Professional Documents
Culture Documents
206 BtoB Brandextensions
206 BtoB Brandextensions
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
ABSTR ACT
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-tobusiness brand extensions, consumers use the transferability of skills and resources, and brand concept consistency with the parent brand category as maj cues to evaluate or 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.
CON TEN TS
Abstract List of tables List of figures 3 8 8
CH APTER 1. IN TRODU CTION 1.1 Background:Leveraging the Brand Across Sectors 1.2 Problem Statement 1.3 Contributions of the Study 1.3.1 Theoreticalcontribution 1.3.2 Practicalcontribution 1.4 Thesis Outline
9 9 11 12 12 13 13
CH APTER 3. BU SIN ESS-TO-BU SIN ESS BRAN DIN G 3.1 Introduction to the Chapter 3.2 Corporate Branding Decisions 3.2.1 Corporate strategy 3.2.2 Corporate brand identity
21 21 21 21 22
3.2.3 Corporate brand strategy 3.2.4 Corporate brand architecture 3.3 Corporate Brand Extensions 3.3.1 I nnovativeness 3.3.2 Corporate SocialResponsibil (CSR) ity 3.3.3 Environm entalconcern 3.4 Summary
23 24 25 25 26 26 27
CH APTER 5. RESEARCH DESIG N 5.1 Introduction to the Chapter 5.2 The H ypothesized Model 5.2.1 The originalm odelof Aaker and Keler l 5.2.2 The current model 5.3 Methodology 5.4 Questionnaire Development 5.4.1 Brands and brand extensions 5.4.2 Scal and m ul ing ti-item scal es 5.5 Data Collection 5.5.1 Pil ot-testing 5.5.2 Sam pl and response ing
36 36 36 36 37 39 39 39 40 42 42 43
BIBLIOG RAPH Y APPEN DIX A. QU ESTION N AIRES APPEN DIX B. SPSS OU TPU TS
71 78 103
LIST OF TABLES
Table 1.1 Companies Formerly Operating in Business-to-Business Markets Only Table 1.2 Expansion to Consumer Markets Table 4.1 H ypotheses and Their References to Literature Table 5.1 Overview of B2B Brands and B2C Extensions Table 5.2 Multiple-scale Items Table 5.3 Response Profile:Sampling Type Table 5.4 Response Profile:G ender Table 5.5 Response Profile:N ationality Table 6.1 Cronbach Alpha Coefficients for Multiple-Item Constructs s Table 6.2 VIF Scores of Regression Variables Before and After Residual Centering Table 6.3 VIF Scores of Regression Variables After Omitting Interaction Term Table 6.4 Comparison Effects of Original and Replication Studies Table 6.5 Regression Results:Main and Interaction Effects Table 6.6 Full Model at Brand Level Table 6.7 Summary of H ypothesis Testing Table 6.8 Extension Level Means Table 6.9 Perceived Quality of Parent Brands 11 11 35 40 42 43 43 43 47 49 50 53 54 55 59 60 60
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. U nderstanding 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 razor s 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 most s 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).
10
Country
Founded
Microsoft IBM Morgan Stanley Merrill Lynch Philips Vattenfall Mitsubishi N okia
BASIC computer programming language Tabulating machines Investment banking Stocks and bonds brokerage Carbon-filament lamps H ydro-electric powerplants Shipping Forestry
Year
N ew offeri ng
B2B Offeri at ti e ng m
Microsoft IBM Morgan Stanley Merrill Lynch Philips Vattenfall Mitsubishi N okia
MS-DOS IBM 5150 Personal computer Private wealth management Cash Management Account Radio Electricity, heating, telephony Automobiles Rubber boots
BASIC computer programming language Mainframes, industrial computing systems Full-service banking Retail brokerage, institutional- and investment banking Lamps, medical X-ray tubes Public utility (domestic electricity) H eavy industries Industrial rubber and cables, radio technology
Innovation Innovation Diversification and growth Diversification and growth Innovation Deregulation Diversification and decentralization Diversification
11
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 (1990) brand extension model is feasible with respect to the current s 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.
That is, it does not use the core brands of Aaker and Keller (1990) study. s
12
empirical generalization holds. Concerning the scope, this study will examine whether Aaker and Keller (1990) predictions are extra-sectoral, that is, whether parent brand s 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 (1990) to confirm Bottomley and Doyle (1996) s s 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 Practicalcontribution 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.
13
14
15
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 rel ated 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 rel ating to the parentbrand 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 revitalizationa new or rej uvenated product can be a
16
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.
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 Spil over and substitution effects l 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 symmetricalthat is, forward spillover effects from advertising of a parent brand on choice of a brand extension are limited. 2.3.3 Categoricaland piecem ealeval uation 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 obj ects to one category rather than another (Kapferer, 1997). Mervis and Rosch (1981, p. 89) propose that a category exists whenever two or more distinguishable obj ects 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
18
evaluation. In the first step, the consumer attempts to match a brand extension (or some other new obj ect) 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 evaluation of a brand extension need s not correspond to evaluation of that brand in its original category (ibid.). Three conclusions can be drawn from Broniarczyk and Alba (1994) research: (1) A perceived s 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 j udgments; 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.
19
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.
20
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 history, its beliefs and philosophy, the nature of its s 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.
22
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-tobusiness 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 employeesview of identity (the internal environment), the marketing communication, and stakeholders (the external environment).
Identi ty
Corporate mission Marketing communications strategy
IM AG E IN TERFACE
Core values
Corporate culture
Personal communications
Source:Abratt(1989)
23
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 adj ustment, in order to achieve a sustainable advantageous position for the organization. Consistent with Abratt (1989) idea of feedback loops is G riffin (2002) view that s s 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 Raj agopal 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 (2000) s 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
24
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.
25
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 perceived fit of the brand extension to the s 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 SocialResponsibil (CSR) ity 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' philosophy to improve the s quality of life in local communities through various activities and programs 3.3.3 Environm entalconcern Keller and Aaker (1997) define environmental concern as a firm policy to sell s "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.
26
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.
27
28
Zeithaml (1988) defines perceived quality as a global assessment of a consumer s j udgment 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 offering. This leads to the second hypothesis: s H ypothesis 2:H igher qual perceptions tow ard the business-to-business parentbrand ity are associated w ith m ore favorabl attitudes tow ard the consum er brand extension. e
4.4 Innovati on
The following hypotheses (H 3, H 4 and H 5) relate to Keller and Aaker (1997) study s 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: H ypothesis 3: H igher perceptions of innovativeness tow ard the business-to-business parent brand are associated w ith m ore favorabl attitudes tow ard the consum er e brand extension.
30
and trustworthiness, but has no significant effect on an extension perceived quality. s This leads to the fourth hypothesis: H ypothesis 4:Perceptions of corporate socialresponsibil of the parent business-toity business brand has no effect the 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 resources and s abilities are useful for making the consumer brand extension, the transfer of positive attitudes is expected to be higher. Therefore, the sixth hypothesis is: H ypothesis 6:The transfer of a business-to-business parent brand perceived qual is s ity enhanced w hen the product cl asses (of the parent brand and the consum er brand extension) in som e w ay fittogether. W hen the fitis w eak, the transfer is inhibited.
31
32
extension is perceived to be too easy or difficult to make, consumersattitude 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: H ypothesis 8:The rel ationship betw een the difficul of m aking the consum er product ty cl of the brand extension and the attitude tow ard the brand extension is positive. ass
H ence, there might be an interaction between brand-specific associations and the similarity of the extension category. This leads to the following: H ypothesis 10:A higher l evelof perceived category fit betw een the parent brand and the brand extension strengthens the im pactof perceived extension brand associations.
34
H1
H igher degrees of know l edge aboutthe business-to-business parentbrand are associated w ith m ore favorabl attitudes tow ard the consum er brand e extension. H igher qual perceptions tow ard the business-to-business parentbrand are ity associated w ith m ore favorabl attitudes tow ard the consum er brand e extension. H igher perceptions of innovativeness tow ard the business-to-business parent brand are associated w ith m ore favorabl attitudes tow ard the consum er e brand extension. Perceptions of corporate socialresponsibil of the parentbusiness-toity business brand has no effectthe on the attitude tow ards the consum er brand extension . Perceptions of environm entalconcern of the parentbusiness-to-business brand has no effecton the attitude tow ards the consum er brand extension. The transfer of a business-to-business parentbrand perceived qual is s ity enhanced w hen the productcl asses (of the parentbrand and the consum er brand extension) in som e w ay fittogether. W hen the fitis w eak, the transfer is inhibited. I the brand associations of the consum er brand extension are consistentw ith f brand conceptof the business-to-business parentbrand, the attitude tow ard the brand extension is positive. The rel ationship betw een the difficul of m aking the consum er productcl ty ass of the brand extension and the attitude tow ard the brand extension is positive. A higher l evelof perceived parentbrand innovativeness w eakens the im pact of perceived difficul (to m ake an extension) on brand extension eval ty uation. A higher l evelof perceived parentbrand innovativeness strengthens the im pactof perceived ease(to m ake an extension) on brand extension eval uation. A higher l evelof perceived extension brand associations strengthens the im pactof perceived category fitbetw een the parentbrand and the brand extension.
H2
H3
H4
Adapted from Keller and Aaker (1997) Adapted from Keller and Aaker (1997) Aaker and Keller (1990)
H5 H6
H7
Adapted from Broniarczyk & Alba (1994) Aaker and Keller (1990)
H8
H 9a
Own construct
H 9b
Own construct
H 10
Own construct
35
S = substitute D = difficult = intercept = 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 KN OW LEDG E 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 BRAN D CON CEPT CON SISTEN CY (hypothesis 4). H ence, there will only be two fit variables in the new model TRAN SFER and BRAN D CON CEPT CON SISTEN CY. 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 N OVATIVEN ESS of the parent brand, hypothesis 7 relates to CORPORATE SOCIAL RESPON SIBILITY, and hypothesis 8 corresponds to EN VIRON MEN TAL CON CERN . This gives the following: (2) Y = + 1K + 2Q + 3T + 4B + 5D + 6I + 7C + 8E + , where the dependent variable Y = attitude towards the brand extension = [ purchase + extension] /2, and where the independent variables K = PAREN T BRAN D KN OW LEDG E B = BRAN D CON CEPT CON SISTEN CY I = IN N OVATIVEN ESS C = CORPORATE SOCIAL RESPON SIBILITY E = EN VIRON MEN TAL CON CERN . Lastly, there might be some interaction effects. The first interaction effect is between IN N OVATIVEN ESS and DIFFICU LT, and the second interaction term is between
37
TRAN SFER and BRAN D CON CEPT CON SISTEN CY, corresponding to H ypothesis 9 and 10 respectively. This leads to: (3) Y = + 1K + 2Q + 3T + 4B + 5D + 6I + 7C + 8E + 9ID + 10BT + , where the additional interaction terms ID = IN N OVATIVEN ESS DIFFICU LT BT = BRAN D CON CEPT CON SISTEN CY TRAN SFER. The hypothesized model is summarized in figure 5.1.
Parent brand knowledge Parent brand perceived quality Parent brand corporate social responsibility Parent brand environmental concern
H1
H2
H7
H8
H4 H 10 Transfer H3
H5
H6
38
39
Servi (S)/ ce Ori nalbrand gi Reuters Product (P) S Ori nalProduct or Servi gi ce Information provider
H ypotheti Extensi cal ons U sed i the Survey n N ews radio station, 24 hour news TV channel
Private tax planning service, Institute of Accountancy (Executive MBA education), off-shore banking service
Advertising
Intel
Portable MP3 music player, notebook PC Chronograph wristwatch, flight simulator computer game, travel luggage
Boeing
Commercial airplanes, military aircraft, satellites, missile defense, human space flight, and launch systems
5.4.2 Scal and m ul ing ti-item scal es 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 obj ects (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 maj disadvantage is that or 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 (QU ALITY) of each parent brand as well as each brand extension, a 7-point Likert scale was used (1 = poor, 7 = outstanding). A second 7point scale measure assessed the attributes of the parent brand (IN N OVATIVE, CSR and
40
EN VIRON MEN TAL CON CERN ) where (1 = not at all, 7 = very). A third 7-point scale measured the three fit variables (TRAN SFER, BRAN D CON CEPT CON SISTEN CY, 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 (KN OW LEDG E) 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 N OVATIVE, CSR, TRAN SFER, BRAN D CON CEPT CON SISTEN CY, and DIFFICU LT by means of 7-point Likertscales. 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.
This solves the problem of the absence of a real zero in the scale (van Riel et al., 2001).
41
42
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 pl and response ing 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 U niversity (U M) 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 U M.
43
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.
44
45
6.2 Rel abi i Col i i l ty, l neari and Regressi Anal ty on yses
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 Rel iabil assessm ent ity 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 al pha, 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.
46
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.
6.2.2 M ul ticol inearity l 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.
47
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.
48
TABLE 6.2 VIF Scores of Regression Variables Before and After Residual Centering
Indep endent vari es abl Before Resi dual Centeri ng After Resi dual Centeri ng
I nteraction effects
Innovativeness Difficult* Brand concept consistency Transfer * 18.91 29.71 13.93* 1.13*
* Formally, the two interaction variables are labeled as IN N OVATIVE DIFFICU LT[ Residual] and TRAN SFER BRAN D CON CEPT CON SISTEN CY[ Residual]after the residual centering has been applied.
Replications of Aaker and Keller (1990) model by N ij s ssen and H artman (1994) and by Bottomley and Doyle (1994) apply Lance (1988) residual centering regression s approach, arguing that their researches offer an improved view on the interaction effects. Following N ij ssen 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 N OVATIVEN ESS DIFFICU LT) was regressed on its two component parts (i.e. IN N OVATIVN ESS and DIFFICU LT in the example) using OLS regression. The resulting residuals (e.g. IN N OVATIVEN 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 N OVATIVE and BRAN D CON CEPT CON SISTEN CY, and the interaction term BRAN D
49
CON CEPT 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 N OVATIVEN 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 N OVATIVEN 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
Indep endent vari es abl
Parent brand-specific effects Parent brand knowledge Parent brand quality Innovative CSR Environmental concern Brand extension-specific effects Transfer Brand concept consistency Difficult I nteraction effects Innovativeness Difficult[ Residual] Brand concept consistency Transfer [ Residual]
removed 1.14
50
6.2.3 Regression anal ysis resul ts After residual centering and omitting one interaction term, the hypothesized model can be summarized as follows: (4) Y = + 1K + 2Q + 3I + 4C + 5E + 6T + 7B + 8D + 9BT[ Residual]+ , where the dependent variable Y = attitude towards the brand extension = [ purchase + extension] /2, and where the independent variables K = PAREN T BRAN D KN OW LEDG E Q = PAREN T BRAN D QU ALITY I = IN N OVATIVEN ESS C = CORPORATE SOCIAL RESPON SIBILITY E = EN VIRON MEN TAL CON CERN . T = TRAN SFER B = BRAN D CON CEPT CON SISTEN CY D = DIFFICU LT BT[ Residual]= BRAN D CON CEPT CON SISTEN CY 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 veral m odeleval l uation 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 adj usted usted for degrees of freedom (ibid.). The adj usted R2 for the current model is R2, i.e. adj
51
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 adj usted R 2 values of these tests are shown in table 6.6b.
52
TABLE 6.4 Comparison Effects of Original and Replication Studies (t values are in parentheses)
Aaker and Keller (1990) Quality Transfer Complement Substitute Difficult Adj usted R2 Sample size N /A 0.24 0.17 0.08 N /A N /A 2,140 0.38 (5.3) 0.21 (3.1) 0.29 (4.2) 0.13 (1.9) 0.00 (0.1) 0.48 1,413 Sunde and Brodie (1993) N ij ssen and H artman (1994) 0.25 (9.1) 0.58 (18.9) 0.01 (0.3) .08 (2.7) Omitted 0.48 693 0.22 (11.2) 0.31 (13.6) 0.31 (13.2) 0.18 (7.67) 0.02 (1.22) 0.47 1,358 .25 (12.91) .26 (12.48) .30 (14.48) .19 (8.86) .03 (1.59) 0.43 1,558 0.15 (9.38) 0.34 (13.5) 0.38 (18.7) 0.06 (3.40) -0.09 (-4.1) 0.57 1,616 0.17 (6.78) 0.41 (13.1) 0.20 (7.59) 0.19 (7.13) 0.16 (5.3) 0.53 808 0.07* (2.39) 0.27* (8.41) NA NA -0.06* (-3.03) .61 1,133 Bottomley and Doyle (1996) U K Bottomley and Doyle (1996) N Z Van Riel et al. (2001) Services Van Riel et al. (2001) G oods Current study1
Beta 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.
53
Std. Error
Beta
T-val ue
Constant
-0.20
0.18
-1.13
I nteraction effects
H 9:Innovativeness Difficult H 10:Transfer Brand concept consistency[ Residual] -/+ + Omitted -0.02 0.01 -0.03 -1.50
Controlvariabl es
Service brand R2 Adj usted R2 Sample size n ****p<.000; ***p<.01; **p<.05; *p<.10 -0.05 0.62 0.61 1133 0.07 -0.02 -0.75
54
TABLE 6.6a Full Model at Brand Level (residually centered, t-values in parentheses)
Parent brand-sp eci c effects fi
Product/ servi ce brand S Parent brand know l edge 0.09 (1.33) -0.11** (-2.07) 0.06 (0.97) -0.01 (-0.23) 0.05 (1.05) -0.06 (-1.60) 0.00 (-0.04) Parent brand qual ty i -0.07 (-0.92) 0.10* (1.78) 0.07 (0.77) 0.15*** (2.27) 0.03 (0.47) 0.05 (1.17) 0.08** (1.98) Envi ronm ental concern 0.06 (0.74) 0.09 (1.62) -0.11 (-1.33) 0.02 (0.28) 0.11*** (1.99) 0.18**** (5.00) 0.13*** (3.21)
Brand Reuters
Innovati ve 0.07 (1.04) 0.18*** (2.81) 0.08 (0.96) 0.11 (1.63) 0.02 (0.37) 0.06* (1.78) 0.06 (1.56)
CSR 0.21*** (2.71) 0.15*** (3.01) 0.38**** (4.96) 0.16*** (2.51) 0.15**** (2.75) 0.14**** (4.03) 0.08* (1.75)
Intel
Boeing
Services
Products
TABLE 6.6b Full Model at Brand Level (residually centered, t-values in parentheses)
Brand extensi on-speci c effects fi
Product/ servi ce brand S Brand concept consi stency 0.35**** (4.37) 0.34**** (4.95) 0.29**** (4.02) 0.25*** (2.85) 0.48**** (8.59) 0.34**** (8.32) 0.45**** (9.83)
Interacti effects on
Transfer brand concept consi stency 0.03 (0.50) 0.00 (-0.05) -0.02 (-0.34) -0.01 (-0.16) -0.13**** (-3.07) 0.01 (0.20) -0.06** (-2.04)
Brand Reuters
Transfer 0.20** (2.42) 0.24**** (3.57) 0.28*** (3.52) 0.36**** (4.00) 0.17**** (3.28) 0.26**** (5.71) 0.26**** (5.69)
Di cul ffi t -0.08 (-1.47) -0.15*** (-3.15) -0.07 (-1.55) -0.05 (-0.89) -0.01 (-0.29) -0.10**** (-3.58) -0.03 (-0.85)
Omitted
206
0.63
Omitted
206
0.60
Intel
Omitted
206
0.57
Boeing
Omitted
309
0.62
Services
Omitted
618
0.58
Products
Omitted
515
0.64
55
56
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 rej ected atboth an aggregate l evel(P<0.000) and a brand l evel(al brands;P<0.01). l 6.2.5 H 5:Environm entalconcern The beta coefficient for EN VIRON MEN TAL CON CERN 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). I sum m ary, the fifth hypothesis is rej n ected at P<0.05 on an aggregate l , and evel rej ected 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. 0.20 for Reuters and 0.36 for Intel. I sum m ary, the third hypothesis is accepted at P=0.00. At a brand l , w e accept H 3 n evel for al brands. l 6.2.7 H 7:Brand concept consistency The beta coefficient for the variable BRAN D CON CEPT CON SISTEN CY is significant (P<0.00) and highly substantial. The beta equals 0.42. Thus, the seventh hypothesis is accepted. BRAN D CON CEPT CON SISTEN CY 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 l , H 4 is evel accepted for al brands (P<0.01). l 6.2.8 H 8:D ifficul t 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 On a brand level, TRAN SFER is significant for all of the five brands tested at P<0.05, although their effect vary between
57
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 l evel w ith exception for Ernst & Young. N everthel because of sign reversalw e cannotacceptthe eighth hypothesis for this brand. ess, 6.2.9 H 10:I nteraction effect betw een Transfer and Brand concept consistency The moderator variable TRAN SFER x BRAN D CON CEPT CON SISTEN CY[ Residual] shows no significance (P>0.10). W e must therefore rej the tenth hypothesis. On ect for which the beta coefficient is negative (-0.13) at P<0.000. I sum m ary, w e rej the tenth hypothesis at an aggregate l . W e al rej the n ect evel so ect hypothesis at a brand l evelfor al brands except for Boeing for w hich w e accept it at a 0.000 l significance l . evel a brand level, the moderator variable shows no significance except for the brand Boeing,
58
Brand l evelresul ts
Accepted for Ernst & Young brand only Accepted for Ernst & Young and Intel only Accepted for Ernst & Young only Rej ected for all brands Rej ected for Boeing only
I nteraction effects
H 9:Innovativeness Difficult H 10:Transfer Brand concept consistency[ Residual] N /A Rej ected N /A Accepted for Boeing brand only
59
Brand extensi on
Innovati ve
CSR
Transfer
Di cul ffi t
Intel N otebook PC Reuters 24-H our N ews TV Channel Intel Portable MP3 player Reuters N ews Radio Station Ernst & Young Private tax planning service Boeing Flight simulator video game Saatchi & Saatchi Institute of Advertising Saatchi & Saatchi Marketing & Advertising books Ernst & Young Institute of Accountancy Boeing "U ltra-tough" travel luggage Boeing Chronograph wristwatch Average Standard deviation
Brand Intel Boeing Reuters Ernst & Young Saatchi & Saatchi Average
60
6.3.2 Parent brand qual ity As expected, parent brand quality plays a role in evaluating the brand extension. H owever, the beta coefficient for the QU ALITY 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 j udged by consumers. N evertheless, the beta coefficient for QU ALITY 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 I nnovative Parent brand innovativeness has a positive effect (0.13) on brand extension evaluation. Thus, Keller and Aaker (1997) prediction also holds in the case of consumer s 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 QU ALITY and IN N OVATIVEN ESS); and (2) the quality of B2B may be too abstract for consumers to evaluate, while innovativeness can be more easily j udged.
61
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 SocialResponsibil ity 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 (1997) s 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--vis B2B customers. 6.3.5 Environm entalconcern 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.
62
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.
63
6.3.8 D ifficul t A positive relationship between extension attitude and perceived difficulty of producing the extension was hypothesized. Although the variable Difficult was significant, the hypothesis was rej ected 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 (2001) study, the Difficult beta coefficients were positive. In the latter study s (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 I nteraction 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 eneraldiscussion ofthe current m odel The current model shows a remarkably high adj usted R2 (0.61). This is an improvement from previous studies, although a direct comparison is unfair since
64
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 adj usted 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 adj usted R2 values of 0.64 and 0.58.
65
66
67
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 adj usted 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 brand portfolio (cfr. Dacin & Smith, s 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).
68
69
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.
70
BIBLIOG R APH Y
Aaker, D. A. (1996). Buil ding 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. Journalof M arketing, 5(1), 63-76. Albaum, G . (1997). The Likert scale revisited and alternative version. Journal of the M arketResearch Society, 39(2), 331-48. Anderson, J. C. and N arus, J. A. (2004). Business M arket M anagem ent - U nderstanding, Creating and D el ivering Val (2nd Ed.). U pper Saddle River:Prentice-H all. ue Balachander, S. and G hose, S. (2003). Reciprocal spillover effects:a strategic benefit of brand extensions. Journalof 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 I Sl M anagem entReview , 44(2), 49-54. T oan
71
Bhattacharya, C. B. and Sen, S. (2003). Consumer-company identification:a framework for understanding consumers'relationships with companies. Journalof 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' model. I s nternational 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. Journalof M arketing Research, 38(4), 494-500. Boush, D. M. and Loken, B. (1991). A process-tracing study of brand extension evaluation. Journalof M arketing Research, 28(1), 16-28. Brody, C. J. and Dietz, J. (1997). On the dimensionality of 2-question format Likert scale attitudes. SocialScience Research, 26(2), 197-204. Broniarczyk, S. M. and Alba, J. W . (1994). The importance of the brand in brand extension. Journalof 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. Journalof M arketing, 61(1), 68-84. Churchill, G . A. Jr. (1979). A paradigm for developing better measures of marketing constructs. Journalof 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.
72
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. Journalof M arketing Research, 31, 229242. 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. Journalof Econom ic Psychol ogy, 21(6), 639660. Douglas, S. P., Craig, C. S. and N ij ssen, E. J. (2001). Executive insights: integrating branding strategy across markets: building international brand architecture. Journalof I nternationalM 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 H andbook 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 ul tivariate D ata Anal ysis (5th Ed.). U pper Saddle River, N J:Prentice H all.
73
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. I nternationalJournalof Research in M arketing, 11(3), 233-249. Ind, N . (1997). The Corporate Brand. London:McMillan Press. Ind, N . (1992). The Corporate I age (revised ed.). London:Kogan Page. m Kapferer, J-N . (2000). Strategic Brand M anagem ent. London:Kogan Page. Keller, K. L. (2003). Brand equity dilution. M I Sl M anagem entReview , 45(1), 12-14. T oan Keller, K. L. (2003). Strategic Brand M anagem ent:Buil ding, 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. Journalof 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?. Journalof Services M arketing, 15(5), 328-342. Lambert, Z. V. (1972). Perceptual patterns, information handling and innovativeness. Journalof M arketing Research, 9(4), 427.
74
Lance, Charles E. (1988), Residual centering, exploratory and confirmatory moderator analysis, and decomposition of effects in path models containing interactions, Appl Psychol ied ogicalM easurem ent, 12(2), 163-175. Leone, R. P. and Schultz, R. L. (1980). A study of Marketing G eneralizations. Journalof M arketing, 44(1), 11-18. Likert, R. (1932). A technique for the measurements of attitudes. Archives of Psychol ogy, 140. Loken, B. and John, D. R. (1993). Diluting brand beliefs:when do brand extensions have a negative impact?. Journalof 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 Appl Approach (3rd Ed.). ied H arlow:Prentice-H all. Martin, E. and Polivka, A. E. (1995). Diagnostics for redesigning survey questionnaires measuring work in the current population survey. Publ O pinion Q uarterl 59(4), ic y, 547-564. McAlexander, J. H ., Schouten, J. W . and Koenig, H . F. (2002). Building brand community. Journalof 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.
75
Mervis, C. B. and Rosch, E. (1981). Categorization of N atural Obj ects. AnnualReview of Psychol ogy, 32, 89-115. Michell, P., King, J. and Reast, J. (2001). Brand values related to industrial products. I ndustrialM arketing M anagem ent, 30(5), 415-425. Moorthi, Y. L. R. (2002). An approach to branding services. Journalof Services M arketing, 16(3), 259-274. Morrin, M. (1999). The impact of brand extensions on parent brand memory structures and retrieval processes. Journalof M arketing Research, 36(4), 517-526. Mudambi, S. (2002). Branding importance in business-to-business markets -Three buyer clusters. I ndustrialM arketing M anagem ent, 31(6), 525-533. N eter, J., W asserman, W . and Kutner, M. H . (1990), Appl Linear Statistical M odel ied s: Regression, Anal of Variance, and Experim entalD esigns. Burr Ridge, IL:Irwin. ysis O' Cass, A. and G race, D. (2003). An exploratory perspective of service brand associations. Journalof 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. Journalof Consum er Research, 1(2), 23-29. Peterson, R. A. (1994). A meta-analysis of Cronbach coefficient alpha, Journal of s Consum er research, 21, 381-91. Porter, M. (1998). Com petitive Advantage -Creating and Sustaining Superior Perform ance. N ew York:Free Press.
76
Raj agopal 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. Journalof Service Research, 3(3), 220-231. Roerich, G . (2004). Consumer innovativeness:concepts and measurements. Journalof Business Research, 57(6), 671-677. Sunde, L. and Brodie, R.J., (1993). Consumer evaluations of brand extensions: Further empirical evidence. I nternationalJournalof 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 I ntroduction. Essex:Addison-W esley. U rde, M. (1999). Brand orientation: a mindset for building brands into strategic resources. Journalof 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. Journalof M arketing, 52, 2-22.
77
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
78
Nationality
Dutch IB
German IM
Study programme
If you are an exchange student: do you study business, economics, etc. in your home country? Yes Year of study 1 No 2 3 4 5 or more
79
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!
80
BRAND 1 OUT OF 5
Consider the following brand:
Now, please write down your immediate associations with this brand in the empty box:
How do you perceive the overall quality of this brand? This brand is modern and up-to-date This brand invests in R&D This brand introduces the latest product/service features This brand is concerned with the environment This brand is involved in helping its community This brand is socially responsible
3 3 3 3 3 3 3
4 4 4 4 4 4 4
5 5 5 5 5 5 5
OUTSTANDING 6 7 6 6 6 6 6 6 VERY 7 7 7 7 7 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
81
Now, please indicate your opinion on the previously mentioned hypothetical brand extensions by marking the appropriate number (1-7).
TOTALLY DISAGREE 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
TOTALLY DISAGREE 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
82
BRAND 2 OUT OF 5
Consider the following brand:
Now, please write down your immediate associations with this brand in the empty box:
How do you perceive the overall quality of this brand? This brand is modern and up-to-date This brand invests in R&D This brand introduces the latest product/service features This brand is concerned with the environment This brand is involved in helping its community This brand is socially responsible
3 3 3 3 3 3 3
4 4 4 4 4 4 4
5 5 5 5 5 5 5
OUTSTANDING 6 7 6 6 6 6 6 6 VERY 7 7 7 7 7 7
Also consider the following brand extensions, and please write down your immediate associations to these in the empty boxes: a) KPMG Private tax planning service
83
Now, please indicate your opinion on the previously mentioned hypothetical brand extensions by marking the appropriate number (1-7).
a) Consider the brand extension KPMG Private tax planning service again.
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it The competences of KPMG are useful to make this extension The resources of KPMG are useful to make this extension This brand extension is consistent with the KPMG brand This brand extension fits with my associations of the KPMG brand This brand extension is too easy to make POOR 1 2 3 4 5 5 5 5 5 5 5 OUTSTANDING 6 7 TOTALLY AGREE 6 7 6 7 6 7 6 7 6 7 6 7
TOTALLY DISAGREE 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
b) Consider the brand extension KPMG Institute of Accountancy (Executive MBA education) again.
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it The competences of KPMG are useful to make this extension The resources of KPMG are useful to make this extension This brand extension is consistent with the KPMG brand This brand extension fits with my associations of the KPMG brand This brand extension is too easy to make POOR 1 2 3 4 5 5 5 5 5 5 5 OUTSTANDING 6 7 TOTALLY AGREE 6 7 6 7 6 7 6 7 6 7 6 7
TOTALLY DISAGREE 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
TOTALLY DISAGREE 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
84
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:
3) How do you perceive the overall quality of this brand? 4) This brand is modern and up-to-date 5) This brand invests in R&D 6) This brand introduces the latest product/service features 7 This brand is concerned with the environment 8) This brand is involved in helping its community 9) This brand is socially responsible
3 3 3 3 3 3 3
4 4 4 4 4 4 4
5 5 5 5 5 5 5
OUTSTANDING 6 7 6 6 6 6 6 6 VERY 7 7 7 7 7 7
Also consider the following brand extensions, and please write down your immediate associations to these in the empty boxes: a) Saatchi & Saatchi marketing & advertising books
85
Now, please indicate your opinion on the previously mentioned hypothetical brand extensions by marking the appropriate number (1-7).
a) Consider the brand extension Saatchi & Saatchi marketing & advertising books again.
How do you perceive the overall quality of this brand extension? POOR 1 2 3 4 5 5 5 5 5 5 5 OUTSTANDING 6 7 TOTALLY AGREE 6 7 6 7 6 7 6 7 6 7 6 7
TOTALLY DISAGREE If this brand extension existed, I would be likely to try it 1 2 3 4 The competences of Saatchi & Saatchi are useful to make this extension 1 2 3 4 The resources of Saatchi & Saatchi are useful to make this extension 1 2 3 4 This brand extension is consistent with the Saatchi & Saatchi brand 1 2 3 4 This brand extension fits with my associations of the Saatchi & Saatchi brand 1 2 3 4 This brand extension is too easy to make 1 2 3 4
b) Consider the brand extension Saatchi & Saatchi Institute of Advertising (executive training courses) again.
How do you perceive the overall quality of this brand extension? POOR 1 2 3 4 5 5 5 5 5 5 5 OUTSTANDING 6 7 TOTALLY AGREE 6 7 6 7 6 7 6 7 6 7 6 7
TOTALLY DISAGREE If this brand extension existed, I would be likely to try it 1 2 3 4 The competences of Saatchi & Saatchi are useful to make this extension 1 2 3 4 The resources of Saatchi & Saatchi are useful to make this extension 1 2 3 4 This brand extension is consistent with the Saatchi & Saatchi brand 1 2 3 4 This brand extension fits with my associations of the Saatchi & Saatchi brand 1 2 3 4 This brand extension is too easy to make 1 2 3 4
c) Consider the brand extension Saatchi & Saatchi Espresso Bar & Caf again.
How do you perceive the overall quality of this brand extension? POOR 1 2 3 4 5 5 5 5 5 5 5 OUTSTANDING 6 7 TOTALLY AGREE 6 7 6 7 6 7 6 7 6 7 6 7
TOTALLY DISAGREE If this brand extension existed, I would be likely to try it 1 2 3 4 The competences of Saatchi & Saatchi are useful to make this extension 1 2 3 4 The resources of Saatchi & Saatchi are useful to make this extension 1 2 3 4 This brand extension is consistent with the Saatchi & Saatchi brand 1 2 3 4 This brand extension fits with my associations of the Saatchi & Saatchi brand 1 2 3 4 This brand extension is too easy to make 1 2 3 4
86
BRAND 4 OUT OF 5
Consider the following brand:
Now, please write down your immediate associations with this brand in the empty box:
How do you perceive the overall quality of this brand? This brand is modern and up-to-date This brand invests in R&D This brand introduces the latest product/service features This brand is concerned with the environment This brand is involved in helping its community This brand is socially responsible
3 3 3 3 3 3 3
4 4 4 4 4 4 4
5 5 5 5 5 5 5
OUTSTANDING 6 7 6 6 6 6 6 6 VERY 7 7 7 7 7 7
Also consider the following brand extensions, and please write down your immediate associations to these in the empty boxes: a) Intel Portable MP3 Music Player
c) Intel Notebook PC
87
Now, please indicate your opinion on the previously mentioned hypothetical brand extensions by marking the appropriate number (1-7).
TOTALLY DISAGREE 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
b) Consider the brand extension Intel Home Wireless (wi-fi) Network Hardware again.
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it The competences of Intel are useful to make this extension The resources of Intel are useful to make this extension This brand extension is consistent with the Intel brand This brand extension fits with my associations of the Intel brand This brand extension is too easy to make POOR 1 2 3 4 5 5 5 5 5 5 5 OUTSTANDING 6 7 TOTALLY AGREE 6 7 6 7 6 7 6 7 6 7 6 7
TOTALLY DISAGREE 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
TOTALLY DISAGREE 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
88
BRAND 5 OUT OF 5
Consider the following brand:
Now, please write down your immediate associations with this brand in the empty box:
3) How do you perceive the overall quality of this brand? 4) This brand is modern and up-to-date 5) This brand invests in R&D 6) This brand introduces the latest product/service features 7 This brand is concerned with the environment 8) This brand is involved in helping its community 9) This brand is socially responsible
3 3 3 3 3 3 3
4 4 4 4 4 4 4
5 5 5 5 5 5 5
OUTSTANDING 6 7 6 6 6 6 6 6 VERY 7 7 7 7 7 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
89
4) Now, please indicate your opinion on the previously mentioned hypothetical brand extensions by placing a tick in the appropriate box.
TOTALLY DISAGREE 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
b) Consider the brand extension Boeing flight simulator computer game again.
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it The competences of Boeing are useful to make this extension The resources of Boeing are useful to make this extension This brand extension is consistent with the Boeing brand This brand extension fits with my associations of the Boeing brand This brand extension is too easy to make POOR 1 2 3 4 5 5 5 5 5 5 5 OUTSTANDING 6 7 TOTALLY AGREE 6 7 6 7 6 7 6 7 6 7 6 7
TOTALLY DISAGREE 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
TOTALLY DISAGREE 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
90
PERSONAL INFORMATION
Name Age
Female
No
91
Content of questions
Sequence of questions
Question difficulty
93
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!
94
BRAND 1 OUT OF 5
Consider the following brand
Please indicate your opinion on this brand by marking the appropriate number (1-7).
How knowledgeable are you about this brand? How do you perceive the overall quality of this brand? This brand is modern and up-to-date This brand invests in R&D This brand introduces the latest product/service features This brand is concerned with the environment This brand is involved in helping its community This brand is socially responsible NOT AT ALL 1 2 3 POOR 1 2 3 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 6 VERY 7
OUTSTANDING 6 7 6 6 6 6 6 6 VERY 7 7 7 7 7 7
NOT AT ALL 1 2 3 1 2 3 1 2 3 1 1 1 2 2 2 3 3 3
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).
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it 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 This brand extension will be difficult to make Reuters is out to make short-term profits with this brand extension
POOR 1 2
5 5 5 5 5 5 5 5
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it 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 This brand extension will be difficult to make Reuters is out to make short-term profits with this brand extension
POOR 1 2
5 5 5 5 5 5 5 5
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
95
BRAND 2 OUT OF 5
Consider the brand
Please indicate your opinion on these brand extensions by marking the appropriate number (1-7).
How knowledgeable are you about this brand? How do you perceive the overall quality of this brand? This brand is modern and up-to-date This brand invests in R&D This brand introduces the latest product/service features This brand is concerned with the environment This brand is involved in helping its community This brand is socially responsible NOT AT ALL 1 2 3 POOR 1 2 3 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 6 VERY 7
OUTSTANDING 6 7 6 6 6 6 6 6 VERY 7 7 7 7 7 7
NOT AT ALL 1 2 3 1 2 3 1 2 3 1 1 1 2 2 2 3 3 3
Three 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).
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
96
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
97
BRAND 3 OUT OF 5
Consider the following brand
Please indicate your opinion on this brand by marking the appropriate number (1-7).
How knowledgeable are you about this brand? How do you perceive the overall quality of this brand? This brand is modern and up-to-date This brand invests in R&D This brand introduces the latest product/service features This brand is concerned with the environment This brand is involved in helping its community This brand is socially responsible NOT AT ALL 1 2 3 POOR 1 2 3 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 6 VERY 7
OUTSTANDING 6 7 6 6 6 6 6 6 VERY 7 7 7 7 7 7
NOT AT ALL 1 2 3 1 2 3 1 2 3 1 1 1 2 2 2 3 3 3
Three 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).
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it The competences of Saatchi & Saatchi are useful to make this extension The resources of Saatchi & Saatchi are useful to make this extension This brand extension is consistent with the Saatchi & Saatchi brand This brand extension fits with my associations of the Saatchi & Saatchi brand This brand extension will be difficult to make Saatchi & Saatchi is out to make short-term profits with this brand extension
POOR 1 2
5 5 5 5 5 5 5 5
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it The competences of Saatchi & Saatchi are useful to make this extension The resources of Saatchi & Saatchi are useful to make this extension This brand extension is consistent with the Saatchi & Saatchi brand This brand extension fits with my associations of the Saatchi & Saatchi brand This brand extension will be difficult to make Saatchi & Saatchi is out to make short-term profits with this brand extension
POOR 1 2
5 5 5 5 5 5 5 5
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
98
BRAND 4 OUT OF 5
Consider the following brand
Please indicate your opinion on this brand by marking the appropriate number (1-7).
How knowledgeable are you about this brand? How do you perceive the overall quality of this brand? This brand is modern and up-to-date This brand invests in R&D This brand introduces the latest product/service features This brand is concerned with the environment This brand is involved in helping its community This brand is socially responsible NOT AT ALL 1 2 3 POOR 1 2 3 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 6 VERY 7
OUTSTANDING 6 7 6 6 6 6 6 6 VERY 7 7 7 7 7 7
NOT AT ALL 1 2 3 1 2 3 1 2 3 1 1 1 2 2 2 3 3 3
Three 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).
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it The competences of Intel are useful to make this extension The resources of Intel are useful to make this extension This brand extension is consistent with the Intel brand This brand extension fits with my associations of the Intel brand This brand extension will be difficult to make Intel is out to make short-term profits with this brand extension
POOR 1 2
5 5 5 5 5 5 5 5
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
99
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it The competences of Intel are useful to make this extension The resources of Intel are useful to make this extension This brand extension is consistent with the Intel brand This brand extension fits with my associations of the Intel brand This brand extension will be difficult to make Intel is out to make short-term profits with this brand extension
POOR 1 2
5 5 5 5 5 5 5 5
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it The competences of Intel are useful to make this extension The resources of Intel are useful to make this extension This brand extension is consistent with the Intel brand This brand extension fits with my associations of the Intel brand This brand extension will be difficult to make Intel is out to make short-term profits with this brand extension
POOR 1 2
5 5 5 5 5 5 5 5
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
100
BRAND 5 OUT OF 5
Consider the following brand
Please indicate your opinion on this brand by marking the appropriate number (1-7).
How knowledgeable are you about this brand? How do you perceive the overall quality of this brand? This brand is modern and up-to-date This brand invests in R&D This brand introduces the latest product/service features This brand is concerned with the environment This brand is involved in helping its community This brand is socially responsible NOT AT ALL 1 2 3 POOR 1 2 3 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 6 VERY 7
OUTSTANDING 6 7 6 6 6 6 6 6 VERY 7 7 7 7 7 7
NOT AT ALL 1 2 3 1 2 3 1 2 3 1 1 1 2 2 2 3 3 3
Three 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).
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it The competences of Boeing are useful to make this extension The resources of Boeing are useful to make this extension This brand extension is consistent with the Boeing brand This brand extension fits with my associations of the Boeing brand This brand extension will be difficult to make Boeing is out to make short-term profits with this brand extension
POOR 1 2
5 5 5 5 5 5 5 5
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it The competences of Boeing are useful to make this extension The resources of Boeing are useful to make this extension This brand extension is consistent with the Boeing brand This brand extension fits with my associations of the Boeing brand This brand extension will be difficult to make Boeing is out to make short-term profits with this brand extension
POOR 1 2
5 5 5 5 5 5 5 5
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
101
How do you perceive the overall quality of this brand extension? If this brand extension existed, I would be likely to try it The competences of Boeing are useful to make this extension The resources of Boeing are useful to make this extension This brand extension is consistent with the Boeing brand This brand extension fits with my associations of the Boeing brand This brand extension will be difficult to make Boeing is out to make short-term profits with this brand extension
POOR 1 2
5 5 5 5 5 5 5 5
TOTALLY DISAGREE 1 2 3 4 1 1 1 1 1 1 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 4
PERSONAL INFORMATION
Name Age
Female
No
102
B.1.2 I nnovativeness
R E L I A B I L I T Y A N A L Y S I S Mean 1. 2. 3. INNOV1 INNOV2 INNOV3 Mean 15.2812 5.0872 5.0916 5.1024 Variance 14.6106 Std Dev 1.3162 1.5555 1.4200 Std Dev 3.8224 S C A L E Cases 1113.0 1113.0 1113.0 N of Variables 3 (A L P H A)
Item-total Statistics Scale Mean if Item Deleted INNOV1 INNOV2 INNOV3 10.1941 10.1896 10.1788 Scale Variance if Item Deleted 7.8148 6.4074 6.5570 Corrected ItemTotal Correlation .6880 .7344 .8301 Alpha if Item Deleted .8647 .8298 .7335
.8667
103
B.1.4 Transfer
R E L I A B I L I T Y A N A L Y S I S Mean 1. 2. TRANSFE1 TRANSFE2 Mean 9.5552 4.7384 4.8167 Variance 8.9507 Std Dev 1.5765 1.5170 Std Dev 2.9918 S C A L E Cases 1124.0 1124.0 N of Variables 2 (A L P H A)
Item-total Statistics Scale Mean if Item Deleted TRANSFE1 TRANSFE2 4.8167 4.7384 Scale Variance if Item Deleted 2.3012 2.4854 Corrected ItemTotal Correlation .8706 .8706 Alpha if Item Deleted . .
.9304
104
Item-total Statistics Scale Mean if Item Deleted BCC1 BCC2 4.3348 4.5009 Scale Variance if Item Deleted 2.9219 2.7066 Corrected ItemTotal Correlation .8790 .8790 Alpha if Item Deleted . .
.9352
B.1.6 D ifficul t
R E L I A B I L I T Y A N A L Y S I S Mean 1. 2. DIFFI1 DIFFI2 Mean 7.4702 3.5067 3.9636 Variance 5.1497 Std Dev 1.5741 1.4635 Std Dev 2.2693 S C A L E Cases 1125.0 1125.0 N of Variables 2 (A L P H A)
Item-total Statistics Scale Mean if Item Deleted DIFFI1 DIFFI2 3.9636 3.5067 Scale Variance if Item Deleted 2.1419 2.4779 Corrected ItemTotal Correlation .1150 .1150 Alpha if Item Deleted . .
.2058
105
a All requested variables entered. b Dependent Variable: Attitude toward brand extension
Model Summary Model 1 R .785 R Square .617 Adjusted R Square .613 Std. Error of the Estimate .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 1 Regression Residual Total Sum of Squares 1429.08 4 887.630 2316.71 4 df 10 1122 1132 Mean Square 142.908 .791 F 180.642 Sig. .000
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
106
Coefficients Unstandardized Coefficients Model 1 (Constant) Parent brand knowledge Parent brand quality Environmental concern Corporate social responsibility Innovative Transfer Brand concept consistency Innovativeness*difficulty Transfer*brand concept consistency Difficult B Std. Error Standardized Coefficients Beta -1.532 .126 -.031 -1.200 .230 .064 .054 .155 .144 .324 .505 -.005 2.314 .021 2.092 .037 5.960 .000 3.242 .001 6.301 .000 7.425 .000 -.058 .954 .524 .448 .513 .507 .173 .129 1.908 2.234 1.951 1.971 5.795 7.733 t Sig. Collinearity Statistics Tolerance VIF
-.521 .340 -2.254E-02 .019 7.587E-02 .033 5.802E-02 .028 .181 .030 .162 .050 .310 .049 .444 .060 -6.840E-04 .012 -1.552E-02 .011 -5.596E-02 .066
107
a All requested variables entered. b Dependent Variable: Attitude toward brand extension
Model Summary Model R R Square Adjusted R Square .613 Std. Error of the Estimate .8894
.785
.617
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 Model 1 Regression Residual Total Sum of Squares 1429.084 887.630 2316.714 df 10 1122 1132 Mean Square 142.908 .791 F 180.642 Sig. .000
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
108
Coefficients Unstandardized Coefficients Model 1 (Constant) Parent brand knowledge Parent brand quality Environmental concern Corporate social responsibility Innovative Transfer Brand concept consistency Difficult Transfer*brand concept consistency (resid) Innovativeness*difficulty (resid) B Std. Error Standardized Coefficients Beta -.730 .466 -.031 -1.200 .230 .064 .054 .155 .142 .263 2.314 .021 2.092 .037 5.960 .000 5.046 .000 6.814 .000 .524 .448 .513 .507 .432 .230 .347 1.908 2.234 1.951 1.971 2.316 4.354 2.879 t Sig. Collinearity Statistics Tolerance VIF
-.254 .348 -2.254E-02 .019 7.587E-02 .033 5.802E-02 .028 .181 .030 .159 .032 .252 .037 .366 .028 -5.596E-02 .066 -1.552E-02 .011 -6.840E-04 .012
.072 13.925
109
a All requested variables entered. b Dependent Variable: Attitude toward brand extension
Model Summary Model R R Square .617 Adjusted R Square .614 Std. Error of the Estimate .8892
.786
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 Model 1 Regression Residual Total Sum of Squares 1429.521 887.193 2316.714 df 10 1122 1132 Mean Square 142.952 .791 F 180.786 Sig. .000
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
110
Coefficients Unstandardized Coefficients Model 1 (Constant) Parent brand knowledge Parent brand quality Innovative Corporate social responsibility Environmental concern Transfer Brand concept consistency Difficult Transfer*brand concept consistency[Residual] Service brand B Std. Error Standardized Coefficients Beta -1.131 .258 -.034 -1.318 .188 .066 .132 .155 .054 .268 2.386 .017 4.306 .000 5.978 .000 2.107 .035 8.412 .000 .512 1.955 .442 2.265 .362 2.766 .507 1.972 .513 1.951 .337 2.964 .348 2.877 .785 1.274 .878 1.139 t Sig. Collinearity Statistics Tolerance VIF
-.200 .177 -2.506E-02 .019 7.875E-02 .033 .148 .034 .181 .030 5.843E-02 .028 .256 .030 .368 .028 -5.757E-02 .019 -1.622E-02 .011
-5.174E-02 .069
-.018
-.746 .456
.585 1.710
111
Model
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 Model 1 Regression Residual Total Sum of Squares 201.360 179.874 381.234 df 9 196 205 Mean Square 22.373 .918 F 24.379 Sig. .000
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
112
Coefficients Unstandardized Coefficients Model 1 (Constant) Parent brand knowledge Parent brand quality Innovative Corporate social responsibility Environmental concern Transfer Brand concept consistency Difficult Transfer*brand concept consistency[Residual] B .684 6.914E-02 -7.240E-02 8.317E-02 .221 5.707E-02 .233 .352 -7.587E-02 1.844E-02 Std. Error .550 .052 .078 .080 .081 .077 .096 .081 .052 .037 .091 -.067 .067 .212 .055 .202 .349 -.082 .033 Standardized Coefficients Beta 1.243 .215 1.329 .186 -.923 .357 1.041 .299 2.710 .007 .744 .458 2.416 .017 4.369 .000 -1.471 .143 .496 .621 .508 1.967 .454 2.200 .584 1.712 .393 2.543 .439 2.278 .343 2.913 .378 2.643 .778 1.285 .556 1.799 t Sig. Collinearity Statistics Tolerance VIF
a Dependent Variable: Attitude toward brand extension b Selecting only cases for which +Brand = Reuters
113
Model
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 Model 1 Regression Residual Total Sum of Squares 248.338 134.551 382.888 df 9 196 205 Mean Square 27.593 .686 F 40.195 Sig. .000
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
114
Coefficients Unstandardized Coefficients Model 1 (Constant) Parent brand knowledge Parent brand quality Innovative Corporate social responsibility Environmental concern Transfer Brand concept consistency Difficult Transfer*brand concept consistency[Residual] B -.293 -9.209E-02 .121 .204 .184 9.772E-02 .287 .301 -.132 -1.581E-03 Std. Error .426 .044 .068 .073 .061 .060 .081 .061 .042 .033 -.110 .104 .176 .149 .086 .243 .337 -.154 -.002 Standardized Coefficients Beta -.687 .493 -2.074 .039 1.777 .077 2.813 .005 3.007 .003 1.618 .107 3.567 .000 4.947 .000 -3.149 .002 -.047 .962 .633 1.579 .527 1.898 .460 2.174 .734 1.362 .635 1.574 .385 2.598 .387 2.586 .749 1.335 .696 1.437 t Sig. Collinearity Statistics Tolerance VIF
a Dependent Variable: Attitude toward brand extension b Selecting only cases for which +Brand = Ernst & Young
115
Model
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 Model 1 Regression Residual Total Sum of Squares 168.389 104.656 273.046 df 9 196 205 Mean Square 18.710 .534 F 35.040 Sig. .000
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
116
Coefficients Unstandardized Coefficients Model 1 (Constant) Parent brand knowledge Parent brand quality Innovative Corporate social responsibility Environmental concern Transfer Brand concept consistency Difficult Transfer*brand concept consistency[Residual] B -.724 3.739E-02 7.194E-02 .101 .509 -.110 .316 .312 -6.868E-02 -1.385E-02 Std. Error .463 .038 .093 .105 .103 .082 .090 .078 .044 .040 .063 .067 .085 .379 -.108 .284 .294 -.074 -.020 Standardized Coefficients Beta -1.564 .119 .973 .332 .773 .441 .963 .337 4.964 .000 -1.334 .184 3.523 .001 4.024 .000 -1.553 .122 -.342 .732 .473 2.115 .258 3.877 .253 3.951 .335 2.983 .297 3.367 .300 3.334 .367 2.722 .866 1.155 .573 1.746 t Sig. Collinearity Statistics Tolerance VIF
a Dependent Variable: Attitude toward brand extension b Selecting only cases for which +Brand = Saatchi & Saatchi
117
Model
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 Model 1 Regression Residual Total Sum of Squares 242.933 166.282 409.215 df 9 196 205 Mean Square 26.993 .848 F 31.817 Sig. .000
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
118
Coefficients Unstandardized Coefficients Model 1 (Constant) Parent brand knowledge Parent brand quality Innovative Corporate social responsibility Environmental concern Transfer Brand concept consistency Difficult Transfer*brand concept consistency[Residual] B -1.092 -1.269E-02 .210 .188 .187 2.237E-02 .375 .226 -4.409E-02 -5.030E-03 Std. Error .565 .056 .093 .115 .074 .080 .094 .079 .049 .032 -.012 .147 .113 .160 .018 .361 .251 -.053 -.008 Standardized Coefficients Beta -1.934 .055 -.226 .822 2.268 .024 1.629 .105 2.508 .013 .281 .779 4.005 .000 2.847 .005 -.893 .373 -.158 .874 .718 1.393 .493 2.029 .433 2.308 .509 1.963 .534 1.873 .254 3.930 .268 3.734 .597 1.676 .736 1.359 t Sig. Collinearity Statistics Tolerance VIF
a Dependent Variable: Attitude toward brand extension b Selecting only cases for which +Brand = Intel
119
Model
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 Model 1 Regression Residual Total Sum of Squares 419.341 243.889 663.230 df 9 299 308 Mean Square 46.593 .816 F 57.122 Sig. .000
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
120
Coefficients Unstandardized Coefficients Model 1 (Constant) Parent brand knowledge Parent brand quality Innovative Corporate social responsibility Environmental concern Transfer Brand concept consistency Difficult Transfer*brand concept consistency[Residual] B .346 4.881E-02 3.492E-02 2.857E-02 .157 9.719E-02 .155 .404 -9.681E-03 -5.631E-02 Std. Error .315 .046 .075 .078 .057 .049 .047 .047 .033 .018 .053 .028 .024 .153 .108 .174 .481 -.011 -.129 Standardized Coefficients Beta 1.099 .273 1.054 .293 .466 .642 .366 .714 2.750 .006 1.993 .047 3.280 .001 8.590 .000 -.291 .771 -3.066 .002 .491 2.038 .337 2.969 .295 3.386 .396 2.526 .417 2.398 .435 2.299 .392 2.551 .868 1.152 .692 1.446 t Sig. Collinearity Statistics Tolerance VIF
a Dependent Variable: Attitude toward brand extension b Selecting only cases for which +Brand = Boeing
121
Model
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 Model 1 Regression Residual Total Sum of Squares 635.010 447.573 1082.583 df 9 608 617 Mean Square 70.557 .736 F 95.847 Sig. .000
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
122
Coefficients Unstandardized Coefficients Model 1 (Constant) Parent brand knowledge Parent brand quality Environmental concern Innovative Corporate social responsibility Transfer Brand concept consistency Difficult Transfer*brand concept consistency[Residual] B -.173 -3.744E-02 4.934E-02 6.725E-02 .176 .210 .287 .334 -9.479E-02 4.065E-03 Std. Error .268 .023 .042 .038 .044 .042 .050 .040 .026 .020 -.057 Standardized Coefficients Beta -.646 .518 - .111 1.596 .526 1.901 .454 2.202 .557 1.794 .530 1.886 .551 1.816 .336 2.972 .396 2.526 .806 1.241 .643 1.556 t Sig. Collinearity Statistics Tolerance VIF
.045 1.172 .242 .062 1.777 .076 .144 4.033 .000 .176 5.003 .000 .256 5.705 .000 .345 8.315 .000 -.104 .007 - .000 3.582 .200 .841
a Dependent Variable: Attitude toward brand extension b Selecting only cases for which Service brand = Service brand
123
Model
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 Model 1 Regression Residual Total Sum of Squares 793.594 430.482 1224.076 df 9 505 514 Mean Square 88.177 .852 F 103.441 Sig. .000
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
124
Coefficients Unstandardized Coefficients Model 1 (Constant) Parent brand knowledge Parent brand quality Environmental concern Innovative Corporate social responsibility Transfer Brand concept consistency Difficult Transfer*brand concept consistency[Residual] B -.219 -1.426E-03 .109 6.479E-02 .104 .144 .236 .388 -2.327E-02 -2.920E-02 Std. Error .262 .035 .055 .041 .059 .045 .041 .039 .028 .014 Standardized Coefficients Beta -.833 .405 -.001 -.041 .967 .079 1.983 .048 .061 1.563 .119 .077 1.747 .081 .126 3.214 .001 .262 5.690 .000 .455 9.829 .000 -.026 -.846 .398 -.057 - .042 2.042 .613 1.632 .444 2.253 .462 2.165 .358 2.790 .456 2.191 .330 3.033 .326 3.071 .761 1.315 .892 1.121 t Sig. Collinearity Statistics Tolerance VIF
a Dependent Variable: Attitude toward brand extension b Selecting only cases for which Service brand = Product brand
125