AN EMPIRICAL STUDY ON CONSUMERS’ ATTITUDES TOWARD CROSSCATEGORY BRAND EXTENSION
by
Chaoling Gan
BEcon, WenZhou University, 2013

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
IN
BUSINESS ADMINISTRATION

UNIVERSITY OF NORTHERN BRITISH COLUMBIA
JUNE 2019
© Chaoling Gan, 2019

i
Internal

Abstract
Consumer evaluations of brand extension are becoming increasingly important to the
consumer market (Kaur & Pandit, 2015); however, little attention has been given to crosscategory specific research in this field. This research examines whether there are correlations
between an iconic product (a product category already occupied by the brand) and its crosscategory extension product and how the user experience on an extension product affects its
iconic product. The findings reveal that consumers have a positive attitude toward an
extended product when they perceive credibility, quality, and innovativeness from its iconic
product. Consumer perceived image-fit and advertisement-match are positively correlated
with consumer attitude. The results support that the post-evaluation on an extended product
affects its iconic product; however, user experience with an extended product does not
correlate with consumers’ evaluations of an iconic product on their evaluations of the
extended product because of the survey limitation.

ii
Internal

TABLE OF CONTENTS
ABSTRACT

II

TABLE OF CONTENTS

III

LIST OF TABLES

IV

CHAPTER ONE: INTRODUCTION

1

CHAPTER TWO: LITERATURE REVIEW

3

CHAPTER THREE: METHODOLOGY

15

Survey Design

15

Sample Description

18

CHAPTER FOUR: RESULTS

19

CHAPTER FIVE: DISCUSSION

40

CHAPTER SIX: LIMITATIONS AND FUTURE RESEARCH

45

REFERENCES

49

APPENDIX 1 SURVEY AND ITS MEASURE RELIABILITIES.

53

APPENDIX 2 VARIABLES FOR CORRELATION ANALYSIS

61

APPENDIX 3 DESCRIPTIVE STATISTICS AND CORRELATION RESULTS FOR H2
AND H3 (COMBINATION SAMPLE)
63

iii
Internal

List of Tables
Table 1: Visual of the hypothesis H4

14

Table 2: Examples used in the survey.

16

Table 3: Demographics of Respondents

19

Table 4: Mean Value and SD of indicators

21

Table 5: Gender differences in response to the brand and the extension related statements 23
Table 6: Cronbach’s  Coefficient for H1

25

Table 7: Correlation analysis results for evaluations between iconic product and crosscategory extensions for H1, H4 and H5 (student sample only).

29

Table 8: Overall correlation analysis results for evaluations between iconic product and
cross-category extensions for H1, H4, and H5.

32

Table 9: Cronbach’s  Coefficient for H2 and H3

34

Table 10: Correlation analysis results for evaluations between iconic product and crosscategory extensions for H2 and H3.

38

Table 11: Nonparametric paired t-test for H4

40

iv
Internal

Chapter One: Introduction
The increasingly fierce competition in the global markets are forcing firms to
differentiate themselves in order to maintain future growth. Many companies are achieving
this differentiation to assure their future growth by leveraging a well-established brand to
enter new markets. This act of leveraging is described as brand extension. According to the
present marketing literature, research on consumers’ attitudes of brand extension has been
found in a number of countries such as United States (Aaker & Keller, 1990), Europe (Dens
& De Pelsmacker, 2010), and India (Punyatoya, 2013; Kaur & Pandit, 2015).
In the past 30 years, the rapid growth of economics has witnessed significant and
empirical changes of consumers’ attitude toward brand extension. Aaker (1991) found that
the majority of companies were still focusing on line extension (extension within the line of
its original product or its iconic product) during the middle of the 20th century; only a
fraction of them were brand extension or new brands. As a result of the pressure of strong
market competition and the high cost of launching a new name brand, firms started taking
brand extension into account. The 1990s saw the emergence of brand extension that caused a
growing number of firms to look at their business differently; hence, an increasing amount of
studies on brand extension have been conducted globally (Aaker & Keller, 1990; Klink &
Smith, 2001; Czellar, 2002; Selvanayagam & Ragel, 2015; Nan, 2006; Aaker, 1996). From
then on, the concept of brand extension continued to receive broad attention (Aaker & Keller,
1990).
The benefits of brand extension were significant. Generally, if the parent brand name
was already well known to consumers, firms could spend less time and energy on the

1
Internal

marketing planning toward the extended product, compared to the parent brand was new to
market, which the company would have to strive to promote the product under the new name.
If consumers have existing relationship with the brand, they were more likely to be aware of
an extension product the first time it is introduced because they were familiar with the iconic
product and the parent brand (Bhat & Reddy, 2001). Riel et al. (2001) pointed out that the
benefits of brand extension were an efficient reduction for advertisement and costs, an
effective reduction of barrier and risk when entering one category into the other, and a
significant tool leveraging brand equity. On the strategic marketing point of view, companies
could not only expand their current market base through favorably evaluated brand
extensions but also expand their future market and revenues by adding brand extension
associations, while additionally solidifying the existing developed market (Barone et al.,
2000).
As indicated above, current brand strategies prefer to use an existing brand to open
new product categories in order to leverage the brand trust from its iconic brand and reduce
several aspects of costs, while also expanding market shares beyond the original market
(Kaur & Pandit, 2015). However, companies need to consider locating feasible categories to
which the extension can be made. When a company extends products into other market areas
that are different from their iconic brand, the company has the opportunity to build a solid
basis with strong market competitiveness and increase their identity across these other
markets. Thus, consumers’ perceptions of a brand can be expanded to create a stronger
appreciation of both the iconic brand and the brand extension, and this appreciation could
allow the company to maintain long-term stability in development for future business in
multiple markets (Aaker, 1996). For example, Caterpillar (CAT) was known as the world’s
2
Internal

largest construction equipment manufacturer; in recent years, the brand has extended its
market to the clothing and footwear industry, where their work boots were recognized as a
bestseller item not only in the work shoe market, but also in the fashion shoe market (Hollis,
2012).
However, according to recent researchers, some concerns have been raised about
extending brands into disparate categories from that of their original market since companies
may face the potential consequence of failing if they are not able to share some attributes or
features in common with the existing products. This concern was also addressed in Taylor
and Bearden’s (2003) theory that when an extension is perceived as dissimilar to the original
product, the brand faces more challenges than when an extension is perceived as a similar
product to the original. Most of the time, consumers are more skeptical and critical of crosscategory extensions as it is more difficult to transfer their attitude toward the original product
to the dissimilar extension.
The majority of research is focusing on the importance and influence of brand
extension while this study considers the relationship between the extension product and its
iconic product. The following article contributes at theoretical and empirical levels by
exploring the correlation between two products. The literature review is established to build
the foundation of hypotheses, this is followed by the methodology, results, discussion,
limitation and future research of this study.
Chapter Two: Literature Review
Monga and John (2010) believed that the characteristics of an iconic product of a
brand drive the results of how far the category of extension could reach. One particular

3
Internal

importance was the prestige associated with the iconic product, which determined the success
ratio when expanding extension products into a variety of product categories, as noted in the
example of CAT, above. A significant number of researchers believed that information is
transferable between related entities. Boisvert and Burton (2011) agreed with this theory and
illustrated that when a consumer held some perceptions of an entity, such as an iconic
product of a parent brand, the perception from the iconic product would be automatically
matched with the new related objective, such as a brand extension. When the match was
successful, at a certain level, the extension is successful. Bhat and Reddy (2001) found that
whether the extension product was unfamiliar or not, consumers tended to evaluate the
extension product against their knowledge of, or acquaintance with, the iconic product.
Therefore, consumer evaluations of an iconic product played a significant role in their initial
opinion of the extension, until familiarity was formed with the extension itself. Boisvert and
Burton (2011) emphasized that an effect of this kind of “transfer of perception” was more
outstanding when the iconic product was more prestigious. That being said, for those brands
who were already well-known in the market could relatively be extended beyond their
original market because the brands had already built their reputation (Aaker, 1991). Park et
al. (1991) also argued that high-prestige brands facilitate to be extended to other product
categories than brands with lower reputations, as the brand name was an identifiable label.
This research predicted that the iconic product needs strong credibility in order to extend the
brand into a different category of market.
Kaur and Pandit (2015) presented that a high degree of fit had been considered as an
important determinant for a successful extension. The fit contained different perspectives,
such as a fit of similarities (Aaker & Keller, 1990), a fit of brand image (Salinas & Perez,
4
Internal

2009), and a fit of quality. Aaker (1996) made a specific assertion that perceived quality
played a positive and significant role in determining consumer evaluation of the extension.
In Aaker’s earlier literature with Keller, the relationship between perceived quality from the
iconic product and consumers’ attitude toward the extension across unrelated categories has
already been explained (Aaker & Keller, 1990). The theory further confirmed that high
quality brands engendered more favourable evaluations for unrelated extensions than low
quality brands (Keller & Aaker, 1992; Dens & De Pelsmacker, 2016). Aaker (1990) also
proposed a significant connection between quality and attitude, such that the perceived
quality could be more easily transferred to the extension with greater congruence between the
iconic product and its cross-category product. In other words, iconic products with higher
quality perceptions were associated with more favourable attitudes toward their extension
category. However, in the current literature there was absence of focus on category
classifications of brand extension.
The innovativeness of a brand product perceived by consumers was a key success
factor when launching a new extension. Boisvert (2012) believed that it was significant and
critical to innovate for any new extensions. The innovativeness should highlight the
functions, features, and benefits of the extension product as its uniqueness. Innovativeness
had been identified as a very efficient element for brands, which significantly impacted
consumers’ evaluation of a product. When the brand initially introduced the cross-category
extension product, consumers perceived this as innovativeness on an iconic product, because
when the extension category was in a different sector, consumers perceived the uniqueness
when comparing the iconic product to the extension product (Boisvert & Burton, 2011).

5
Internal

However, whether the innovativeness perceived by consumers from an iconic product was
transferrable to a cross-category extension product had never been tested.
H1: Consumers’ attitudes toward an iconic product correlates with consumers’
attitudes toward the cross-category extension product.
H1a: The credibility of an iconic product in its market correlates with consumers’
attitudes toward its cross-category extension.
H1b: The perceived quality of an iconic product correlates with consumers’ attitudes
toward its cross-category extension.
H1c: The innovativeness of an iconic product correlates with consumers’ attitudes
toward its cross-category extension.

Marketers believed that consumers could favourably evaluate brand extensions by
transferring attitudes or preferences for the iconic product to their extensions. However, this
kind of transfer may depend upon two important factors: the similarity and fit that the
consumer perceives between the iconic product and its extension (Aaker & Keller, 1990;
Barone et al., 2000).
Based on previous research, the product-level similarity between the iconic product
and its extension has been identified as an important factor in the evaluation of the extension.
Aaker (1990) indicated that the key to successfully transferring positive associations was
closely related to the degree of similarity the consumer perceived between the iconic product
category and its unrelated extension sector. Selvanayagam and Ragel (2015) also indicated
that the new extension product was judged by consumers based on their perceptions of

6
Internal

similarity between the iconic product and its extension. Bhat and Reddy (2001) summarized
that the more similar an extension, the more likely the consumer will infer the iconic
product’s features and characteristics onto the extension. Some research brought up a
categorization theory, which indicates that when consumers were exposed to a new extension
in a new category, they evaluated the degree to which this new extension was consistent or
inconsistent with the original brand associations, such as an iconic product (Park et al.,
1991).
Furthermore, Bhat and Reddy (2001) found that the similarity between an iconic
product and the cross-category product was different compared to the similarity of the
product images between the iconic product and its extension; the product category similarity
was irrelevant in the evaluation for extension. To understand this theory, Aaker (1996)
asserted the fundamental principle of brand image rests in the assumption that it was a
multidimensional concept which related to consumers’ overall perception of a brand, the
associations of the brand, and extensions of the brand, consisting of consumers’ product
knowledge, consumers’ perceived value of the product, consumers’ perceived product
features (such as an expertise, a benefit, a component, a lifestyle, etc.) and consumers’ usage
of the product. This explained Bhat and Reddy’s (2001) finding that consumers tended to see
the fit between an iconic product and its extension in terms of the similarity of the images of
both products and irrelevant of their product categories. Bhat and Reddy (2001) further
concluded that fit might contain two dimensions: one related to the product (product category
fit) and the other related to the brand (brand image fit). Product category fit deals with
consumers’ perceptions of the similarity between product categories for the iconic product
and its extension, while brand image fit refers to consumers’ perceptions of the similarity
7
Internal

between the image of the iconic product and the initial image of the extension. A large
amount of evidence pointed to high product category fit impacting consumers to the effect
that, when this situation exists, consumers transferred positive evaluations from the iconic
product to its extension, as there was a large amount of evidence to prove that high product
category fit impacts consumers to transfer positive evaluations from the iconic product to its
extension (Riel et al., 2001; Aaker & Keller, 1990; Park et al., 1991); however, brand image
fit does matter when product category fit was not found in the brand extension (crosscategory or irrelevant category), because when brand image fit exists, consumers were more
likely to treat the cross-category extension as a typical member under iconic product’s
category, which would enhance the positive attitudes transferred to its extension. For
instance, as opposed to non-category fit product, Colgate established a toothbrush line after
its toothpaste was well-known in the market: product category fit existed between these two
products as both were under the oral care market. Another example is the Honda lawnmower,
which can lay claim to superior engine technology based on the company’s long-standing
engine performance in the motor vehicle industry: this extension into the lawnmower market
was possible because consumers recognized the brand image fit (expertise in engines)
between Honda’s vehicles/power engines and Honda’s lawnmower(Chi, 2019).
Broniarczyk and Alba (1994) explained that consumers were likely to spend more
time and attention evaluating the extended product, in terms of discovering the similarity of
the total feature overlap between the iconic product and the extension, when the category of
the extension was in a different sector than within a similar category. Thus, the image-fit, as
part of the similarity, between iconic product and its extension seemed to play a significant

8
Internal

role in making the extension more “natural” for the brand, especially when the category of
extension was a different sector.
Bhat and Reddy (2001) agreed with Broniarczyk and Alba’s statement and further
emphasized that the fit between an iconic product and its extension will be based on the
congruity, similarity, and common attributes between two products. The brand image was
made up of a series of associations that consumers perceived, which may not fall under the
same image-fit as the iconic product ’s category. Other than the typical extensions, the key to
making an extension across categories successful may depend on transferring the image-fit
from an iconic product to the extension. For example, Michelin tires had strong
characteristics of high-end quality, high reputation, travel, tourism, ideas and innovations.
These brand specific attributes distinguished the Michelin brand from other tire brands, and
had been successfully transferred to the Michelin Guide, a hallmark of fine dining restaurants
around the world awarded by the Michelin tire, which was recognized as same characteristics
as its iconic product and had been maintained for more than a century (Liu et al., 2015).
Broniarczyk and Alba (1994) also substantiated this effect as they asserted that consumers’
assessment of fit between the iconic product and its extension depended not only on the
similarity between the two product categories, but also on the brand specific attributes of the
extension.
A large body of evidence from previous studies showed that an extension with a higher
fit received more positive evaluations toward the extension (Bhat & Reddy, 2001; Boush &
Loken, 1991; Aaker & Keller, 1990). However, when product categories were in different
sectors, the more image-fit between iconic product and its extension, the more positive the
consumers’ attitude toward its cross-category extension will be. As long as the brand image
9
Internal

was tied to the specific category or was associated with some relevant associations, the
company may find it easier to extend beyond the original category. For instance, Google
extended its business from a search engine website category into a cell phone category,
because some of the symbolic associations of Google can match the symbolic associations of
the cell phone category, such as “technology,” “fast,” “innovation,” “easy access,”
etc.(Taylor et al., 2017). As another example, if the functional brand image of CAT was
described as “heavy equipment,” “power systems,” and “construction machinery,” the brand
would only make excavators and dozers. But, with the symbolic brand image of “high
quality,” “durability,” and “professionalism,” it is easy to understand that why CAT can
successfully extend into shoes and clothing lines (Hollis, 2012).
H2：Consumers’ attitudes toward cross-category extension correlates with the
image-fit consumers perceive, particularly, when the image-fit of the extended product
matches that of the iconic product, the image-fit positively correlates to consumers’
attitudes toward the extended product.
When the extension product is extended into a new market area, firms may be more
aware of their need to use the advertisement plan effectively in order to fit consumers’
perceptions between the iconic product and its extension because advertising can improve
consumers’ fit perceptions directly (Czellar, 2003). Taylor and Bearden (2003) insisted that
the advertisement expenses and purchase intentions of consumers after advertisement had
significant positive impacts on consumers’ quality evaluations when the brand extension was
in a similar category, while the evaluations for a dissimilar category of brand extension was
somewhat lower. However, there was no specific differences between categories when the
iconic product had a high level of perceived quality. Bei et al. (2011) also found that
10
Internal

promoting the brand extension with a similar brand in a similar category could lead to an
assimilation effect on consumer evaluation. However, when a brand tried to extend to a
dissimilar category, the company may want to use a comparative advertisement to create a
visual and/or sensory link to the iconic product because consumers had already memorized
the style of such advertisements from iconic product advertising. Consumers may find it
easier to absorb information delivered by the extension product’s advertising if there are
consistencies in the style of advertisements for both the iconic product and its extension. For
example, Apple may assess consumers’ attitudes toward iPhone ads, as a comparative
reference, before launching their iWatch to the market, because consumers may transfer the
main features of iPhone (good quality, innovative, etc.) to the iWatch. Positioning the
advertisement successfully by employing the same style as the iconic product seems to be the
key element, especially when a company is launching an extension into a different category
(Dens & De Pelsmacker, 2016). Furthermore, Bambauer-Sachse et al. (2011) believed that
pictures, celebrity endorsers, and slogans were the key elements of advertisements for the
iconic product. And these elements were also believed to be the significant links between an
iconic product and its extension when using the same elements. Because consumers
perceived fit as those elements had been used in advertisements for the iconic product for a
certain period and thus can be perceived by consumers as a typical model of the brand. For
instance, the Apple brand portrayed its brand consistency between the iconic product and its
extension by ensuring their advertising remained consistent between the two products; the
advertising for both shared the same simplicity of clean lines, sharply contrasting colours,
and an expensive-looking sleek design on advertising posters. According to Dens and De
Pelsmacker’s (2016) findings, with the example illustrated, the Apple brand used advertising

11
Internal

typically associated with iPhone to establish the link between iPhone and iWatch, which
enhanced the fit between the two products. Therefore, one could assume that the links would
be transferred from the iconic product’s advertisement to the advertisement of the extension.
This transfer would shrink the limitation of category differences, because transferring the ads
links successfully would help in building the relationship between two products.
H3：Consumers’ attitudes toward cross-category brand extension correlates with
their attitudes toward the advertisement, particularly, when the style of the advertisement
for the extended product matches that of the iconic product, the congruent style positively
correlates with consumers’ attitudes toward the extended product.
Dewey (1986) proposed a theoretical foundation that people’s experiences were
disconnected from one to another and not necessarily connected each other, while each one
experience stands its own opinion or feedback. The author further indicated that since each
experience had its own uniqueness, this fact led to the possible consequence of scattering or
blurring the connections between two experiences and, therefore, losing control the of
direction of any future experience.
Dewey (1986) further focused on classifying experiences. Specifically, the study
identified that experiences had a continuity feature, assuming that each recent or future
experience was influenced by past experiences (this theory can also interpret hypotheses 1
and 2, above). Dewey (1986) further claimed that an experience with an existing product or
service had the effect of impacting the growth of further experiences, more particularly, after
people had an experience with a certain product or service, this experience would lead people
in a specific direction, which imperceptibly narrowed the scale of further experiences. This
12
Internal

finding has been recently applied in the brand and marketing field: Lee et al. (2018) declared
that experiences impact the interactivity between a consumer and the product or service,
which would influence a subsequent change to the consumer’s attitude and behaviour toward
a future experience with another product or service.
Bhat and Reddy (2001) argued that consumers become more acquainted with the
brand extension itself after their first encounter with the extension. Furthermore, they also
discovered that recent research was limited to analyzing whether or not the consumers’
evaluations of a brand extension were successful in transferring the attributes from the iconic
product. Huber et al. (2001) mentioned in their study that a brand extension success was
related to the satisfaction of consumers’ evaluations by ensuring that a certain group of
consumers had perceived the value they expected when using the extended product.
However, there was very limited research targeting a group of respondents who had
established user experience on an extended product; most of the research asks general
questions of their respondents without knowing their actual experiences with the extended
product. The present study proposed that it was more accurate to evaluate consumers’
attitudes toward an extended product when they had specifically used the extended product,
rather than evaluating on a general basis, which required companies to rely on consumers’
perceptions without actual experiences. Therefore, the Table 1 proposed that: before
consumers used the extended product, they are evaluating it solely on the basis of the iconic
product; after consumers used the extended product, they are evaluating it solely on the basis
of the extended product itself.

13
Internal

Table 1: Visual of the hypothesis H4
Consumers’

Consumers’ evaluations

evaluations of the

of the extended product

iconic product
Before UE

**

*

After UE

*

**

Note: UE stands for user experience with the extended product, “*” stands for the degree of
influences.
Hypothesis 4：UE (user experience) with the extended product correlates with the
consumers’ evaluations of the iconic product on their evaluations of the extended
product. In particular, the influence of the iconic product on the evaluations of the
extended product weakens as UE increases.
Some researchers were aware of a crucial fact that failure of an extension will
eventually affect the brand name and the extension will “dilute” the iconic product in many
aspects, such as reputation, perceived quality, and even the brand equity (Loken & John,
1993). Aaker (1996) and Boisvert (2012) indicated that the iconic product could be damaged
or diluted as a result of a negative impression of the extension product. This effect was
defined as extension reciprocity, which describes any changes in evaluations and attitudes
caused by the extension toward the iconic product. A more serious result was that the iconic
product may not have any chance to develop another extension. Boush and Loken (1991)
agreed on the same point as Aaker’s, and they indicated that the future value of the iconic
product was also affected due to the negative impact on the iconic product.
14
Internal

Park et al. (1991) concluded that when an extension product was perceived to be
unrelated and dissimilar to existing brand products, the perception would lead to negative
impacts, which may potentially harm the brand. They also suggested that a sub-branding
strategy was deemed to be an essential method to mitigate the risk of extending a crosscategory product and to ensure a solid bridge between iconic product and extended product.
Followed by this theory, the majority of research (Park et al. 1991; Boisvert, 2012;
Punyatoya, 2013; Bhat & Reddy, 2001) had developed a series of surveys to test their
hypotheses, which showed that consumers’ feedback on an extended product may drive a
positive or negative effect back to an iconic product; however, the participating group of
respondents were randomly selected without filtering the group for those consumers who had
actually had user experience with the extended product. There were only limited researchers
targeting those experienced individuals, which may result in inaccurate outcomes (similar
condition to H4). Therefore, the hypothesis was proposed below:
Hypothesis 5： Consumers’ post-evaluations of a cross-category extension correlates
with their evaluations of its iconic product.
Chapter Three: Methodology
Survey Design
To test the hypotheses, a paper-based survey was established. The purpose of the
survey was to test consumers’ attitudes toward cross-category brand extensions against
established brands. In order to preserve the meaningful significance on theory and practice,
real brands were used as examples since consumers already have perceptions of those
established brands. The survey provided a short illustration and, after that, the survey was
15
Internal

divided into three sections with three different brands: Apple, Google, and Michelin. These
well-recognized brand names in different industries (technology and manufacturing) were
chosen as the iconic products in this study. Each section had a brief description of the
designated brand that stated that the particular brand had introduced a cross-category
product. This was followed by multiple statements measured on an eleven-point Likert scale
(where a -5 corresponds with strongly disagree, a 0 equals a neutral response, and a 5
corresponds with strongly agree) for which participants were requested to point out their
level of agreement or disagreement with the given statements. The participants were also
asked to answer questions relating to demographic information, such as gender, age, and
household income (See Appendix 1).
In section 1, Apple was initially famous for its iPhone, which was categorized as a
smartphone. This iconic product, the iPhone, was put into the test in correlation with the
extended product, iWatch, which was categorized as a watch. By that analogy, Section 2 and
Section 3 follow the same model as Section 1 (See Table 2 for examples used in the survey).
Table 2: Examples used in the survey.
Brand

Iconic Product

Category

Extended
Product

Category

Apple

iPhone

Smartphone

iWatch

Watch

Google

Google Search

Search Engine

Google
Smartphone
(Pixel)

Smartphone

Michelin

Michelin Tire

Tire

Michelin Guide

Book

16
Internal

The familiarity of each brand was tested on the first statement in the survey to obtain
the key indicator for determining the reliability of the entire survey. Those respondents who
were not familiar with the brand were filtered out. The ensuing statement tested whether or
not participants’ familiarity with the brand came from the iconic product. The hypotheses
H1a, H1b, and H1c were tested, thereafter, by stating the iconic product and its extended
product had good credibility, quality, and innovations, respectively. A sample item read,
“The Apple iconic product, iPhone, has good credibility.” Along with “The Apple extended
product, iWatch, has good credibility.” The hypothesis H1a was measured by participants’
responses on iPhone and iWatch. In the next few statements, the participants were tested on
the correlation of image fit for hypotheses H2 and H3. As per hypothesis H4, the participants
were tested on their user experience with the extended product; participants were asked to
answer whether or not they had used the extended product prior to moving to the next
statement. For participants who had used the extended product, they were directed to the next
statements in order to test hypothesis H4. Participants who had never used the extended
product were redirected to a series of other statements for a general perception result, as a
referential point of view. Participants, taken from university students (refer to the section
below for details), completed this survey in exchange for one participation score for their
classes.
When examining the data that was collected from the survey, not all of the data were
analyzed as some of the statements were only offering the reference basis for the study, but
not directly related to the hypothesises. For instance, the survey requested participants who
had no user experience with the extended products to point out their level of agreement with
the general opinions of the extended products to check if the evaluations of the extended
17
Internal

products were on the basis of the brand, the iconic product, or the extended product itself.
These data were not directly related to the study, as the participants who had user experience
with the extended product were the focus of this study.
Sample Description
The survey (N=76) was conducted by utilizing two groups of participants. The first
group consisted of two classes of marketing students at the University of Northern British
Columbia (N=56). In addition to university students, for the second group, surveys were also
randomly conducted among residents of Prince George (N=20) to receive a broader sample,
other than the simple and convenient sample of students. This method was recognized in
some of the previous literature (Keller & Aaker, 1990; Punyatoya, 2013). The full sample
consisted of 76 participants, of whom 55.26% were female, 42.11% were male, and 2.6%
were unknown (no responses). The age of participants ranged from 18 to 53 years old
(M=21), where a majority of the respondents were between 20 and 30 years old. About a
third of total participants reported their annual household income, of which the median size
was CA$90,000.00, and the majority of the reported household incomes were between
CA$50,000.00 and CA$100,000.00. Specifically, the sample was relatively representative of
the population in terms of gender, but not age and household income. Table 3 details the
demographics of the sample.

18
Internal

Table 3: Demographics of Respondents
Variables

No.

Percent

Sex

Variables

No.

Percent

Age
Male

32

42.11%

<20

22

28.95%

Female

44

55.26%

20-30

41

53.95%

Unknown

2

2.63%

31-40

8

10.53%

>40

3

3.95%

No Response

2

2.62%

Household Income
<$50,000

8

10.53%

$50,000 - $100,000

9

11.84%

>$100,000

6

7.89%

No Response

53

69.74%

Chapter Four: Results
Univariate analysis was used in order to evaluate the significance of dimensions and
variables individually according to the responses from the key statements in the survey.
Mean value and standard deviation were the main factors for measurement.
Table 4 shows the responses pertaining to the variables ‘brand familiarity,’ ‘relation
between iconic product and its brand,’ ‘consumer attitudes toward cross-category extended
product,’ ‘brand image consistencies toward cross-category extended product,’ etc., with its
dimension for each of the three given brands. As per the results obtained, all mean values
ranged from 0.63 to 4.73 on the eleven-point Likert scale and all were acceptably positive.
However, there were significant differences between brands on these measures. Apple had
the higher mean value for nearly all of the indicators (except Table 4.7 shows the mean value
19
Internal

of Apple was lower than Google), followed by Google and Michelin, while Michelin showed
relatively lower values in consumer attitudes, brand image consistencies, and advertisement
when compared to Apple and Google. The strongest level of familiarity with the brands
tested was for Google (M = 4.73), followed by Apple (M = 4.32), and then Michelin (M =
2.44), where Michelin showed a notably lower level of familiarity compared to the others.
Overall, the results proved that the Apple, Google, and Michelin brands as familiar brands.
Targeted products iPhone, Google Search, and Michelin Tires were confirmed as iconic
products of their brands, which was supported by the results in Table 4.2, where the Google
brand (M = 4.45) performed as highest, followed by Apple (M = 3.41), and then Michelin (M
= 3.24). The standard deviation of all indicators was within a range of 0.62 – 3.20 on an
eleven-point Likert scale basis.

20
Internal

Table 4: Mean Value and SD of indicators
Brand

Mean

SD

Brand

Mean

SD

4.1 Indicators of brand familiarity

4.2 Indicators of relation between iconic
product and its brand extension

Apple

4.32

1.65

Apple

3.41

2.42

Google

4.73

0.62

Google

4.45

1.71

Michelin

2.44

3.09

Michelin

3.24

2.44

4.3 Overall indicators of consumer attitudes
toward cross-category extended product

4.4 Overall indicators of image-fit toward
cross-category extended product

Apple

2.13

2.71

Apple

3.69

1.73

Google

1.56

3.02

Google

2.21

2.49

Michelin

0.90

3.20

Michelin

0.8

3.14

4.5 Overall indicators of advertisement
toward cross-category extended product

4.6 Overall indicators of user experience
toward cross-category extended product

Apple

3.60

1.61

Apple

3.38

1.75

Google

1.3

2.36

Google

3.88

1.35

Michelin

0.625

2.74

Michelin

3.5

1.43

4.7 Overall indicators of consumers’ postevaluations toward iconic product
Apple

2.5

2.20

Google

3.75

1.83

Michelin

2.11

2.42

Because the three well-recognized brand examples in different industries were chosen
(Apple represented consumer electronics and services, Google represented the technology

21
Internal

industry, Michelin represented a manufacturing business), a Chi-square analysis was used to
test differences between male and female responses to key measures (brand familiarity,
definition of iconic product, the extension product related variables: credibility, quality,
innovativeness, consumer’s overall attitude, image-fit, advertisement-match, user experience
(After I used the extended product, I evaluate it on the basis of the extended product itself)
and post-evaluation (After I used the extended product, my experience with the product is
positive)). The Chi-square analysis was computed using a 2 x 3 design (male/female x
agree/neutral/disagree), where in certain circumstances the analysis was computed with a 2 x
2 design when any one of the three values was not applicable. Some results became
unavailable when two of the three values were not applicable. According to the results in
Table 5, there were no statistically significant differences in the responses to the brand and
the extension related statements between male and female respondents, except for the
response to the innovativeness of the Apple brand, which means that the male and female
respondents perceived the innovativeness of the iWatch differently.

22
Internal

Table 5: Gender differences in response to the brand and the extension related statements
Gender

DF

Chi-square test
(p-value)

Male

Female

(N=32) (N=44)
Brand familiarity

Apple

31

42

1

0.05 (0.83)

(Neutral=0)
Google

31

42

N/A

N/A

Michelin 31

40

2

1.65 (0.44)

Relation between

Apple

32

41

2

0.09 (0.96)

iconic product and its

Google

31

41

2

2.39 (0.30)

brand extension

Michelin 30

38

2

0.67 (0.72)

Credibility (Extended

Apple

31

38

2

0.10 (0.95)

product)

Google

31

41

2

2.39 (0.30)

Michelin 28

31

2

2.29 (0.32)

Quality (Extended

Apple

30

36

2

2.13 (0.34)

product)

Google

30

34

2

1.16 (0.57)

Michelin 27

30

2

3.66 (0.16)

Innovativeness

Apple

31

37

2

6.54 (0.04*)

(Extended product)

Google

30

34

2

2.34 (0.31)

Michelin 28

29

2

3.18 (0.20)

Apple

28

1

0.25 (0.62)

14

(Disagree=0)
23
Internal

Gender

DF

Chi-square test
(p-value)

Consumers’ overall

Google

20

24

2

3.21 (0.20)

attitude toward

Michelin 17

21

2

1.27 (0.53)

Apple

30

33

2

3.53 (0.17)

Google

19

32

2

5.00 (0.08)

Michelin 22

21

2

2.18 (0.34)

Apple

35

1

0.53 (0.47)

extended product
Image-fit

Advertisement-match

28

(Disagree=0)
Google

User-experience

22

25

2

1.77 (0.41)

Michelin 21

19

2

0.95

Apple

5

1

0.13 (0.72)

8

(Disagree=0)
Google

5

3

N/A

N/A

Michelin 7

4

1

0.63 (0.43)

(Disagree=0)
Post-evaluation

Apple

9

5

2

3.11 (0.21)

(Extended product)

Google

5

3

1

0.69 (0.41)

(Neutral=0)
Michelin 4

4

N/A

N/A

Note: * significant at p < .05.

24
Internal

Before testing H1, the reliability test was first examined to measure the consistency of
the statements for testing consumers’ attitudes toward the iconic product and the crosscategory product (credibility, quality, and innovativeness). For instance, the respondents
were asked to share the degree to which they agreed to the following statements:
“The [brand name] iconic product, [an iconic product], has good credibility.”
“The [brand name] iconic product, [an iconic product], has superior quality.”
“The [brand name] iconic product, [an iconic product], strives to introduce innovations.”
SPSS 1.0.0.1275 was used to obtain Cronbach’s  coefficient. According to the
results in Table 6, the Cronbach’s  coefficients for all the measures of the three brands
exceeded the commonly accepted cut-off point (.70), indicating acceptable reliability and
consistency for all measures. The one exception was the measure for the Google iconic
product, where the Cronbach’s  coefficient was .54, this meant that the statement measuring
consumers’ attitudes toward Google search were not representative.
Table 6: Cronbach’s  Coefficient for H1
Brand
Apple

Google

Michelin

Variables

Items Reliability
Coefficient

Valid N (%)

Iconic product (H1)

3

.78

72 (94.7%)

Cross-category product (H1)

3

.78

68 (89.5%)

Iconic product (H1)

3

.54

70 (92.1%)

Cross-category product (H1)

3

.87

65 (85.5%)

Iconic product (H1)

3

.74

63 (82.9%)

Cross-category product (H1)

3

.87

56 (73.7%)

Note: Total N = 76

25
Internal

After the results of the reliability test were obtained, the Pearson's product-moment
correlation analysis was used to measure the strength and direction of any association that
exists in the tested variables between the iconic products and the extended products, which
revealed the best fit and correlation. R 3.4.4 was used for analysis. The data cleaning was
completed by removing the sets of responses that were answered with “Not Applicable” in
either statements for the iconic product or statements for the extended product, and the sets of
the responses that were only answered partially, thus entire rows of data were removed.
Therefore, the degree of freedom for each group of data varied. For H4 and H5, in addition to
the data cleaning mentioned, the data was also manually cleaned by filtering to the group of
responses that had user experiences on the extended product (the respondents were asked if
they had used the extended product and only the group of respondents who had user
experience were asked to answer statements for H4 and H5, it was these respondents who
composed the sample used to test H4 and H5).
After completing the correlation analysis for H1a, H1b, and H1c, the analysis for H1
was further conducted by testing the relationship between consumers’ overall attitudes
toward the extended product and the iconic product of H1 (Consumers’ attitudes toward the
extended product = (ExtH1a + ExtH1b + ExtH1c)/3, consumers’ attitudes toward the iconic
product = (IcoH1a + IcoH1b + IcoH1c)/3). The data cleaning was also completed by manually
eliminating the entire sets of data containing missing values.
According to the samples collected, the sample of university students was tested
independently. As shown in Table 7, testing results for H1, H4, and H5 were obtained. The
Apple and Google brands received the most degree of freedom other than the Michelin
brand, which positively correlated with their level of familiarity of the brands noted in Table
26
Internal

4.1. Based on the values of the Apple brand given in Table 7.1, the p-values of H1 and H5
clearly indicated that the variables ‘Credibility’ (r = 0.53, p < .001), ‘Quality’ (r = 0.54, p
< .001), ‘Innovativeness’ (r = 0.72, p < .001), and ‘Customer evaluation’ (r = 0.93, p < .001)
were significant. Moreover, all the variables were highly and positively correlated, and all the
p-values were less than .001, further supporting the correlation proposed in the hypotheses.
However, the result of H4 showed that the user experience was clearly uncorrelated, with the
p-value as high as 0.99 and the number of respondents as low as 6.
Based on the results shown in Table 7.2, H1c and H5 of the Google brand were
successful (‘Innovativeness’ (r = 0.29, p < .05) and ‘Customer evaluation’ (r = 0.98, p
< .001). However, both results for H1a and H1b showed negative correlations with p-values
at 0.17 and 0.14, respectively. H5 showed results as strongly significant with p-values that
were less than 0.001. Similarly to that of the Apple brand, H4 for the Google brand also
failed with a p-value of 0.13 (DF=5).
As illustrated in Table 7.3, the p-values of the Michelin brand indicated that H1a,
H1b, and H1c showed significant results in supporting the hypotheses (‘Credibility’ (r =
0.31, p < .05), ‘Quality’ (r = 0.44, p < .01), and ‘Innovativeness’ (r = 0.75, p < .001)), while
the results of both H4 and H5 failed to support the hypotheses (‘User Experience’ (r = -0.04,
DF = 3, p = 0.95), ‘Post Evaluation’ (r = 0.50, DF = 3, p = 0.38)).
Finally, the results of the correlation analysis for consumers’ overall attitudes
between the extended product and the iconic product were also obtained and presented in
Table 7. The results illustrated that all three of the brands were responded to as strongly
significant (Apple: r = 0.65, p < .001; Google: r = 0.48, p < .01; Michelin: r = 0.54, p < .001),

27
Internal

which suggests a positive correlation between consumers’ attitudes toward the cross-category
extension and the iconic product as evaluated by the combination of consumers’ perceived
credibility, quality, and innovativeness of two products.

28
Internal

Table 7: Correlation analysis results for evaluations between iconic product and crosscategory extensions for H1, H4 and H5 (student sample only).
7.1 Brand: Apple

7.2 Brand: Google

Variables

Cor

DF PValue

Result

Cor

DF

PValue

Result

Credibility
(H1a)

0.53*** 52

0.00

Support

0.19

49

0.17

Not
Support

Quality (H1b)

0.54*** 50

0.00

Support

0.21

49

0.14

Not
Support

Innovativeness 0.72*** 50
(H1c)

0.00

Support

0.29*

48

0.04

Support

H1

0.65*** 51

0.00

Support

0.48**

48

0.00

Support

User
Experience
(H4)

-0.01

6

0.99

Not
Support

0.63

5

0.13

Not
Support

PostEvaluation
(H5)

0.93*** 6

0.00

Support

0.98*** 5

0.00

Support

7.3 Brand: Michelin
Variables

Cor

DF PValue

Result

Credibility
(H1a)

0.31*

43

0.04

Support

Quality (H1b)

0.44**

42

0.00

Support

Innovativeness 0.75*** 39
(H1c)

0.00

Support

H1

0.54*** 39

0.00

Support

User
Experience
(H4)

-0.04

0.95

Not
support

3

29
Internal

Variables

Cor

DF PValue

Result

PostEvaluation
(H5)

0.50

3

Not
Support

0.38

Notes: *** significant at p < .001; ** significant at p < .01; * significant at p < .05; Cor is
Correlation.
The correlation test was further conducted by using the combination sample,
including the student sample and the general population, to infer whether or not the analysis
results listed in Table 7 would be influenced by the added sample. As shown in Table 8, the
Apple brand and Google brand continued to receive most of the responses, while the
Michelin brand continued to receive relatively smaller responses. This fact potentially
illustrated that the low responses for H4 and H5 from targeted participants were not affected
by sample types but by the geographic limitation of the sample. According to the correlation
results for the Apple brand (see Table 8.1 for details), ‘Credibility’ (r = 0.54, p < .001),
‘Quality’ (r = 0.53, p < .001), ‘Innovativeness’ (r = 0.73, p < .001), and ‘Customer
evaluation’ (r = 0.89, p < .001) showed positive and strong significant results in further
support of H1 and H5. However, the results did not appear to support H4, which showed the
user experience was clearly uncorrelated with the p-value as high as 0.80 and the number of
respondents as low as 11.
Based on the results shown in Table 8.2, H1c and H5 for the Google brand were
successful (‘Innovativeness’ (r = 0.31, p < .05) and ‘Customer evaluation’ (r = 0.96, p
< .001)), but H1a and H1b were not (‘Credibility’ (r = 0.18, p = 0.14), ‘Quality’ (r = 0.17, p =
0.17)). Matching the results of the Apple brand, H4 for the Google brand also failed with a pvalue of 0.08 (DF=6).
30
Internal

The results for the Michelin brand incurred the exact same situation (See Table 8.3
for details) where the p-values of H1a, H1b, and H1c showed significant results in supporting
the hypotheses (‘Credibility’ (r = 0.37, p < .01), ‘Quality’ (r = 0.44, p < .001),
‘Innovativeness’ (r = 0.74, p < .001)). The testing results for both H4 and H5 were not
successful (‘User Experience’ (r = 0.22, DF = 7, p = 0.57), ‘Post Evaluation’ (r = 0.62, DF =
7, p = 0.08)).
Lastly, the overall test of H1 was also analyzed to obtain whether a positive
correlation exists between consumers’ attitudes toward the two products, using the same
method as the one used in the student sample. The results illustrated that all three of the
brands were continuously responded to as strongly significant (Apple: r = 0.65, p < .001;
Google: r = 0.41, p < .01; Michelin: r = 0.39, p < .01), which has further demonstrated the
relationship between the two products.
To summarize the results obtained in Table 7 and Table 8, the results from the student
only sample and the results from the combination of student and general population resulted
in a correlation between the results of either grouping, so the combined group results were
used for this study: The overall evaluation of H1 was supported for both samples.
Meanwhile, the variables ‘Credibility’ and ‘Quality’ were significant to both the Apple brand
and the Michelin brand, but were not significant to the Google brand. The variable
‘Innovativeness’ showed significance to all three tested brands, while the variable ‘UserExperience’ was not significant to any of the tested brands. The variable ‘Post-Evaluation’
partially supported the Apple brand and the Google brand, but not the Michelin brand.
Because the two tests revealed the same results, this study used the results of Table 7 as the
final testing results.
31
Internal

Table 8: Overall correlation analysis results for evaluations between iconic product and
cross-category extensions for H1, H4, and H5.
8.1 Brand: Apple

8.2 Brand: Google

Variables

Cor

DF PResult
Value

Credibility
(H1a)

0.54*** 69

0.00

Quality (H1b)

0.53*** 66

Innovativeness
(H1c)

Cor

DF PValue

Result

Support 0.18

65

0.14

Not
Support

0.00

Support 0.17

64

0.17

Not
Support

0.73*** 67

0.00

Support 0.31*

62

0.01

Support

H1

0.65*** 67

0.00

Support 0.41**

62

0.00

Support

User
Experience
(H4)

0.77

11

0.80

Not
0.66
Support

6

0.08

Not
Support

PostEvaluation
(H5)

0.89*** 10

0.00

Support 0.96*** 6

0.00

Support

8.3 Brand: Michelin
Variables

Cor

DF PResult
Value

Credibility
(H1a)

0.37**

57

0.00

Support

Quality (H1b)

0.44*** 55

0.00

Support

Innovativeness
(H1c)

0.74*** 54

0.00

Support

H1

0.39**

53

0.00

Support

User
Experience
(H4)

0.22

7

0.57

Not
support

32
Internal

Variables

Cor

DF PResult
Value

PostEvaluation
(H5)

0.62

7

0.08

Not
Support

Notes: *** significant at p < .001; ** significant at p < .01; * significant at p < .05; Cor is Correlation.

H2 and H3 were tested separately, because consumers perceived “Image-fit” and
“Advertisement-match” were influenced by their attitude toward the iconic product, as
proposed in the hypothesises. Before testing the H2 and H3, the reliability test was first
examined to measure the scales set up for H2 and H3, because the respondents were asked to
provide their degree of agreement to two consistent statements that associated with the
hypothesises. For instance, the respondents were asked to share the degree to which they
agreed to the following statements:
“The image of the [brand name] extended product, [an extended product], agrees with
the image of the [brand name] iconic product, [an iconic product].”
“Even though [an iconic product] and [an extended product] are two different types of
products, they represent the same brand image of [brand name].”
The results of both statements contributed to the variable “Image-fit.” The same
situation applies to the variable “Advertisement,” where the respondents were asked to share
the degree to which they agreed with the following statements, and the results of both
statements contributed to the variable “Advertisement-fit”:
“When I was watching [an extended product] advertisement, I can tell it is a typical
[brand name] advertisement.”

33
Internal

“The style of [an extended product] is similar to that of [an iconic product]
advertisement.”
As shown in Table 9, the Cronbach’s  coefficients for all measures of the three
brands exceeded the commonly accepted cut-off point (0.70), ranging from 0.76 to 0.88, with
the Image-fit of the Apple brand being the lowest (0.76) and the Advertisement of the
Michelin brand being the highest (0.88), indicating acceptable reliability and consistency for
all measures. (Note: The survey also provided a reserve statement for H2 by asserting: “The
image of the [brand name] extended product, [an extended product], deviates from the image
of the [brand name] iconic product, [an iconic product].” However, this measure was
discarded in view of the potential confusion with the statement which may mislead the
participants, with MApple = -1, SDApple = 2.7; MGoogle = -1.51, SDGoogle= 2.62; MMichelin = -1.86,
SDMichelin = 2.71, where the items were reverse scored, so that the opposite was true.)
Table 9: Cronbach’s  Coefficient for H2 and H3
Brand
Apple

Google

Michelin

Variables

Items

Reliability
Coefficient

Valid N (%)

Image-fit (H2)

2

.76

71 (93.4%)

Advertisement-match (H3)

2

.84

67 (88.2%)

Image-fit (H2)

2

.79

71 (93.4%)

Advertisement-match (H3)

2

.82

60 (78%)

Image-fit (H2)

2

.83

59 (77.6%)

Advertisement-match (H3)

2

.88

47 (61.8%)

Note: Total N = 76

34
Internal

After obtaining the acceptable reliability for all measures of H2 and H3, the
measurement of the consumers’ attitudes toward the extended product was collected by the
same method illustrated in H1 (Consumers’ attitudes toward extended product = (Ext H1a +
ExtH1b + ExtH1c)/3). Furthermore, the measurement of image-fit and advertisement were
collected by obtaining the average data of the two consistent statements that associated with
the hypothesises respectively (image-fit = (image-fit1 + image-fit2)/2; advertisement-match =
(advertisement-match1 + advertisement-match2)/2).
Next, the data cleaning was completed before running the Pearson’s correlation test
by using the same method as previously stated. Thereafter, the correlation results were
obtained in Table 10 by observing the relationship between H2/H3 and the respondents’
overall evaluation on the extended product that associated with their attitudes toward the
iconic product.
According to the correlation results for the Apple brand for the student sample (see
Table 10.1 for details), the image-fit showed strong significant results with r = 0.40, p < .01,
which postulated that consumers’ attitudes toward the cross-category extension had a
positive effect on the image-fit that consumers perceived. However, the opposite result was
found for the advertisement where the p-value was observed to be slightly more than 0.05 (r
= 0.28, p = 0.07). The results of the Google brand showed that the image-fit that consumers’
perceived highly correlated with consumers’ evaluations on the extended product (r = 0.41, p
< .01), further supporting H2. Similar to the results of the advertisement-match for the Apple
brand, the results of the advertisement-match for the Google brand showed there was no
positive or significant correlation between the advertisement-match and consumers’
evaluations of the extended product, with the p-value as high as 0.60. Table 10.3 provided
35
Internal

the results for the Michelin brand, both H2 and H3 were found to be not significant (Imagefit: r = 0.28, p = 0.08; advertisement-match: r = 0.24, p = 0.17).
As with the method for the previous hypotheses, the correlation tests for H2 and H3
were conducted using the combination sample, including the students and the general
population, to infer whether or not the analysis results would be influenced by the added
sample. According to the correlation results showed in Table 10.4 for the Apple brand, the
image-fit continued to show strong significant results (with r = 0.49, p < .001), where the
degree of significance became stronger when the additional sample was added into the
measurement (the correlation coefficient increased from 0.28 to 0.43, the degree of freedom
increased from 47 to 66, and the p-value showed even more significant value from p < .01 to
p < .001). Opposite to the failed result of the advertisement-match with the student sample,
the correlation result of the combined sample was found to be a significant and positive
relationship, where the correlation coefficient increased from 0.28 to 0.42 and the p-value
reduced from p = 0.07 to p < .001.
The results of the Google brand with the combined sample, as shown in Table 10.5,
aligned with the results for the student sample, which showed the image-fit that consumers
perceived highly correlated with consumers’ evaluations of the extended product, further
supporting H2 (r = 0.42, p < .001). Although the result of advertisement-match was still not
significant for the Google brand, the correlation coefficient increased from 0.08 to 0.20 and
the p-value reduced from 0.60 to 0.14.
Table 10.6 provides the results of the Michelin brand with the combined sample. Both
H2 and H3 were significantly supported. The correlation coefficient of image-fit (H2)

36
Internal

increased from 0.28 to 0.49, the degree of freedom increased from 41 to 53, and the p-value
reduced from p =0.08 to p < .01. And the correlation coefficient of advertisement-match (H3)
increased from 0.24 to 0.32, the degree of freedom increased from 33 to 44, and the p-value
reduced from p = 0.17 to p < .05.
To summarize Table 10, the analysis results from the student only sample and the
results from the combination of student and general population respondents resulted in a
correlation between the results of both grouping, except for the Michelin brand and the
advertisement-match (H3) of the Apple brand: neither the image-fit nor the advertisementmatch of the Michelin brand were supported by the student sample, but they were supported
by the combined sample. The advertisement-match of the Apple brand was not supported by
the student sample but was significant to the combined sample. Furthermore, according to the
comparison between the student sample and the combined sample, all of the correlation
coefficients became higher and the p-values became more significant when the degree of
freedom increased. This addition particularly improved the results for H3 of the Apple brand
and both H2 and H3 of the Michelin brand, where the p-values were found to be significant
for the combined sample. It could be assumed that, with this observation, the advertisementmatch of the Google brand could have obtained a significant result for the combined sample
if there had been a high degree of freedom because the p-value was significantly reduced
from 0.60 to 0.14. Therefore, it was considered reasonable to use the results from the
combined group for this study: Image-fit was found to be significant for all three brands and
advertisement-match showed a significance for the Apple and Michelin brands, but not the
Google brand.

37
Internal

Table 10: Correlation analysis results for evaluations between iconic product and crosscategory extensions for H2 and H3.
Student Sample Only
10.1 Brand: Apple

10.2 Brand: Google

Variables

Cor

DF PValue

Result

Cor

DF

PValue

Result

Image-Fit
(H2)

0.40**

47

0.00

Support

0.41**

51

0.00

Support

Advertisem
ent-Match
(H3)

0.28

44

0.07

Not
Support

0.08

41

0.60

Not
Support

10.3 Brand: Michelin
Image-Fit
(H2)

0.28

41

0.08

Not
Support

Advertisem
ent-Match
(H3)

0.24

33

0.17

Not
Support

Combination Sample (Student Sample and General population)
10.4 Brand: Apple

10.5 Brand: Google

Variables

Cor

DF PValue

Result

Image-Fit
(H2)

0.49***

66

0.00

Advertisem
ent-Match
(H3)

0.42***

60

53

Cor

DF

PValue

Result

Support 0.42***

64

0.00

Support

0.00

Support 0.20

54

0.14

Not
Support

0.00

Support

10.6 Brand: Michelin
Image-Fit
(H2)

0.40**

38
Internal

Variables

Cor

DF PValue

Result

Advertisem
ent-Match
(H3)

0.32*

44

Support

0.04

Notes: *** significant at p < .001; ** significant at p < .01; * significant at p < .05; Cor is
Correlation.
According to the results obtained in Table 8, the correlation test for H4 failed for all
three brands, which illustrated that there was no correlation relationship between consumers’
evaluations of the extended product on the basis of the iconic product before UE and the
consumers’ evaluations of the extended product on the basis of the extended product itself
after UE. Therefore, a supplemental test was conducted to compare the two related samples
and determine whether their population mean ranks differed. The null hypothesis and the
alternative hypothesis are listed below:
Hnull: Both before and after UE on the extended product, consumers evaluate the extended
product on the basis of the iconic product. H null = 0
Ha: Before UE with the extended product, consumers evaluate the extended product on the
basis of the iconic product; after UE with the extended product, consumers’ evaluations on
the basis of the iconic product decrease. Ha > 0
In this study, the data obtained was on the basis of an eleven-point Likert Scale; the
purpose of this hypothesis was to measure how consumers’ evaluations of the iconic product
on their evaluations of the extended product are affected by the UE with the extended
product. In other words, the consumers’ evaluations of the iconic product on their evaluations
of the extended product exist before the user experience with the extended product. The same
39
Internal

group sample was used for each brand to test the significance of the before and after effect. A
nonparametric paired t-test was used. The data cleaning was completed by removing all the
invalid data (“Not Applicable” or missing values) in each row.
According to Table 11 (below), all three brands received limited responses (N Apple =
13, NGoogle = 8, NMichelin = 9); the direction of the hypothesis (Ha < 0) for the Apple and
Michelin brands were identified with mean differences at +0.07 and +0.33, respectively;
however, the differences were not obviously significant. The paired t-test for the three brands
also failed compared to the commonly accepted significant levels of p < .05 (p Apple= 0.44,
pGoogle = 0.50, pMichelin = 0.50). Therefore, Ha was not supported.
Table 11: Nonparametric paired t-test for H4
Brand

N

M

SD

Min

Max

Sig. (1-tailed)
.44

Apple

Before UE

13

2.38

1.90

0

5

13

2.31

2.29

-1

5

Google

After UE
Before UE
After UE
Before UE

After UE
Before UE
After UE
Michelin Before UE
After UE
Before UE
After UE

+.07

.44

8

1.38

2.77

-4

5

8

1.38

2.32

-4

3

0

.50

9

.89

3.55

-5

5

9

.56

4.15

-5

5

+.33

.50

Chapter Five: Discussion

40
Internal

This study concentrated on the cross-category brand extension from an iconic product
angle. In order to better analyze the subject, the topic was hypothesized using a diversified
portfolio of companies. Three iconic products from three brands under different market
conditions were selected, which differentiated from prior research that used a parent brand or
parent brands from the same category. This diversification may account for the results of this
study providing more meaningful insights into extension evaluations.
Another highlight discovered during this study was that the evaluated extensions, in
the view of the iconic product, provided more accuracy when analyzing the results, whereas
prior research often observed the extensions from the consumer evaluations of the parent
brand. It was believed that the influence of the parent brand may have led to misleading
results in extension evaluation because when consumers associated the extension with a
parent brand, the consumers often associated the extension with a specific and existing
product from the parent brand. Therefore, consumers’ evaluations on an extension product
may have come from different product associations, rather than the parent brand, itself.
This study was able to locate the iconic product of the brand and provided a relatively
clear direction for consumer evaluations of the extension products. The selected crosscategory products were tested to assess the leading factors of consumer attitudes on these
types of extensions, and it was shown that there were no gender differences when consumers
were evaluating the extensions. Additionally, this study investigated the reverse effect of
consumer evaluation against iconic products. According to the results, previous hypotheses
could be illustrated as follows:

41
Internal

The findings of this research further confirmed the theory by Chun et al. (2015) when
the results demonstrated that innovativeness was an absolutely significant predictor for
brands entering into cross-category markets. This attribute established a consumer attitude
toward an iconic product because innovativeness positively affected their acceptance of the
cross-category extension and eliminated the “incongruity feeling”: The degree to which
consumers felt innovativeness in the extension was the same degree to which the consumer
was willing to accept the extension product, as it was congruent with the iconic product. On
the other hand, a consumer who perceived innovativeness from an iconic product was more
likely to buy an extension product that largely differed from the iconic product, because the
category ‘mismatch’ was not a consideration when a consumer was evaluating the extension
product.
Credibility and Quality were considered strong factors since the Apple and Michelin
brands reported them as strongly significant factors; however, the results for the Google
brand was not successful. A potential cause for the failure of H1a and H1b could involve the
fact that consumers did not perceive the credibility and quality of the iconic product (the
search engine) and/or the extended product (the smartphone). This was part of a trend that
was seen across the failure of the reliability test for H1a and H1b for the Google brand. The
Google iconic product was a virtual product and the extended product lacked adequate
“market testing” given it was just recently launched. Since so little was known about this
relatively new product, consumers had limited user experiences to base their evaluations for
this extension product.
Overall, results may differ based on consumers’ perceptions of different brand
concepts, class, etc. (Park et al., 1991). With that being said, it could be predicted that when
42
Internal

the iconic product was a real object, consumers were able to transmit credibility and quality
from one product to another. However, when the iconic product was a virtual object (a search
engine), then the extension product may not necessarily have the same perceived credibility
and quality translate from the iconic product, because there was no visual way to hold and
use the virtual product the same way as a non-virtual product. One cannot hold a search
engine and claim that it was made of good quality material, nor can one say definitively that
one search engine was measurably better than another, so how can one translate an
evaluation from this iconic product to their extension. One could assume the credibility, but
not necessarily the quality, because one can measure quick response time and the accuracy of
the word associations, but that is not easily a confirmation of quality. Other variables were
needed to be found when measuring, because quality cannot be determined in exact
quantities with a virtual product like the Google search engine.
The results obtained indicate that the image-fit was a strong determinant of consumer
attitude toward cross-category extension, which supported H2. This finding was significant
because it further confirmed consumers’ acceptance of a cross-category product was not
limited to categorization if an image-fit was perceived between the iconic product and its
extension. On the other hand, the higher image-fit that consumers perceived would dilute the
“incongruity feeling” toward the cross-category extension. Therefore, a major
recommendation for companies was to pre-test the level of image-fit before launching the
cross-category extension to avoid failure when extending.
Apart from image-fit, the results also confirmed the significance of the role that
advertisement-match played when launching an extension into a new category. Respondents
agreed that they perceived the similarities in the style of advertisement between an iconic
43
Internal

product and its extended product, which intangibly increased the connections between two
products, and reduced the “incongruity feeling” when the extended product was unrelated to
its iconic product. However, the testing toward the Google brand was an exception, where
the results did not support in either sample.
According to the results presented in Table 8 and Table 11, both the correlation test
and the t-test for H4 across the three brands all failed. The expectation was that before
respondents used the extended product, they would evaluate it on the basis of their evaluation
of its iconic product; whereas, after the consumer purchased and used the extension (user
experience), they would evaluate it based on the extended product itself, which would
translate the evaluation from an iconic product basis to an extended product basis, so that the
extension became its own independent experience and evaluation. As the user experience
with the extended product creates a strong familiarity and a more personal evaluation of the
extension product, the original evaluation of the iconic product dissipates in importance to
the extension and may also disappear entirely from the consumer’s evaluation, making the
extension evaluation independent from the iconic product. However, one reason that may
cause the failure was a result of limited respondents for this particular hypothesis. This
failure was critical to the hypothesis because less than 15% of total respondents had user
experience, which did not provide enough valid data for H4 and H5, as the sample size was
too small. A number of potential factors for this failure exist, such as geographical limitations
of the tested area, the popularity of the extended products, and the challenge of extended
products recently released into the market (iWatch was introduced to the market in 2015,
similarly, Google Smartphone was just launched in late 2016). As with H4, H5 also received
limited responses as a result of the same conditions.
44
Internal

H5 was supported, which illustrated that, after the UE with a cross-category
extension, consumers’ evaluations of the cross-category product had a positive correlation
with their evaluations of its iconic product. This finding confirmed Park et al.’s (1991)
theory. Furthermore, this study used a group of respondents who had user experience with
the extended product, which provided references for companies. Overall, the findings
provided the best guidelines to what a company should take when the brand would like to
develop cross-category extensions with more suitability and less risk.
Chapter Six: Limitations and Future Research
As pointed out previously, additional work might consider collecting a larger number
of responses to obtain bigger sample sizes against H4 and H5. Based on this study, the
research still strongly corroborates that both H4 and H5 could validly be supported by a
series of previous research in other categories and their extensions if the geographic
limitation could be addressed.
The findings of this study were also limited by the convenience nature of the sample,
that using university students was a common approach in the research but doing so limited
the generalizability to different groups. Particularly, the limitation of the age distribution of
the sample (over 80% of the sample were under 30 years old) could potentially have resulted
in a certain level of bias on the results. Because younger people were more likely to take risk,
to try a new product or experience, and to explore completely different fields, this consumer
behaviour would change depending on the age and stage of life of respondents (Figner et al.,
2009). That being said, when a brand launches a new product in a new market, consumers
who are in early adulthood tend to be more willing to try and adapt to the product than
45
Internal

consumers who are elderly, therefore, it was not a surprise that the hypotheses of this study
were supported. However, it could be assumed that the final results obtained from the data
analysis may be effected if the age distribution of the participants was more diversified.
In addition to the age limitation, the other demographic limitation was the household
income of the participants. Nearly 70% of the participants’ household income could not be
obtained and 10% of participants’ household income was below $50,000.00. As the three
extension products provided to the participants were relatively expensive and not related to
daily necessities (iWatch, Google smartphone, and Michelin Guide), it was demonstrated that
demographic factors had a significant impact on consumer behaviour, including the
differences in household income (Laoviwat et al., 2014). The income level affects what
consumers are able to afford and their purchase intentions. Individuals from a lower income
level are more likely to spend money buying products that are necessary for daily living,
whereas wealthy people have relatively more financial allowance and willingness to purchase
new products, which could potentially lead to a relatively bigger sample size for H4 and H5
of this study. However, the influence of household income on the results of this study is
unknown due to the fact that the majority of the participants were university students and did
not disclose their household income. Future work could utilize online survey methods with
the view of providing a more private environment for disclosure and reaching a broader
target audience of participants in a fast and accurate fashion.
The results of this study were examined by taking three extensions from upscale
consumption brands with relatively higher than average purchases. Future studies could
choose brands in daily necessities categories, such as apparel, fast-moving consumer goods
(FMCG), etc., because the consumer feedback in such categories is expected to be higher and
46
Internal

broader compared to the upscale consumption categories. More importantly, when daily
necessities brands extended into an entirely different category, it was hypothesized that
additional risks must be taken into consideration, such as the demographic differentiation, a
relatively diversified consumer group, greater transaction repetition, etc. (Laoviwat et al.,
2014).
This study raised the importance of the image-fit that consumers perceived when they
were evaluating a cross-category extension; however, the study only proved that the imagefit exists between the iconic product and the cross-category extension; Specifics about the
dimensions or aspects of the image-fit that were involved in such extensions was not
addressed, nor was the extent to which the image-fit was needed (level and measurement of
the image-fit) to be able to successfully extend a new product in a different category.
According to the results specifically for the Google brand, H1a, H1b, H3, and H4 all
failed. A potential cause of the failure may be the fact that the iconic product was a virtual
product and the survey designed for this study could not efficiently apply to such
merchandise. Future surveys could focus on designing the questionnaire exclusively for
virtual products and specifically target virtual products, such as search engines, online
games, social networking sites, etc. It was also interesting to study that how a virtual brand
could extend its business to a tangible product.
In addition, the study used real market conditions with real examples of iconic
products and their extensions. It was interesting that the results may differ by providing a
fictional extension product to respondents to test whether the relationships still remain under
the same variables, which helps companies navigate a general orientation of consumer

47
Internal

preferences on extended products and contributes to companies who are at the proposal and
testing stage for the extended product. Furthermore, if an empirical, long-term study was
done to examine an original proposal, which provided a follow-up survey to the same group
of participants after the fictional extended product was lunched in the market, this future
study would need to concentrate on consumer feedback to highlight the differences between
consumer reactions to a fictional extension product and their reactions to a real extension
product. This empirical study would be able to map the differences between the consumers’
initial responses to the fictional extension product and their responses to the actual product
after its launch, which may show a surprising amount of contrast.
Lastly, this study had been performed with a geographical limitation; future studies
could extend the research internationally, which will benefit companies who want to
introduce cross-category products overseas. Therefore, they will have to pay more attention
to added variables, such as a particular country’s politics and culture, as there are several
risks associated these unchangeable conditions. Language and religion are also major factors
(Kyambalesa, 1993). Unfamiliarity with language context creates misunderstandings and can
cause major losses when translating advertising statements, as the expected idea was not
delivered to consumers properly. Religion has always been tied to people’s way of thinking,
their type of lifestyle, how they process commercial messages, and their decisions about
purchasing certain products. Rare literature provides sufficient evidence that extension
products can be affected by culture and religion; these interactions are still undeveloped
concepts that could provide future fodder for brand extension research (Kyambalesa, 1993).

48
Internal

References
Aaker, D. A., & Keller, K.L. (1990). Consumer evaluations of brand extensions. Journal of
Marketing 54 (1), 27-41. Business Source Complete. Web. 9 Jan. 2016.
Aaker, D. A. (1991). Managing brand equity: Capitalizing on the value of a brand name.
New York,NY; Toronto, ON: Free Press.
Aaker, D. A. (1996). Building strong brands. New York: The Free Press.
Bambauer-Sachse, S., Hüttl, V., & Gierl, H. (2011). Can advertising elements improve
consumer evaluations of brand extensions with a moderate or low fit? Psychology &
Marketing, 28 (2), 205-218. Business Source Complete. https://doiorg.prxy.lib.unbc.ca/10.1002/mar.20387
Barone, M. J., Miniard, P. W., & Romeo, J. B. (2000). The Influence of Positive Mood on
Brand Extension Evaluations. Journal of Consumer Research, 26(4), 386–400.
https://search-ebscohostcom.prxy.lib.unbc.ca/login.aspx?direct=true&db=bth&AN=3087034&site=ehostlive&scope=site
Bei, L.-T., Chu, C.-H., & Shen, Y.-C. (2011). Positioning brand extensions in comparative
advertising: An assessment of the roles of comparative brand similarity, comparative
claims, and consumer product knowledge. Journal of Marketing
Communications, 17(4), 229–244. https://doiorg.prxy.lib.unbc.ca/10.1080/13527260903478367
Bhat, S., & Reddy, S. K. (2001). The Impact of parent brand attribute associations and affect
on brand extension evaluation. Journal of Business Research, 53 (3), 111-122.
Business Source Complete. https://doi-org.prxy.lib.unbc.ca/10.1016/S01482963(99)00115-0
Boush, D. M., & Loken, B. (1991). A Process-tracing study of brand extension evaluation.
Journal of Marketing Research (JMR), 28 (1), 16-28. Business Source Complete.
https://doi-org.prxy.lib.unbc.ca/10.2307/3172723
Boisvert, J. (2012). The Reciprocal impact of vertical services line extensions on parent
brands. Emerald Group Publishing Limited, 22 (6), 546-564.
Boisvert, J., & Burton, S. (2011). Towards a better understanding of factors affecting transfer
of brand associations. Journal of Consumer Marketing, 28 (1), 57-66. https://doiorg.prxy.lib.unbc.ca/10.1108/07363761111101958
Broniarczyk, S. M., & Alba, J. W. (1994). The Importance of the brand in brand extension.
Journal of Marketing Research (JMR,) 31 (2), 214-228. Business Source Complete.
https://doi-org.prxy.lib.unbc.ca/10.2307/3152195
49
Internal

Chi, C. (2019, April 17). 10 of the Most (& Least) Successful Brand Extensions to Inspire
Your Own. [Web log post]. Hubspot. Retrieved from
https://blog.hubspot.com/marketing/brand-extensions
Chun, H. H., Park, C. W., Eisingerich, A. B., & MacInnis, D. J. (2015). Strategic benefits of
low fit brand extensions: When and why? Journal of Consumer Psychology (John
Wiley & Sons, Inc. ), 25(4), 577–595. https://doiorg.prxy.lib.unbc.ca/10.1016/j.jcps.2014.12.003
Czellar, S. (2003). Consumer attitude toward brand extensions: An Integrative model and
research propositions. International Journal of Research in Marketing, 20 (1), 97.
Business Source Complete. Web. 25 July 2016. https://doiorg.prxy.lib.unbc.ca/10.1016/S0167-8116(02)00124-6
Dens, N., & De Pelsmacker, P. (May 2016). Does poor fit always lead to negative
evaluations?: Extension advertising and perceived brand quality. International
Journal of Advertising, 35 (3), 465–485. https://doiorg.prxy.lib.unbc.ca/10.1080/02650487.2015.1057924
Dewey, J. (1986). Experience and Education. Educational Forum, 50(3), 241–52.
https://search-ebscohostcom.prxy.lib.unbc.ca/login.aspx?direct=true&db=eric&AN=EJ332588&site=ehostlive&scope=site
Figner, B., Weber, E. U., Mackinlay, R. J., & Wilkening, F. (2009). Affective and
Deliberative Processes in Risky Choice: Age Differences in Risk Taking in the
Columbia Card Task. Journal of Experimental Psychology. Learning, Memory &
Cognition, 35(3), 709–730. https://doi-org.prxy.lib.unbc.ca/10.1037/a0014983
Hollis, N. (2012, April 16). Leveraging the power of a strong brand by extension. [Web log
post]. Kantar. Retrieved from http://www.millwardbrown.com/globalnavigation/blogs/post/mb-blog/2012/04/16/Leveraging-the-power-of-a-strong-brandby-extension.aspx
Huber, F., Herrmann, A., & Morgan, R. E. (2001). Gaining competitive advantage through
customer value oriented management. Journal of Consumer Marketing, 18(1), 41–53.
https://doi-org.prxy.lib.unbc.ca/10.1108/07363760110365796
Kaur, H., & Pandit, A. (2015). Modelling Consumer Evaluation of Brand Extensions:
Empirical Evidence from India. Vision (09722629), 19(1), 37–48. https://doiorg.prxy.lib.unbc.ca/10.1177/0972262914565557
Klink, R. R., & Smith, D. C. (2001). Threats to the External Validity of Brand Extension
Research. Journal of Marketing Research, 38 (3), 326-335.

50
Internal

Kyambalesa, H. (1993). Business innovation and competitiveness in developing countries.
Aldershot, Hants, England; Brookfield, Vt., USA: Avebury, 1993.
Laoviwat, P., Suppapanya, P., & Yousapronpaiboon, K. (2014). A Study of Demographics
Influencing on Consumer Behavior and Attitude towards Brand Equity of Optical
Business in Thailand. International Journal of Trade, Economics and Finance.
5(395), 347-350. http://doi.org/10.7763/ijtef.2014.v5.395
Lee, H., Lee, K., & Choi, J. (2018). A Structural Model for Unity of Experience: Connecting
User Experience, Customer Experience, and Brand Experience. Journal of Usability
Studies, 11(1), 8–34. https://search-ebscohostcom.prxy.lib.unbc.ca/login.aspx?direct=true&db=a9h&AN=133658487&site=ehostlive&scope=site
Loken, B., & Roedder J. D. (1993). Diluting brand beliefs: When do brand extensions have a
negative impact? Journal of Marketing, 57(3), 71–84.
Liu, C.-H., Chou, S.-F., Gan, B., & Tu, J.-H. (2015). How “quality” determines customer
satisfaction. TQM Journal, 27(5), 576–590. https://doiorg.prxy.lib.unbc.ca/10.1108/TQM-01-2013-0004
Martínez Salinas, E., & Pina Pérez, J. M. (2009). Modeling the brand extensions’ influence
on brand image. Journal of Business Research, 62(1), 50–60. https://doiorg.prxy.lib.unbc.ca/10.1016/j.jbusres.2008.01.006
Monga, A. B., & John, D. R. (2010). What Makes Brands Elastic? The Influence of Brand
Concept and Styles of Thinking on Brand Extension Evaluation. Journal of
Marketing, 74(3), 80–92. https://doi-org.prxy.lib.unbc.ca/10.1509/jmkg.74.3.080
Nan, X. (2006). Affective cues and brand-extension evaluation: Exploring the influence of
attitude toward the parent brand and attitude toward the extension ad." Psychology &
Marketing, 23 (7), 597-616. Business Source Complete. https://doiorg.prxy.lib.unbc.ca/10.1002/mar.20148
Park, C. W., Milberg, S., & Lawson, R. (1991). Evaluation of Brand Extensions: The Role of
Product Feature Similarity and Brand Concept Consistency. Journal of Consumer
Research, 18(2), 185–193. Retrieved from https://search-ebscohostcom.prxy.lib.unbc.ca/login.aspx?direct=true&db=bth&AN=4657871&site=ehostlive&scope=site
Punyatoya, P. (May 2013). Consumer evaluation of brand extension for global and local
brands: The Moderating role of product similarity. Journal of International Consumer
Marketing, 25 (3), 198–215.
Selvanayagam, J. E., & Ragel, V. R. (2015). Consumer Acceptability of Brand Extensions:
The Role of Brand Reputation and Perceived Similarity. IUP Journal of Brand
Management, 12(3), 18–29. https://search-ebscohost51
Internal

com.prxy.lib.unbc.ca/login.aspx?direct=true&db=bth&AN=111807910&site=ehostlive&scope=site
Taylor, K. H., Takeuchi, L., & Stevens, R. (2017). Mapping the daily media round: novel
methods for understanding families’ mobile technology use. Learning, Media and
Technology, 43(1), 70–84. doi: 10.1080/17439884.2017.1391286
Taylor, V. A., & Bearden, W. O. (2003). Ad spending on brand extensions: Does similarity
matter? Journal of Brand Management 11 (1), 63. Business Source Complete.
https://doi-org.prxy.lib.unbc.ca/10.1057/palgrave.bm.2540148
van Riel, A. C. R., Lemmink, J., & Ouwersloot, H. (2001) Consumer Evaluations of Service
Brand Extensions. Journal of Service Research, 3 (3), 220. Business Source
Complete. Web. https://doi-org.prxy.lib.unbc.ca/10.1177/109467050133003

52
Internal

Appendix 1 Survey and its measure reliabilities.
Survey
Many companies seek to capitalize on their current brand by introducing brand extensions,
which are new products introduced under an existing brand name. The purpose of this survey
is to understand consumer attitude toward brand extension of three well-known brands, i.e.,
Apple, Michelin, and Google.
For each of the three brands, you will be presented with a set of statements. Please read these
statements carefully, and indicate to which degree you agree with each of the statement using
the following scale. (Circle the number that corresponds with your degree of agreement.)
Scale
Strongly disagree
-5

-4

-3

Neutral
-2

-1

0

Strongly Agree
1

2

3

4

5

N/A

What’s your age? __________
Please indicate your gender. Male
Female
Other
What’s your household income? __________ I don’t want to answer

Brand: Apple (consumer electric brand)
In 2007, Apple first launched its iconic product, iPhone. More recently, Apple introduced an
extended product, iWatch, in 2015.
1

2
3
4
5
6
7

Statement
I’m familiar with Apple.
I began to know Apple because of
its iconic products, iPhone.
The Apple iconic product, iPhone,
has good credibility.
The Apple extended product,
iWatch, has good credibility.
The Apple iconic product, iPhone,
has superior quality.
The Apple extended product,
iWatch, has superior quality.
The Apple iconic product, iPhone,
strives to introduce innovations.

Scale
-5

-4

-3

-2

-1

0

1

2

3

4

5

N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

53
Internal

8

9

10

11

12

13
14

The Apple extended product,
iWatch, strives to introduce
innovations.
Overall, my evaluation of the
Apple extended product, iWatch,
is influenced by my attitude
toward the Apple iconic product,
iPhone.
The image of the Apple extended
product, iWatch, deviates from
the image of the Apple iconic
product, iPhone. (*)
The image of the Apple extended
product, iWatch, agrees with the
image of the Apple iconic
product, iPhone.
Even though iPhone and iWatch
are two different types of
products, they represent the same
brand image of Apple.
When I was watching Apple
iWatch advertisement, I can tell it
is a typical Apple advertisement.
The style of iWatch advertisement
is similar to that of iPhone
advertisement.

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

Have you used iWatch?

Yes

No

If you have used iWatch, please answer Questions 15;
If you haven’t used iWatch, please skip Question 15 and answer Question 16.

15

15.1) Before I used it, I evaluated
iWatch on the basis of the Apple
brand.
15.2) Before I used it, I evaluated
iWatch on the basis of the Apple
iconic product, iPhone.
15.3) Before I used it, I evaluated
iWatch on the basis of the iWatch
product features.

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

54
Internal

15.4) After I used it, I evaluated
iWatch on the basis of the Apple
brand.
15.5) After I used it, I evaluated
iWatch on the basis of the Apple
iconic product, iPhone.
15.6) After I used it, I evaluated
iWatch on the basis of the iWatch
product features.
15.7) After I used iWatch, my
experience with iWatch is
positive.

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

3

4

5 N/A

3

4

5 N/A

15.8) After I used iWatch, I feel
-5 -4 -3 -2 -1 0 1 2
positive about iPhone.
15.9) After I used iWatch, I feel
-5 -4 -3 -2 -1 0 1 2
positive about the Apple brand.
Answer Question 16 if you haven’t used iWatch.
I evaluate iWatch on the basis of
the Apple brand.

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

I evaluate iWatch on the basis of
the Apple iconic product, iPhone.

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

16

I evaluate iWatch on the basis of
the iWatch product features.

Brand: Google (technology brand)
Google is well known for its iconic product, Google Search — a search engine service. In
recent years, Google expanded its business into hardware products, and launched an extended
product, Google Smartphone.
1

2

Statement
I’m familiar with Google.
I began to know Google because of
its iconic products, Google Search.

Scale
-5

-4

-3

-2

-1

0

1

2

3

4

5

N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A
55

Internal

3
4
5
6
7
8

9

10

11

12

13

14

The Google iconic product, Google
Search, has good credibility.
The Google extended product,
Google Smartphone, has good
credibility.
The Google iconic product, Google
Search, has superior quality.
The Google extended product,
Google Smartphone, has superior
quality.
The Google iconic product, Google
Search, strives to introduce
innovations.
The Google extended product,
Google Smartphone, strives to
introduce innovations.
Overall, my evaluation of the
Google extended product, Google
Smartphone, is influenced by my
attitude toward the Google iconic
product, Google Search.
The image of the Google extended
product, Google Smartphone,
deviates from the image of the
Google iconic product, Google
Search. (*)
The image of Google extended
product, Google Smartphone,
agrees with the image of the Google
iconic product, Google Search.
Even though Google Search and
Google Smartphone are two
different types of products, they
represent the same brand image of
Google.
When I was watching Google
Smartphone advertisement, I can
tell it is typical Google
advertisement.
The style of Google Smartphone
advertisement is similar to that of
Google Search advertisement.

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

56
Internal

Have you used Google Smartphone?

Yes

No

If you have used Google Smartphone, please answer Questions 15;
If you haven’t used Google Smartphone, please skip Question 15 and answer
Question 16.
15.1) Before I used it, I evaluated
Google Smartphone on the basis of
-5 -4 -3 -2 -1 0 1 2 3 4 5 N/A
the Google brand.
15.2) Before I used it, I evaluated
Google Smartphone on the basis of
the Google iconic product, Google
Search.
15.3) Before I used it, I evaluated
Google Smartphone on the basis of
the Google Smartphone product
features.

15

15.4) After I used it, I evaluated
Google Smartphone on the basis of
the Google brand.
15.5) After I used it, I evaluated
Google Smartphone on the basis of
the Google iconic product, Google
Search.
15.6) After I used it, I evaluated
Google Smartphone on the basis of
the Google Smartphone product
features.
15.7) After I used Google
Smartphone, my experience with
Google Smartphone is positive.

16

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

15.8) After I used Google
Smartphone, I feel positive about
-5 -4 -3 -2 -1 0 1 2 3 4
Google Search.
15.9) After I used Google
Smartphone, I feel positive about
-5 -4 -3 -2 -1 0 1 2 3 4
the Google brand.
Answer Question 16 if you haven’t used Google Smartphone.

5 N/A
5 N/A

57
Internal

I evaluate Google Smartphone on
the basis of the Google brand.
I evaluate Google Smartphone on
the basis of Google iconic product,
Google Search.
I evaluate Google Smartphone on
the basis of the Google Smartphone
product features.

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

Brand: Michelin (manufacture brand)
Michelin is well known for its iconic product, Michelin Tire. It also produces an extended
product, Michelin Guide — a series of rating books that award Michelin stars for excellence
to a select few restaurants and hotels.
1

2
3
4
5
6
7
8

9

Statement
I’m familiar with Michelin.
I began to know Michelin because of
its iconic products, Michelin Tire.
The Michelin iconic product,
Michelin Tire, has good credibility.
The Michelin extended product,
Michelin Guide, has good
credibility.
The Michelin iconic product,
Michelin Tire, has superior quality.
The Michelin extended product,
Michelin Guide, has superior
quality.
The Michelin iconic product,
Michelin Tire, strives to introduce
innovations.
The Michelin extended product,
Michelin Guide, strives to introduce
innovations.
Overall, my evaluation of the
Michelin extended product, Michelin
Guide, is influenced by my attitude
toward the Michelin iconic product,
Michelin Tire.

Scale
-5

-4

-3

-2

-1

0

1

2

3

4

5

N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

58
Internal

10

11

12

13
14

The image of the Michelin extended
product, Michelin Guide, deviates
from the image of Michelin iconic
product, Michelin Tire. (*)
The image of the Michelin extended
product, Michelin Guide, agrees
with the image of the Michelin
iconic product, Michelin Tire.
Even though Michelin Tire and
Michelin Guide are two different
types of products, they represent the
same brand image of Michelin.
When I was watching Michelin
Guide advertisement, I can tell it is
typical Michelin advertisement.
The style of Michelin Guide
advertisement is similar to that of
Michelin Tire advertisement.

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

Have you used Michelin Guide?

Yes

No

If you have used Michelin Guide, please answer Questions 15;
If you haven’t used Michelin Guide, please skip Question 15 and answer Question 16.
15.1) Before I used it, I evaluated
Michelin Guide on the basis of the
Michelin brand.

15

15.2) Before I used it, I evaluated
Michelin Guide on the basis of the
Michelin iconic product, Michelin
Tire.
15.3) Before I used it, I evaluated
Michelin Guide on the basis of the
Michelin Guide product features.
15.4) After I used it, I evaluated
Michelin Guide on the basis of the
Michelin brand.
15.5) After I used it, I evaluated
Michelin Guide on the basis of the
Michelin iconic product, Michelin
Tire.

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

59
Internal

15.6) After I used it, I evaluated
Michelin Guide on the basis of the
Michelin Guide product features.
15.7) After I used Michelin Guide,
my experience with Michelin Guide
is positive.

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

4

5 N/A

4

5 N/A

15.8) After I used Michelin Guide, I
-5 -4 -3 -2 -1 0 1 2 3
feel positive about Michelin Tire.
15.9) After I used Michelin Guide, I
feel positive about the Michelin
-5 -4 -3 -2 -1 0 1 2 3
brand.
Answer Question 16 if you haven’t used Michelin Guide.
I evaluate Michelin Guide on the
basis of the Michelin brand.

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

-5

-4

-3

-2

-1

0

1

2

3

4

5 N/A

16
I evaluate Michelin Guide on the
basis of the Michelin iconic product,
Michelin Tire.

I evaluate Michelin Guide on the
basis of the Michelin Guide product
-5 -4 -3 -2 -1 0 1 2 3 4 5 N/A
features.
Note: * Asterisked items are reverse scored, so that the opposite is true (i.e. -5 = Strongly
agree and 5 = Strongly disagree)
Statement
Measures
Statement

1
Familiarity of brand
5-6

2
Iconic product relation
7-8

Measures

Quality

Innovativeness

Statement
Measures
Statement
Measures

(4+6+8) / 3, (11+12) / 2 (4+6+8) / 3, (13+14) / 2
Image-fit
Advertisement
15.7-15.8
Post-evaluation

3-4
Credibility
(4+6+8) / 3
Proxy of consumers’
attitude toward extension
product
15.2-15.6
User experience

60
Internal

Appendix 2 Variables for Correlation Analysis
Statements

H1a

H1b

H1c

H2

H3

H4

The Michelin iconic product,
Michelin Tire, has good
credibility.
The Michelin extended product,
Michelin Guide, has good
credibility.
The Michelin iconic product,
Michelin Tire, has superior
quality.
The Michelin extended product,
Michelin Guide, has superior
quality.
The Michelin iconic product,
Michelin Tire, strives to
introduce innovations.
The Michelin extended product,
Michelin Guide, strives to
introduce innovations.
The image of the Michelin
extended product, Michelin
Guide, agrees with the image of
the Michelin iconic product,
Michelin Tire.
Even though Michelin Tire and
Michelin Guide are two different
types of products, they represent
the same brand image of
Michelin.
Overall, my evaluation of the
Michelin extended product,
Michelin Guide, is influenced by
my attitude toward the Michelin
iconic product, Michelin Tire.
When I was watching Michelin
Guide advertisement, I can tell it
is typical Michelin
advertisement.
The style of Michelin Guide
advertisement is similar to that
of Michelin Tire advertisement.
Overall, my evaluation of the
Michelin extended product,
Michelin Guide, is influenced by
my attitude toward the Michelin
iconic product, Michelin Tire.
Before I used it, I evaluated
Michelin Guide on the basis of
the Michelin iconic product,
Michelin Tire.
After I used it, I evaluated
Michelin Guide on the basis of

Apple
iconApplCred

Variables
Google
iconggleCred

Michelin
iconmchlCred

extApplCred

extggleCred

extmchlCred

iconApplQlty

iconggleQlty

iconmchlQlty

extApplQlty

extggleQlty

extmchlQlty

iconApplInnvtn

iconggleInnvtn

iconmchlInnvtn

extApplInnvtn

extggleInnvtn

extmchlInnvtn

iconExtImageAgr
eeAppl

iconExtImageAg
reeggle

iconExtImageAgr
eemchl

sameImageiconex
tAppl

iconExtImageAg
reeggle

iconExtImageAgr
eemchl

ovaliconExtAppl

ovaliconExtggle

ovaliconExtmchl

extAdAppl

extAdggle

extAdmchl

extAdIconAdsml
appl

extAdIconAdsml
ggle

extAdIconAdsml
mchl

ovaliconExtAppl

ovaliconExtggle

ovaliconExtmchl

bfextEvliconappl

bfextEvliconggle

bfextEvliconmchl

afextWvlextappl

afextWvlextggle

afextWvlextmchl

61
Internal

H5

the Michelin Guide product
features.
After I used Michelin Guide, my
experience with Michelin Guide
is positive.
After I used Michelin Guide, I
feel positive about Michelin
Tire.

afextextappl

afextextggle

afextextmchl

afexticonappl

afexticonggle

afexticonmchl

62
Internal

Appendix 3 Descriptive Statistics and Correlation Results for H2 and H3 (Combination
Sample)
Descriptive Statistics
Mean
Std. Deviation
iconExtImageAgreeAppl
2.75
1.81
sameImageiconextAppl
3.79
1.68
ovaliconExtAppl
2.22
2.73
iconExtImageAgreeggle
1.59
2.52
sameImageiconextggle
2.15
2.56
ovaliconExtggle
1.55
3.04
iconExtImageAgreemchl
.53
2.86
sameImageiconextmchl
.86
3.15
ovaliconExtmchl
.86
3.30
extAdAppl
3.74
1.58
extAdIconAdsmlappl
3.52
1.64
ovaliconExtAppl
2.32
2.64
extAdggle
2.02
2.29
extAdIconAdsmlggle
1.15
2.35
ovaliconExtggle
1.60
2.86
extAdmchl
.81
2.72
extAdIconAdsmlmchl
.64
2.77
ovaliconExtmchl
1.26
3.27

N
68
68
68
66
66
66
58
58
58
62
62
62
55
55
55
47
47
47

H2 of Apple:

63
Internal

H2 of Google:

H2 of Michelin:

H3 of Apple:

64
Internal

H3 of Google:

H3 of Michelin:

65
Internal

66
Internal