1. Consumer Evaluations of Brand Extentions B2B to B2C

�

�

�

�

�

�

Consumer Evaluations of

Brand Extensions:

Can B2B Brands be Extended

into the Consum er M arket?

�

�

Maastricht U niversity

Faculty of Economics and Business Administration

September 2004

Master Thesis International Management Studies Author: Leon Phang ID number: 982261 Supervisor: Dr. H ans Ouwersloot

Can B2B Brands be Extended into the Consumer Market?

2

“Branding adds spirit and a soul to what would

otherwise be a robotic, automated, generic

price-value proposition. If branding is ultimately

about the creation of human meaning, it follows

logically that it is the humans who must ultimately

provide it.” – David Aaker


3

ABSTRACT

Brand extensions allow companies to leverage the equity in established brands, and

thereby reducing risk associated with launching new products. A plethora of brand

extension studies have been done in recent years. H owever, there is a paucity of

research investigating business-to-business brand extensions. This thesis examines

whether business-to-business brands can leverage their brands in the consumer market

through brand extensions. A new model is developed by combining Aaker and Keller’s

brand extension model with theories from business-to-business branding as well as

other consumer branding concepts, and tested quantitatively to understand how

consumer evaluate brand extensions. The results of the present study show support for

this new model. More specifically, the results indicate that in the context of business-to-

business brand extensions, consumers use the transferability of skills and resources, and

brand concept consistency with the parent brand category as major cues to evaluate

extensions. Innovativeness and corporate social responsibility are also relevant cues. As

a consequence of these findings, branding strategies that stretch business-to-business

brands into the domain of consumer markets can be successful in cases where

consumers perceive a fit with respect to skills and resources, and brand concept, and

when the parent brand is perceived as being innovative and socially responsible.


4

CON TEN TS

Abstract 3

List of tables 8

List of figures 8

CH APTER 1. IN TRODU CTION 9

1.1 Background: Leveraging the Brand Across Sectors 9

1.2 Problem Statement 11

1.3 Contributions of the Study 12

1.3.1 Theoretical contribution 12

1.3.2 Practical contribution 13

1.4 Thesis Outline 13

PART I. LITERATURE REVIEW

CH APTER 2. CON SU M ER BRAN D EXTEN SION S 15

2.1 Introduction to the Chapter 15

2.2 Consumer Brand Extensions 15

2.2.1 Brand Extension D efinition 15

2.2.2 Benefits 15

2.2.3 D raw backs 17

2.3 Concepts for Evaluating Brand Extensions 17

2.3.1 Extension reaction 17

2.3.2 Spillover and substitution effects 18

2.3.3 Categorical and piecem eal evaluation processes 18

2.3.4 Brand-specific associations 19

2.4 Summary 20

CH APTER 3. BU SIN ESS-TO-BU SIN ESS BRAN DIN G 21


3.2 Corporate Branding Decisions 21

3.2.1 Corporate strategy 21

3.2.2 Corporate brand identity 22


5

3.2.3 Corporate brand strategy 23

3.2.4 Corporate brand architecture 24

3.3 Corporate Brand Extensions 25

3.3.1 Innovativeness 25

3.3.2 Corporate Social Responsibility (CSR) 26

3.3.3 Environm ental concern 26

3.4 Summary 27

PART II. CONCEPTUAL FRAMEW ORK

CH APTER 4. H YPOTH ESES 29


4.2 Knowledge About the Parent Brand 29

4.3 Attitude Towards the Parent Brand 29

4.4 Innovation 30

4.5 Corporate Social Responsibility 30

4.6 Environmental Concern 31

4.7 Transfer 31

4.8 Brand Associations 32

4.9 Perceived Difficulty 32

4.10 Moderating Factors 33

4.11 Summary 35

CH APTER 5. RESEARCH DESIG N 36


5.2 The H ypothesized Model 36

5.2.1 The original m odel of Aaker and Keller 36

5.2.2 The current model 37

5.3 Methodology 39

5.4 Questionnaire Development 39

5.4.1 Brands and brand extensions 39

5.4.2 Scaling and m ulti-item scales 40

5.5 Data Collection 42

5.5.1 Pilot-testing 42

5.5.2 Sam pling and response 43


6

PART III. RESEARCH FINDIN G S

CH APTER 6. AN ALYSIS, RESU LTS AN D IN TERPRETATION S 46


6.2 Reliability, Collinearity and Regression Analyses 46

6.2.1 Reliability assessm ent 46

6.2.2 M ulticollinearity 47

6.2.3 Regression analysis results 51

6.2 H ypothesis Testing 56

6.2.1 H1: Parent brand know ledge 56

6.2.2 H2: Parent brand quality 56

6.2.3 H3: Innovative 56

6.2.4 H4: Corporate Social Responsibility 56

6.2.5 H5: Environm ental concern 57

6.2.6 H6: Transfer 57

6.2.7 H7: Brand concept consistency 57

6.2.8 H8: D ifficult 57

6.2.9 H10: Interaction effect betw een Transfer and Brand concept consistency 58

6.3 Discussion and Interpretation 61

6.3.1 Parent brand know ledge 61

6.3.2 Parent brand quality 61

6.3.3 Innovative 61

6.3.4 Corporate Social Responsibility 62

6.3.5 Environm ental concern 62

6.3.6 Transfer 63

6.3.7 Brand concept consistency 63

6.3.8 D ifficult 64

6.3.9 Interaction effects 64

6.3.10 General discussion of the current m odel 64


7

PART IV. CLOSIN G PERSPECTIVES

7. CON CLU SION S 67

7.1 Summary of the Findings 67

7.2 Conclusions 67

7.3 Limitations and Directions for Future Research 67

7.4 Implications 69

7.4.1 Theoretical implications 69

7.4.2 Managerial implications 69

BIBLIOGRAPHY 71

APPENDIX A. QU ESTION NAIRES 78

APPENDIX B. SPSS OU TPUTS 103


8

LIST OF TABLES

Table 1.1 Companies Formerly Operating in Business-to-Business Markets Only 11

Table 1.2 Expansion to Consumer Markets 11

Table 4.1 H ypotheses and Their References to Literature 35

Table 5.1 Overview of B2B Brands and B2C Extensions 40

Table 5.2 Multiple-scale Items 42

Table 5.3 Response Profile: Sampling Type 43

Table 5.4 Response Profile: G ender 43

Table 5.5 Response Profile: N ationality 43

Table 6.1 Cronbach’s Alpha Coefficients for Multiple-Item Constructs 47

Table 6.2 VIF Scores of Regression Variables Before and After Residual Centering 49

Table 6.3 VIF Scores of Regression Variables After Omitting Interaction Term 50

Table 6.4 Comparison Effects of Original and Replication Studies 53

Table 6.5 Regression Results: Main and Interaction Effects 54

Table 6.6 Full Model at Brand Level 55

Table 6.7 Summary of H ypothesis Testing 59

Table 6.8 Extension Level Means 60

Table 6.9 Perceived Quality of Parent Brands 60

LIST OF FIG U RES

Figure 3.1 An Integrated Approach to Communications 23

Figure 5.1 H ypothesized Model 38


9

CH APTER 1.

IN TRODU CTION

1.1 Background: Leveraging the Brand Across Sectors

The changing market dynamics and heightened competition of the global economy has

amplified the role of brands to an unsurpassed level. Brand marketers seek ways to

achieve growth while reducing both the cost of new product introductions as well as the

risk of new product failure. A popular way of launching new products has therefore been

to leverage the equity of an existing brand into a new sector, market, or product

category – a so-called brand extension.

A brand extension strategy can be beneficial because it reduces the new product

introduction costs and also increases the chance of success (Kapferer, 1994). The

rationale behind brand extensions is simple: when a strong brand has been established,

the brand has moved beyond the functional product into a realm of values. It makes

economical sense to try to deliver the same emotional benefits in a different market

(Mortimer, 2003). Since awareness of a certain brand already exists, costs of launching a

new product will, ceteris paribus, be lower than in the absence of a strong brand. The

main objective of brand extensions is hence to leverage the intangible qualities of a

brand since the functional benefits can generally be imitated (U rde, 1999).

Since brand extensions imply launching new products, a key issue is to what

extent these extensions are successful. A theoretical starting point to this discussion is

the concept of brand orientation, which can be defined as “an approach in which the

processes of the organization revolve around the creation, development, and protection

of brand identity in an ongoing interaction with target customers with the aim of

achieving lasting competitive advantages in the form of brands” (ibid., p. 117-118).

Brand orientation is an additional degree of sophistication to market orientation. In

other words, what is demanded by customers at any given moment in time is not

necessarily the same as what will reinforce the brand as a strategic resource. Market

orientation can therefore be a source of conflict to a brand’s long-term development

when achieving competitive advantage is the aim. Ideally, a firm is both market-oriented


10

and brand-oriented. Keller (2003) argues along the same line concerning brand

extensions –it is not a question of whether a brand should be extended, but rather

where, when, and how it should be extended. Simply put: extend the brand –if it is

possible. Reality shows that companies do this to a large extent: in the 1990s, 81 percent

of new products were brand extensions (Mortimer, 2003), for previously reasons

mentioned. This is not to say that brand extensions are risk-free –it is crucial to know

where the ‘boundaries’ of the brand are. Understanding these limits is not a simple

matter, however. As an example, the deodorant brand Lynx made an unsuccessful

attempt to stretch itself into hair care products. On the other hand, U nilever’s razor

brand G illette successfully stretched into after shaves and deodorants. Thus, even if the

product category of the extension is intuitively related to the product category of the

parent brand, there can still be a lack of fit. On the other hand, brand extensions do not

necessarily have to stick to their parent category. The department store chain Marks &

Spencer launched financial services, although it was a totally different area than

retailing. N evertheless, it worked well, because its customers associated the parent

brand and the financial services with trust.

Virtually all discussions of branding are structured in a consumer marketing

context (Aaker, 1996). That is not to say that industrial or business-to-business branding

is not as important and valuable as consumer branding. Some of the world’s most

powerful brands are B2B brands: ABB, Caterpillar, Cisco, Dupont, FedEx, G E, H ewlett

Packard, Intel, and Siemens1 (W ebster & Keller, 2004). An interesting question then is:

what if the business-to-business brand wants to make a stretch into the consumer

market? This may seem confusing, but many now-famous consumer brands have once

been business-to-business brands and now serving both sectors (see tables 1.1 and 1.2).

The global mobile phone brand N okia started out in forestry (B2B) in 1865, and then

began selling rubber boots in the 1960s, and it was not until the 1980s when it started

making mobile phones, for which it is now famous. Other examples include Philips,

Mitsubishi, Microsoft and IBM, three companies that through new innovations began

selling new products to consumers. This historical perspective highlights one fact:

brands are in constant flux. A stretch from B2B to the consumer market is perhaps not

that far-fetched or uncommon.

1 N ote that some of these brands are promoted to end-customers, giving them some characteristics of consumer brands. Yet, they can be considered as industrial brands with respect to the development of purchase specifications, the buying process and the actual purchase (which usually made by a B2B buyer) (W ebster & Keller, 2004).


11

TABLE 1.1 Companies Formerly Operating in Business-to-Business Markets Only

Com pany nam e Country Founded Initial business at year 1

Microsoft U S 1975 BASIC computer programming language

IBM US 1911* Tabulating machines

Morgan Stanley U S 1935 Investment banking

Merrill Lynch U S 1907 Stocks and bonds brokerage

Philips N etherlands 1891 Carbon-filament lamps

Vattenfall Sweden 1909 H ydro-electric powerplants

Mitsubishi Japan 1870 Shipping

N okia

Finland 1865 Forestry

Source: Com pany w ebsites

TABLE 1.2 Expansion to Consumer Markets

Com pany nam e Year N ew offering B2B Offering at tim e

Driver for strategic expansion to B2C m arket

Microsoft 1981 MS-DOS BASIC computer programming

language

Innovation

IBM 1981 IBM 5150 Personal computer

Mainframes, industrial computing systems

Innovation

Morgan Stanley 1977 Private wealth management

Full-service banking Diversification and growth

Merrill Lynch 1977 Cash Management Account®

Retail brokerage, institutional- and investment banking

Diversification and growth

Philips 1927 Radio Lamps, medical X-ray tubes Innovation

Vattenfall 1996 Electricity, heating, telephony

Public utility (domestic electricity) Deregulation

Mitsubishi 1917 Automobiles H eavy industries Diversification and decentralization

N okia 1960s Rubber boots Industrial rubber and cables, radio technology

Diversification

Source: Com pany w ebsites

1.2 Problem Statem ent

In order to determine whether a brand extension is able to capitalize on its parent

brand whilst avoiding or minimizing potential disadvantages, it is crucial to understand

how the extensions are evaluated by consumers. The success of a brand extension is

largely determined by how customers evaluate the extension (Klink & Smith, 2001).

N umerous academic studies on brand extensions have been made. A landmark study in


12

this area was made by Aaker and Keller in 1990. In the current study, the issue of brand

extension evaluation will be investigated in a different context, namely that discussed in

the previous section. In other words, this paper tries to answer the problem statement:

Can business-to-business brands be extended into the consum er m arket?

The purpose of this research is also to: (1) determine whether a broad replication of

Aaker and Keller’s (1990) brand extension model is feasible with respect to the current

context, (2) link theory from consumer branding to corporate branding; (3) examine

whether concepts of evaluating brand extensions can be successfully combined to form

an effective model for predicting extension acceptance in the research context; and (4)

determine the relative importance of these concepts.

1.3 Contributions of the Study

1.3.1 Theoretical contribution

Aaker and Keller’s (1990) framework has been subject to substantial scrutiny as a

widely replicated study (Bottomley & H olden, 2001), and has also been confirmed in

marketing textbooks. W hile the series of so-called close replications confirm the

generalizability of Aaker and Keller’s (1990) predictions (Bottomley & H olden, 2001),

Barwise (1995) argues that a model which does not stretch beyond Aaker and Keller’s

(1990) choice of brand extensions is of limited scope and therefore of negligible value.

Replications are seen as crucial for empirical generalization (Leone & Schultz, 1980;

Barwise, 1995) and knowledge development (H ubbard et al. 1994; Bottomley & H olden,

2001). Furthermore, it can be argued that those academic research publications that are

peer-reviewed are worthy of replication (Bottomley & Doyle, 1996). Barwise (1995)

contends that a high-quality empirical generalization should be “characterized by its

scope, precision, usefulness and link with the theory”, and usefully defines a domain

which excludes most marketing practice, teaching and exercise.

The current study is a broad replication of Aaker and Keller’s (1990) model in the

sense that it does not generalize across brand extensions2, but does attempt to make a

generalization of the model by expanding the scope of which it has previously studied.

Scope can be defined as the domain (e.g. sectors, countries, situations) of which an

2 That is, it does not use the core brands of Aaker and Keller’s (1990) study.


13

empirical generalization holds. Concerning the scope, this study will examine whether

Aaker and Keller’s (1990) predictions are extra-sectoral, that is, whether parent brand

equity built up in one sector can be leveraged through brand extensions in another

sector.

Besides empirical replication, an additional contribution of the present study is

model development. Bottomley and Doyle (1996) point out “brand concept consistency”

as a better facilitator of brand extension acceptance than product-related similarity

(used in the A&K model). The present study will therefore integrate “brand concept

consistency” with Aaker and Keller’s (1990) to confirm Bottomley and Doyle’s (1996)

proposal.

A last contribution of the present study is to add to the relatively small but

growing supply of business-to-business/industrial branding theory.

1.3.2 Practical contribution

It may be of particular interest for managers to know whether their B2B parent

brands can be extended into the consumer market, especially when a strategic

opportunity arises, and consumer brand value can be created and delivered. In such an

event, the findings of the present study could tell managers whether brand extension

would be accepted by consumers. The findings of the study would also pinpoint which

of the brand extension factors they should focus on if an extension strategy is pursued.

1.4 Thesis Outline

The thesis is structured as follows. The first part is a literature review on brand

extension concepts and theories from both a consumer and business-to-business

perspective. This sets the stage for the second part, the conceptual framework. In this

part, hypotheses that are built on the theories of the preceding chapters are presented.

This is followed by the research design of the current study. The third part contains the

research findings, where the empirical data is analyzed, hypotheses tested and

interpreted. The final parts of the thesis are the closing perspectives, where a summary

and conclusion of the findings are given. Implications of the study, both theoretical and

managerial are mentioned, as well as limitations and directions for future research.


14

PART I.

LITERATU RE REV IEW

Chapter 2. Consumer Brand Extensions 15

Chapter 3. Business-to-Business Branding 21


15

CH APTER 2.

CON SU M ER BRAN D EXTEN SION S

2.1 Introduction to the Chapter

The discussion on brand extensions begins with a consideration of a firm’s sources of

growth. As a useful perspective, Keller (2003) proposes the so-called Ansoff’s

product/market expansion grid, which categorizes various growth strategies “according

to whether they involve existing or new products and whether they target existing or

new customers or markets (ibid., p. 577)”. An important point in the context of growth

strategies is that new-product introductions are crucial for a firm to sustain its long-term

competitive advantage (Keller, 2003). N ew-product introductions can, in turn, be

executed in three general ways: (1) as a new brand that is individually designed for the

new product; (2) as an application of an existing brand; and (3) as a combination of a

new brand with an existing brand. H ence, brand extension is a managerial topic.

N evertheless, for the purposes of the current study, the literature review in this chapter

will be of more theoretical nature, focusing on potential advantages and disadvantages,

as well as different theoretical concepts for how consumers evaluate brand extensions.

2.2 Consum er Brand Extensions

2.2.1 Brand Extension D efinition

According to Keller (2003, p. 577), a brand extension is defined as “when a firm

uses an established brand name to introduce a new product”. This is analogous to

approach (2) and (3). Brand extensions are made on an ad hoc basis or according to a

strategy to create a range brand (Aaker, 1996).

2.2.2 Benefits

The benefits of brand extensions is a much-discussed subject. Keller (2003)

distinguishes two kinds of benefits: (1) benefits that relate to the acceptance of the

brand extension, and (2) benefits that relate to the parent brand image. Kapferer (1997)


16

also makes a distinction between brand extensions and their benefits from an

operational point of view, and proposes that brand extensions that are intended to

boost sales should be distinguished from new products that carry brand image and exist

to fuel the brand.

2.2.2.1 Benefits related to brand extension acceptance

Brand extensions allow consumers to draw conclusions and form expectations

about the potential performance of a new product (i.e. the brand extension) based on

their existing knowledge about the brand (Keller, 2003). Provided that a strong brand

name is present, the perceived risk by consumers is substantially reduced when

familiarity and knowledge about the parent brand is present (Keller, 2003; Aaker & Keller,

1990). Benefits of introducing new products also include different ways of achieving

operational efficiencies. A favorable parent brand reduces costs associated with gaining

distribution since retailers are more positive to stock and promote a brand extension

(ibid.). Another benefit relates to marketing communications: since brand awareness

already exists, promotional activities (including introductory and follow-up advertising

and other marketing programs) of a brand extension can be less intensive and thus less

costly than those of a totally new brand and product (Keller, 2003; Kapferer, 1997). Other

efficiencies includes avoiding costly development of brand names, logos, symbols,

packages, characters, slogans, etc. (Keller, 2003).

2.2.2.2 Benefits relating to the parent brand im age

Brand extensions also have positive spillover effects on the parent brand. Firstly,

extensions can clarify the brand m eaning to consumers and define the boundaries of the

domain in which it competes (Keller, 2003). Second, by improving the favorability of an

existing brand association, adding a new brand association, or a combination of these, a

brand extension can enhance the parent brand image (ibid.). Consistent with this view

are the findings of Morrin (1999), which propose that consumer exposure to brand

extensions will increase parent brand awareness in terms of recognition and recall.

Similarly, Balachander and G hose (2003) find evidence of beneficial spillover effects of

advertising of a child brand, for example a brand extension, on choice of a parent brand.

A third benefit involves brand revitalization—a new or rejuvenated product can be a


17

mean to renew interest and improve attitude towards the parent brand (Keller, 2003;

Kapferer, 1997).

2.2.3 D raw backs

Keller (2003) mentions several drawbacks of brand extensions. First, the image of

the parent brand can be hurt irrespective of the success or failure of the extension. This

happens when the attributes of the extension are seen as inconsistent or conflicting

with the corresponding attributes of the parent brand. Second, brand extensions may

obscure the identification of the brand with its original categories, reducing brand

awareness (Morrin, 1999) and/or diluting the brand meaning. Third, brand extensions

can lead to problems of practical nature, for example a large number of extensions

might confuse or frustrate customers, and there might be problems with retailers being

unwilling to shelf/store all the different extensions. Similarly, Loken and John (1993, p.

79) suggest that “unsuccessful brand extensions can dilute brand names by diminishing

the favorable attitudes that consumers have learned to associate with the family brand

name”.

2.3 Concepts for Evaluating Brand Extensions

2.3.1 Extension reaction

A principle study in the field of brand extensions is Aaker and Keller’s (1990) study

on how consumers evaluate brand extensions. The authors hypothesize that

“evaluations of brand extensions are based on the quality of the original brand, the fit

between the parent and extension categories and the interaction between the two”

(Bottomley & H olden, 2001, p. 494). Despite the fact that this study per se provides no

evidence that a direct relationship between the quality of the parent brand and the

consumer evaluation of the brand extension exist (Aaker & Keller, 1990; ibid.), the

empirical generalizability of Aaker and Keller’s (1990) model is well supported in

Bottomley and H olden’s (2001) secondary analysis, which examines seven replication

studies. Based on their findings, Bottomley and H olden (2001) draw three general

conclusions: (1) The quality of the parent brand and the fit between the parent brand

and the brand extension are key determinants of consumer evaluations of brand

extensions; (2) Consumer’s brand extension evaluations are also determined by (a) the

dimensions of fit (i.e. the complementarity and transferability of assets and skills)


18

between the parent brand and the brand extension, and (b) to what extent consumers

perceive the brand extension is difficult to produce; (3) Cultural differences influence

how brand extensions are evaluated with respect to relative measurement factors.

2.3.2 Spillover and substitution effects

W hile Aaker and Keller (1990) and consequent replication studies provide a

rationale for leveraging parent brand equity through brand extensions, from which

economic profits can be extruded, Balachander and G hose (2003) examine the reciprocal

effect of brand extensions on the parent brand. This effect or productivity is measured

by “brand-choice elasticities”, which measure the increase in choice probability that

results from increase in exposure (ibid., p. 11). The findings of Balachander and G hose

(2003) provide strong support to positive spillover effects from advertising of a brand

extension on choice of a parent brand. This reciprocal spillover effect does, however, not

seem to be symmetrical—that is, forward spillover effects from advertising of a parent

brand on choice of a brand extension are limited.

2.3.3 Categorical and piecem eal evaluation processes

To understand how consumers evaluate new brand extensions, categorization

theory is a useful concept. It aims at identifying the processes by which consumers form

categories, and assigns certain objects to one category rather than another (Kapferer,

1997). Mervis and Rosch (1981, p. 89) propose that “a category exists whenever two or

more distinguishable objects are treated equivalently”. W hen a new brand extension is

launched, a set of attributes or beliefs in addition to the already existing family or parent

brand image is introduced (Loken & John, 1993). If these attributes or beliefs are

consistent with the parent brand image, an extension is considered to be acceptable

(Kapferer, 1997) or perceived to “fit” the category (Boush & Loken, 1991).

Another concept for attitude formation towards brand extensions is by so-called

“piecemeal”, “analytical” or “computational” processing (Fiske, 1982; Cohen, 1982;

Brooks, 1978), where attitude is “computed” from specific brand extension attributes.

This type of model does not aim to describe conscious evaluation processes (Boush &

Loken, 1991).

Categorical and computational evaluation processes are not mutually exclusive in

any given affective reaction. Fiske and Pavelchak (1986) propose a two-step process of


19

evaluation. In the first step, the consumer attempts to match a brand extension (or some

other new object) with the current category. If categorization is successful, in other

words, if there is a match, the affect that is associated with the category type is applied

to the brand extension and so the evaluation process is complete. If there on the other

hand is a poor match between the category and the brand extension, piecemeal

processes are initiated. Affect is then evaluated through a weighted combination of

attributes.

Even if inconsistency implies that the extension is not “integrated” in the parent

category, an inconsistent brand extension can have a negative impact on the parent

brand by “diluting” specific attribute beliefs that consumers have come to hold about an

established brand name, rather than “diluting” the global affect associated with the

established brand name (Loken & John, 1993). The negative impact of an inconsistent

extension depends on the typicality of the brand attribute at stake. H ence, brand

dilution is an important issue when launching new brand or category extensions.

2.3.4 Brand-specific associations

A brand-specific association is defined as an attribute or benefit that differentiates

a brand from competing brands (MacInnis & N akamoto, 1990). This means that a brand

can be associated with a salient attribute, but this association is per se not strongly

associated with competing brands or the product class as a whole (Broniarczyk & Alba,

1994).

Since the brand association varies depending on the benefits that are sought

within a particular product category, a consumer’s evaluation of a brand extension need

not correspond to evaluation of that brand in its original category (ibid.). Three

conclusions can be drawn from Broniarczyk and Alba’s (1994) research: (1) A perceived

lack of fit between the product category of the parent brand and the proposed

extension category can be overcome if key parent brand associations are salient and

relevant in the extension category; (2) brand-specific associations allow for brand

extensions to unrelated product categories. Brand-specific associations moderate the

role of product category similarity in brand extension judgments; a brand extension is

more preferred in an unrelated category that valued its association than in a similar

category that does not value its associations; and (3) the boundaries for the

appropriateness of a certain brand extension were determined by knowledge about the

incumbent brand.


20

2.4 Sum m ary

A brand can be leveraged by means of brand extensions. Two types of benefits

arise from an extension strategy: (1) an extension is easier accepted among consumers if

the parent brand is known; and (2) an extension can increase parent brand awareness

and positive attitudes. The success or failure of an extension strategy depends highly on

how consumers evaluate the extension. In general, extension acceptance is higher when

there is a perceived “fit” between the parent brand and the extension. This fit can be

based on associations of the extension product category of the, or on the consistency of

the extension with the parent “brand concept”. This, in turn, depends on how the

evaluation process of the consumer is structured.


21

CH APTER 3.

BU SIN ESS-TO-BU SIN ESS BRAN DIN G


W hile consumer branding tends to focus much on marketing areas such as consumer

behavior where emotions, perceptions and behaviors are central, B2B branding is less

concerned with the buyer-seller interface and more focused on functionality and

performance. Brand elements (such as brand names) are associated with offerings that

consistently deliver superior functionality and are hence valuable resources (Anderson &

N arus, 2004). B2B branding and brand building are concepts that are gaining in

popularity.

This chapter begins with a brief introduction to corporate strategy, which provides

the starting point to branding topics such as corporate brand strategy, brand identity

and brand architecture. Following this is a review of theory on corporate brand

extensions.

3.2 Corporate Branding Decisions

3.2.1 Corporate strategy

Ind (1997) proposes that successful corporate branding is derived from Porter’s

well-established analytical model of competitive strategy. In this model, corporate

strategy is defined as the search for a favorable competitive position in an industry in

order to establish a profitable and sustainable position against competitive forces

(Porter, 1998). The core of the Porter model is that strategy is defined according to (1)

competitive scope, and (2) competitive advantage, the main point being that firms must

choose between being different and being the lowest cost producer in a certain industry

(Ind, 1997).

An important concept of the Porter (1998) model is the value chain. Porter (ibid.)

argues that a firm’s competitive advantage cannot be understood by looking at a firm as

a whole. Instead, a firm should be disaggregated into activities –that are performed to


22

design, produce, market, deliver and support its offering (e.g. logistics, operations,

marketing and sales, and service). These activities represent the value chain, and if

deconstructed and analyzed properly can constitute a source of competitive advantage.

To link the Porter model of corporate business strategy to corporate branding

strategy, it is useful to first discuss the concept of corporate brand image.

3.2.2 Corporate brand identity

Brand identity is, in general, a set of brand associations that the brand strategist

aspires to create or maintain (Aaker & Joachimsthaler, 2000). Ind (1992, p. 19) states, with

respect to corporate brand identity, that:

“Identity is formed by an organization’s history, its beliefs and philosophy, the nature of its

technology, its ownership, its people, the personality of its leaders, its ethical and cultural values

and its strategies.”

In addition, Keller and Aaker (1997) argue that corporate brands may be more likely to

possess intangible attributes or organizational characteristics that span product classes,

compared to product brands, whose associations are more likely to be product specific.

Anderson and N arus (2004), on the contrary, state that corporate brand elements

may not enclose any intangible meaning –instead they are charged with meaning

through the performance of the supplier and its market offering over time. The value of

the brand, i.e. the brand equity, hence lies in the association that business customers or

suppliers have with offerings that consistently deliver superior functionality and

performance (Anderson & N arus, 2004).

The issue of tangibility versus intangibility is clarified by Mudambi (2002). H er

findings conclude that the importance of intangible or tangible industrial brand

attributes, respectively, is situation- and buyer-dependent. In addition, Mudambi (2002)

proposes that B2B branding in general is more important than commonly believed.

Michell et al. (2001) conclude that tenets of consumer brand equity theory such as

perceived quality, a recognizable image, market leadership, and a differentiated position

in the marketplace are also relevant among industrial buyers. Factors that are associated

with building brand loyalty in consumer markets (i.e. quality, reliability, and

performance) are also relevant in B2B markets.


23

3.2.3 Corporate brand strategy

The choice of business-to-business brand strategy is derived from the fact hat

industrial marketing and buying is increasingly focused on relationships rather than

individual transactions (W ebster & Keller, 2004). Industrial customers want ongoing

relationships with reliable suppliers of quality products and services. In a business-to-

business context, a brand is a relationship between buyer and seller.

Abratt (1989) presents a model (figure 3.1) which provides with a structure for

looking at a corporate brand. The underlying thought behind the model is to view

corporate brand management in a holistic point way: external and internal marketing

communications are interrelated. In other words, there must be a “fit” between the

corporate brand identity, the employees’ view of identity (the internal environment), the

marketing communication, and stakeholders (the external environment).

Corporate mission

Corporate philosophy

Core values

Corporate culture

FIG U RE 3.1 An Integrated Approach to Communications

Identity

Marketing communications

strategy

Employees’ view of identity

Products and services

Customers

G overnment

Local communities

Financial

Suppliers

Buyers

Media

Influential groups

Im age

IMA

GE

INT

ER

FA

CE

AU

DIE

NC

ES

M arketing com m unications

Personal communications

Internal marketing

Source: Abratt (1989)


24

An important aspect of this model are the so-called ‘feedback-loops’, which

suggest that the “marketing communications strategy, although based on the reality of

the identity, is a dynamic force” (Ind, 1997, p. 51). This implies that although the

corporate image and employees’ view of the corporate brand identity are transmitted

from the corporate identity via marketing communications, there are also backflows of

information (‘feedback’) from customers and other audiences to employees and the

marketing communications. The corporate identity and image are hence in a constant

flux and adjustment, in order to achieve a sustainable advantageous position for the

organization.

Consistent with Abratt’s (1989) idea of feedback loops is G riffin’s (2002) view that

all corporate branding strategy decisions stem both external and internal factors.

External factors are, theoretically, any factors that affect a corporate brand. Brand

strategy decisions try to reduce the uncertainty that is caused by external factors – brand

strategy is ‘outside-in’. Internal factors are the corporate values and culture, as well as

business exposure3. The more exposure a company has, the greater is the need for

consistent communications to stakeholders. H ence, brand strategy is also ‘inside-out’.

3.2.4 Corporate brand architecture

Brand architecture is defined by Rajagopal and Sanchez (2004, p. 236) as “the

organizing structure of the brand portfolio that specifies brand roles and the nature of

relationships between brands and different product-market brand contexts”. Aaker and

Joachimsthaler (2000) describe it as the medium by which the brand team functions as a

unit to create synergy, clarity, and leverage. A brand portfolio includes all the brands and

sub-brands attached to product-market offerings, including co-brands with other firms

(ibid.).

Very little academic research has been conducted with respect to brand

architecture, and there is to date no research in the context of business-to-business

branding. W ebster and Keller (2004) do however suggest that through observation,

industrial brands often use branding strategies that organize their brand portfolio under

a company “blanket brand” consistent with Aaker and Joachimsthaler’s (2000)

definitions of sub-brands and branded house strategies. Sub-brands are brands

connected to a master or parent brand and augment or modify the associations of that

master brand. H ence, the parent brand is the primary frame of reference. In a branded

3 The extent to which management and employees are ‘visible’.


25

house strategy, the role of the parent brand is to be a dominant driver across multiple

offerings. The difference between a sub-brand strategy and branded house strategy is

that in the later the role of the parent brand is more prominent. As Aaker and

Joachimsthaler (2000) describe it, it is like “putting a lot of eggs in one basket”.

3.3 Corporate Brand Extensions

Keller and Aaker (1997) examined how various types of corporate marketing

activities (communication activities portraying a firm as innovative, environmentally

concerned, and involved with the community) would influence corporate credibility (i.e.

perceived expertise, trustworthiness, and likeability) and thus have a positive effect on

brand extension evaluation. In their study, four hypothetical corporate brand extensions

outside the current brand offering were presented alongside corporate descriptions that

emphasized one of the following three types of attributes: (1) A firm’s reputation of

being innovative and philosophy of launching technologically advanced products; (2) a

firm’s policy to offer “environm entally friendly” products and to manufacture products in

an environmentally safe fashion; and (3) a firm's philosophy to im prove the quality of life

in local com m unities through various activities and programs.

The findings of Keller and Aaker (1997) suggest that corporate marketing efforts

can be beneficial as it improves perceptions and evaluations of a corporate brand

extension. Creating a positive corporate image and executing a corporate brand

strategy can thus facilitate new product acceptance. The three types of brand attributes

mentioned above can be categorized into innovativeness, corporate social responsibility

and environm ental concern.

3.3.1 Innovativeness

An innovative brand image involves being perceived as being modern and up-to-

date, investing in research and development, utilizing state-of-the-art manufacturing

technologies, and introducing the latest product features (Keller, 2003). Though this

suggests that positive brand attributes that signal innovativeness are important,

surprisingly little research has been done in this field. Indeed, most research in the topic

of innovativeness in marketing has been in the area of consumer innovativeness and the

innovation diffusion4 (Roerich, 2004). In this line of research it is argued that different

4 For example, the rate of adopting new technologies.


26

individuals have different predispositions to adopt or buy new products (ibid; Ostlund,

1974). An important point to make is that although innovative brand attributes is

favorable for building brand equity, the extent to which this is successful also depends

on how “innovative” the target audience is.

Marketing activities that emphasized innovation have a significant impact on

corporate brand extension evaluation as it leads to the favorable perceptions of

corporate expertise and to presumptions that the corporate brand extension will also be

innovative (Keller & Aaker, 1997). Emphasis on innovation is the only type of marketing

activity that enhances the customer’s perceived “fit” of the brand extension to the

parent brand, as well as evaluations of specific product attributes. H ence, marketing

efforts to emphasize innovation significantly increases both perceived quality and

purchase likelihood score for the brand extension.

3.3.2 Corporate Social Responsibility (CSR)

Much has been written about corporate social responsibility (CSR) in recent years.

Kitchen (2003) argues that companies undeniably have responsibilities within their

surrounding community, and that these responsibilities must be clarified and aligned

with the companies’ core businesses. Since these responsibilities are relationships and

promises, CSR is ultimately a function of the brand (ibid.). H ence, Kitchen (2003) defines

CSR as:

“The brand-specific duties and resultant actions of commercial organizations in relation to their

communities of need – defined and delivered outside the core transactional context of the

business” (ibid, p. 10).

Similarly, Keller and Aaker (1997) define CSR as “a firm's philosophy to improve the

quality of life in local communities through various activities and programs”

3.3.3 Environm ental concern

Keller and Aaker (1997) define environmental concern as “a firm’s policy to sell

"environmentally friendly" products and to manufacture products in an environmentally

safe fashion”. Corporate marketing efforts that emphasize environmental concern

enhance perceptions of corporate trustworthiness and likeability as well as inferences

that the corporate brand extension is environmentally aware.


27

3.4 Sum m ary

Business-to-business branding is buyer-focused and corporate brand image is

more likely to span product classes. H ence, much focus of current literature is on

corporate brand identity and communication of intangible brand attributes.

Observations indicate that product or service portfolios of B2B companies are often built

around one “blanket brand” consistent with branded-house and sub-brand strategies.

Virtually no academic research in the field of corporate brand extensions has been

published to date. Keller and Aaker (1997) did however examine in their working paper

how corporate marketing activities portraying a firm as innovative, environmentally

concern, and involved with the community might lead to positive corporate credibility

and thus have a positive effect on brand extension evaluation. Only perceived

innovativeness had a significant impact on corporate brand extension evaluation,

because emphasis on innovation was the only type of marketing activity that enhanced

the perceived fit between the corporate parent brand and the extension.


28

PART II.

CON CEPTU AL FRAM EW ORK

Chapter 4. H ypotheses 29

Chapter 5. Research Design 36


29

CH APTER 4.

H YPOTH ESES


Since this study is a broad replication of Aaker and Keller’s (1990), the original model will

be modified in order to be consistent with the scope of the study.

The dependent variable of the model is the overall attitude towards the business-

to-consumer brand extension, operationalized as the average of perceived quality of the

business-to-consumer brand extension and the likelihood of trial.

4.2 Know ledge About the Parent Brand

A prerequisite for drawing conclusions about the quality of a brand is to be familiar

with it. H ence, there has to exist a brand node in the consumer’s memory with a variety

of associations linked to it–this is conceptualized as brand knowledge (Keller, 2003).

Information stored in the memory network can be verbal, visual, abstract, or contextual

in nature. Brand knowledge can be characterized in terms of two components: brand

awareness and brand image (ibid.). If consumers are to appreciate the appropriateness

of the brand extension, knowledge of the brand-specific association is required

(Broniarczyk & Alba, 1994). Thus the first hypothesis is:

H ypothesis 1: H igher degrees of know ledge about the business-to-business parent

brand are associated w ith m ore favorable attitudes tow ard the consum er brand

extension.

4.3 Attitude Tow ards the Parent Brand

Attitude is conceptualized in this study as consumer’s perception of the brand’s

overall quality. Aaker and Keller (1990) propose a relation between perceived quality of

parent brand and consumers’ attitude toward the extensions in unrelated product

categories. As the perceived quality (termed QUALITY) of the parent brand is higher, the

transfer of positive attitudes toward the extension is also higher.


30

Zeithaml (1988) defines perceived quality as a global assessment of a consumer’s

judgment about the superiority or excellence of a product, and also concludes that

perceived quality is a construct that is on a higher level of abstraction compared to a

specific product attribute. Following this last finding, it could then be argued that

despite the absence of any prior experience with a business-to-business brand,

consumer attitudes of that brand might still be present. H ence, it will be expected to find

similar results for business-to-business parent brands with business-to-consumer brand

extensions. Keller and Aaker (1997) mention that corporate brand equity lies in the

association of consistent delivery of superior functionality and performance that

customers or suppliers have with a firm’s offering. This leads to the second hypothesis:

H ypothesis 2: H igher quality perceptions tow ard the business-to-business parent brand

are associated w ith m ore favorable attitudes tow ard the consum er brand extension.

4.4 Innovation

The following hypotheses (H 3, H 4 and H 5) relate to Keller and Aaker’s (1997) study

on corporate brand extensions. Since their study examines the impact of different types

of corporate descriptions on brand extension evaluation, the hypotheses will be

formulated in order to fit the original model of Aaker and Keller (1990).

Keller and Aaker (1997) propose that marketing efforts that emphasize innovation

leads to favorable perceptions of corporate expertise and thus has a positive impact on

corporate brand extension evaluation. For the purposes of the current study, it is then

reasonable to assume that a parent brand that is perceived to be innovative will lead to

a more favorable brand extension evaluation compared to a brand which is not

perceived as innovative. The third hypothesis then becomes:

Hypothesis 3: H igher perceptions of innovativeness tow ard the business-to-business

parent brand are associated w ith m ore favorable attitudes tow ard the consum er

brand extension.

4.5 Corporate Social Responsibility

Keller and Aaker (1997) propose that marketing activities directed towards

environmental awareness and community involvement increase the perceived likeability


31

and trustworthiness, but has no significant effect on an extension’s perceived quality.

This leads to the fourth hypothesis:

Hypothesis 4: Perceptions of corporate social responsibility of the parent business-to-

business brand has no effect the on the attitude tow ards the consum er brand

extension.

4.6 Environm ental Concern

Following Keller and Aaker’s (1997) findings that marketing efforts emphasizing

environmental concern only has a modest impact on an extension’s perceived quality,

the fifth hypothesis is:

H ypothesis 5: Perceptions of environm ental concern of the parent business-to-business

brand has no effect on the attitude tow ards the consum er brand extension.

4.7 Transfer

As proposed by Aaker and Keller (1990), the transfer of positive attitudes is also

influenced by the similarity between the corporate brand and the extension. This follows

the categorization theory and category-based processing, where consumers evaluate a

new brand extension as to how well this “fits” with the parent brand. Similarly, Boush

and Loken (1991) propose that affect associated with the original brand is transferred to

the extension when similarity between the two products is high. Following this, if

consumers perceive a “fit” between a business-to-business brand and a consumer

product class, they will transfer perceptions of quality to the new brand extension.

The first dimension of fit is the variable TRAN SFER, which is defined as the degree

in which the overall skills of the corporate brand can be helpful for the development of

the brand extension. In this case, if the parent brand organization’s resources and

abilities are useful for making the consumer brand extension, the transfer of positive

attitudes is expected to be higher. Therefore, the sixth hypothesis is:

Hypothesis 6: The transfer of a business-to-business parent brand’s perceived quality is

enhanced w hen the product classes (of the parent brand and the consum er brand

extension) in som e w ay fit together. W hen the fit is w eak, the transfer is inhibited.


32

4.8 Brand Associations

Aaker and Keller (1990) identify two additional dimensions of fit: the first one is

COMPLEMEN T, which is related to the product; the second one is SU BSTITU TE, which is

related to the producer. W hen the product of the parent brand and the brand extension

can be consumed jointly to satisfy some need, it is said that they complement each

other (H enderson & Quandt, 1980). On the other hand, when the brand extension can be

used instead of the parent brand product, they are substitutes.

Substitutability is in this model a generally weak predictor, since relatively few

brand extensions represent true substitutes (Bottomley & H olden, 2001). In the case of

an extra-sectoral movement, the brand extension can by definition not substitute for the

product of the parent brand. The Substitute dimension will hence be omitted from

further analyses. The original microeconomic definition of complementarity is limited for

current purposes, as according to above given definition it involves joint consumption

of two product classes. In the case of an extra-sectoral movement, it is not possible for a

brand extension to complement the original brand, since the customers for the original

brand and brand extension respectively are different (i.e. business-to-business and

consumers). H ence, the Complement dimension has to be replaced in order to fit the

model.

A useful alternative measure for complementarity is the concept of brand

associations (Broniarczyk & Alba, 1994; Section 2.3.4), which proposes that consumer do

not only evaluate the brand extension based on the perceived product category fit (c.f.

H ypothesis 6), but that their assessment are driven primarily by the associations of the

brand. In other words, if a person considers a brand extension to be relevant with the

original brand concept, the attitude towards the extension will be positive.

The seventh hypothesis is:

H ypothesis 7: If the brand associations of the consum er brand extension are consistent

w ith brand concept of the business-to-business parent brand, the attitude tow ard the

brand extension is positive.

4.9 Perceived Difficulty

Aaker and Keller (1990) introduce a last variable: DIFFICU LT. This denotes the

perceived difficulty in designing or producing the brand extension product. If the


33

extension is perceived to be too easy or difficult to make, consumers’ attitude towards

the extension will be negative. This is because a quality brand combined with a trivial

product class is viewed as either incompatible with the parent brand or exploitative by

means of a premium pricing strategy (Aaker & Keller, 1990; Bottomley & H olden, 2001).

Intermediate levels of perceived difficulty lead to a favorable attitude, as suggested by

Kapferer (1997).

In the current study, it is reasonable too assume that consumers will evaluate the

difficulty of designing and making consumer brand extensions of business-to-business

parent brands as they would with “normal” brand extensions. Thus, the eighth

hypothesis is:

Hypothesis 8: The relationship betw een the difficulty of m aking the consum er product

class of the brand extension and the attitude tow ard the brand extension is positive.

4.10 M oderating Factors

In addition for testing for the effects described so far, some interaction effects will

be assessed.

First, it is expected that the effect of difficulty to make a brand extension will be

weakened in the case of higher levels of perceived innovativeness of the parent brand.

This is because an innovative parent brand might be perceived as being more capable of

producing brand extensions that are difficult to make. Reasoning in the same way, an

innovative parent brand might be perceived as being less capable of producing brand

extensions that are easy to make (easy is assumed to be the linear opposite of difficult).

H ence, the following will be tested:

Hypothesis 9a: A higher level of perceived parent brand innovativeness w eakens the

im pact of perceived difficulty(to produce an extension) on brand extension evaluation.

Hypothesis 9b: A higher level of perceived parent brand innovativeness strengthens the

im pact of perceived ease (to produce an extension) on brand extension evaluation.

Second, evaluation of a brand extension need not correspond to evaluation of that

brand in its original category because the value of a brand association differs depending

on the benefits sought within a particular product category (Broniarczyk & Alba, 1994).


34

H ence, there might be an interaction between brand-specific associations and the

similarity of the extension category. This leads to the following:

Hypothesis 10: A higher level of perceived category fit betw een the parent brand and

the brand extension strengthens the im pact of perceived extension brand associations.


35

4.11 Sum m ary

Exhibit 4.1 below lists all hypotheses and also refers to its academic source from

which the concept was drawn to form each specific hypothesis.

TABLE 4.1 H ypotheses and Their References to Literature

H ypothesis Literature reference

H 1 H igher degrees of know ledge about the business-to-business parent brand are associated w ith m ore favorable attitudes tow ard the consum er brand extension.

Broniarczyk & Alba (1994)

H 2 H igher quality perceptions tow ard the business-to-business parent brand are associated w ith m ore favorable attitudes tow ard the consum er brand extension.

Aaker and Keller (1990)

H 3 H igher perceptions of innovativeness tow ard the business-to-business parent brand are associated w ith m ore favorable attitudes tow ard the consum er brand extension.

Adapted from Keller and Aaker (1997)

H 4 Perceptions of corporate social responsibility of the parent business-to-business brand has no effect the on the attitude tow ards the consum er brand extension .


H 5 Perceptions of environm ental concern of the parent business-to-business brand has no effect on the attitude tow ards the consum er brand extension.


H 6 The transfer of a business-to-business parent brand’s perceived quality is enhanced w hen the product classes (of the parent brand and the consum er brand extension) in som e w ay fit together. W hen the fit is w eak, the transfer is inhibited.


H 7 If the brand associations of the consum er brand extension are consistent w ith brand concept of the business-to-business parent brand, the attitude tow ard the brand extension is positive.

Adapted from Broniarczyk & Alba (1994)

H 8 The relationship betw een the difficulty of m aking the consum er product class of the brand extension and the attitude tow ard the brand extension is positive.


H 9a A higher level of perceived parent brand innovativeness w eakens the im pact of perceived difficulty (to m ake an extension) on brand extension evaluation.

Own construct

H 9b A higher level of perceived parent brand innovativeness strengthens the im pact of perceived ease(to m ake an extension) on brand extension evaluation.

Own construct

H 10 A higher level of perceived extension brand associations strengthens the im pact of perceived category fit betw een the parent brand and the brand extension.

Own construct


36

CH APTER 5.

RESEARCH DESIG N


This chapter will provide with the research design of the present study. First, the

hypotheses brought forward in the previous chapter are operationalized through the

development of a new model. An argument for the choice of method for executing the

new model is then provided. A description of questionnaire development and sampling

procedure follows.

5.2 The H ypothesized M odel

5.2.1 The original m odel of Aaker and Keller

Aaker and Keller (1990) hypothesized that “the consumer’s attitude towards the brand

extension is a positive function of the quality of parent brand, the fit between the

parent’s brand category and the extension category (measured in terms of the

transferability of skills and expertise from one category to the other and the

complementarity and substitutability of one category and the other), the interactions of

quality with three fit variables, and the degree of difficulty in designing and making a

product in the extension category” (Bottomley & H olden, 2001, p. 495). Formally, the

following model was tested:

Y = α + β1Q + β2T + β3C + β4S + β5QT + β6QC + β7QS + β8D + ε,

where the dependent variable

Y = attitude towards the brand extension = [purchase + extension]/2,

and where the independent variables

Q = quality

T = transfer

C = complement


37

S = substitute

D = difficult

α = intercept

ε = error term.

5.2.2 The current m odel

Aaker and Keller (1990) and subsequent replications operationalize the

hypotheses by means of a linear regression model. The model in its original state cannot

be used in the current context, but has to be changed. First, following Section 4.2 and

hypothesis 1, the variable KNOW LEDGE will be added. Second, following the reasoning

from Section 4.3, the independent variables SU BSTITU TE and COMPLEMEN T will be

omitted. These will be replaced by the new variable BRAND CONCEPT CON SISTENCY

(hypothesis 4). H ence, there will only be two “fit” variables in the new model –

TRAN SFER and BRAND CONCEPT CON SISTENCY. Furthermore, all three original

interaction variables will be omitted from the original model. Third, three independent

variables corresponding to hypotheses 6, 7 and 8 will be added to the new model.

H ypothesis 6 relates to the IN NOVATIVEN ESS of the parent brand, hypothesis 7 relates to

CORPORATE SOCIAL RESPON SIBILITY, and hypothesis 8 corresponds to

EN VIRONMEN TAL CONCERN . This gives the following:

(2) Y = α + β1K + β2Q + β3T + β4B + β5D + β6I + β7C + β8E + ε,




K = PAREN T BRAND KNOW LEDGE

B = BRAND CONCEPT CON SISTENCY

I = IN NOVATIVEN ESS

C = CORPORATE SOCIAL RESPON SIBILITY

E = EN VIRONMEN TAL CONCERN .

Lastly, there might be some interaction effects. The first interaction effect is

between IN NOVATIVEN ESS and DIFFICU LT, and the second interaction term is between


38

TRAN SFER and BRAND CONCEPT CON SISTENCY, corresponding to H ypothesis 9 and 10

respectively. This leads to:

(3) Y = α + β1K + β2Q + β3T + β4B + β5D + β6I + β7C + β8E + β9ID + β10BT + ε,

where the additional interaction terms

ID = IN NOVATIVEN ESS × DIFFICU LT

BT = BRAND CONCEPT CON SISTENCY × TRAN SFER.

The hypothesized model is summarized in figure 5.1.

Parent brand innovativeness

Brand concept consistency

Attitude towards brand

extension

Parent brand perceived quality

Transfer

Parent brand environmental

concern

Parent brand corporate social responsibility

Difficult

Parent brand knowledge

H 1

H 2

H 7

H 8

H 4

H 5

H 3

H 6

H 10

H 11

FIG U RE 5.1 H ypothesized Model


39

5.3 M ethodology

The current research follows the approach used in the quantitative analysis in

Aaker and Keller’s (1990) study and replication studies. In these studies, qualitative

analyses are also performed prior to the quantitative analysis. Since the current study

aims to validate the Aaker and Keller (1990) in a B2B context for the sake of

generalizability and reproducibility, the current study will exclusively be of quantitative

nature.

The quantitative part tries to formally assess the consumer’s evaluation of brand

extensions through measuring attitude for different variables. In other words,

hypotheses 1 to 10 will be operationalized through a new model adapted from Aaker

and Keller (1990) and other theories discussed in chapter two and three.

5.4 Questionnaire Developm ent

5.4.1 Brands and brand extensions

To test the hypothesized model, a questionnaire was developed. This

questionnaire consisted of various questions on five well-known global B2B brands and

11 hypothetical consumer brand extensions (see table 5.1). Out of these five B2B brands,

three were distinct service brands with little product features. The other two brands

were distinct product brands with minimal service features. All brands were chosen to

maximize the spread regarding industry in order to elicit specific associations per brand.

To avoid any confounding reactions, only the logos of each brand were presented

without any additional information.

The brand extensions used were chosen had to be reasonable and not illogical,

but aimed to provide with heterogeneity on the three fit measures of the hypothesized

model. To achieve this, some extensions were deliberately chosen to be of “lower

quality”, thus allowing for some variance with respect to the perceived quality of the

brand extension.


40

Table 5.1 Overview of B2B Brands and B2C Extensions

Original brand

Service (S)/

Product (P) Original Product or Service

H ypothetical Extensions

U sed in the Survey

Reuters

S

Information provider

N ews radio station,

24 hour news TV channel

S Assurance, tax and legal, and financial

advisory services

Private tax planning service,

Institute of Accountancy

(Executive MBA education),

off-shore banking service

Saatchi & Saatchi S Advertising Marketing/advertising books,

Institute of Advertising (executive

training courses)

Intel P Chips, boards, systems, and software

components

Portable MP3 music player,

notebook PC

Boeing P Commercial airplanes, military aircraft,

satellites, missile defense, human space

flight, and launch systems

Chronograph wristwatch,

flight simulator computer game,

travel luggage

5.4.2 Scaling and m ulti-item scales

The Likert scale is a widely used rating scale that requires the respondents to

indicate a degree of agreement or disagreement with each of a series of statements

about the stimulus objects (Albaum, 1997; Brody & Dietz, 1997; Likert, 1932). The

advantages of Likert-type scales are that they are easy to construct and administer, and

respondents are familiar about how to use them. This makes Likert-type scales suitable

for Internet surveys, mail, telephone or personal interviews. A major disadvantage is that

it takes longer to complete Likert-type scales than other itemized rating scales because

the respondents have to read and fully reflect upon each statement (Malhotra and Birks,

2003).

To assess the overall quality (QUALITY) of each parent brand as well as each

brand extension, a 7-point Likert scale was used (1 = poor, 7 = outstanding). A second 7-

point scale measure assessed the attributes of the parent brand (IN NOVATIVE, CSR and


41

EN VIRONMEN TAL CONCERN ) where (1 = not at all, 7 = very). A third 7-point scale

measured the three fit variables (TRAN SFER, BRAND CONCEPT CON SISTENCY, and

DIFFICU LT) where (1= totally disagree, and 7 = totally agree). This scale was also used to

assess the likelihood of brand extension trial (TRIAL). A fourth 7-point scale measured

the familiarity about the parent brand (KNOW LEDGE) where (1 = not at all, and 7 = very).

In Likert-type scales, the neutral points (4) can be considered to be natural

origins5. As in the original Aaker and Keller (1990) model and consequent replication

studies, the measured ordinal scales can be interpreted as interval scales. This allows for

using multiple regression as an analytical method.

Multiple-item rating scales are often use in social sciences research to measure

abstract constructs (Aaker et al., 1998). By having several items that measure the same

so-called construct, the problem of having single unrepresentative questions is solved

(ibid.). The greater the number of initial items generated, the better will be the final

scale. The larger the scale, the greater is the reliability, but shorter scales are easier for

respondents to answer. H ence, a balance between brevity and reliability has to be struck

to determine the optimal scale.

Multiple-item scales were developed for the variables IN NOVATIVE, CSR,

TRAN SFER, BRAND CONCEPT CON SISTENCY, and DIFFICU LT by means of 7-point Likert-

scales. Several items relevant for each variable were generated to form constructs. The

multiple-item scales were evaluated in the pilot-testing of the questionnaire, after which

some items were rephrased for clarity and some were omitted for brevity with respect to

the final length of the questionnaire. The final set of multiple-item constructs can be

seen in table 5.2.

5 This solves the problem of the absence of a real zero in the scale (van Riel et al., 2001).


42

TABLE 5.2 Multiple-scale items

Innovative

This brand is modern and up-to-date

This brand invests in R&D

This brand introduces the latest product/service features

Corporate Social Responsibility

This brand is involved in helping its community

This brand is socially responsible

Transfer

The competences of Reuters are useful to make this extension

The resources of Reuters are useful to make this extension


This brand extension is consistent with the Reuter brand

This brand extension fits with my associations of the Reuter brand

D ifficult

This brand extension will be difficult to make

Reuters is out to make short-term profits with this brand extension

5.5 Data Collection

5.5.1 Pilot-testing

Pilot-testing can be used to test the questionnaire on a small sample of

respondents to identify and eliminate potential problems (Martin & Polivka, 1995). A

paper version of the questionnaire along with additional pages with open-ended

questions concerning question content, wording, sequence, form, layout, question

difficult and instructions were handed out to ten graduate students enrolled at the

Faculty of Economics and Business Administration of Maastricht U niversity in the

N etherlands. The respondents of the pilot-test were similar to those who were included

in the subsequent survey, i.e. they were drawn from the same population.

Some changes were made to the first version of the questionnaire, including

removal of some brand extensions to make the questionnaire length shorter. Some


43

questions were also rephrased for sake of clarity. The pilot-testing as well as the final

version of the questionnaire can be found in appendix A.

5.5.2 Sam pling and response

In the original Aaker and Keller (1990) study as well as in most of the replication

studies, the samples were drawn from student populations. The survey of the current

study was conducted both online and via regular paper questionnaires. Both survey

types sampled graduate and undergraduate students enrolled at the Faculty of

Economics and Business Administration of Maastricht University (UM) in the

N etherlands. The online (Internet) questionnaire was arranged by the Marketing

Department of the Faculty of Economics and Business Administration of Maastricht

U niversity by which e-mail invitations were sent out to graduate and undergraduate

faculty students who were registered as willing to participate in online surveys

conducted by the faculty. The regular (paper) questionnaires were handed out among

faculty students at UM.

TABLE 5.3 Response Profile: Sampling Type

Frequency Percentage

Paper 35 34.0 %

Online 68 66.0 %

Total 103 100.0 %

TABLE 5.4 Response Profile: G ender


M ale 59 57.3 %

Fem ale 44 42.7 %

Total 103 100.0 %

TABLE 5.5 Response Profile: N ationality


Dutch 49 47.6 %

G erm an 20 19.4 %

Other 34 33.0 %

Total 103 100.0 %


44

A total of 103 students were sampled out of which 66% were sampled via the

Internet and 34% sampled via regular paper questionnaires. Respondents varied in age

between 19 and 29 years old, 57% were male and 43% were female. In terms of

nationality, two thirds of the sample consisted of two large groups: 47% were Dutch and

19% were G erman.


45

PART III.

RESEARCH FIN DIN G S

Chapter 6. Analysis, Results and Interpretation 46


46

CH APTER 6.

AN ALYSIS, RESU LTS AN D IN TERPRETATION S

�

6.1 Introduction to the Chapter In this chapter, the findings of the empirical research are presented. First are the results

of the three types of analyses in the order in which they were performed. The first two

analyses aim to increase internal consistency and reduce collinearity, and thus enable a

more solid regression analysis. The regression results pave the way for the hypothesis

testing, the next section of this chapter. Following this is an interpretation and

discussion of each hypothesis.

6.2 Reliability, Collinearity and Regression Analyses The data provided by the questionnaires was analyzed using the statistical

software application SPSS (Statistical Package for the Social Sciences). The reliability of

multiple-item constructs were assessed by measuring internal consistency. The problem

of multicollinearity was dealt with using the method of residual centering. To test the

hypothesized model, multiple regression analyses were used.

6.2.1 Reliability assessm ent

Prior to performing the regression, reliability analyses were performed on the

following constructs: (1) the dependent variable, operationalized by the average of the

perceived quality of the extension and the likelihood of trial, and (2) variables

operationalized through multiple-item scales.

Because the two items of the DIFFICU LT construct were the same measure but in

two different directions6, the item that measured “ease” was reversed as necessary step

to perform the reliability analysis as well regression analyses.

Reliability was assessed by measuring the reliability coefficient, or Cronbach’s

alpha, which is defined as the average of all possible split-half coefficients resulting from

6 One item measured the difficulty to make an extension, while the other item measured the ease to make an extension.


47

different ways of splitting the scale items (Cronbach, 1951). This coefficient varies

between 0 and 1, where a value of 0.6 or less generally indicates unsatisfactory internal

consistency reliability (Malhotra & Birks, 2003). An important property of Cronbach’s

alpha is that its value tends to increase as the number of scale items increases. Therefore,

the coefficient alpha may be artificially inflated by including several redundant scale

items (Peterson, 1994).

The reliability analysis for the dependent variable and the five independent

variable constructs reveled that all measures had an alpha coefficient exceeding 0.7

except for the construct DIFFICU LT (�=0.21) (see table 6.1). The results from the

reliability analysis indicated that the scale variance would be reduced the most if the

item corresponding to the ease to make an extension (labeled DIFF2; see appendix B.1)

was removed. Consequently, this item of the DIFFICU LT construct was eliminated from

further analyses.

Because the multiple-item scales only included two or three items per construct,

the problem of artificially inflated coefficients is considered not to be an issue in the

current study.

TABLE 6.1 Cronbach’s Alpha Coefficients for Multiple-Item Constructs

D ependent variable construct

Attitude towards brand extension 0.74

Independent variable constructs

Innovative 0.87

Corporate social responsibility 0.85

Transfer 0.93

Brand concept consistency 0.94

Difficult 0.21

6.2.2 M ulticollinearity

Multicollinearity is when two or more independent variables are correlated with

each other. W hen highly correlated independent variables are present in a regression

model, the results are confusing and interpretation of estimated variables is difficult

(McClave et al., 1998). The original Aaker and Keller study (1990) and replication studies

(Sunde & Brodie, 1993; Bottomley & Doyle, 1996; van Riel et al., 2001) have all noted a

presence of multicollinearity between main effects and interaction terms.


48

In the current model, multicollinearity caused by a high correlation between

main effects and the interaction term was also present. This was indicated by the

Variance Inflation Factors (VIFs) of the independent variables, as shown in table 6.2. The

VIF scores indicate to what extent each independent variable is explained by the other

independent variables (H air et al., 1998). A VIF score of four or higher is considered to be

indicative of severe collinearity (Olsen, n.d.).

A method was developed by Lance (1988) to deal with the effects of

multicollinearity. This method, called “residual centering” is a two-step regression that

“serves to substantially reduce multicollinearity among first-order terms in polynomial

regression equations for any given independent variable” (N eter et al., 1990, p. 411).

Lance (1988) argues that the benefits of the “residual centering” approach over

conventional OLS regression analysis in the presence of high degrees of multicollinearity

between the main and interaction terms are fourfold: (1) multicollinearity between the

exogenous variables is substantially reduced; (2) standard errors are reduced; (3) main

and interaction effects are separated; and (4) the relative importance of the main and

interaction variables can easier be identified.


49

TABLE 6.2 VIF Scores of Regression Variables Before and After Residual Centering

Independent variables Before

Residual Centering

After Residual

Centering

Parent brand-specific effects

Parent brand knowledge 1.91 1.91

Parent brand quality 2.23 2.23

Innovative 5.80 2.32

CSR 1.97 1.97

Environmental concern 1.95 1.95

Brand extension-specific effects

Transfer 7.73 4.35

Brand concept consistency 13.54 2.89

Difficult 15.38 15.38

Interaction effects

Innovativeness × Difficult* 18.91 13.93*

Brand concept consistency × Transfer * 29.71 1.13*

* Formally, the two interaction variables are labeled as “IN NOVATIVE ×

DIFFICULT[Residual]” and “TRAN SFER × BRAND CONCEPT

CON SISTENCY[Residual] after the residual centering has been applied.

Replications of Aaker and Keller’s (1990) model by N ijssen and H artman (1994) and by

Bottomley and Doyle (1994) apply Lance’s (1988) residual centering regression

approach, arguing that their researches offer an improved view on the interaction

effects.

Following N ijssen and H artman (1994) and like Bottomley and Doyle (1994) as well

as Riel et al. (2001), the residual centering technique was applied in the current study. In

this two-stage regression procedure each interaction variable (e.g. IN NOVATIVEN ESS ×

DIFFICU LT) was regressed on its two component parts (i.e. IN NOVATIVN ESS and

DIFFICU LT in the example) using OLS regression. The resulting residuals (e.g.

IN NOVATIVEN ESS × DIFFICU LT[Residual]) were then used in place of the respective

interaction term when estimating the full effects model using OLS (cfr. Bottomley &

Doyle, 1994).

Multicollinearity was successfully eliminated for independent variables

IN NOVATIVE and BRAND CONCEPT CON SISTENCY, and the interaction term BRAND


50

CONCEPT CON SISCEN TY × TRAN SFER, whose VIF scores fell below 2.89 or lower after

residual centering. For the independent variable TRAN SFER, and for the interaction term

IN NOVATIVEN ESS × DIFFICU LT, multicollinearity was only partially eliminated (VIF scores

remain above four). N o multicollinearity could be eliminated for the variable DIFFICU LT.

Exhibit 6.4 summarizes the VIF scores of regression variables before and after residual

centering.

As a consequence of the high VIF scores, the interaction term IN NOVATIVEN ESS ×

DIFFICU LT was omitted from further analysis, after which multicollinearity was totally

eliminated (VIF < 2.89). The variable DIFFICU LT was kept in the model. N ote that its VIF

score after the removal of the interaction term is well below four. Exhibit 6.5 shows the

results of this action.

TABLE 6.3 VIF Scores of Regression Variables After Omitting Interaction Term

Independent variables


Parent brand knowledge 1.96

Parent brand quality 2.27

Innovative 2.77

CSR 1.97

Environmental concern 1.95


Transfer 2.96

Brand concept consistency 2.88

Difficult 1.27

Interaction effects

Innovativeness × Difficult[Residual] removed

Brand concept consistency × Transfer [Residual] 1.14


51

6.2.3 Regression analysis results

After residual centering and omitting one interaction term, the hypothesized

model can be summarized as follows:

(4) Y = α + β1K + β2Q + β3I + β4C + β5E + β6T + β7B + β8D + β9BT[Residual] + ε,




K = PAREN T BRAND KNOW LEDGE

Q = PAREN T BRAND QUALITY

I = IN NOVATIVEN ESS

C = CORPORATE SOCIAL RESPON SIBILITY

E = EN VIRONMEN TAL CONCERN .

T = TRAN SFER

B = BRAND CONCEPT CON SISTENCY

D = DIFFICU LT

BT[Residual] =BRAND CONCEPT CON SISTENCY × TRAN SFER [Residual].

The hypothesized model (4) was formally tested by means of linear regression. The

analysis included the data from the 103 respondents, giving a total sample size of 1,133.

6.1.3.1 O verall m odel evaluation

The significance of the regression model as a whole was tested automatically by

the SPSS software. The computed F statistic equals 180.77, which is significant at

p=0.000. Theoretically, this indicates that one or more regression coefficients have a

value different from zero (Thomas, 1997), i.e. at least one variable is relevant. Section 6.2

will assess the significance of specific coefficients.

The “closeness of fit” of the regression model is determined by its R2. It captures the

percentage deviation from the mean in the dependent variables that could be explained

by the model. It is always possible to increase R2 by adding extra explanatory variables,

regardless of the true significance of these. A fairer measure of closeness of fit is adjusted

R2, i.e. adjusted for degrees of freedom (ibid.). The adjusted R2 for the current model is


52

0.62. This compares favorably with the original Aaker and Keller (1990) model and

replications studies. The regression model was also tested for each brand as well as for

services and products. The adjusted R2 values of these tests are shown in table 6.6b.


53

TABLE 6.4 Comparison Effects of Original and Replication Studies (t values are in parentheses)

Aaker

and

Keller

(1990)

Sunde and

Brodie

(1993)

N ijssen

and

H artman

(1994)

Bottomley

and Doyle

(1996) U K

Bottomley

and Doyle

(1996) N Z

Van Riel et

al. (2001)

Services

Van Riel et

al. (2001)

Goods

Current study1

Quality N /A 0.38 (5.3) 0.25 (9.1) 0.22 (11.2) .25 (12.91) 0.15 (9.38) 0.17 (6.78) 0.07* (2.39)

Transfer 0.24 0.21 (3.1) 0.58 (18.9) 0.31 (13.6) .26 (12.48) 0.34 (13.5) 0.41 (13.1) 0.27* (8.41)

Complement 0.17 0.29 (4.2) 0.01 (0.3) 0.31 (13.2) .30 (14.48) 0.38 (18.7) 0.20 (7.59) N A

Substitute 0.08 0.13 (1.9) .08 (2.7) 0.18 (7.67) .19 (8.86) 0.06 (3.40) 0.19 (7.13) N A

Difficult N /A 0.00 (0.1) Omitted 0.02 (1.22) .03 (1.59) -0.09 (-4.1) –0.16 (–5.3) -0.06* (-3.03)

Adjusted R2 N /A 0.48 0.48 0.47 0.43 0.57 0.53 .61

Sample size 2,140 1,413 693 1,358 1,558 1,616 808 1,133

*Significant at 0.000 level.

1Beta coefficients are taken from the full model. Since the variables of the current model are different from the variables of the

original and replication studies, a formal comparison is not suitable. The comparison is nevertheless interesting on an intuitive level.


54

TABLE 6.5 Regression Results: Main and Interaction Effects (residually centered)

Independent variables Full m odel

Predicted

direction of hypotheses b

Std. Error Beta T-value

Constant -0.20 0.18 -1.13


H 1: Parent brand knowledge + -0.03 0.02 -0.03 -1.32

H 2: Parent brand quality + 0.08** 0.03 0.07 2.39

H 3: Innovative + 0.15**** 0.03 0.13 4.31

H 4: CSR 0 0.18**** 0.03 0.16 5.98

H 5: Environmental concern 0 0.06** 0.03 0.05 2.11


H 6: Transfer + 0.26**** 0.03 0.27 8.41

H 7: Brand concept consistency + 0.37**** 0.03 0.42 13.33

H 8: Difficult + -0.06*** 0.02 -0.06 -3.03

Interaction effects

H 9: Innovativeness × Difficult -/+ Omitted

H 10: Transfer × Brand concept consistency[Residual] + -0.02 0.01 -0.03 -1.50

Control variables

Service brand -0.05 0.07 -0.02 -0.75

R2 0.62

Adjusted R2 0.61

Sample size n 1133

****p<.000; ***p<.01; **p<.05; *p<.10


55

TABLE 6.6a Full Model at Brand Level (residually centered, t-values in parentheses)


Brand

Product/ service brand

Parent brand

know ledge

Parent brand quality Innovative CSR

Environ-m ental concern

Reuters S 0.09 -0.07 0.07 0.21*** 0.06

(1.33) (-0.92) (1.04) (2.71) (0.74)

Ernst & Young S -0.11** 0.10* 0.18*** 0.15*** 0.09

(-2.07) (1.78) (2.81) (3.01) (1.62)

Saatchi & Saatchi S 0.06 0.07 0.08 0.38**** -0.11

(0.97) (0.77) (0.96) (4.96) (-1.33)

Intel P -0.01 0.15*** 0.11 0.16*** 0.02

(-0.23) (2.27) (1.63) (2.51) (0.28)

Boeing P 0.05 0.03 0.02 0.15**** 0.11***

(1.05) (0.47) (0.37) (2.75) (1.99)

Services S -0.06 0.05 0.06* 0.14**** 0.18****

(-1.60) (1.17) (1.78) (4.03) (5.00)

Products P 0.00 0.08** 0.06 0.08* 0.13***

(-0.04) (1.98) (1.56) (1.75) (3.21)

****p<.000; ***p<.01; **p<.05; *p<.10

TABLE 6.6b Full Model at Brand Level (residually centered, t-values in parentheses)

Brand extension-specific effects Interaction effects

Brand

Product/service brand Transfer

Brand concept

consistency Difficult

Innovativeness

×××× difficulty

Transfer ×××× brand

concept consistency

Sam ple Size n

Adjusted R2

Reuters S 0.20** 0.35**** -0.08 Omitted 0.03 206 0.51

(2.42) (4.37) (-1.47) (0.50)

Ernst & Young S 0.24**** 0.34**** -0.15*** Omitted 0.00 206 0.63

(3.57) (4.95) (-3.15) (-0.05)

Saatchi & Saatchi S 0.28*** 0.29**** -0.07 Omitted -0.02 206 0.60

(3.52) (4.02) (-1.55) (-0.34)

Intel P 0.36**** 0.25*** -0.05 Omitted -0.01 206 0.57

(4.00) (2.85) (-0.89) (-0.16)

Boeing P 0.17**** 0.48**** -0.01 Omitted -0.13**** 309 0.62

(3.28) (8.59) (-0.29) (-3.07)

Services S 0.26**** 0.34**** -0.10**** Omitted 0.01 618 0.58

(5.71) (8.32) (-3.58) (0.20)

Products P 0.26**** 0.45**** -0.03 Omitted -0.06** 515 0.64

(5.69) (9.83) (-0.85) (-2.04)

****p<.000; ***p<.01; **p<.05; *p<.10


56

6.2 H ypothesis Testing

6.2.1 H 1: Parent brand know ledge

The beta coefficient for the variable KNOW LEDGE is not significant at any level (P >

0.1), thus the first hypothesis has to be rejected. The KNOW LEDGE variable also is

insignificant on a brand level, except for cases where only the Ernst & Young brand

analyzed. For these cases, the KNOW LEDGE coefficient is both significant (P < 0.05) and

substantial (-0.11). It is noteworthy, however, that the coefficient has a negative sign as

opposed to the predicted direction of the first hypothesis.

To sum m arize, w e can reject the first hypothesis at a 0.05 significance level. At a brand

level w e reject H1 for all brands except Ernst & Young for w hich w e accept it at P<0.05.

6.2.2 H 2: Parent brand quality

The beta coefficient for the variable QUALITY is significant (P < 0.05). The effect of

QUALITY is relatively weak (0.07) compared to the original and replication studies. The

second hypothesis is henceforth accepted. On a brand level, QUALITY is significant at

P>0.10 for the Ernst & Young brand, and at P<0.01 for Intel. The coefficients for these

brands are 0.10 and 0.15, respectively. For all other brands we reject the second

hypothesis.

In sum m ary, the second hypothesis is accepted at P=0.05. At a brand level w e m ust

reject H2 except for the tw o brands Ernst & Young (P<0.10) and Intel (P<0.01.

6.2.3 H 3: Innovative

The beta coefficient for the variable IN NOVATIVE is significant at P<0.000 with a

moderate effect (0.13). The third hypothesis is therefore accepted. On a brand level, we

reject H 3 except for the Ernst & Young brand for which the beta coefficient is 0.18

(P<0.01).

In sum m ary, the third hypothesis is accepted at P=0.00. At a brand level, w e m ust reject

H3 w ith exception of the brand Ernst & Young.

6.2.4 H 4: Corporate Social Responsibility

The beta coefficient for the variable CORPORATE SOCIAL RESPON SIBILITY (CSR) is

significant at P<0.00, with a value of 0.18. Since the CSR effect was hypothesized to be

zero, the fourth hypothesis is rejected. H 4 was also rejected for all brands at P<0.01. The


57

CSR beta coefficient for brands are inline with the aggregate beta with exception for the

beta for the Saatchi & Saatchi brand, which is 0.38 (P<0.000).

To sum up, the fourth hypothesis is rejected at both an aggregate level (P<0.000) and a

brand level (all brands; P<0.01).

6.2.5 H 5: Environm ental concern

The beta coefficient for EN VIRONMEN TAL CONCERN is significant at P<0.05,

although its effect is rather unsubstantial (0.05). On a brand level, only the beta

coefficient for the brand Boeing is significant (P<0.01) with a slightly larger effect

compared with the aggregate level beta (0.11).

In sum m ary, the fifth hypothesis is rejected at P<0.05 on an aggregate level, and

rejected for the brand Boeing.

6.2.6 H 6: Transfer

The beta coefficient for the variable TRAN SFER is significant (P<0.00) and highly

substantial (0.27). Thus, the sixth hypothesis is accepted. The effect of TRAN SFER is inline

with findings from previous replication studies. On a brand level, TRAN SFER is

significant for all of the five brands tested at P<0.05, although their effect vary between

0.20 for Reuters and 0.36 for Intel.

In sum m ary, the third hypothesis is accepted at P=0.00. At a brand level, w e accept H3

for all brands.

6.2.7 H 7: Brand concept consistency

The beta coefficient for the variable BRAND CONCEPT CON SISTENCY is significant

(P<0.00) and highly substantial. The beta equals 0.42. Thus, the seventh hypothesis is

accepted. BRAND CONCEPT CON SISTENCY is also relevant for all brands at a 0.01

significance level. W hat is noteworthy is that the beta coefficient for Boeing is 0.48,

which is much higher than for the other brands.

To sum m arize, w e accept the seventh hypothesis at P=0.00. At a brand level, H4 is

accepted for all brands (P<0.01).

6.2.8 H 8: D ifficult

The beta coefficient for the variable DIFFICU LT is significant at P<0.05. H owever,

the coefficient indicates a negative effect (-0.06), which is contradictory to the direction


58

of the fifth hypothesis. W e can therefore not accept the eighth hypothesis. On a brand

level, only the beta coefficient for the brand Ernst & Young is significant (P<0.01). It too is

negative, with a beta of -0.15.

To sum m arize, w e cannot accept the eighth hypothesis because of sign reversal. The

beta coefficient is insignificant on a brand level w ith exception for Ernst & Young.

Nevertheless, because of sign reversal w e cannot accept the eighth hypothesis for this brand.

6.2.9 H 10: Interaction effect betw een Transfer and Brand concept consistency

The moderator variable TRAN SFER x BRAND CONCEPT CON SISTENCY[Residual]

shows no significance (P>0.10). W e must therefore reject the tenth hypothesis. On a

brand level, the moderator variable shows no significance except for the brand Boeing,

for which the beta coefficient is negative (-0.13) at P<0.000.

In sum m ary, w e reject the tenth hypothesis at an aggregate level. W e also reject the

hypothesis at a brand level for all brands except for Boeing for w hich w e accept it at a 0.000

significance level.


59

TABLE 6.7 Summary of H ypothesis Testing

H ypothesis

Aggregate level

results Brand level results


H 1: Parent brand knowledge Rejected Accepted for Ernst & Young brand only

H 2: Parent brand quality Accepted Accepted for Ernst & Young and Intel only

H 3: Innovative Accepted Accepted for Ernst & Young only

H 4: CSR Rejected Rejected for all brands

H 5: Environmental concern Rejected Rejected for Boeing only


H 6: Transfer Accepted Accepted for all brands

H 7: Brand concept consistency Accepted Accepted for all brands

H 8: Difficult Rejected Rejected for all brands

Interaction effects

H 9: Innovativeness × Difficult N /A N /A

H 10: Transfer × Brand concept consistency[Residual] Rejected Accepted for Boeing brand only


60

TABLE 6.8 Extension Level Means

Brand extension

Attitude tow ard brand

extension

Parent brand

know ledge

Parent brand quality Innovative

Environ-m ental concern CSR Transfer Difficult

Brand concept

consistency

Intel N otebook PC 5.08 5.58 5.67 6.22 3.62 3.90 5.57 2.85 5.31

Reuters 24-H our N ews TV Channel 5.02 4.21 5.31 4.55 3.45 3.89 5.49 3.26 5.28

Intel Portable MP3 player 4.82 5.58 5.67 6.22 3.62 3.90 5.00 3.33 4.41

Reuters N ews Radio Station 4.68 4.21 5.31 4.55 3.45 3.89 5.40 3.33 5.00

Ernst & Young Private tax planning service 4.61 4.99 5.17 4.61 3.31 3.81 5.40 3.12 5.15

Boeing Flight simulator video game 4.30 5.27 5.55 5.46 3.62 4.01 4.83 3.62 4.42

Saatchi & Saatchi Institute of Advertising 4.24 2.69 4.55 4.38 3.52 3.73 4.67 3.77 4.40

Saatchi & Saatchi Marketing & Advertising books 4.22 2.69 4.55 4.38 3.52 3.73 4.50 3.84 4.36

Ernst & Young Institute of Accountancy 4.06 4.99 5.17 4.61 3.31 3.81 5.16 3.89 4.44

Boeing "U ltra-tough" travel luggage 4.06 5.27 5.55 5.46 3.62 4.01 3.53 3.50 3.56

Boeing Chronograph wristwatch 3.16 5.27 5.55 5.46 3.62 4.01 2.95 4.06 2.40

Average 4.39 4.61 5.28 5.08 3.52 3.88 4.77 3.51 4.43

Standard deviation 0.55 1.06 0.40 071 0.12 0.11 0.84 0.37 0.85

TABLE 6.9 Perceived Quality of Parent Brands

Brand M ean Standard Deviation

Intel 5.67 1.00

Boeing 5.55 1.18

Reuters 5.31 1.27

Ernst & Young 5.17 1.17

Saatchi & Saatchi 4.55 1.08

Average 5.25 1.14


61

6.3 Discussion and Interpretation 6.3.1 Parent brand know ledge

Surprisingly, knowledge about the parent brand had no significant effect on the

dependent variable. Possible reasons for this could be that brand familiarity is an

insufficient condition for evaluating brand extensions. This may suggest that pure

affective responses do not take place, but that cognitive processing is necessary in order

to evaluate the extension. This is inline with Fiske and Pavelchak’s (1986) proposed

concept of piecemeal processing.

Interestingly, the KNOW LEDGE beta coefficient was negative (-0.11) on a brand

level for the brand Ernst & Young. A possible explanation for this may be that the sample

population (i.e. business students) in part possess so much knowledge about Ernst &

Young that they feel that the brand extensions would be unrealistic.

6.3.2 Parent brand quality

As expected, parent brand quality plays a role in evaluating the brand extension.

H owever, the beta coefficient for the QUALITY variable of the current study is

considerably lower than in previous replication studies (0.07). In other words, the

perceived quality of a B2B brand has less spillover effects than a B2C brand, ceteris

paribus. Brand quality may not be the optimal way to measure brand equity in the

context of B2B brands judged by consumers. N evertheless, the beta coefficient for

QUALITY is positive. Thus, even though consumers lack experience of B2B brands they

still are able to make some inferences on their overall quality albeit in a cautious way.

6.3.3 Innovative

Parent brand innovativeness has a positive effect (0.13) on brand extension

evaluation. Thus, Keller and Aaker’s (1997) prediction also holds in the case of consumer

brand extensions of B2B brands. As mentioned in Section 3.3.1, not much research has

been done with respect to brand innovativeness (most research has been done in the

are of consumer innovativeness). The results show that innovativeness is a stronger

predictor of the dependent variable than parent brand quality. There might be two

reasons for this phenomenon: (1) Innovativeness may be a salient attribute of brand

quality, at least partially (i.e. there is an interaction effect between QUALITY and

IN NOVATIVEN ESS); and (2) the quality of B2B may be too abstract for consumers to

evaluate, while innovativeness can be more easily judged.


62

On a brand level, only the Ernst & Young brand showed significance for the

innovativeness coefficient (0.18), indicating that innovativeness is brand-specific.

6.3.4 Corporate Social Responsibility

Contrary to expectations, parent brand corporate social responsibility has a

positive effect (0.16) on extension evaluation. In addition, this is larger than both parent

brand quality and innovativeness.

On a brand level, the beta for CSR was consistently high on corporate social

responsibility, with exception of the Saatchi & Saatchi brand, which scored exceptionally

high.

There is a noteworthy difference in the importance of CSR when comparing

service and product brands. The beta coefficient for CSR is higher for service brands

(0.14) than for product brands (0.08). Although this may be brand specific, another

reason for this can be the drawn from the recent corporate scandals involving e.g.

accounting firms and other service companies (e.g. Enron).

Although the importance of this coefficient may have been inflated by the fact

that all brands included in the survey were global brands with high degree of perceived

CSR and codes of conduct, the effect of corporate social responsibility on consumer

extension evaluations cannot be discounted for. A likely explanation for the positive and

relatively large effect of the CSR coefficient compared to Keller and Aaker ‘s (1997)

findings is that consumers values corporate social responsibility significantly more than

industrial buyers do. This highlights the difference of perceived social responsibility

among end-consumers vis-à-vis B2B customers.

6.3.5 Environm ental concern

Environmental concern has a small but significant effect on extension evaluation

(0.05). On a brand level, only the beta coefficient for the brand Boeing was significant.

This might be seen as logical since Boeing might be seen as a company that produces

heavy equipment and airplanes with environmental impact. This is in contrast to the

other product brand Intel, which manufactures microprocessors, and can be considered

as being “clean” and having less impact on the environment.

There is a notable difference between services and products –service brands have

a higher beta coefficient than product brands. This is remarkable since services have no

“production processes” that impact the environment.


63

The effect of environmental concern was hypothesized to be zero, but turned out

to be positive. This fact alongside the positive effect of corporate social responsibility

indicates that consumers to a certain extent take ethical considerations when evaluating

parent brands.

6.3.6 Transfer

The first fit variable, TRAN SFER is significant and its effect is substantial (0.27). This

is an important finding that indicates that current skills and resources of a company are

perceived as being transferable from business markets to consumer markets.

This finding was also relevant across all brands with a notably higher beta

coefficient for the Intel brand (0.36) compared to the other brands. A likely explanation

for the higher beta of the Intel brand may be the high extension attitude scores for the

two Intel brand extensions (see table 6.8).

The beta coefficients are the same (0.26) for product and service brands, implying

that consumers do not distinguish between skills and resources in making services or

goods respectively.

6.3.7 Brand concept consistency

The second fit variable, brand concept consistency, is the coefficient with the

most explanatory power (0.42) in the model. Although this variable was introduced in

this study and has not been used in previous studies, it has proven to be a highly

relevant fit variable compared to the variables of fit used in the original and replication

studies (see table 6.4) –adding the fact that there were no interaction effects between

the two fit variables.

Brand concept consistency is also highly relevant on a brand level, with an

exceptionally high beta for the Boeing brand. A likely explanation might be the

interaction effect between transfer and brand concept consistency. This phenomenon

will be further discussed in Section 6.3.7.

There is also an observable difference in beta coefficient weight between service

and product brands. Brand concept consistency has a higher beta for products (0.45)

than services (0.34). A reason for this may be the intangible nature of services, which

makes it more difficult to evaluate their respective brand concepts when no consumer

service offering exists.


64

6.3.8 D ifficult

A positive relationship between extension attitude and perceived difficulty of

producing the extension was hypothesized. Although the variable Difficult was

significant, the hypothesis was rejected because of sign reversal, i.e. the beta coefficient

turned out to be negative (-0.06). This might indicate that some brand extensions were

perceived to be too difficult to make and that this leads to negative extension

evaluation. This might be an important distinction in comparison to “traditional” brand

extensions that operate in the same market etc. For example, in the original Aaker and

Keller (1990) study and replication studies with exception from van Riel, Lemmink and

Ouwersloot’s (2001) study, the Difficult beta coefficients were positive. In the latter study

(van Riel et al., 2001), which tested service brand extensions, the Difficult beta was

negative. In other words, the more difficult an extension is to produce, the higher will be

the evaluation, except in cases when the extension is a service and the parent brand is a

product brand, or –as concluded in the current study –when the extension is a consumer

product/service and the parent brand is a business-to-business brand. This suggests that

there is an invisible “barrier” that brands face when extending brands to unrelated

markets.

6.3.9 Interaction effects

The first hypothesized interaction effect between Innovativeness and Difficult

was omitted because of multicollinearity, hence this effect cannot be measured. The

second hypothesized interaction effect between Transfer and Brand concept

consistency was insignificant on an aggregate level, and on a brand level only significant

for one brand, Boeing. The beta coefficient for this interaction effect was negative in this

case (-0.13), implying that no synergy effect is achieved (i.e. extension evaluation does

not increase even though brand concept consistency is perceived to be high when there

is no category fit). A reasonable explanation for this is that the brand extensions under

the Boeing brand were perceived to be of low quality (see table 6.8).

The beta coefficient of the second interaction effect was also significant for

product brands, although its direction was in opposite of the hypothesis.

6.3.10 G eneral discussion of the current m odel

The current model shows a remarkably high adjusted R2 (0.61). This is an

improvement from previous studies, although a direct comparison is unfair since


65

different variables were used in the regression analysis of the current model. Van Riel et

al. (2001) propose that the possible explanation for the high adjusted R2 of their model

could be explained by the fact that brand extensions have become more common over

the years and the explanatory power of the used constructs have been reinforced by

successful extensions (and thus positive extension evaluations). This seems like a

plausible explanation in the case of consumer brand extensions of business brands:

consumers are able to “make sense” out of these type of extensions.

The model of the current study also seems to have a better fit for products than

services, with respective adjusted R2 values of 0.64 and 0.58.


66

PART IV . CLOSIN G PERSPECTIV ES

Chapter 7. Conclusions 67


67

CH APTER 7.

CON CLU SION S

7.1 Sum m ary of the Findings

In general, the variables that by far have the largest effects in explaining extension

attitude are the two fit variables Transfer and Brand concept consistency. This is inline

with previous studies where the effect of fit variables to a great extent superseded those

of other variables (see table 6.4). Parent brand quality has, compared to previous studies,

surprisingly little effect on the dependent variable. Variables corresponding to Keller and

Aaker’s (1997) study on corporate brand extension play, contrary to expectations, a

significant part in explaining extension evaluation.

7.2 Conclusions

The current study has shown that it indeed is possible to extend business-to-

business brands into the consumer market. It presented the brand extension model of

Aaker and Keller (1990), which was modified to fit the current context by drawing

theories from Broniarczyk and Alba (1994) and Keller and Aaker (1997). More specifically,

this study provided evidence that in the context of business-to-business brand

extensions, consumers use the transferability of skills and resources and brand concept

consistency with the parent brand category as major cues to evaluate extensions. H ence,

abstract factors facilitate extension acceptance beyond the limitations of product-

related similarity. Corporate branding attributes such as innovativeness and corporate

social responsibility also play a large role. The goodness of fit of the current model also

seems to be higher for product brands than services.

7.3 Lim itations and Directions for Future Research

There are some limitations to the present study. A first concern relates to the way

parent brand equity was measured –by measuring parent brand quality with a single-

item measure. Single-item measures are sometimes subject to criticism with respect to


68

their unreliability and low validity (Churchill, 1979). Multiple-item scales is one solution,

but the emphasis should be on constructing better measures of brand equity, in order to

fully assess the extent to which a brand can be leveraged.

A second concern relates to the one-sidedness of the current study. It measured

only consumer acceptance of the brand extensions. A relevant question in the context of

the current research would also address the attitudes of existing B2B customers when

launching consumer brand extensions. In other words, the reciprocal impacts of

consumer brand extensions on brand equity can be measured with respect to both

consumers and business-to-business customers.

A third concern relates to the fit variables. Because of the non-applicability of the

fit variables Substitute and Complement used in previous replication studies, only two fit

variables were used (Transfer and Brand concept consistency). Brand concept

consistency proved to be a useful factor, probably because of its abstractness. Future

studies on brand extensions could include Brand concept consistency as well as the

original fit variables Substitute and Complement to further examine whether the

abstractness of the former or the concreteness of the latter two are superior in attitude

formation.

A fourth limitation relates to the number of brands used in the study. Only five

brands were used, all of which were well-known global brands. Differences in adjusted

R2 on a brand level suggest that there are attributes unique to each brand, which are not

measured by Brand concept consistency. A more detailed study on brand extensions

could take into account numerous factors such as previous extensions (cfr. Keller &

Aaker, 1992), effects of extensions on a company’s brand portfolio (cfr. Dacin & Smith,

1994), or brand architecture (e.g. how brand architecture facilitates brand extension

acceptance).

A fifth limitation concerns the way the brand extension were presented. Since

each brand extension was presented only as a non-branded generic product and

without any accompanying text or visual cues, the extent to which a true assessment of

the quality and likelihood of trial by the consumer might have been limited. Related to

the brand extension presentation is the absence of pricing. Van Riel et al. (2001) suggest

that consumers may use “price clues” to assess [service] quality (Zeithaml, 1988).


69

7.4 Im plications

7.4.1 Theoretical im plications

Bottomley and Doyle (1996) called for further research on the role of Brand

concept consistency as an important factor in determining how consumers form

attitudes towards brand extensions. The current study has proven, at least in the current

context, that brand-specific associations are more important than category similarity, in

consumer attitude formation of B2B-brand extensions. H ence, Broniarczyk and Alba’s

(1994) claim that the “brand” in brand extension is superior to category-based similarity

is supported by the current study.

The present study has also proven that variables of Keller and Aaker’s (1997)

corporate brand extension model can be used in the original A&K model, showing

further evidence that the original A&K model can be contextually adapted.

By replacing irrelevant variables of the A&K model with contextually relevant

concepts and theories as discussed above, the present study shows that a broad

empirical replication of Aaker and Keller’s (1990) is both possible and valuable for

additional explanatory power and insights in more complex or specific cases.

7.4.2 M anagerial im plications

W hether to extend a business-to-business brand to the consumer market or not

remains a predominantly managerial topic. W hat the current study has shown is that it is

possible to do so.

Brand extension strategies in the current context will be most successful when

there is a “fit” between the parent brand and the extension. This fit is determined by the

extent to which consumers perceive that the skills and resources of a company are

useful in making the extension, and more importantly whether the extension is

consistent with the brand concept of the parent brand. Any extension must therefore

begin with examining the parent brand itself, and what it stands for.

The quality of the parent brand plays a lesser role in brand extension acceptance;

although quality should not be discounted since consumers can assess brand equity in

different ways than measured by the current study.

Corporate brand attributes such as corporate social responsibility (CSR) is highly

important and can be achieved by e.g. sponsoring local communities through various

activities and programs, especially for service companies. It is, however, unclear whether


70

CSR is a trend or whether it will remain as an important attribute. N evertheless, a B2B

company should be aware of the difference in ethical values among consumers and

industrial buyers.

Another corporate brand attribute that facilitates brand extension acceptance is

innovativeness. A company should therefore strive to build an innovative reputation and

form a philosophy of constantly launching advanced products or services.

An important practical constraint with respect to the current model is that it is

only tested on B2B brands. The validity or importance of the above is hence not

confirmed in cases after a brand makes the transition from B2B to both B2B and

consumer brand. Some time after a transition is made, i.e. when the former B2B brand is

both a B2B and B2C brand, consumers may evaluate the brand extensions according to

the original model by Aaker and Keller (1990). This pinpoints the context-specificity of

the current model.


71

BIBLIOG RAPH Y

Aaker, D. A. (1996). Building Strong Brands. N ew York: The Free Press.

Aaker, D. A. and Joachimsthaler, E. (2000). Brand Leadership. London: Free Press Business.

Aaker, D. A. and Keller, K. L. (1990). ”Consumer evaluations of brand extensions”. Journal

of M arketing Research, 54(1), 27-41.

Aaker, D. A., Kumar, V. and Day, G . S. (1998) Marketing Research (6th Ed.). N ew York: John

W iley & Sons.

Aaker, J. L. (1997). "Dimensions of brand personality". Journal of M arketing Research,

34(3), 347-356.

Abratt, R. (1989). “A new approach to the corporate image management process”.

Journal of M arketing, 5(1), 63-76.

Albaum, G . (1997). “The Likert scale revisited – and alternative version”. Journal of the

M arket Research Society, 39(2), 331-48.

Anderson, J. C. and N arus, J. A. (2004). Business M arket M anagem ent - Understanding,

Creating and D elivering Value (2nd Ed.). U pper Saddle River: Prentice-H all.

Balachander, S. and G hose, S. (2003). “Reciprocal spillover effects: a strategic benefit of

brand extensions”. Journal of M arketing, 67(1), 4-13.

Barwise, P. (1995). “G ood empirical generalizations”. M arketing Science, 14(3), 29-36.

Berthon, P., H olbrook, M.B. and H ulbert. J.M. (2003). “U nderstanding and managing the

brand space”. M IT Sloan M anagem ent Review , 44(2), 49-54.


72

Bhattacharya, C. B. and Sen, S. (2003). “Consumer-company identification: a framework

for understanding consumers' relationships with companies”. Journal of M arketing,

67(2), 76-88.

Bottomley, P. A. and Doyle, J. R. (1996). “The formation of attitudes towards extensions:

testing and generalising Aaker and Keller's model”. International Journal of

Research in M arketing, 13(4), 365-377.

Bottomley, P. A. and H olden, S. J. S. (2001). “Do we really know how consumers evaluate

brand extensions? Empirical generalizations based on secondary analysis of eight

studies”. Journal of M arketing Research, 38(4), 494-500.

Boush, D. M. and Loken, B. (1991). “A process-tracing study of brand extension

evaluation”. Journal of M arketing Research, 28(1), 16-28.

Brody, C. J. and Dietz, J. (1997). “On the dimensionality of 2-question format Likert scale

attitudes”. Social Science Research, 26(2), 197-204.

Broniarczyk, S. M. and Alba, J. W . (1994). “The importance of the brand in brand

extension”. Journal of M arketing Research, 31(2), 214-239.

Brooks, L. (1978). “N onanalytic concept formation and memory for instances”, in

Cognition and Categorization, Rosch, E. and Lloyd B. B., eds. H illsdale, N J: Lawrence

H illbaum Associates, 169-215.

Brown, T. J. and Dacin, P. A. (1997). “The company and the product: corporate

associations and consumer product responses”. Journal of M arketing, 61(1), 68-84.

Churchill, G . A. Jr. (1979). “A paradigm for developing better measures of marketing

constructs”. Journal of M arketing Research, 16, 64-73.

Cohen, J. B. (1982). “The role of affect in categorization: towards a reconsideration of the

concept of attitude”, in Advances in Consum er research, 9, Mitchel, A., ed. Ann

Arbor, MI: Association for Consumer Research, 94-100.


73

Cronbach, L. J. (1951) “Coefficient alpha and the internal structure of tests”,

Psychom etrica, 16, 297-334.

Dacin, P. A. and Smith, D. C. (1994). “The effect of brand portfolio characteristics on

consumer evaluations of brand extensions”. Journal of M arketing Research, 31, 229-

242.

de Chernatony, L. and Segal-H orn, S. (2001). “Building on services' characteristics to

develop successful services brands”. Journal of M arketing M anagem ent, 17(7/8),

645-669.

de Ruyter, K. and W etzels, M. (2000). “The role of corporate image and extension

similarity in service brand extensions”. Journal of Econom ic Psychology, 21(6), 639-

660.

Douglas, S. P., Craig, C. S. and N ijssen, E. J. (2001). “Executive insights: integrating

branding strategy across markets: building international brand architecture”.

Journal of International M arketing, 9(2), 97-114.

Fiske, S. T. (1982). “Schema-triggered affect: applications to social perception” in Affect

and Cognition: The 17th Annual Carnegie Sym posium on Cognition, Clark, M. S. and

Fiske, S. T., eds. H illsdale, N J: Lawrence Erlbaum Associates, 171-190.

Fiske, S. T. and Pavelchak, M. A. (1986). “Category-based versus piecemeal-based

affective responses: developments in schema-triggered affect”, in The Handbook of

M otivation and Cognition: Foundation for Social Behaviour, Sorrentino, R. W . and

H iggins, E. T., eds. N ew York: G uilford Press, 167-203.

G riffin, J. J. (2002). “Branding as strategy: new forms of decision-processes”. Corporate

Reputation Review , 5(2/3), 227-240.

H air, J. F., Anderson, R. E., Tatham, R. L. and Black, W . C. (1998). M ultivariate D ata Analysis

(5th Ed.). U pper Saddle River, N J: Prentice H all.


74

H enderson, J. M. and Quandt, R. E. (1980). M icroeconom ic theory: a m athem atical

approach. N ew York: McG raw-H ill.

H ubbard, R. and Armstrong, J. S. (1994). “Replications and extensions in marketing: rarely

published but quite contrary”. International Journal of Research in M arketing, 11(3),

233-249.

Ind, N . (1997). The Corporate Brand. London: McMillan Press.

Ind, N . (1992). The Corporate Im age (revised ed.). London: Kogan Page.

Kapferer, J-N . (2000). Strategic Brand M anagem ent. London: Kogan Page.

Keller, K. L. (2003). “Brand equity dilution”. M IT Sloan M anagem ent Review , 45(1), 12-14.

Keller, K. L. (2003). Strategic Brand M anagem ent: Building, M easuring, and M anaging Brand

Equity. U pper Saddle River: Prentice H all.

Keller, K. L. and Aaker, D. A. (1992). “The effects of sequential introduction of brand

extensions”. Journal of M arketing Research, 29, 35-50.

Keller, K. L. and Aaker, D. A. (1997). “Managing the corporate brand: the effect of

corporate marketing activity on consumer evaluations of brand extensions”.

Cambridge, MA: Marketing Science Institute.

Kitchin, T. (2003). “Corporate social responsibility: a brand explanation”. Brand

M anagem ent, 10(4-5), 312-326.

Krishnan, B. C. and H artline, M. D. (2001). “Brand equity: is it more important in services?”.

Journal of Services M arketing, 15(5), 328-342.

Lambert, Z. V. (1972). “Perceptual patterns, information handling and innovativeness”.

Journal of M arketing Research, 9(4), 427.


75

Lance, Charles E. (1988), “Residual centering, exploratory and confirmatory moderator

analysis, and decomposition of effects in path models containing interactions,”

Applied Psychological M easurem ent, 12(2), 163-175.

Leone, R. P. and Schultz, R. L. (1980). “A study of Marketing G eneralizations”. Journal of

M arketing, 44(1), 11-18.

Likert, R. (1932). “A technique for the measurements of attitudes”. Archives of Psychology,

140.

Loken, B. and John, D. R. (1993). “Diluting brand beliefs: when do brand extensions have

a negative impact?”. Journal of M arketing, 57(3), 71-85.

MacInnis, D. J. and N akamoto, K. (1990). “Examining factors that influence the perceived

goodness of brand extensions”. W orking Paper #54, Karl Eller G raduate School of

Management, U niversity of Arizona.

Mackay, M. M. (2001). “Application of brand equity measures in service markets”. Journal

of Services M arketing, 15(3), 210-221.

Malhotra, N . K. and Birks, D. F. (2003). M arketing Research: An Applied Approach (3rd Ed.).

H arlow: Prentice-H all.

Martin, E. and Polivka, A. E. (1995). “Diagnostics for redesigning survey questionnaires –

measuring work in the current population survey”. Public O pinion Q uarterly, 59(4),

547-564.

McAlexander, J. H ., Schouten, J. W . and Koenig, H . F. (2002). “Building brand community”.

Journal of M arketing, 66(1), 38-54.

McClave, J. T., Benson, P. G . and Sinich, T. (1998). Statistics for Business and Econom ics (7th

Ed.). U pper Saddle River: Prentice-H all.


76

Mervis, C. B. and Rosch, E. (1981). “Categorization of N atural Objects”. Annual Review of

Psychology, 32, 89-115.

Michell, P., King, J. and Reast, J. (2001). “Brand values related to industrial products”.

Industrial M arketing M anagem ent, 30(5), 415-425.

Moorthi, Y. L. R. (2002). “An approach to branding services”. Journal of Services M arketing,

16(3), 259-274.

Morrin, M. (1999). “The impact of brand extensions on parent brand memory structures

and retrieval processes”. Journal of M arketing Research, 36(4), 517-526.

Mudambi, S. (2002). “Branding importance in business-to-business markets -Three buyer

clusters”. Industrial M arketing M anagem ent, 31(6), 525-533.

N eter, J., W asserman, W . and Kutner, M. H . (1990), Applied Linear Statistical M odels:

Regression, Analysis of Variance, and Experim ental D esigns. Burr Ridge, IL: Irwin.

O'Cass, A. and G race, D. (2003). “An exploratory perspective of service brand

associations”. Journal of Services M arketing, 17(5), 452-475.

Olsen, B. L. (n.d.) “Regression Analysis: Issues of Multicollinearity too often Overlooked”

[PDF document], retrieved 7th July 2004 from

http://www.iaca.net/resources/Articles/multicollinearity_article.pdf

Ostlund, L. E. (1974). “Perceived innovation attributes as predictors of innovativeness”.

Journal of Consum er Research, 1(2), 23-29.

Peterson, R. A. (1994). “A meta-analysis of Cronbach’s coefficient alpha”, Journal of

Consum er research, 21, 381-91.

Porter, M. (1998). Com petitive Advantage -Creating and Sustaining Superior Perform ance.

N ew York: Free Press.


77

Rajagopal and Sanchez, R. (2004). “Conceptual analysis of brand architecture and

relationships within product categories”. Brand M anagem ent, 11(3), 233-247.

Riel, A.C.R. van, Lemmink, J. and Ouwersloot, H . (2001). “Consumer evaluations of service

brand extensions”. Journal of Service Research, 3(3), 220-231.

Roerich, G . (2004). “Consumer innovativeness: concepts and measurements”. Journal of

Business Research, 57(6), 671-677.

Sunde, L. and Brodie, R.J., (1993). “Consumer evaluations of brand extensions: Further

empirical evidence”. International Journal of Research in M arketing, 10, 47-53.

The 100 Top Brands (2004, August). Business W eek, 3864, 68-71.

Thomas, R. L. (1997). M odern Econom etrics –an Introduction. Essex: Addison-W esley.

U rde, M. (1999). “Brand orientation: a mindset for building brands into strategic

resources”. Journal of M arketing M anagem ent, 15(1-3), 117-133.

U rde, M. (2003). “Core value-based corporate brand building”. European Journal of

M arketing, 37(7/8), 1017-1040.

W ebster, F. E. Jr. and Keller, K. L. (2004). “A roadmap for branding in industrial markets”.

Brand M anagem ent, 11(5), 388-402.

Yoo, B., Donthu, N . and Lee, S. (2000). “An examination of selected marketing mix

elements and brand equity”. Journal of the Academ y of M arketing Science, 28(2),

195-211.

Zeithaml, V. A. (1988). “Consumer perceptions of price, quality, and value: a means-end

model and synthesis of evidence”. Journal of M arketing, 52, 2-22.


78

APPEN DIX A. QU ESTION N AIRES A.1 Pilot-testing Questionnaire

THESIS SURVEY: PILOT-TESTING Hello,

I am conducting a survey for my final thesis on brands. To make my survey as robust as possible, I

am performing a pilot-test (of the questionnaire I will be using later in my research) for which you

have been invited. Instructions are as follows:

• Your task is to fill out the questionnaire like you usually do, but in addition to the

questionnaire questions I would like you to also think about the quality of the questionnaire

itself.

• You can write down your opinions on the questionnaire on the FEEDBACK FORM which is

found on the last two pages.

• You are also allowed to make notes/markings on the questionnaire itself, but if you do,

please make sure that these are readable.

• Since the pilot-testing only samples a small number of people, it is important that you do

your best to answer all questions of the feedback form.

• Please be as thorough and clear as possible!

The “real” questionnaire starts on the next page.

Please return this questionnaire to me as soon you can. Thanks for your help!

Leon

Please write down your personal details (you can skip this part in the actual questionnaire).

Name Age Gender � Male � Female


79

Nationality � Dutch � German � Other, namely

Study programme � IB � IM � IES � FE � INF � Exchange student

If you are an exchange student: do you study business, economics, etc. in your home country?

� Yes � No

Year of study � 1 � 2 � 3 � 4 � 5 or more


80

SURVEY ON BRAND EXTENSIONS

Dear participant,

My name is Leon Phang, and I am conducting a survey for my final thesis on

brands.

I am curious about what you think about different business-to-business brands and

their hypothetical brand extensions.

Please help me do this survey by taking the time to fill out this questionnaire. All

information will be treated in the strictest confidence and results will be produced in

the form of aggregated data only.

Thank you for your time and help.

INSTRUCTIONS

As mentioned, this questionnaire is about business-to-business (B2B) brands and hypothetical

brand extensions. A brand extension is a new product that uses an already existing brand name.

You will be asked about your opinion about 5 different brands. The questionnaire will take some

time and effort to fill out, so please try to do your best and make both your and my time worthwhile!


81

BRAND 1 OUT OF 5 Consider the following brand:

Now, please write down your immediate associations with this brand in the empty box:

POOR OUTSTANDING How do you perceive the overall quality of this brand? 1 2 3 4 5 6 7 NOT AT ALL VERY This brand is modern and up-to-date 1 2 3 4 5 6 7 This brand invests in R&D 1 2 3 4 5 6 7 This brand introduces the latest product/service features 1 2 3 4 5 6 7 This brand is concerned with the environment 1 2 3 4 5 6 7 This brand is involved in helping its community 1 2 3 4 5 6 7 This brand is socially responsible 1 2 3 4 5 6 7

Also consider the following brand extensions, and please write down your immediate associations to

these in the empty box:

a) Reuters 24 hour news TV channel

b) Reuters news radio station


82

Now, please indicate your opinion on the previously mentioned hypothetical brand extensions by

marking the appropriate number (1-7). a) Consider the brand extension Reuters 24 hour news TV channel again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Reuters are useful to make this extension 1 2 3 4 5 6 7 The resources of Reuters are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Reuter brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Reuter brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7

b) Consider the brand extension Reuters news radio station again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Reuters are useful to make this extension 1 2 3 4 5 6 7 The resources of Reuters are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Reuter brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Reuter brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7


83





these in the empty boxes:

a) KPMG Private tax planning service

b) KPMG Institute of Accountancy (Executive MBA education)

c) KPMG Offshore banking service


84


marking the appropriate number (1-7). a) Consider the brand extension KPMG Private tax planning service again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of KPMG are useful to make this extension 1 2 3 4 5 6 7 The resources of KPMG are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the KPMG brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the KPMG brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7

b) Consider the brand extension KPMG Institute of Accountancy (Executive MBA education)

again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of KPMG are useful to make this extension 1 2 3 4 5 6 7 The resources of KPMG are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the KPMG brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the KPMG brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7

c) Consider the brand extension KPMG Offshore banking service again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of KPMG are useful to make this extension 1 2 3 4 5 6 7 The resources of KPMG are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the KPMG brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the KPMG brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7


85

BRAND 3 OUT OF 5 1) Consider the following brand:

2) Now, please write down your immediate associations with this brand in the empty box:

POOR OUTSTANDING 3) How do you perceive the overall quality of this brand? 1 2 3 4 5 6 7 NOT AT ALL VERY 4) This brand is modern and up-to-date 1 2 3 4 5 6 7 5) This brand invests in R&D 1 2 3 4 5 6 7 6) This brand introduces the latest product/service features 1 2 3 4 5 6 7 7 This brand is concerned with the environment 1 2 3 4 5 6 7 8) This brand is involved in helping its community 1 2 3 4 5 6 7 9) This brand is socially responsible 1 2 3 4 5 6 7



a) Saatchi & Saatchi marketing & advertising books

b) Saatchi & Saatchi Institute of Advertising (executive training courses)

c) Saatchi & Saatchi Espresso Bar & Café


86


marking the appropriate number (1-7).

a) Consider the brand extension Saatchi & Saatchi marketing & advertising books again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Saatchi & Saatchi are useful to make this extension 1 2 3 4 5 6 7 The resources of Saatchi & Saatchi are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Saatchi & Saatchi brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Saatchi & Saatchi brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7

b) Consider the brand extension Saatchi & Saatchi Institute of Advertising (executive

training courses) again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Saatchi & Saatchi are useful to make this extension 1 2 3 4 5 6 7 The resources of Saatchi & Saatchi are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Saatchi & Saatchi brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Saatchi & Saatchi brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7

c) Consider the brand extension Saatchi & Saatchi Espresso Bar & Café again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Saatchi & Saatchi are useful to make this extension 1 2 3 4 5 6 7 The resources of Saatchi & Saatchi are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Saatchi & Saatchi brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Saatchi & Saatchi brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7


87






a) Intel® Portable MP3 Music Player

b) Intel® Home Wireless (wi-fi) Network Hardware

c) Intel® Notebook PC


88


marking the appropriate number (1-7).

a) Consider the brand extension Intel® Portable MP3 Player again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Intel are useful to make this extension 1 2 3 4 5 6 7 The resources of Intel are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Intel brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Intel brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7

b) Consider the brand extension Intel® Home Wireless (wi-fi) Network Hardware again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Intel are useful to make this extension 1 2 3 4 5 6 7 The resources of Intel are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Intel brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Intel brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7

c) Consider the brand extension Intel® Notebook PC again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Intel are useful to make this extension 1 2 3 4 5 6 7 The resources of Intel are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Intel brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Intel brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7


89



POOR OUTSTANDING 3) How do you perceive the overall quality of this brand? 1 2 3 4 5 6 7 NOT AT ALL VERY 4) This brand is modern and up-to-date 1 2 3 4 5 6 7 5) This brand invests in R&D 1 2 3 4 5 6 7 6) This brand introduces the latest product/service features 1 2 3 4 5 6 7 7 This brand is concerned with the environment 1 2 3 4 5 6 7 8) This brand is involved in helping its community 1 2 3 4 5 6 7 9) This brand is socially responsible 1 2 3 4 5 6 7

3) Also consider the following brand extensions, and please write down your immediate association

to these in the empty box:

a) Boeing chronograph wristwatch

b) Boeing flight simulator computer game

c) Boeing travel luggage


90

4) Now, please indicate your opinion on the previously mentioned hypothetical brand

extensions by placing a tick in the appropriate box. a) Consider the brand extension Boeing chronograph wristwatch again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Boeing are useful to make this extension 1 2 3 4 5 6 7 The resources of Boeing are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Boeing brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Boeing brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7

b) Consider the brand extension Boeing flight simulator computer game again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Boeing are useful to make this extension 1 2 3 4 5 6 7 The resources of Boeing are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Boeing brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Boeing brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7

c) Consider the brand extension Boeing travel luggage again. POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Boeing are useful to make this extension 1 2 3 4 5 6 7 The resources of Boeing are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Boeing brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Boeing brand 1 2 3 4 5 6 7 This brand extension is too easy to make 1 2 3 4 5 6 7


91

PERSONAL INFORMATION

Name (this will be used for administrational purposes only)

Age Gender: � Male � Female

Do you study business or economics? � Yes � No

Nationality

� Dutch

� German

� Other, namely:


92

PILOT-TESTING FEEDBACK FORM

Please give your opinions, suggestions, and possible corrections concerning the following aspects: Instructions Content of questions Wording (spelling, grammar, choice of words etc.) Sequence of questions Form and layout

Please turn to the next page >>


93

Question difficulty Any other comments, remarks, suggestions, etc?


94

A.2 Questionnaire (Final version)

SURVEY ON BRAND EXTENSIONS

Dear participant,

My name is Leon Phang, and I am conducting a survey for my final thesis in brand

management.

I am curious about what you think about various business-to-business brands and

their hypothetical brand extensions for the consumer market.

Please help me do this survey by taking the time to fill out this questionnaire, and I

shall be eternally grateful.

All information will be treated in the strictest confidence and results will be

produced in the form of aggregated data only.

Thank you for your time and help.

INSTRUCTIONS

As mentioned, this questionnaire is about strictly B2B brands and hypothetical B2C brand

extensions. A brand extension is a new product that uses an already existing brand name.

You will be asked about your opinion about 5 different brands. The questionnaire will take some

time and effort to fill out, so please try to do your best and make both your and my time worthwhile!


95

BRAND 1 OUT OF 5

• Consider the following brand

Please indicate your opinion on this brand by marking the appropriate number (1-7).

NOT AT ALL VERY How knowledgeable are you about this brand? 1 2 3 4 5 6 7 POOR OUTSTANDING How do you perceive the overall quality of this brand? 1 2 3 4 5 6 7 NOT AT ALL VERY This brand is modern and up-to-date 1 2 3 4 5 6 7 This brand invests in R&D 1 2 3 4 5 6 7 This brand introduces the latest product/service features 1 2 3 4 5 6 7 This brand is concerned with the environment 1 2 3 4 5 6 7 This brand is involved in helping its community 1 2 3 4 5 6 7 This brand is socially responsible 1 2 3 4 5 6 7

Two hypothetical B2C brand extensions of the above brand will be mentioned.

Please indicate your opinion on these brand extensions by marking the appropriate number (1-7).

• Consider the brand extension

“24-Hour News TV Channel” POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Reuters are useful to make this extension 1 2 3 4 5 6 7 The resources of Reuters are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Reuter brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Reuter brand 1 2 3 4 5 6 7 This brand extension will be difficult to make 1 2 3 4 5 6 7 Reuters is out to make short-term profits with this brand extension 1 2 3 4 5 6 7


”News Radio Station”

POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Reuters are useful to make this extension 1 2 3 4 5 6 7 The resources of Reuters are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Reuter brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Reuter brand 1 2 3 4 5 6 7 This brand extension will be difficult to make 1 2 3 4 5 6 7 Reuters is out to make short-term profits with this brand extension 1 2 3 4 5 6 7


96

BRAND 2 OUT OF 5

• Consider the brand



Three hypothetical B2C brand extensions of the above brand will be mentioned.



“Private tax planning service”.

POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Ernst & Young are useful to make this extension 1 2 3 4 5 6 7 The resources of Ernst & Young are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Ernst & Young brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Ernst & Young brand 1 2 3 4 5 6 7 This brand extension will be difficult to make 1 2 3 4 5 6 7 Ernst & Young is out to make short-term profits with this brand extension 1 2 3 4 5 6 7


97


“Institute of Accountancy” (Executive MBA education).



“Offshore banking service”.



98

BRAND 3 OUT OF 5







“Marketing & Advertising Books” POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Saatchi & Saatchi are useful to make this extension 1 2 3 4 5 6 7 The resources of Saatchi & Saatchi are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Saatchi & Saatchi brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Saatchi & Saatchi brand 1 2 3 4 5 6 7 This brand extension will be difficult to make 1 2 3 4 5 6 7 Saatchi & Saatchi is out to make short-term profits with this brand extension 1 2 3 4 5 6 7


“Institute of Advertising” (executive training courses) POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Saatchi & Saatchi are useful to make this extension 1 2 3 4 5 6 7 The resources of Saatchi & Saatchi are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Saatchi & Saatchi brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Saatchi & Saatchi brand 1 2 3 4 5 6 7 This brand extension will be difficult to make 1 2 3 4 5 6 7 Saatchi & Saatchi is out to make short-term profits with this brand extension 1 2 3 4 5 6 7


99

BRAND 4 OUT OF 5







”Portable MP3 Player” POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Intel are useful to make this extension 1 2 3 4 5 6 7 The resources of Intel are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Intel brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Intel brand 1 2 3 4 5 6 7 This brand extension will be difficult to make 1 2 3 4 5 6 7 Intel is out to make short-term profits with this brand extension 1 2 3 4 5 6 7


100


”Home Wireless (wi-fi) Network Hardware” POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Intel are useful to make this extension 1 2 3 4 5 6 7 The resources of Intel are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Intel brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Intel brand 1 2 3 4 5 6 7 This brand extension will be difficult to make 1 2 3 4 5 6 7 Intel is out to make short-term profits with this brand extension 1 2 3 4 5 6 7


”Notebook PC” POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Intel are useful to make this extension 1 2 3 4 5 6 7 The resources of Intel are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Intel brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Intel brand 1 2 3 4 5 6 7 This brand extension will be difficult to make 1 2 3 4 5 6 7 Intel is out to make short-term profits with this brand extension 1 2 3 4 5 6 7


101

BRAND 5 OUT OF 5







“Chronograph wristwatch” POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Boeing are useful to make this extension 1 2 3 4 5 6 7 The resources of Boeing are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Boeing brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Boeing brand 1 2 3 4 5 6 7 This brand extension will be difficult to make 1 2 3 4 5 6 7 Boeing is out to make short-term profits with this brand extension 1 2 3 4 5 6 7


“Flight simulator computer game” POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Boeing are useful to make this extension 1 2 3 4 5 6 7 The resources of Boeing are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Boeing brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Boeing brand 1 2 3 4 5 6 7 This brand extension will be difficult to make 1 2 3 4 5 6 7 Boeing is out to make short-term profits with this brand extension 1 2 3 4 5 6 7


102


“Travel luggage” POOR OUTSTANDING How do you perceive the overall quality of this brand extension? 1 2 3 4 5 6 7 TOTALLY DISAGREE TOTALLY AGREE If this brand extension existed, I would be likely to try it 1 2 3 4 5 6 7 The competences of Boeing are useful to make this extension 1 2 3 4 5 6 7 The resources of Boeing are useful to make this extension 1 2 3 4 5 6 7 This brand extension is consistent with the Boeing brand 1 2 3 4 5 6 7 This brand extension fits with my associations of the Boeing brand 1 2 3 4 5 6 7 This brand extension will be difficult to make 1 2 3 4 5 6 7 Boeing is out to make short-term profits with this brand extension 1 2 3 4 5 6 7

PERSONAL INFORMATION

Name (this will be used for administrational purposes only)

Age Gender: � Male � Female

Do you study business or economics? � Yes � No

Nationality

� Dutch

� German

� Other, namely:


103

APPEN DIX B. SPSS OU TPU TS B.1 Reliability Analyses B.1.1 Attitude tow ards brand extension R E L I A B I L I T Y A N A L Y S I S - S C A L E (A L P H A) Mean Std Dev Cases 1. EXTQUAL 4.7767 1.4177 1124.0 2. EXTTRIAL 3.9929 1.7649 1124.0 N of Cases = 1124.0 N of Statistics for Mean Variance Std Dev Variables Scale 8.7696 8.1597 2.8565 2 Item Means Mean Minimum Maximum Range Max/Min Variance 4.3848 3.9929 4.7767 .7838 1.1963 .3072 Reliability Coefficients 2 items Alpha = .7439 Standardized item alpha = .7551

B.1.2 Innovativeness R E L I A B I L I T Y A N A L Y S I S - S C A L E (A L P H A) Mean Std Dev Cases 1. INNOV1 5.0872 1.3162 1113.0 2. INNOV2 5.0916 1.5555 1113.0 3. INNOV3 5.1024 1.4200 1113.0 N of Statistics for Mean Variance Std Dev Variables SCALE 15.2812 14.6106 3.8224 3 Item-total Statistics Scale Scale Corrected Mean Variance Item- Alpha if Item if Item Total if Item Deleted Deleted Correlation Deleted INNOV1 10.1941 7.8148 .6880 .8647 INNOV2 10.1896 6.4074 .7344 .8298 INNOV3 10.1788 6.5570 .8301 .7335 Reliability Coefficients N of Cases = 1113.0 N of Items = 3 Alpha = .8667


104

B.1.3 Corporate Social Responsibility R E L I A B I L I T Y A N A L Y S I S - S C A L E (A L P H A) Mean Std Dev Cases 1. SOCIAL 3.8978 1.3330 1125.0 2. COMMUN 3.8667 1.3036 1125.0 N of Cases = 1125.0 N of Statistics for Mean Variance Std Dev Variables Scale 7.7644 6.0290 2.4554 2 Item Means Mean Minimum Maximum Range Max/Min Variance 3.8822 3.8667 3.8978 .0311 1.0080 .0005 Reliability Coefficients 2 items Alpha = .8468 Standardized item alpha = .8470

B.1.4 Transfer R E L I A B I L I T Y A N A L Y S I S - S C A L E (A L P H A) Mean Std Dev Cases 1. TRANSFE1 4.7384 1.5765 1124.0 2. TRANSFE2 4.8167 1.5170 1124.0 N of Statistics for Mean Variance Std Dev Variables SCALE 9.5552 8.9507 2.9918 2 Item-total Statistics Scale Scale Corrected Mean Variance Item- Alpha if Item if Item Total if Item Deleted Deleted Correlation Deleted TRANSFE1 4.8167 2.3012 .8706 . TRANSFE2 4.7384 2.4854 .8706 . Reliability Coefficients N of Cases = 1124.0 N of Items = 2 Alpha = .9304


105

B.1.5 Brand concept consistency R E L I A B I L I T Y A N A L Y S I S - S C A L E (A L P H A) Mean Std Dev Cases 1. BCC1 4.5009 1.6452 1114.0 2. BCC2 4.3348 1.7094 1114.0 N of Statistics for Mean Variance Std Dev Variables SCALE 8.8357 10.5723 3.2515 2 Item-total Statistics Scale Scale Corrected Mean Variance Item- Alpha if Item if Item Total if Item Deleted Deleted Correlation Deleted BCC1 4.3348 2.9219 .8790 . BCC2 4.5009 2.7066 .8790 . Reliability Coefficients N of Cases = 1114.0 N of Items = 2 Alpha = .9352

B.1.6 D ifficult R E L I A B I L I T Y A N A L Y S I S - S C A L E (A L P H A) Mean Std Dev Cases 1. DIFFI1 3.5067 1.5741 1125.0 2. DIFFI2 3.9636 1.4635 1125.0 N of Statistics for Mean Variance Std Dev Variables SCALE 7.4702 5.1497 2.2693 2 Item-total Statistics Scale Scale Corrected Mean Variance Item- Alpha if Item if Item Total if Item Deleted Deleted Correlation Deleted DIFFI1 3.9636 2.1419 .1150 . DIFFI2 3.5067 2.4779 .1150 . Reliability Coefficients N of Cases = 1125.0 N of Items = 2 Alpha = .2058


106

B.2 Regression Analyses B.2.1 Full m odel before residual centering Variables Entered/Removed

Model Variables Entered Variables Removed

Method

1 Difficult, Environmental concern, Parent brand knowledge, Transfer, Parent brand quality, Corporate social responsibility, Innovative, Brand concept consistency, Innovativeness*difficulty, Transfer*brand concept consistency

. Enter

a All requested variables entered. b Dependent Variable: Attitude toward brand extension Model Summary

Model R R Square

Adjusted R Square

Std. Error of the Estimate

1 .785 .617 .613 .8894

a Predictors: (Constant), Difficult, Environmental concern, Parent brand knowledge, Transfer, Parent brand quality, Corporate social responsibility, Innovative, Brand concept consistency, Innovativeness*difficulty, Transfer*brand concept consistency ANOVA

Model Sum of Squares

df Mean Square

F Sig.

1 Regression 1429.084

10 142.908 180.642 .000

Residual 887.630 1122 .791

Total 2316.714

1132

a Predictors: (Constant), Difficult, Environmental concern, Parent brand knowledge, Transfer, Parent brand quality, Corporate social responsibility, Innovative, Brand concept consistency, Innovativeness*difficulty, Transfer*brand concept consistency b Dependent Variable: Attitude toward brand extension


107

Coefficients

Unstandardized Coefficients

Standardized Coefficients

t Sig. Collinearity Statistics

Model B Std. Error

Beta Tolerance VIF

1 (Constant) -.521 .340 -1.532 .126


-2.254E-02 .019 -.031 -1.200 .230 .524 1.908

Parent brand quality 7.587E-02 .033 .064 2.314 .021 .448 2.234

Environmental concern 5.802E-02 .028 .054 2.092 .037 .513 1.951

Corporate social responsibility

.181 .030 .155 5.960 .000 .507 1.971

Innovative .162 .050 .144 3.242 .001 .173 5.795

Transfer .310 .049 .324 6.301 .000 .129 7.733


.444 .060 .505 7.425 .000 .074 13.540

Innovativeness*difficulty -6.840E-04 .012 -.005 -.058 .954 .053 18.912

Transfer*brand concept consistency

-1.552E-02 .011 -.145 -1.444 .149 .034 29.714

Difficult -5.596E-02 .066 -.061 -.847 .397 .065 15.384 a Dependent Variable: Attitude toward brand extension


108

B.2.2 Full m odel after residual centering Variables Entered/Removed


Method

1 Innovativeness*difficulty (resid), Transfer, Environmental concern, Transfer*brand concept consistency (resid), Parent brand knowledge, Parent brand quality, Corporate social responsibility, Innovative, Brand concept consistency, Difficult

. Enter


Model R R Square

Adjusted R

Square

Std. Error of

the Estimate

1 .785 .617 .613 .8894

a Predictors: (Constant), Innovativeness*difficulty (resid), Transfer, Environmental concern, Transfer*brand concept consistency (resid), Parent brand knowledge, Parent brand quality, Corporate social responsibility, Innovative, Brand concept consistency, Difficult ANOVA


df Mean Square

F Sig.

1 Regression 1429.084 10 142.908 180.642 .000

Residual 887.630 1122 .791

Total 2316.714 1132

a Predictors: (Constant), Innovativeness*difficulty (resid), Transfer, Environmental concern, Transfer*brand concept consistency (resid), Parent brand knowledge, Parent brand quality, Corporate social responsibility, Innovative, Brand concept consistency, Difficult b Dependent Variable: Attitude toward brand extension


109

Coefficients




Model B Std. Error

Beta Tolerance VIF

1 (Constant) -.254 .348 -.730 .466


-2.254E-02 .019 -.031 -1.200 .230 .524 1.908


Environmental concern 5.802E-02 .028 .054 2.092 .037 .513 1.951


.181 .030 .155 5.960 .000 .507 1.971

Innovative .159 .032 .142 5.046 .000 .432 2.316

Transfer .252 .037 .263 6.814 .000 .230 4.354


.366 .028 .416 13.283 .000 .347 2.879

Difficult -5.596E-02 .066 -.061 -.847 .397 .065 15.384

Transfer*brand concept consistency (resid)

-1.552E-02 .011 -.028 -1.444 .149 .884 1.131

Innovativeness*difficulty (resid)

-6.840E-04 .012 -.004 -.058 .954 .072 13.925

a Dependent Variable: Attitude toward brand extension


110

B.2.3 Full m odel after residual centering and om itted m oderator variable Variables Entered/Removed


Method

1 Service brand, Difficult, Environmental concern, Transfer*brand concept consistency[Residual], Parent brand quality, Transfer, Parent brand knowledge, Corporate social responsibility, Innovative, Brand concept consistency

. Enter


Model R R Square

Adjusted R Square

Std. Error of the

Estimate

1 .786 .617 .614 .8892

a Predictors: (Constant), Service brand, Difficult, Environmental concern, Transfer*brand concept consistency[Residual], Parent brand quality, Transfer, Parent brand knowledge, Corporate social responsibility, Innovative, Brand concept consistency ANOVA


df Mean Square

F Sig.

1 Regression 1429.521 10 142.952 180.786 .000

Residual 887.193 1122 .791

Total 2316.714 1132

a Predictors: (Constant), Service brand, Difficult, Environmental concern, Transfer*brand concept consistency[Residual], Parent brand quality, Transfer, Parent brand knowledge, Corporate social responsibility, Innovative, Brand concept consistency b Dependent Variable: Attitude toward brand extension


111

Coefficients




Model B Std. Error

Beta Tolerance VIF

1 (Constant) -.200 .177 -1.131 .258


-2.506E-02 .019 -.034 -1.318 .188 .512 1.955


Innovative .148 .034 .132 4.306 .000 .362 2.766


.181 .030 .155 5.978 .000 .507 1.972

Environmental concern

5.843E-02 .028 .054 2.107 .035 .513 1.951

Transfer .256 .030 .268 8.412 .000 .337 2.964


.368 .028 .418 13.335 .000 .348 2.877

Difficult -5.757E-02 .019 -.063 -3.029 .003 .785 1.274

Transfer*brand concept consistency[Residual]

-1.622E-02 .011 -.030 -1.505 .133 .878 1.139

Service brand -5.174E-02 .069 -.018 -.746 .456 .585 1.710 a Dependent Variable: Attitude toward brand extension


112

B.2.4 Regression at brand level: brand 1 Model Summary

R R Square Adjusted R Square

Std. Error of

the Estimate

Model +Brand = Reuters

(Selected)

1 .727 .528 .507 .9580

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Environmental concern, Difficult, Parent brand knowledge, Innovative, Parent brand quality, Brand concept consistency, Corporate social responsibility, Transfer ANOVA


df Mean Square

F Sig.

1 Regression 201.360 9 22.373 24.379 .000

Residual 179.874 196 .918

Total 381.234 205

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Environmental concern, Difficult, Parent brand knowledge, Innovative, Parent brand quality, Brand concept consistency, Corporate social responsibility, Transfer b Dependent Variable: Attitude toward brand extension c Selecting only cases for which +Brand = Reuters


113

Coefficients




Model B Std. Error

Beta Tolerance VIF

1 (Constant) .684 .550 1.243 .215


6.914E-02 .052 .091 1.329 .186 .508 1.967

Parent brand quality -7.240E-02 .078 -.067 -.923 .357 .454 2.200

Innovative 8.317E-02 .080 .067 1.041 .299 .584 1.712


.221 .081 .212 2.710 .007 .393 2.543


5.707E-02 .077 .055 .744 .458 .439 2.278

Transfer .233 .096 .202 2.416 .017 .343 2.913


.352 .081 .349 4.369 .000 .378 2.643

Difficult -7.587E-02 .052 -.082 -1.471 .143 .778 1.285


1.844E-02 .037 .033 .496 .621 .556 1.799

a Dependent Variable: Attitude toward brand extension b Selecting only cases for which +Brand = Reuters


114



Std. Error of the

Estimate

Model +Brand = Ernst & Young

(Selected)

1 .805 .649 .632 .8285

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Innovative, Difficult, Corporate social responsibility, Parent brand knowledge, Environmental concern, Brand concept consistency, Parent brand quality, Transfer ANOVA


df Mean Square

F Sig.

1 Regression 248.338 9 27.593 40.195 .000

Residual 134.551 196 .686

Total 382.888 205

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Innovative, Difficult, Corporate social responsibility, Parent brand knowledge, Environmental concern, Brand concept consistency, Parent brand quality, Transfer b Dependent Variable: Attitude toward brand extension c Selecting only cases for which +Brand = Ernst & Young


115

Coefficients




Model B Std. Error

Beta Tolerance VIF

1 (Constant) -.293 .426 -.687 .493


-9.209E-02 .044 -.110 -2.074 .039 .633 1.579

Parent brand quality .121 .068 .104 1.777 .077 .527 1.898

Innovative .204 .073 .176 2.813 .005 .460 2.174


.184 .061 .149 3.007 .003 .734 1.362


9.772E-02 .060 .086 1.618 .107 .635 1.574

Transfer .287 .081 .243 3.567 .000 .385 2.598


.301 .061 .337 4.947 .000 .387 2.586

Difficult -.132 .042 -.154 -3.149 .002 .749 1.335


-1.581E-03 .033 -.002 -.047 .962 .696 1.437

a Dependent Variable: Attitude toward brand extension b Selecting only cases for which +Brand = Ernst & Young


116



Std. Error of the

Estimate

Model +Brand = Saatchi &

Saatchi (Selected)

1 .785 .617 .599 .7307

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Corporate social responsibility, Difficult, Parent brand knowledge, Brand concept consistency, Parent brand quality, Environmental concern, Transfer, Innovative ANOVA


df Mean Square

F Sig.

1 Regression 168.389 9 18.710 35.040 .000

Residual 104.656 196 .534

Total 273.046 205

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Corporate social responsibility, Difficult, Parent brand knowledge, Brand concept consistency, Parent brand quality, Environmental concern, Transfer, Innovative b Dependent Variable: Attitude toward brand extension c Selecting only cases for which +Brand = Saatchi & Saatchi


117

Coefficients




Model B Std. Error

Beta Tolerance VIF

1 (Constant) -.724 .463 -1.564 .119


3.739E-02 .038 .063 .973 .332 .473 2.115

Parent brand quality 7.194E-02 .093 .067 .773 .441 .258 3.877

Innovative .101 .105 .085 .963 .337 .253 3.951


.509 .103 .379 4.964 .000 .335 2.983


-.110 .082 -.108 -1.334 .184 .297 3.367

Transfer .316 .090 .284 3.523 .001 .300 3.334


.312 .078 .294 4.024 .000 .367 2.722

Difficult -6.868E-02 .044 -.074 -1.553 .122 .866 1.155


-1.385E-02 .040 -.020 -.342 .732 .573 1.746

a Dependent Variable: Attitude toward brand extension b Selecting only cases for which +Brand = Saatchi & Saatchi


118



Std. Error of the

Estimate

Model +Brand = Intel

(Selected)

1 .770 .594 .575 .9211

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Environmental concern, Parent brand knowledge, Parent brand quality, Difficult, Brand concept consistency, Corporate social responsibility, Innovative, Transfer ANOVA


df Mean Square

F Sig.

1 Regression 242.933 9 26.993 31.817 .000

Residual 166.282 196 .848

Total 409.215 205

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Environmental concern, Parent brand knowledge, Parent brand quality, Difficult, Brand concept consistency, Corporate social responsibility, Innovative, Transfer b Dependent Variable: Attitude toward brand extension c Selecting only cases for which +Brand = Intel


119

Coefficients




Model B Std. Error

Beta Tolerance VIF

1 (Constant) -1.092 .565 -1.934 .055


-1.269E-02 .056 -.012 -.226 .822 .718 1.393


Innovative .188 .115 .113 1.629 .105 .433 2.308


.187 .074 .160 2.508 .013 .509 1.963


2.237E-02 .080 .018 .281 .779 .534 1.873

Transfer .375 .094 .361 4.005 .000 .254 3.930


.226 .079 .251 2.847 .005 .268 3.734

Difficult -4.409E-02 .049 -.053 -.893 .373 .597 1.676


-5.030E-03 .032 -.008 -.158 .874 .736 1.359

a Dependent Variable: Attitude toward brand extension b Selecting only cases for which +Brand = Intel


120



Std. Error of the

Estimate

Model +Brand = Boeing

(Selected)

1 .795 .632 .621 .9032

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Parent brand quality, Difficult, Environmental concern, Transfer, Parent brand knowledge, Brand concept consistency, Corporate social responsibility, Innovative ANOVA


df Mean Square

F Sig.

1 Regression 419.341 9 46.593 57.122 .000

Residual 243.889 299 .816

Total 663.230 308

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Parent brand quality, Difficult, Environmental concern, Transfer, Parent brand knowledge, Brand concept consistency, Corporate social responsibility, Innovative b Dependent Variable: Attitude toward brand extension c Selecting only cases for which +Brand = Boeing


121

Coefficients




Model B Std. Error

Beta Tolerance VIF

1 (Constant) .346 .315 1.099 .273


4.881E-02 .046 .053 1.054 .293 .491 2.038

Parent brand quality 3.492E-02 .075 .028 .466 .642 .337 2.969

Innovative 2.857E-02 .078 .024 .366 .714 .295 3.386


.157 .057 .153 2.750 .006 .396 2.526


9.719E-02 .049 .108 1.993 .047 .417 2.398

Transfer .155 .047 .174 3.280 .001 .435 2.299


.404 .047 .481 8.590 .000 .392 2.551

Difficult -9.681E-03 .033 -.011 -.291 .771 .868 1.152


-5.631E-02 .018 -.129 -3.066 .002 .692 1.446

a Dependent Variable: Attitude toward brand extension b Selecting only cases for which +Brand = Boeing


122

B.2.9 Regression: Service brands Model Summary


Std. Error of the

Estimate

Model Service brand = Service

brand (Selected)

1 .766 .587 .580 .8580

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Environmental concern, Difficult, Parent brand knowledge, Innovative, Brand concept consistency, Corporate social responsibility, Parent brand quality, Transfer ANOVA


df Mean Square

F Sig.

1 Regression 635.010 9 70.557 95.847 .000

Residual 447.573 608 .736

Total 1082.583 617

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Environmental concern, Difficult, Parent brand knowledge, Innovative, Brand concept consistency, Corporate social responsibility, Parent brand quality, Transfer b Dependent Variable: Attitude toward brand extension c Selecting only cases for which Service brand = Service brand


123

Coefficients




Model B Std. Error

Beta Tolerance VIF

1 (Constant) -.173 .268 -.646 .518


-3.744E-02 .023 -.057 -1.596

.111 .526 1.901



6.725E-02 .038 .062 1.777 .076 .557 1.794

Innovative .176 .044 .144 4.033 .000 .530 1.886


.210 .042 .176 5.003 .000 .551 1.816

Transfer .287 .050 .256 5.705 .000 .336 2.972


.334 .040 .345 8.315 .000 .396 2.526

Difficult -9.479E-02 .026 -.104 -3.582

.000 .806 1.241


4.065E-03 .020 .007 .200 .841 .643 1.556

a Dependent Variable: Attitude toward brand extension b Selecting only cases for which Service brand = Service brand


124

B.2.10 Regression: Product brands Model Summary


Std. Error of the

Estimate

Model Service brand = Product

brand (Selected)

1 .805 .648 .642 .9233

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Corporate social responsibility, Difficult, Parent brand knowledge, Transfer, Parent brand quality, Environmental concern, Innovative, Brand concept consistency ANOVA


df Mean Square

F Sig.

1 Regression 793.594 9 88.177 103.441 .000

Residual 430.482 505 .852

Total 1224.076 514

a Predictors: (Constant), Transfer*brand concept consistency[Residual], Corporate social responsibility, Difficult, Parent brand knowledge, Transfer, Parent brand quality, Environmental concern, Innovative, Brand concept consistency b Dependent Variable: Attitude toward brand extension c Selecting only cases for which Service brand = Product brand


125

Coefficients




Model B Std. Error

Beta Tolerance VIF

1 (Constant) -.219 .262 -.833 .405


-1.426E-03 .035 -.001 -.041 .967 .613 1.632



6.479E-02 .041 .061 1.563 .119 .462 2.165

Innovative .104 .059 .077 1.747 .081 .358 2.790


.144 .045 .126 3.214 .001 .456 2.191

Transfer .236 .041 .262 5.690 .000 .330 3.033


.388 .039 .455 9.829 .000 .326 3.071

Difficult -2.327E-02 .028 -.026 -.846 .398 .761 1.315


-2.920E-02 .014 -.057 -2.042

.042 .892 1.121

a Dependent Variable: Attitude toward brand extension b Selecting only cases for which Service brand = Product brand

1. Consumer Evaluations of Brand Extentions B2B to B2C

Documents

Transcript of 1. Consumer Evaluations of Brand Extentions B2B to B2C