miller, 1995). The
most important factor transforming the structure of the Internet has been the
commercialization of the Internet through advertising on the WWW (PR Newswire,
1996). Among the many segments of the Internet, advertising is becoming the one with
the greatest growth, with 129.5 million online in the first quarter, an increase of 18%
from the preceding quarter (IAB 1997, at URL: http://www.iab.net). Jupiter
Communications (URL://www.webtrack.com) has recently projected the advertising
expenditure for the Internet to grow to $5 billion by the year 2000. With the advent of
the new Internet World Wide Web (WWW) as an advertising medium, measuring
advertising effectiveness on the WWW has become the critical demand of Web
advertisers. But there has been little research on copytesting systems of advertising on
the WWW.
Copytesting faces new challenges with the advent of the interactive medium, i.e.,
the Internet, and the issue arises as to the applicability of the standard copytesting
systems for existing mass media, as applied to this new medium. Technology is changing
rapidly, so developing a copytesting system may be impractical or soon obsolete. One
inherent problem with copytesting advertising on the WWW is that the message is partly
controlled by the consumer, and the nature of the communication is not one-way but twoway,
interactive.
The present method of pricing online ad campaigns is based mostly on clickthrough
rates (PR Newswire, 1997), even though a debate is arising over whether or not
click-through tells advertisers anything worth knowing and whether it should be a factor
in pricing Web advertising. This click-through-based pricing system is reasonable for
those advertisers who recognize that interactivity is the most distinguishing and important
characteristic of the Internet, and that click-through is the first gate to entering the world
of interactivity in Web advertising. That is, the clicking of a banner initiate users’
interactivity with Web advertising. Therefore, click-through is one of the many possible
measures of users’ interactivity in Web advertising.
Another reason for the prevailing use of click-through as a pricing base for Web
advertising is that it is a concrete measure of users’ actual clicking behavior and that it is
relatively easy to measure with the aid of innovative technology. However, the question
arises as to how the click-through directly relates to the effectiveness of Web advertising;
that is, what is the impact of the click-through on advertising effectiveness‘? In other
words, the assumption that a banner ad yielding the higher click-through is more effective
4
than that with low click-through has never been empirically tested. This study first tries
to test this assumption.
There are also many other important questions unanswered: 1) why certain banner
ads perform better than others; 2) why users click one banner more than the other; and 3)
why different creative executions of the same basic message lead to different clickthroughs.
In an attempt to answer these questions, this paper first conceptualizes
voluntary exposure (i.e., clicking of banner ads) and then applies its concept to the
development of copytesting systems for Web advertising. In the process of developing
the new copytesting method, the researchers empirically test the evaluative and
diagnostic functions of the copytesting method by quantifying and qualifying voluntary
exposure (i.e., click-through).
Copytesting of Advertising on the WWW
Copytesting research has been one of the hottest areas of advertising research.
Copytesting started in the early 193Os, when Starch developed a recognition measure for
print ads, called the Starch score (Baldinger, 1991 ARF). Since then, many different
methods have been developed. Most existing copytesting methods are based on the
theories of the one-way communication process of the traditional mass media, i.e., simple
involuntary exposure to advertising with no interactivity. With the advent of the new
interactive medium, i.e., the Internet, however, the applicability of these existing
copytesting methods to the Internet is questionable because of the different advertising
process of the Internet from that of other traditional media. Then how is the advertising
process of the Internet different from that of traditional mass media, and how can these
differences be used to develop a new copytesting method?
5
Voluntary vs. Involuntary Advertising Exposure
Traditional hierarchy-of-effects models assume that the very tirst stage of the
persuasion process is awareness through advertising exposure (Preston, 1982). Here,
advertising exposure is involuntary and/or incidental because individuals involuntarily
just happen to come across an ad in traditional media. In contrast to advertising exposure
in traditional media, advertising exposure in the Internet can be either involuntary or
voluntary, depending on the types of Web advertising.
For banner ads, the traditional involuntary exposure concept can be applied; that
is, banner ads on the Web are nothing but the traditional passive form of noninteractive
advertising unless they are clicked and move users into the separate interactivity site. If
the users are only exposed to the banner ads but do not click them to open to see linked
interactivity ads, it can be said that they are not interacting with the advertising messages
or the advertisers, i.e., this is traditional one-way, involuntary communication from
advertisers to consumers.
One recent study actually confirmed this rationale-the similarity of the
advertising effects of a banner ad to those of traditional mass advertising, if the banner is
not clicked. The study by Mbinteractive, commissioned by the Internet Advertising
Bureau, found that simple involuntary exposure to a banner ad without a click-through
generated increases in advertisement awareness, brand awareness, and purchase intention
(MBinteractive, 1997 at URL: http://www.mbinteractive.corn/site/iab/exec.html). This is
the same as the effects on consumers’ awareness and attitudes of simple involuntary
exposure to advertising in mass media; this has been supported by many other previous
6
studies. That is, the MBinteractive study just reconfirmed those advertising effects on
awareness and attitude of simple involuntary exposure to banner ads.
However, this study did not consider the real power of banner ads, i.e., the
initiation of interactivity by clicking them. This type of advertising exposure can be
called voluntary or sought-out exposure to a linked interactivity ad. This voluntary
exposure requires users to voluntarily perform an action (i.e., clicking banners) to see the
content of advertising messages, which will yield more active and intensive information
processing than passive exposure without voluntary action. This voluntary exposure will
draw more attention to the messages and activate the consumer learning processes more
intensively than will involuntary exposure. After the initial action (i.e., clicking banners),
consumers have the choice to perform more actions for further active information
processing by interacting with messages (e.g., clicking to deeper sites, searching contents,
providing feedback, bookmarking advertising sites for future reference, purchasing
products on-line, etc.).
Therefore, when we consider the difference in advertising processing between
traditional mass media (involuntary exposure) and the Internet (either voluntary or
involuntary exposure), it is necessary to develop a new copytesting method for testing
advertising effectiveness on the WWW, a method which differentiates between the two
different types of advertising exposure.
Intention of Voluntary Exposure as a Copytesting Method of Advertising
Then what are the different effects of voluntary vs. involuntary advertising
exposure in Web advertising? As mentioned earlier, the voluntary exposure is more
likely to yield active and intensive information processing than is passive exposure
7
without voluntary action. That is, more interaction between consumers and messages or
between consumers and advertisers may yield more intensive and active advertising
processing, and this active advertising processing through voluntary interaction with
advertising messages or advertisers may result in more favorable and desirable consumer
attitudes and actions (e.g., favorable attitude toward the ad, favorable attitude toward the
brand, high purchase intention, and actual purchasing). That is, we can say that active
advertising processing through voluntary exposure may yield more positive advertising
effects than involuntary exposure. Therefore, it is worthwhile to compare the advertising
effectiveness of involuntary exposure to a banner ad without a click-through with that of
voluntary exposure to a linked interactivity site by clicking the banner ad. It is assumed
that there will be significantly different advertising effects between voluntary exposure
by clicking a banner and simple involuntary exposure to the banner. As a matter of fact,
this assumption works as the underlying presupposition that most Web advertisers and
publishers are relying on when they evaluate and buy banner ads based on their clickthrough
rates. Therefore, testing this basic assumption is very worthwhile.
Testing this assumption has another implication in terms of the validity of the new
copytesting method, i.e., checking whether the copytesting method using click-through or
voluntary exposure is valid or not. Looking at the actual relationship between clickthrough
and other existing advertising effectiveness measures can check this validity.
Among the many existing measures of advertising effectiveness, consumers attitude
measures are found to be superior to memory-based measures, such as recall measures, in
terms of reliability (Clancy and Ostlund, 1976) and validity (Gibson, 1983). These
attitude measures are now popular, and most copytesting fiims have added attitude
measures to their regimen of tests (Haley, 1994). For the purpose of checking the
8
validity of using click-through as a copytesting method and testing the assumption that
high click-through will yield better advertising effects, the following hypotheses are
generated:
H1.l: Clicking banner ads will yield a more favorable attitude toward the linked ad.
H1.2: Clicking banner ads will yield a more favorable attitude toward the brand.
H1.3: Clicking banner ads will yield a higher purchase intention.
The above hypotheses will test the construct validity and convergent validity (i.e.,
relationship to other measures) of using click-through as a copytesting method for Web
advertising. However, a good copytesting method should also serve two different
functions simultaneously: 1) evaluative or global function and 2) diagnostic function
(Leckenby and Plummer, 1983). That is, a good copytesting method should be able to
quantify and qualify consumers’ subjective responses to advertising, i.e., the voluntary
exposure in our study. Here, the evaluative function or quantification ability ofa
copytesting method means its ability to differentiate advertising stimuli (i.e., banner ads)
in terms of its copytesting measure, i.e., voluntary exposure or click-through. In other
words, a good copytesting method should be able to detect and quantify the differences,
in terms of its measure (i.e., click-through), among many different versions of advertising
for the same brand.
Meanwhile, diagnostic function or qualification ability implies diagnosing the
underlying reasons for yielding the different copytesting measures, i.e., different clickthrough
rates. That is, it serves to find out why users click a banner ad or what
motivations lead them to click the banner ad. This diagnostic function can offer
advertisers the ability to learn why certain banner ads perform better and can allow them
9
to refine their online campaigns for the target market. Therefore, exploring various
consumers’ motivations underlying the clicking of banner ads will serve this diagnostic
function of the copytesting method.
The clicking of a banner can be interpreted as a voluntary activity of requesting
more information, i.e., voluntary action of seeing more detailed description of the ad.
Consumers’ motivations and reasons underlying this voluntary exposure will vary,
depending on many different motivation factors. These various motivations underlying
the clicking of a banner can be profiled, based on several important factors affecting
advertising effectiveness. This prolile is called the CMP (Clicking Motivation Profile),
which will help us to understand why people click certain banners more than others.
Figure 1 shows the detailed description of the CMP.
The first motivation factor in CMP, which influences the motivation underlying
voluntary exposure (i.e., clicking of banner ads), is the advertising values. This factor
includes information motivation, entertainment motivation, and usefulness motivation.
That is, users may click a banner ad to obtain more information about the product, to get
entertainment, and to get useful value from the ad. One recent study actually found out
that the perceived values of Web advertising among Internet users were its
informativeness, entertainment, and usefulness (Ducoffe, 1996).
The second motivation factor in the CMP (Clicking Motivation Profile) is the
advertising characteristics. That is, users may click a banner ad because of its attentiongetting
and curiosity-evoking creatives. The last motivation factor is the user
characteristics, which intluence the degree of the clicking of banner ads. That is, people
are more likely to click a banner ad if they feel that they are involved with the product,
10
that the banner has something to do with them or their needs, or that they want to learn
more about the product.
To check empirically whether the CMP (Clicking Motivation Profile) serves well
the two previously mentioned functions of a good advertising copytesting method (i.e.,
evaluative function and diagnostic function), in other words, to test whether the CMP can
differentiate two different versions of banner ads for the same brand, the following
hypotheses can be generated:
H2.1: Average clicking of one banner ad is significantly different from that
of another banner ad for the same brand
H2.2: Individual CMP items of one banner ad are significantly different
from those of another banner ad for the same brand.
Methodology
To test the above hypotheses, a between-group experimental design was used.
The experiment used an off-line method with forced exposure manipulation. Here, it is
worthwhile to describe the operational definition of the clicking of banner ads.
According to Preston (15X35), the perfect advertising effectiveness measure should be
related to the actual behavior. Similarly, the most concrete measure of clicking of banner
ads is looking at users’ actual behavior, i.e., click-through data. However, recognizing
the difficulty of getting the actual click-through data, this study employed a mental
measure of clicking, i.e., people’s self-reported intention to click banner ads. Many
previous research studies on advertising effectiveness have used various mental
measures, such as recall, self-reported attitude toward the ad and the brand, and purchase
intention. In addition, this self-reported intention to click banner ads is more useful and
11
valuable to the current study, which tries to find out the relationship between the clicking
of banner ads and other mental advertising effectiveness measures (H 1.1, H 1.2 and
H1.3).
A total of 203 undergraduate students in a large southwestern university, divided.
into two experimental groups, participated in the experiment. The experiment employed
a between-group subject design, where each subject was exposed to only one of two
experimental treatments (I or II). Each subject was exposed to a set of banner ads and
Web sites based on his/her experimental group (I or II).
Sample Materials
According to Mitchell (1986), professionally developed ads rather than mock ads
are encouraged to be used in experimental research in order to elicit a more natural
response from the subjects, Following this suggestion, professionally developed Web
sites and banner ads were used in this experiment. A total of eight banner ads were used,
two different versions of four products. Each experimental group was exposed to four
banner ads and four linked ad sites. Four banner ads were selected, based on the four
popular product categories of Web advertising, which include financial services,
consumer brands, retailers, and travel-related products (WebTrack, 1997). Two different
versions of one brand in each product category were randomly selected from the three
most popular search engines on the WWW, i.e., Yahoo, Infoseek, and Excite. These four
banner ads included the American Express Card (financial services), Kodak film
(consumer brands), JCPenny (retailers), and American Airlines (travel-related products).
All eight banner ads, i.e., two banner ads for each brand, were placed at the top of the
same Web site, Infoseek (URL: http://www.infoseek.com). The size and the location of
12
each banner ad were identical. Two banner ads for the same product were linked to the
same advertising site. Figure 2 shows the four experimental banner ads for each
experimental group.
Procedure
First, each subject saw the very first banner ad located at the top of the Infoseek
site and was asked to answer questions concerning his/her intention to click the banner
ad. Then, each subject saw the linked advertising site from the banner ad s/he had just
seen. After browsing this linked advertising-site for a while, each subject was asked to
respond to the questions concerning his/her attitude toward the linked site, attitude
toward the brand, and purchase intention. After completing the question items for the
first banner ad and the linked ad site from the banner, each subject was exposed to the
second banner ad located at the top of Infoseek site and then again asked to fill out the
question items measuring his/her intention to click the banner ad; that was followed by
the browsing of the linked site from the banner ad and answering the questions measuring
attitude toward the ad, attitude toward the brand, and purchase intention. Each subject
followed the same procedure for the remaining two banner ads and two linked sites from
the banners. The participation for each subject took approximately 25 minutes.
Results
This study used a between-group experimental design because of its advantage;
that is, there is no chance of one treatment’s contaminating the other, since the same
subject never receives both treatments. However, the between-subject design must
eliminate the possibility that the subjects in the two groups are different enough to
13
influence the effects of the treatment. To guarantee that as few differences as possible
existed between the two groups, the researcher compared the groups in terms of their
demographics and Internet usage. Table 1 indicates that the two groups are very similar
in terms of age, gender, major, Internet-surlmg hours, and the purpose of surfing the
Internet. Therefore, the results eliminate the possibility that the subjects in the two
groups are different enough to influence the effects of the treatments.
The first stream of hypotheses states that users who have higher intention to click
a banner ad are more likely to have a favorable attitude toward the linked ad (Hl. l), a
favorable attitude toward the brand (H1.2), and high purchase intention (H1.3). To test
these hypotheses, simple correlation and between-group t-tests were used. Table 2 shows
that, for American Express, JCPenny and American Airlines, intention to click a banner
ad is moderately correlated with attitude toward the linked ad, attitude toward the brand,
and purchase intention. Pearson correlation ranges from .30 to .45. But there were weak
relationships for Kodak. All results were statistically significant (p I .Ol).
Table 3 shows the detailed relationship between intention to click a banner ad and
attitude toward the linked ad. The results show that people who have high intention to
click a banner ad are more likely to have a favorable attitude toward the linked ad than
those who have low intention to click the banner ad (M=3.3 vs. M=2.9 for American
Express, M=3.7 vs. M=3.S for Kodak, M=3.2 vs. M=3.0 for JCPenny, and M=3.6 vs.
M=3.4 for American Airlines). The results were statistically significant (p I .OS), and
thus H 1.1 is supported. Table 4 shows that, except for American Airlines, people who
have high intention to click a banner ad are more likely to have a favorable attitude
toward the brand than those who have low intention to click the banner ad (M=3.4 vs.
M=3.1 for American Express, M=4.2 vs. M=3.8 for Kodak, and M=3.2 vs. M=2.8 for
14
JCPenny). These results were statistically significant (p I .OS). Therefore, H1.2 is
mostly supported. As shown in Table 5, people who have high intention to click a banner
ad are more likely to have higher purchase intention than those who have low intention to
click the banner ad, but only for American Express (M=3.4 vs. M=2.9) and JCPenny
(M=3.3 vs. M=2.8). These results were statistically significant (p I .05). Therefore,
H 1.3 is partly supported.
To determine whether people who were exposed to one version of a banner ad had
a significantly different intention to click the banner ad from those who were exposed to
another version for the same brand, the researchers used between-group t-tests. Table 6
shows that the mean of the probability to click a banner ad for the first group is
significantly different from that for the second group for all four products (M=50.4% vs.
M=27.0% for American Express, M=38.9% vs. M=55.4% for Kodak, M=23.8% vs.
M=41.9% for JCPenny, and M=56.1% vs. M=42.8% for American Airlines). The results
were statistically significant (p I .Ol). Therefore, H2.1 is supported.
To determine whether people in different groups show significant differences in
individual items of the CMP (Clicking Motivation Profile), a series of between-group ttests
was used for each product. As shown in Tables 7 and 9, the means of all 10 CMP
items for the first group were significantly different from those for the second group, for
American Express and JCPenny, respectively. All results were statistically significant (p
I .Ol). Table 8 shows that, for Kodak, the means of 9 CMP items for the first group were
significantly different from those for the second group. The results were statistically
significant (p I .Ol). As shown in Table 10, for American Airlines, only 5 CMP items
show significant differences between the two experimental groups. These include “I
15
would click this banner ad to learn more about the product, ” “I would click this banner ad
to get entertainment,” “I would click this banner ad because of its reward program,” “I
would click this banner ad because it draws my attention,” and “I would click this banner
ad because it is unusual and unique.” Although the differentiability of individual CMP
items varied across the four products, most of them could successfully differentiate two
different versions of banner ads for the same brand in terms of motivations to click the
banner ad. Therefore, H2.2 is also supported.
Discussion’
In summary, it is concluded that the CMP (Clicking Motivation Profile) can work
well as a copytesting method of advertising on the WWW because it has construct
validity and convergent validity (H 1.1, H 1.2, and H1.3), and because it serves well both
the evaluative and diagnostic functions of advertising copytesting systems.
This study is pioneering in the sense that it is the first formal research on
copytesting of advertising on the WWW. However, the greatest weakness of this study is
that the student samples are not representative of the general population, even though
college students are one of the largest segments of Internet users. The picture might have
been different if the researchers drew the samples from the general population. Another
weakness of this study is that mental measures (i.e., intention to click to banner ads) do
not usually represent actual behavioral measures (i.e., actual click-throughs). This can be
demonstrated by comparing the average click-throughs in the industry with the average
mental click-throughs of this study. There is a big gap between these two measures, i.e.,
the average industry click-throughs of 2% vs. the average mental click-throughs of 42%.
This kind of gap has been always a problem in advertising research.
16
Another weakness of this study is that the CMP (Clicking Motivation Profile)
considers only one aspect of interactivity, i.e., intention to click banner ads. As
mentioned earlier, interactivity is considered the single most important characteristic of
the Internet (Hoffman and Novak, 1996), and there are many other types of interactivity
such as clicking into deeper sites, searching contents, providing feedback, bookmarking
advertising sites for future reference, and so on. Therefore, it would be valuable to study
the relationships between these different types of interactivity and other measures of
advertising effectiveness.
Despite the increasing significance of Internet advertising, there has been little
research on copytesting of Web advertising. In this sense, this paper provides some
groundwork in this field. In conclusion, copytesting of Internet advertising is too
important to leave unstudied; therefore, more future studies in this area are strongly
recommended.
17
Figure 1
CMP (Clicking Motivation Profile)
1. Information value: Clicking a banner ad
to obtain relevant news and information
about the product.
Advertising Values Motivations 2. Entertainment value: Clicking a banner ad
to get entertainment.
3. Usefulness value: Clicking a banner ad
because it is useful/rewarding.
Advertising Characteristics
Motivations
1. Attention-generating characteristics:
Clicking a banner ad because it draws
attention/unique/unusual.
2. Curiosity-generating characteristics:
Clicking a banner ad because it evokes
curiosity.
User Characteristics Motivations
1. Consumer needs: Clicking a banner ad
because it has something to do with him
or his needs.
2. Involvement: Clicking a banner ad
because the user is involved with the
advertised product.
3. Learning: Clicking a banner ad to learn
more about the product.
18
Figure 2
Banner Ads for each experimental group located at the top of Infoseek
Group I
Group II
**All banners except American Airlines banner in Group II are animated. To see the
complete sample materials, please visit httD://uts.cc.utexas.edu/-ccho/samnlead.html
19
Table I
The comparison of two experimental groups
Group I (n= 102) Group II (n=lOl)
Mean age 21.0 21.9
Gender (female / male) 60142 55146
Major (advertising / others) 96 I 6 90111
Average surfing hours per week 3.01 3.78
Information (34) Information (48)
Purpose of surfing Entertainment (11) Entertainment (16)
Product Service (4) Product Service (2)
N=203
Table 2
Correlation between intention to click banner ad and attitude measures
Intention to click Intention to click Intention to click
and attitude toward and attitude and purchase
the linked ad toward the brand intention
American Express .42** .40** .30**
Kodak .38** .22”” .09
JCPenny .30** .33** .39**
American Airlines .43** .30** .45**
N=203
** p I .Ol, * p I .05
Intention to click was measured by the self-reported probability to click a banner ad, out of 100%
Attitude toward the linked ad was measured by the average of fifteen Spoint Likert items measuring
attitude toward the linked ad (this ad is personal; this ad satisfies what I expected from the banner ad;
this ad is useful to me; this ad provides relevant news and information; this ad is easy to follow and
understand; I like this ad; this ad is unpleasant; this ad is involving; this ad is annoying; this ad is
informative; this ad is boring: this ad is good; this ad is entertaining; I would enjoying seeing this ad
again; and I learned something from the ad).
Attitude toward the brand was measured by the average of two Spoint Likert items measuring attitude
toward the brand (I like the brand and this brand is good).
Purchase intention was measured by one S-point Likert item (I would purchase the advertised brand if 1
were in the market for the advertised brand).
20
Table 3
The relationship between intention to click a banner ad and attitude toward the linked ad
Intention
to click a
banner
American
Express
Kodak
JCPenny
American
Airlines
Low
High
Low
High
Low
High
Low
High
Case# Attitude toward the linked ad
Mean (Std. Dev)
t-value
127 2.9 (5) 4.93**
69 3.3 (.5)
127 3.5 (.6) 1.69”
69 3.7 (.6)
127 3.0 (.7) 2.44””
68 3.2 (.7)
128 3.4 (7) 2.53””
69 3.6 (.5)
““pl.01, *p1.05
. Intention to click was divided into low (less than 50%) and high (more than 50%) probability to click a
banner ad, out of 100%.
. Attitude toward the linked ad was measured by the average of fifteen 5-point Likert items measuring
attitude toward the linked ad (please refer to the footnote of Table 2).
Table 4
The relationship between intention to click a banner ad and attitude toward the brand
American
Express
Kodak
JCPenny
American
Airlines
Intention
to click a
banner
Low
High
Low
High
Low
High
Low
High
CaSe# Index score of attitude toward the
brand
Mean (Std. Dev)
129 3.1 (.7)
68 3.4 (.7)
129 3.8 (-6)
68 4.2 (.7)
128 2.8 (.9)
69 3.2 (.9)
128 3.6 (.9)
69 3.7 (.8)
** p I .Ol, * p I .05
t-value
3.32””
1.76””
2.18”
-64
. Intention to click was divided into low (less than 50%) and high (more than 50%) probability to click a
banner ad, out of 100%.
. Attitude toward the brand was measured by the average of two 5-point Likert items measuring attitude
toward the brand (I like the brand and this brand is good).
21
Table 5
The relationship between intention to click a banner ad and attitude toward the brand
American
Express
Kodak
JCPenny
American
Airlines
Intention
to click a
banner
Low
High
Low
High
Low
High
Low
High
Case#
129
69
128
69
128 2.8 (1.0)
69 3.3 (.9)
128
69
Purchase intention
Mean (Std. Dev)
2.9 (.9)
3.4 (8)
3.9 (.8)
3.9 (.7)
3.6 (.9) .96
3.7 (.8)
p I .Ol, * p I .05
t-value
3.43””
.29
3.29””
. Intention to click was divided into low (less than 50%) and high (more than 50%) probability to click a
bamler ad, out of 100%.
. Purchase intention was measured by one Spoint Likert item (I would purchase the advertised brand if I
were in the market for the advertised brand)
Table 6
Mean differences of the probability to click a banner ad
between two different experimental groups
Group
American I
Exuress II
Kodak
JCPenny
I
II
I
II
Case#
101
98
102
98
102
Intention to click a banner
Mean (Std. Dev)
50.4 (27.5)
27.0 (28.2)
38.9 (27.1)
55.4 (29.6)
23.8 (24.7)
t-value
5.93””
4.13**
4.51””
99 41.9 (31.9)
102 56.1 (29.3) 3.03””
99 42.8 (32.7)
** p2 . Ol
. Probability to click was measured by the self-reported probability to click a banner ad, out of 100%.
22
Table 7
Mean differences in CMP (Clicking Motivation Profile) between two different
experimental groups for American Express Card
Group Intention to click a banner t-value
Mean Std. Dev
I would click this banner ad to lea-n more I 3.2 .9 4.91**
about the product. II 2.5 1.1
I would click this banner ad to get I 2.5 .9 5.32””
entertainment. II 1.8 .9
I would click this banner ad to obtain relevant I 3.5 1.0 4.52**
news and information about the product. II 2.8 1.0
I would click this banner ad because of its I 3.5 1.2 10.06”
reward program. II 2.0 0.9 *
I would click this banner ad because it seems I 3.1 1.1 2.56””
useful to me. II 2.6 1.2
I would click this banner ad because it draws I 3.0 1.1 3.16””
my attentiou. II 2.5 1.2
I would click this brumer ad because it is I 2.5 .9 3.71**
unusual and unique. II 2.0 1.0
I would click this banner ad because it evokes I 3.1 1.0 3.89””
my curiosity. II 2.5 1.2
I would click this banner ad because it has I 3.1 1.1 2.97**
something to do with me or my needs. II 2.6 1.2
I would click this banner ad because I’m I 2.7 1.2 2.14”
involved with the product. II 2.4 1.2
** p 2 .Ol, * p < .05 (one-tailed)
. All CMP items were measured by a 5-point Likert scale.
23
Table 8
Mean differences in CMP (Clicking Motivation Profile) between two different
experimental groups for Kodak
I would click this banner ad to lean1 more
about the product.
I would click this banner ad to get
entertainment.
I would click this banner ad to obtain relevant
news and informatiou about the product.
I would click this banner ad because of its
reward program.
I would click this banner ad because it seems
useful to me.
I would click this banner ad because it draws
my attention.
I would click this banuer ad because it is
unusual and unique.
I would click this barmer ad because it evokes
my curiosity.
I would click this banuer ad because it has
something to do with me or my needs.
I would click this banner ad because I’m
involved with the product.
** p 2 .Ol,
Group
I
II
I
II
I
II
I
II
I
II
I
II
I
II
I
II
I
II
I
II
c p I .05
Intention to click a banner
Mean Std. Dev
3.0 1.1
3.7 .9
2.7 1.2
3.1 1.1
3.0 1.1
3.6 1.0
2.3 .9
2.6 1.0
2.6 1.1
3.2 .9
2.9 1.2
3.7 1.0
2.6 1.1
3.4 1.2
3.0 1.1
3.8 1.0
2.7 1.1
3.2 1.0
2.7 1.1
2.9 1.0
one-tailed)
t-value
5.00””
2.23**
4.07””
2.10”
4.26””
5.44””
4.59””
4.93**
3.16””
1.51
. All CMP items were measured by a S-poht Likert scale.
24
Table 9
Mean differences in CMP (Clicking Motivation Profile) between two different
experimental groups for JCPenny
Group Intention to click a banner t-value
Mean Std. Dev
I would click this banner ad to learu more I 2.5 1.1 3.86””
about the product. II 3.1 1.1
I would click this banner ad to get I 2.0 .8 2.06”
entertainment. II 2.2 1.0
I would click this banner ad to obtain relevaut I 2.5 1.1 3.94””
news and iuformatiou about the product. II 3.2 1.2
I would click this banner ad because of its I 2.0 -8 2.76””
reward program. II 2.4 1.0
I would click this banner ad because it seems I 2.3 1.0 3.99””
useful to me. II 3.0 1.2
I would click this banuer ad because it draws I 2.1 1.0 6.63””
my attention. II 3.1 1.2
I would click this banner ad because it is I 1.8 .8 5.54””
unusual aud unique. II 2.6 1.1
I would click this banner ad because it evokes I 2.4 1.0 4.03””
my curiosity. II 2.9 1.1
I would click this banner ad because it has I 2.4 1.0 4.70””
something to do with me or my needs. II 3.1 1.1
I would click this bauner ad because I’m I 2.2 .9 5.41””
iuvolved with the product. II 2.9 1.1
*-** p _5 .UAI. , .a-. p_ 5 n.ur 3 (Ione-t.a il.e a).\
. All CMP items were measured by a 5-point Likert scale.
25
Table 10
Mean differences in CMP (Clicking Motivation Profile) between two different
experimental groups for American Airlines
Group Intention to click a banner t-value
Mean Std. Dev
I would click this banner ad to learn more I 3.6 .9 3.13””
about the product. II 3.1 1.1
I would click this banner ad-to get I 2.5 .9 2.92””
entertainment. II 2.1 .9
I would click this banuer ad to obtain relevaut I 3.6 1.1 1.22
news and information about the product. II 3.4 1.1
I would click this banner ad because of its I 3.5 1.1 2.01*
reward program. II 3.2 1.2
I would click this banner ad because it seems I 3.5 1.1 .86
useful to me. II 3.3 1.2
I would click this bauner ad because it draws I 3.0 1.1 2.39**
my attention. II 2.7 1.2
I would click this bauner ad because it is I 2.7 1.0 4.71””
unusual and unique. II 2.1 .9
I would click this banner ad because it evokes I 4.0 1.0 1.31
my curiosity. II 2.8 1.2
I would click this barmer ad because it has I 3.4 1.0 .63
something to do with me or my needs. II 3.3 1.2
I would click this banner ad because I’m I 3.2 1.1 .89
involved with the product. II 3.1 1.2
** p S .Ol, * p 2.05 (one-tailed)
. All CMP items were measured by a S-point Likert scale.
26
References
Baldinger, Allan L. (1991). “The Case for Multiple Measures: Is One Ever Enough‘?”
8th Annual ARF Research Workshop. In Understanding Copy Pretesting,
published by The Advertising Research Foundation.
Clancy, K. J. and L. E. Ostlund (1976). “Commercial Effectiveness Measures.” Journal
OfAdvertising Research. 16(l), Feb.
Cyber Atlas (1996), [URL: http://www.cyberatlas.com/news.html].
Ducoffe, H. Robert (1996). “Advertising Value and Advertising on the Web.” Journal of
Advertising Research. Sep/Oct: 2 l-34.
Forrester Research (1996, 1997), [URL: http://www.forrester.com/].
Gibson, D. Lawrence (1983). “If the question is copy testing, the answer is not Recall.”
Journal of Advertising Research. 23(l), Feb/Mar.
Haley, Russell, J. Staffaroni, and A. Fox (1994). “The missing measures of copy
testing.” Journal of Advertising Research. 34(3): 46-56.
Hoffman, Donna and Thomas Novak (1996). “New Metrics for New Media: Toward the
Development of Web Measurement Standards,”
[URL:http://www2000.ogsm.vanderbilt.edu/nov~web.standard/webstand.html].
IAB (Internet Advertising Bureau) (1997), [URL: http://www.iab.net/].
Jupiter Communication (1996), [URL: http://www.webtrack.com/].
Leckenby, John D. and Joseph T. Plummer (1983). “A Review of Advertising Testing
Methods.” Current Issues and Research in Advertising.
MBinteractive (1997). “IAB Ad Effectiveness Study,” [URL:
http://www.mbinteractive.com/site/iab/exec.html]
McKenna, Regis (1995), “Real-Time Marketing,” Harvard Business Review, July, 87-95.
Miller, Steven (1996). Civilizing Cyberspuce : Policy, Power, and the Information
Superhighwuy. New York. Addison-Wesley Publishing Company.
Mitchell, A. A. (1986). “Theoretical and Methodological Issues in Developing an
Individual Level Model of Advertising Effects.” In Olson, J.C., and K. Sentis
(eds.). Advertising and Consumer Psychology. New York: Praeger Publishing
Co. 172- 196.
27
PR Newswire (1996), “MapQuest Launches Revolutionary Geocentric Advertising
Program Enabling Businesses to Provide Geographically Sensitive Advertisements
as Users’ Online Advertisers Can Extend Narrowcasting With a True Geographic
Dimension as Users Interact with MapQuest Maps.” November 4.
PR Newswire (1997), “Infoseek Launches Ultramatch: Advanced Behavioral Targeting of
Internet Advertising.” May 27.
PR Newswire (1997), “World Online Ad Revenue Well Behind U.S. -- But Growing;
Jupiter to Release World Advertising Numbers at MILIA Today.” February 11.
Preston, Ivan L. (1982). “The Association Model of the Advertising Communication
Process.” Journal ofAdvertising. 11(2):3- 15.
Preston, L. Ivan (1985). “The Developing Detachment of Advertising Research from the
Study of Advertisers’ Goals.” Current Issues and Research in Advertising.
WebTrack (1997). “WebTrack Advertisers Scoreboard.” WebTrack: Online Advertising
Monthly. August. Jupiter Communications [URL: http://www.jup.com].
28