Original Article
Consumer evaluation of brand extensions: Comparing goods to goods brand extensions with goods to services Received (in revised form): 09th October 2015
Jayasankar Ramanathan is Visiting Assistant Professor of Marketing at the Indian Institute of Management, Indore. His research interests include brand extensions and virtual worlds. His research is published in journals including Journal of Product and Brand Management.
Sanal Kumar Velayudhan Fellow IIM (Ahmedabad), Professor of Marketing at IIMK. Industry experience of 8 years prior to teaching at XLRI, ASCI and visiting assignments at IIMA, IIM Ranchi, MDI and BITS Pilani. Handled consultancy assignments in marketing for manufacturing and service sector. Publications are in Journal of Small Business Economics, Asian Case Research Journal, Energy Policy, Asia Pacific Journal of Marketing and Logistics and IJRDM. Couple of cases hosted by Ivey. Published book on Rural Marketing by Sage and two readings in Rural Marketing.
ABSTRACT Studies in literature on consumer attitude toward the brand extension, and revision of attitude toward parent brand have focused on the influence of a variety of factors concerning parent brand, fit between the parent brand and extension, extension product category and consumer characteristics. In all such studies, the brand that is extended and the extension product category are both either goods-intensive offerings (GG brand extension) or services-intensive offerings (SS brand extension), and researchers have observed paucity of studies on brand extension from goods-intensive offerings to services-intensive offerings (GS brand extension), and services-intensive offerings to goods-intensive offerings (SG brand extension). In this study, the influence of factors concerning parent brand and fit between the parent brand and extension are contrasted between GG and GS brand extensions. Using survey design, data was collected from individual consumers recruited from households in two Indian cities, and analysis of the data was performed using a structural equation modeling approach. The findings from this study suggest that brand extensions from goods to goods is preferable to brand extensions from goods to services under some conditions, such as when quality variance across brand portfolio is high and when product category fit is present. Correspondence: Jayasankar Ramanathan, Department of Marketing, Indian Institute of Management Indore, Office Room A211, IIM Indore Campus, Prabandh Shikhar, RauPithampur Road, Indore, Madhya Pradesh 453556, India
Journal of Brand Management (2015) 22, 778–801. doi:10.1057/bm.2015.37; published online 6 November 2015 Keywords:
brands; extensions; attitudes; goods; services
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
www.palgrave-journals.com/bm/
Consumer evaluation of brand extensions
INTRODUCTION Brands are valuable assets, as indicated by the valuation of top 10 global brands, which range from US$118.863 billion to $34.338 billion (Interbrand, 2014).1 Brands can be acquired (Mann and Kohli, 2012), licensed (Bass, 2004) and hypothecated (Sanjai and Rebello, 2014) by firms. The brand derives its value from the utility it provides to the business, and is an important aid to the marketer to gain consumer acceptance for new offerings. Marketers aim for better consumer acceptance of new products by naming the new products with existing brand names or by naming these using a combination of existing brand name and new brand name (Milberg et al, 1997). Marketers experience reduced marketing expenditure (Tauber, 1988; Aaker and Keller, 1990; Smith and Park, 1992) and lesser risk of new product failure (Tauber, 1988; Aaker and Keller, 1990) through brand extensions. Careful management of brand extension is a critical task for brand managers, to increase market acceptance of extension as well as market acceptance of other products in the brand portfolio. Studies in the literature on consumer evaluation of brand extensions include brand extensions from goods-intensive offerings to goods-intensive offerings (for example, Britannia brand extended from biscuits to packaged milk), and brand extensions from services-intensive offerings to services-intensive offerings (for example, ICICI brand extended from banking to insurance). The field of brand extension is well researched, but there are only a few
studies in the published literature on brand extensions from goods-intensive offerings to services-intensive offerings (for example, Kingfisher brand extended from beer to airline), and brand extensions from servicesintensive offerings to goods-intensive offerings (for example, Croma brand extended from retailing to appliances). This is observed in Arslan and Altuna (2011), Brown et al (2011), Estelle (2010), Lei et al (2004) and Ramanathan (2013). Though there is a study (Lei et al, 2004) that examines the strength of a few relationships between ‘brand extensions from moderately services-intensive offerings to low services-intensive offerings’ and ‘brand extensions from moderately services-intensive offerings to high services-intensive offerings’, there is no published study that examines the brand extension from ‘low services-intensive offerings to high servicesintensive offerings’. The less explored contexts are highlighted in Figure 1. The scope of this study is restricted to goods to goods brand extensions and goods to services brand extensions. Specifically, this study compares ‘brand extensions from low services-intensive offerings to low servicesintensive offerings’ with ‘brand extensions from low services-intensive offerings to high services-intensive offerings’. In the remainder of this article, goods-intensive offerings will be referred to as simply ‘goods’ and servicesintensive offerings will be referred to as simply ‘services’ to conserve space. The research question addressed in this study is therefore as follows: between goods to goods (GG) and goods to services (GS), how different will the strength of effects of
Goods intensive offerings
Extension Services intensive offerings
Goods intensive offerings
‘Goods to Goods Brand Extensions’
‘Goods to Services Brand Extensions’
Services intensive offerings
‘Services to Goods Brand Extensions’
‘Services to Services Brand Extensions’
Parent Brand
Figure 1: A typology of brand extensions. Source: Ramanathan (2013).
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
779
Ramanathan and Velayudhan
established determinants be on attitude toward the extension of, and revision of attitude toward, the parent brand? In the next section, a review of the literature on consumer attitude toward extension, and revision of attitude toward parent brand, is presented and hypotheses are developed.
LITERATURE REVIEW AND HYPOTHESES The literature reviewed is organized as follows: discussion of theoretical perspective; the factors affecting consumer attitude toward extension; and factors affecting consumer revision of attitude toward parent brand.
Theoretical perspective Categorization theory is an extensively adopted theoretical perspective in brand extensions literature (for example, Song et al, 2010; Keaveney et al, 2012; Sood and Keller, 2012; Spiggle et al, 2012; Sichtmann and Diamantopoulos, 2013; Song et al, 2013; Kalamas et al, 2015) and this research is informed by the same theoretical perspective. Mervis and Rosch (1981, p. 1) specified the concept of category as existing ‘whenever two or more objects are treated equivalently’ and the equivalent treatment ‘may take any number of forms, such as labelling distinct objects with the same name, or performing the same action on different objects’. Given that distinct products carry the same brand name in the real world, a brand can be conceptualized as a category. In this sense, any brand is a category consisting of its products (Boush and Loken, 1991; Loken and John, 1993). When a brand is conceptualized as a category, a consumer is said to have knowledge on this category in his/her memory (Keller, 1993). Mitchell (1982) reviewed the studies on the organization of knowledge in a consumer’s memory and reported two views of how this knowledge is represented: prototype
780
(an abstraction of a particular class of objects) and exemplar (a specific instance that is considered representative of a set of objects). These two views do apply for brand category, and consumers may have a prototype of a brand just as they may have an exemplar of a brand (Boush and Loken, 1991). Consumer knowledge of brand is suggested to be held as a mesh of nodes as per the associative network model of memory (Keller, 1993; Chapman and Aylesworth, 1999). According to this view, nodes are pieces of information and any node is said to be linked to other nodes. The strength of the link between the nodes determines whether activation of one node will lead to the activation of another. If a new instance (extension) is identified by the consumer as belonging to a known category (parent brand), then the effect linked with that category is extended to the new instance as well (Sujan, 1985; Fiske and Pavelchak, 1986). Perception of a new instance can activate a node and its associated affect, which in turn is a distinct node (Chapman and Aylesworth, 1999). However, such a transfer of affect occurs only when the consumer finds the instance as ‘fitting’ with the known category. Furthermore, when a purportedly new instance of a category is perceived by a consumer, the existing knowledge of that category may undergo revision (Weber and Crocker, 1983; Sujan and Bettman, 1989; Gurhan-Canli and Maheswaran, 1998). According to the bookkeeping model, every instance, when perceived as fully or remotely fitting with an existing knowledge of a category, leads to its accommodation with the parent category knowledge structure. However, according to the subtyping model, remotely fitting instances would lead to the formation of separate knowledge structures. In this study, brand extensions from goods to goods are contrasted with brand extensions from goods to services. Drawing from categorization theory, it follows that brand extensions from goods to goods and
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
Consumer evaluation of brand extensions
brand extensions from goods to services are of different kinds, and this study seeks to examine the extent to which the perception of difference influences consumer evaluations. In the following sections, factors affecting consumer’s attitude toward extension, and factors affecting consumer’s revision of attitude toward parent brand, are considered, and their differential effect across ‘goods to goods’ brand extension context and ‘goods to services’ brand extension context are hypothesized.
Factors affecting consumer’s attitude toward extension The factors affecting consumer’s attitude toward extension, identified for this study from the literature review, include: attitude toward the parent brand, number of products in brand portfolio, quality variance across products in brand portfolio and product category fit.
Attitude toward the parent brand Attitude toward the parent brand was largely found to be positively related to attitude toward the extension for goods to goods brand extensions (Boush et al, 1987; Sunde and Brodie, 1993; Bottomley and Doyle, 1996; Bhat and Reddy, 2001; Bottomley and Holden, 2001; Volckner and Sattler, 2007; Dens and Pelsmacker, 2010a, b; Milberg et al, 2010; Sattler et al, 2010; Iyer et al, 2011; Hem et al, 2013; Milberg et al, 2013) and goods to services brand extensions (Dwivedi and Merrilees, 2013). Studies of Bottomley and Doyle (1996), Bottomley and Holden (2001), Sunde and Brodie (1993), Van Riel et al (2001) and Van Riel and Ouwersloot (2005) are all replications of Aaker and Keller’s (1990) seminal study in different geographic locations, and evidence was consistently found for the relationship
between attitude toward the parent brand and attitude toward the extension. According to categorization perspective (cf. Sujan, 1985), attitude transfer is enhanced from parent brand to extension when there is increased fit between parent brand and extension. Except for Sunde and Brodie’s (1993) study, the effect of attitude toward the parent brand on attitude toward the extension was found to be stronger when there was better perceived fit between parent brand and extension (Boush et al, 1987; Bottomley and Doyle, 1996; Bottomley and Holden, 2001; Van Riel et al, 2001; Van Riel and Ouwersloot, 2005; Echambadi et al, 2006; Volckner and Sattler, 2007). Consumers may perceive better fit in GG than in GS because they may categorize services to be different from goods on intangibility, heterogeneity, inseparability, perishability (Zeithaml et al, 1985; Iacobucci, 1992, 1998) and lack of ownership (Lovelock and Gummesson, 2004). Since consumers may perceive better fit in GG than in GS, attitude transfer may be stronger for GG than for GS. Thus the following is proposed. Hypothesis 1: The positive relationship between attitude toward the parent brand and attitude toward the extension is stronger for goods to goods brand extensions but weaker for goods to services brand extensions.
Number of products in brand portfolio The number of products in brand portfolio was found to be positively related to attitude toward the future extension for goods to goods (Dacin and Smith, 1994). Dacin and Smith (1994) argued that (i) the number of products in brand portfolio acts as a collateral or bond for consumers to judge extensions (Wernerfelt, 1988); (ii) if the extension is of poor quality, it will affect
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
781
Ramanathan and Velayudhan
the sales of both the extension as well as other products of the brand (Sullivan, 1990); and (iii) consumers can be expected to recognize that the bigger the size of collateral, the lower the likelihood of extension being of poor quality. Given goods–services categorization (Zeithaml et al, 1985; Iacobucci, 1992, 1998; Lovelock and Gummesson, 2004), consumers can be expected to recognize that the chance of a brand’s goods sales being affected because of poor quality of a ‘services’ extension is relatively low, and so the collateral becomes relatively less credible for GS than GG. The large number of products with the same brand name would cause identification to be sharper on common characteristics and if all the products of a brand are goods, the common characteristic that is sharply identified could be the association of the brand with goods. The consumer may not find the extension of the brand to services as appropriate for such a brand category. Since the collateral becomes less credible for GS than GG, the number of products in brand portfolio may have weaker impact in GS than in GG. Thus the following is proposed. Hypothesis 2: The positive relationship between the number of products in brand portfolio and attitude toward extension is stronger for goods to goods brand extensions but weaker for goods to services brand extensions.
Quality variance across products in brand portfolio Quality variance across products in brand portfolio was found to be negatively related to the attitude toward the extension for goods to goods (Dacin and Smith, 1994; Volckner and Sattler, 2007). Dacin and Smith (1994) argued that (i) quality variance across products in brand portfolio affects the extent to which
782
consumers perceive the brand to be a source of reliability; (ii) if there is more variance in quality across products of a brand, then consumers will be uncertain as to whether overall brand quality is generalizable to the extension; and (iii) with more uncertainty, consumers are found to give more importance to less favorable information and judge the extension unfavorably. When quality variance across a brand’s goods is high, extension of such an unreliable brand – especially into services, given goods–services categorization (Zeithaml et al, 1985; Iacobucci, 1992, 1998; Lovelock and Gummesson, 2004) – could be perceived by consumers as extremely humorous or ridiculous.2 Thus the following is proposed. Hypothesis 3: The negative relationship between quality variance across brand portfolio and attitude toward the extension is weaker for goods to goods brand extensions but stronger for goods to services brand extensions.
Product category fit
Product category fit was largely found to be positively related to attitude toward the extension for goods to goods brand extensions (Boush et al, 1987; Aaker and Keller, 1990; Boush and Loken, 1991; Bottomley and Doyle, 1996; Bottomley and Holden, 2001; Milberg et al, 2010; Iyer et al, 2011; Milberg et al, 2013; cf. Park et al, 1991; Sunde and Brodie, 1993; Nijssen and Agustin, 2005; Volckner and Sattler, 2007). Some studies reported that transferability and complementarity are more important influencers of attitude toward extension than substitutability (Aaker and Keller, 1990; Sunde and Brodie, 1993; Bottomley and Doyle, 1996; Bottomley and Holden, 2001). Bottomley and Doyle (1996) and Bottomley and Holden (2001) observed that this could be because of few
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
Consumer evaluation of brand extensions
brand extensions being substitutes in the stimuli used by researchers. Aaker and Keller (1990) observed that a fit on either transferability or complementarity may be adequate and a high fit on both is not necessary. Similarly, Bottomley and Holden (2001) observed that an extension does not need high fit on multiple dimensions of fit to be favorably received. When product category fit is less, consumers may doubt the firm’s ability to produce the extension and may find it less credible (Aaker and Keller, 1990; Bhat and Reddy, 2001). Given goods–services categorization (Zeithaml et al, 1985; Iacobucci, 1992, 1998; Lovelock and Gummesson, 2004), such doubts about a firm’s ability could heighten from a consumer’s perspective when a brand is extended from goods to services. Therefore, the effect of product category fit on attitude toward the extension may be weaker in the context of goods to services brand extensions than in goods to goods brand extensions.
1993; Chen and Chen, 2000; Dens and Pelsmacker, 2010a; Albrecht et al, 2013) and goods to services brand extensions (Dwivedi and Merrilees, 2013). According to typicality-based perspective, the impact or weight given to information on beliefs about a family brand name depends on whether the information pertains to more or less typical members of the category. Thus when information about the brand extension suggests that the extension is atypical of products marketed under the family brand name, then consumers will be less likely to make an inference from the individual extension to the family brand beliefs (Loken and John, 1993; cf. Weber and Crocker, 1983; Martinez and Pina, 2010). Given goods–services categorization (Zeithaml et al, 1985; Iacobucci, 1992, 1998; Lovelock and Gummesson, 2004), consumers will be less likely to make an inference from the individual extension to the family brand beliefs in the context of GS than in the context of GG. Thus the following is proposed.
Hypothesis 4: The positive relationship between product category fit and attitude toward extension is stronger for goods to goods brand extensions but weaker for goods to services brand extensions.
Hypothesis 5: The positive relationship between attitude toward the extension and revision of attitude toward the parent brand is stronger for goods to goods brand extensions but weaker for goods to services brand extensions.
Factors affecting consumer’s revision of attitude toward parent brand
Product category fit
The factors affecting consumer’s revision of attitude toward parent brand identified for this study include attitude toward the extension and product category fit.
Attitude toward the extension Attitude toward the extension was found to be positively related to the revision of attitude toward the parent brand for goods to goods brand extensions (Loken and John,
Product category fit was found to be positively related to the revision of attitude toward the parent brand for goods to goods brand extensions (Romeo, 1991; Keller and Aaker, 1992; Choi et al, 2010; Dwivedi et al, 2010). When product category fit is high, a firm’s ability to produce is reinforced in the minds of consumers. However, when product category fit is less, consumers may doubt the firm’s ability to produce the extension and may find it exploitative (Keller and Aaker, 1992). Given goods–services
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
783
Ramanathan and Velayudhan
categorization (Zeithaml et al, 1985; Iacobucci, 1992, 1998; Lovelock and Gummesson, 2004), such doubts about a firm’s ability to produce the extension, and skepticism of exploitation, could heighten when a brand is extended from goods to services. Therefore, the effect of product category fit on revision of attitude toward the parent brand may be weaker in the context of goods to services brand extensions than in goods to goods brand extensions. Hypothesis 6: The positive relationship between product category fit and revision of attitude toward the parent brand is stronger for goods to goods brand extensions but weaker for goods to services brand extensions.
Sattler, 2007; Iyer et al, 2011; Viot, 2011; Goh et al, 2013), extension product category involvement (Nijssen and Bucklin, 1998 and Nijssen et al, 1995 as cited by Nkowcha et al, 2005; Nijssen and Bucklin, 1998 as cited by Volckner and Sattler, 2007), quality variance across brands in the extension product category (Kardes and Allen, 1991), consumer knowledge of the extension product category (Smith and Park, 1992; Dacin and Smith, 1994 as cited by Volckner and Sattler, 2007), consumer innovativeness (Klink and Smith, 2001; Klink and Athaide, 2010) and consumer brand consciousness (Ramanathan, 2013). The hypotheses are shown in Figure 2.
RESEARCH DESIGN Control factors
FIT BETWEEN PARENT BRAND AND EXTENSION
PARENT BRAND CHARACTERISTICS
Some of the factors from earlier studies are not hypothesized in this study but are measured and controlled within each context during data analysis, so as to reduce omitted variable bias (Woolridge, 2007). These control factors include brand image fit (Park et al, 1991; Broniarczyk and Alba, 1994; Bhat and Reddy, 2001; Volckner and
Published research studies examining brand extension from goods to goods have generally used experiments. In this study, survey design is used since (i) studies on goods to services brand extensions are scarce and factors to be controlled are not fully known; (ii) number of factors examined in this study is quite large (cf. Volckner and Sattler, 2007); (iii) researchers have issued calls for broadening methodological perspectives
Attitude toward the parent brand (+) H1: GG > GS
Number of products in brand portfolio (+) H2: GG > GS
Attitude toward the extension (+) H5: GG > GS
Quality variance across brand portfolio (-) H3: GG < GS Revision of attitude toward the parent brand
Product category fit (+) H4: GG > GS H6: GG > GS
Figure 2: Conceptual model. Note: H1 in Figure 2 reads that the positive effect of attitude toward the parent brand on attitude toward the extension is stronger for brand extensions from goods to goods but weaker for brand extensions from goods to services.
784
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
Consumer evaluation of brand extensions
(Czellar, 2003); and (iv) the specific relationships found in this study can be confirmed using experiments in future studies (Winer, 1999).
Chennai and Patna – were selected for the study, one with a high index and another with a low index, making the aggregate data more representative of the Indian context.
Stimuli
Questionnaire design
Selection of stimuli was based on the criteria followed by Aaker and Keller (1990) such as original brands being relevant to respondents and generally perceived as high quality. One popular Brand X (known for antiseptic liquid) was selected from the list of ‘most trusted brands 2011’ published by a leading Indian business newspaper. Extension product categories were selected based on their expected widespread usage across most households in India. As is prevalent in the literature (for example, Aaker and Keller, 1990), hypothetical extensions of wellknown brands were used as stimuli. Hypothetical brand extensions selected for the study include Brand X Washing Powder (GG) and Brand X Dispensary (GS). The mean scores on product category fit for the extensions were found to be 3.83 and 3.78, respectively, on a 5-point scale, which is suggestive of extensions being perceived by respondents as reasonable and not illogical.
A questionnaire was designed that included both goods to goods and goods to services brand extensions. To reduce order effects, two versions of the questionnaire were generated to ensure that the sequence of brand extensions was counterbalanced and the sequence of items within each section of the questionnaire was rotated. The questionnaires were translated to local language (Tamil for Chennai and Hindi for Patna) through translation and back-translation procedure, and discussions were held with typical respondents to assess the face validity of the questionnaire.
Measures Measures for the concepts included in this study are given in Table 1. The measures in the table are contextualized for Brand X Washing Powder.
Location of data collection Using 2001 census data, cities in India with a population of at least 1 million were indexed on number of households, average household size, sex ratio, literacy rate and work participation rate. For each city, the corresponding score on each of the above variables was divided by maximum score, and the ratios thus obtained were then summed to form the index. Two cities –
Eligibility of respondent for participation For the questionnaire, which included both goods to goods and goods to services brand extensions, the following eligibility criteria were set for participation of respondents so as to reduce heterogeneity in terms of familiarity and knowledge of parent brand (cf. Fu et al, 2009). The respondent should have bought, within the past year, any product of Brand X either for personal sake or for the sake of other members of the household. Alternatively, the respondent should have used, within the past year, any product of Brand X.
Sampling method Simple random sampling was attempted in Chennai with local telephone directory of the most widely subscribed landline telephone service provider as the sampling frame but the response rate was less than 5 per cent. So area sampling was adopted and list of pin codes, obtained from the
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
785
786
Ramanathan and Velayudhan
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
Table 1: Measures Concept
Measure type
Item label
Measure
Source of measure
Attitude toward the parent brand
Reflective
APB1
Section 1: Respond to the following about X brand X is a brand of (1 = very low quality; 5 = very high quality)
Aaker and Keller (1990)
Protypicality of the parent brand Number of products in brand portfolio Brand breadth
Reflective
APB2 PTB4
My opinion about X brand is (1 = very bad; 5 = very good) When I think of antiseptic liquid, I think of X (1 = strongly disagree; 5 = strongly agree)
Boush and Loken (1991) Zimmer and Bhat (2004)
Reflective
NPBP2
The number of products with X brand name is (1 = so few; 5 = so many)
Volckner and Sattler (2006)
The products with X brand name are substitutes to one another in terms of usage (1 = strongly disagree; 5 = strongly agree)
Aaker and Keller (1990)
Formative (Index = Sum of BB_F1 measures) Low score indicates broad brand breadth and high score indicates narrow brand breadth BB_F2
QVBP1
The products with X brand name are complements to one another in terms of usage Aaker and Keller (1990) (1 = strongly disagree; 5 = strongly agree) The products with X brand name are similar to one another in terms of the people, facilities Aaker and Keller (1990) and skills used in making them (1 = strongly disagree; 5 = strongly agree) Quality across products of X (1 = does not vary at all; 5 = varies a lot) Dacin and Smith (1994)
AE1
Section 2: Imagine that X’s washing powder has been recently introduced in the market and respond to the following X’s washing powder must be of (1 = very low quality; 5 = very high quality) Aaker and Keller (1990)
BB_F3 Quality variance across products in brand portfolio
Attitude toward the extension
Reflective
Reflective
AE2 AE3 Revision of attitude toward the parent brand
Reflective
RAPB1
RAPB2 Product category fit
Formative
a
PCF_F1 PCF_F2 PCF_F3
My opinion about X’s washing powder is (1 = very bad; 5 = very good) Boush and Loken (1991) Assuming that I have planned for buying a washing powder, my likelihood of trying X’s washing Aaker and Keller (1990) powder is (1 = not at all likely; 5 = very likely) Given that X’s washing powder was recently introduced, my usual attitude toward X brand Dwivedi et al (2010), Lei et al (2004) has now become (1 = more strongly unfavorable; 5 = more strongly favorable) Assuming that X’s washing powder was recently introduced, my usual opinion about X brand has now become (1 = more strongly negative; 5 = more strongly positive) Washing powder is a substitute to X brand’s previous products in terms of usage (1 = strongly disagree; 5 = strongly agree) Washing powder is a complement to X brand’s previous products in terms of usage (1 = strongly disagree; 5 = strongly agree) Would the people, facilities and skills used in making X’s previous products be helpful for making washing powder? (1 = not at all helpful; 5 = very helpful)
Dwivedi et al (2010), Lei et al (2004) Aaker and Keller (1990) Aaker and Keller (1990) Aaker and Keller (1990)
Brand image fit (nonproduct image fit)
Formative (Index = Sum of measures)
BIF_F1
Quality variance across Reflective brands of the extension product category Extension product Reflective category involvement
QVBEPC1
EPCI1 EPCI2 EPCI3
Consumer knowledge of the extension product category
Reflective
CKEPC1
CKEPC2 Consumer innovativeness
Reflective
CI1 CI2 CI3 CI4
Consumer brand consciousness
Reflective
CBC1 CBC2 CBC3
a
How important would it be for me to make a right choice of a washing powder? (1 = not at all important; 5 = extremely important) In making my selection of a washing powder to buy, how concerned would I be about the outcome of my choice? (1 = not at all concerned; 5 = very much concerned) In selecting from the many types and brands of washing powder available in the market (1 = I would not care at all as to which one I select; 5 = I would care a great deal as to which one I select) Section 3: Respond to the following about your buying behavior I feel very knowledgeable about washing powder. (1 = strongly disagree; 5 = strongly agree) If a friend asked me about washing powder, I could give him advice about different brands. (1 = strongly disagree; 5 = strongly agree) I like buying any latest brand of product (1 = strongly disagree; 5 = strongly agree)
On the basis of Keller (1993) On the basis of Keller (1993) Volckner and Sattler (2007)
Mittal (1995) Mittal (1995) Mittal (1995)
Smith and Park (1992), Volckner and Sattler (2007) Smith and Park (1992)
Klink and Smith (2001), Klink and Athaide (2010) If I needed to buy a product, I would buy any latest brand available (1 = strongly disagree; Klink and Smith (2001), Klink and 5 = strongly agree) Athaide (2010) When I see a new brand of product, I often buy it because it is new (1 = strongly disagree; Klink and Smith (2001), Klink and 5 = strongly agree) Athaide (2010) I like to buy any latest brand of product before others do (1 = strongly disagree; 5 = strongly Klink and Smith (2001), Klink and agree) Athaide (2010) I believe that the brands I buy are a reflection of who I am (1 = strongly disagree; 5 = strongly Nan and Heo (2007) agree) Sometimes I am willing to pay more money for a product because of its brand name Nan and Heo (2007) (1 = strongly disagree; 5 = strongly agree) I pay attention to the brand names of products I buy (1 = strongly disagree; 5 = strongly Nan and Heo (2007) agree)
Extension cannot be both substitute and complement to the parent brand at the same time. Therefore, only one of the two can be chosen as a measure. If extension is neither substitute nor complement to the parent brand, still transferability of people, facilities and skills from parent brand to extension may be possible. Therefore, one way of forming product category fit for a brand extension could be to choose the score for substitute or complement whichever turns out to be higher on mean and add that with the score for transferability. However, in this study, only ‘transferability’ was chosen as a measure for product category fit so as to enable comparison of results (obtained from SEM) across different brand extensions.
Consumer evaluation of brand extensions
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
BIF_F2
X’s brand image (in terms of price) fits with washing powder. (1 = strongly disagree; 5 = strongly agree) X’s brand image (in terms of who typically uses X brand) fits with washing powder (1 = strongly disagree; 5 = strongly agree) Quality across brands of washing powder (1 = does not vary at all; 5 = varies a lot)
787
Ramanathan and Velayudhan
Indian Department of Posts, was used as a sampling frame in each city of Chennai and Patna. About half the pin codes in each city were randomly chosen. In the chosen pin codes, one starting household address was selected in Chennai and two starting household addresses were selected in Patna. Subsequent addresses were attempted based on right-hand rule and three households were skipped after each successful attempt. Potential respondents from attempted households were briefed on the study, and questionnaires were administered to willing respondents through interviews. Sample size was estimated using Cochran’s (1977) formula, that is, sample size, n = (tα/2 s/d)2 where t is the value of selected α level in each tail, s is the estimate of standard deviation in the population and d is the acceptable margin of error for the mean being estimated (as cited in Bartlett et al, 2001). A total of 306 questionnaires were administered in Chennai and a total of 326 questionnaires were administered in Patna. For the sample sizes reported above, the margin of error (that is, tα/2 * s/√n; α = 0.05) for population mean of each measure in the questionnaire was found to be less than 5 per cent. Sample mean and standard deviation of measures are provided in the next section in Tables 2 and 3. Having presented the research design, data analysis and results are reported in the next section.
DATA ANALYSIS AND RESULTS Data preparation From the data collected in each location, cases that had the same score across all items were deleted, as done by Volckner et al (2010). Further examination of data for each item revealed minimal evidence of serious outliers, supporting researcher’s recognition that consumers can vary
788
naturally in their perception of brands, products and traits (cf. Hair et al, 2006). The extreme values in the data were not observed to be present owing to data entry error. Since the extreme values were believed to be representative of the population these were retained. For each item, the number of missing values was less than 5 per cent of the cases and these were replaced with sample mean as given in Hair et al (2006). Examination of data for each item using box plot and interquartile range suggested the presence of skewness but the degree of skewness was found to be between 0 and −1.620, which is within acceptable limits of −3 to +3 suggested by Kline (2011). The sample size used for analysis of GG and GS extension is 630. The aggregate sample size achieved is considered acceptable for structural equation model (SEM)-based analysis (exceeds 500, which is recommended when the number of factors in SEM is greater than 6, as per Hair et al, 2006; exceeds the ‘item-to-response’ ratio of 1:10, as per Hinkin, 1995; exceeds the ‘number of cases to number of model parameters estimated’ ratio of 5:1, as per Kline, 2011).
Respondent profile The respondents are 55.2 per cent female and 44.8 per cent male. More than 88 per cent of the respondents are in the age group 20–59 years. Only 12.4 per cent of the respondents studied until ninth standard or less and 33.7 per cent of the respondents completed undergraduate studies. Thus, the sample is relatively biased toward females, relatively young, and more educated people. In general, gender, age and education have not been reported in literature as factors influencing consumer’s attitude toward brand extensions and revision of attitude toward parent brand. Thus, the samples obtained for this study are considered suitable for analysis.
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
Table 2: Mean and Cronbach’s α of measures for goods to goods brand extension Item label
Attitude toward parent brand
APB1
APB2 PTB4 NPBP2 QVBP1 BB_F1 BB_F2 BB_F3 Product category fit PCF_F1 PCF_F2 PCF_F3 Brand image fit BIF_F1 BIF_F2 Attitude toward the extension AE1 AE2 AE3 Revision of attitude toward the parent brand RAPB1 RAPB2 Extension product category involvement EPCI1
Prototypicality of the brand Number of products in brand portfolio Quality variance across brand portfolio Brand breadth
Quality variance across brands in the extension product category Consumer knowledge of extension product category Consumer innovativeness
Consumer brand consciousness
Mean (SD)
4.33 (0.630) 4.57 (0.499) 4.58 (0.534) 3.99 (0.863) 3.79 (1.109) 3.46 (1.179) 3.59 (1.185) 3.66 (1.107) 3.42 (1.147) 3.56 (1.097) 3.83 (0.991) 3.98 (0.687) 3.96 (0.730) 4.11 (0.603) 4.25 (0.571) 4.19 (0.612) 4.22 (0.627) 4.14 (0.632) 4.36 (0.616)
EPCI2 3.99 (0.994) EPCI3 4.24 (0.746) QVBEPC1 4.07 (0.794) CKEPC1
3.88 (0.936)
CKEPC2 CI1 CI2 CI3 CI4 CBC1
3.95 (0.847) 3.53 (1.046) 3.23 (1.099) 3.38 (1.075) 3.46 (1.039) 4.14 (0.713)
CBC2 CBC3
3.63 (0.997) 4.35 (0.622)
Cronbach’s Comments α 0.457
Owing to less than adequate α value, the two-item combination was not pursued. Instead, the single item used by Aaker and Keller (1990) – APB1 – is used in SEM
— — — — — — — — — — — 0.671
The single item used by Zimmer and Bhat (2004) – PTB4 – is used in SEM The single item used by Volckner and Sattler (2006) – NPBP2 – is used in SEM The single item used by Dacin and Smith (1994) – QVBP1 – is used in SEM Formative and so Cronbach’s α not applicable
Owing to adequate α value, all items are selected for CFA
0.588
Owing to adequate α value, all items are selected for CFA
0.531
Owing to less than adequate α value, the three-item combination was not pursued. Instead, the single item used by Volckner and Sattler (2007) – EPCI1 – is used in SEM
—
PCF_F1, PCF_F2, PCF_F3 are formative and so Cronbach’s α not applicable BIF_F1 and BIF_F2 are formative and so Cronbach’s α not applicable
The single item used by Volckner and Sattler (2007) – QVBEPC1 – is used in SEM
0.489
Owing to less than adequate α value, the two-item combination was not pursued. Instead, the single item used by Volckner and Sattler (2007) – CKEPC1 – is used in SEM
0.791
Owing to adequate α value, all items are selected for CFA
0.458
Owing to less than adequate α value, the three-item combination was not pursued. Instead, a single item that was easily understood by respondents – CBC3 – is used in SEM
Consumer evaluation of brand extensions
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
Concept label
789
790 © 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
Concept label
Item label
Attitude toward parent brand
APB1
APB2 PTB4 NPBP2 QVBP1 BB_F1 BB_F2 BB_F3 Product category fit PCF_F1 PCF_F2 PCF_F3 Brand image fit BIF_F1 BIF_F2 Attitude toward the extension AE1 AE2 AE3 Revision of attitude toward the parent brand RAPB1 RAPB2 Extension product category involvement EPCI1 Prototypicality of the brand Number of products in brand portfolio Quality variance across brand portfolio Brand breadth
Quality variance across brands in the extension product category Consumer knowledge of extension product category Consumer innovativeness
Consumer brand consciousness
Mean (SD)
4.33 (0.630) 4.57 (0.499) 4.58 (0.534) 3.99 (0.863) 3.79 (1.109) 3.46 (1.179) 3.59 (1.185) 3.66 (1.107) 3.45 (1.161) 3.71 (1.062) 3.78 (0.980) 4.00 (0.735) 3.94 (0.795) 4.21 (0.609) 4.29 (0.560) 4.25 (0.607) 4.19 (0.623) 4.17 (0.597) 4.44 (0.604)
EPCI2 4.09 (0.891) EPCI3 4.18 (0.732) QVBEPC1 4.04 (0.867) CKEPC1
3.80 (1.028)
CKEPC2 CI1 CI2 CI3 CI4 CBC1
4.01 (0.812) 3.53 (1.046) 3.23 (1.099) 3.38 (1.075) 3.46 (1.039) 4.14 (0.713)
CBC2 CBC3
3.63 (0.997) 4.35 (0.622)
Cronbach’s Comments α 0.457
Owing to less than adequate α value, the two-item combination was not pursued. Instead, the single item used by Aaker and Keller (1990) – APB1 – is used in SEM
— — — —
The single item used by Zimmer and Bhat (2004) – PTB4 – is used in SEM The single item used by Volckner and Sattler (2006) – NPBP2 – is used in SEM The single item used by Dacin and Smith (1994) – QVBP1 – is used in SEM Formative and so Cronbach’s α not applicable
—
PCF_F1, PCF_F2, PCF_F3 are formative and so Cronbach’s α not applicable
—
BIF_F1 and BIF_F2 are formative and so Cronbach’s α not applicable
0.637
Owing to adequate α value, all items are selected for CFA
0.639
Owing to adequate α value, all items are selected for CFA
0.559
Owing to less than adequate α value, the three-item combination was not pursued. Instead, the single item used by Volckner and Sattler (2007) – EPCI1 – is used in SEM
—
The single item used by Volckner and Sattler (2007) – QVBEPC1 – is used in SEM
0.442
Owing to less than adequate α value, the two-item combination was not pursued. Instead, the single item used by Volckner and Sattler (2007) – CKEPC1 – is used in SEM
0.791
Owing to adequate α value, all items are selected for CFA
0.458
Owing to less than adequate α value, the three-item combination was not pursued. Instead, a single item that was easily understood by respondents – CBC3 – is used in SEM
Ramanathan and Velayudhan
Table 3: Mean and Cronbach’s α of measures for goods to services brand extension
Consumer evaluation of brand extensions
Reliability assessment and validation of measures There are 14 constructs used in the analysis as given in Table 1. Out of the 14 constructs, 3 constructs (that is, product category fit, brand breadth and brand image fit) are non-reflective. As indicated in Table 1, product category fit is being included in analysis only using single item (PCF_F3/ ‘transferability’) and does not present the issue of multicollinearity within indicators of a non-reflective construct. For the other non-reflective constructs (brand breadth and brand image fit), the value of collinearity between their respective indicators was observed to be less than 0.5, in the contexts of GG and GS brand extensions, suggesting little evidence of multicollinearity within indicators of a non-reflective construct (Hair et al, 2006). The mean and reliability (Cronbach’s α) of measures for each context (that is, GG and GS) are given in Tables 2 and 3. As seen in Tables 2 and 3, the Cronbach’s α for the measures of four constructs (that is, attitude toward parent brand, extension product category involvement, consumer knowledge of the extension product category and consumer brand consciousness) was much less than 0.7 in all contexts (that is, GG and GS). As has been practiced by researchers (for example, Aaker and Keller, 1990; Volckner and Sattler, 2007), these constructs were therefore modeled in SEM using single item. As seen in Tables 2 and 3, the Cronbach’s α was 0.671 in GG context and 0.637 in GS context for ‘attitude toward the extension’, 0.588 in GG context and 0.639 in GS context for ‘revision of attitude toward the
Figure 3:
parent brand’ and 0.791 in both GG and GS contexts for ‘consumer innovativeness’. These three constructs were selected for confirmatory factor analysis (CFA) because literature-based multi-item measures were available for these constructs, and Cronbach’s α between 0.6 and 0.7 may be acceptable provided that other indicators of a model’s construct validity are good (Hair et al, 2006). CFA, using maximum likelihood estimation method in AMOS 20, was used to validate the multi-item measures. It was found that the average variance extracted (AVE) for each of the constructs ‘revision of attitude toward the parent brand’ and ‘attitude toward the extension’ were less than the squared inter-construct correlation. This suggested a lack of discriminant validity since a construct should explain its own measures better than it explains another construct (Hair et al, 2006). The common remedy suggested for items not performing well with respect to model integrity, model fit or construct validity in CFA is item deletion (Hair et al, 2006). In this case, it was decided that ‘revision of attitude toward the parent brand’ would be dropped from CFA and be modeled as single item in SEM. The item selected for SEM was the one found relatively easier to comprehend by respondents. CFA of two constructs (that is, attitude toward the extension and consumer innovativeness) was conducted. The CFA model is given in Figure 3 and the results of CFA are given in Table 4. As seen in Table 4, the AVE for ‘attitude toward the extension’ is 0.413 for GG context and 0.374 for GS context, and the AVE for ‘consumer innovativeness’ is 0.509 for
CFA model.
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
791
Ramanathan and Velayudhan
Table 4: CFA results for goods to goods and goods to services brand extensions Brand extension
Composite reliability
Goods to goods Goods to services
AE: 0.676 CI: 0.799 AE: 0.640 CI: 0.799
AVE
AE: 0.413 CI: 0.509 AE: 0.374 CI: 0.509
χ2
Squared inter construct correlation
DF
χ2/ DF
IFI
GFI
TLI
CFI
RMSEA SRMR
0.015
32.231** 13 2.479 0.983 0.986 0.972 0.983 0.048 0.0438
0.145
17.823
13 1.371 0.996 0.992 0.993 0.996 0.024 0.0242
**P < 0.01. Note: CI: Consumer innovativeness; AE: Attitude toward the extension.
both GG and GS contexts. Hair et al (2006) state that AVE of 0.5 or higher is a good rule of thumb suggesting adequate convergence, and AVE of less than 0.5 indicates that on average more error remains in the items than variance explained by the latent factor structure imposed on the measure. The composite reliability for ‘attitude toward the extension’ is 0.676 for GG context and 0.640 for GS context. The composite reliability for ‘consumer innovativeness’ is 0.799 for both GG and GS contexts. Hair et al (2006) observe that the rule of thumb for reliability estimate is that 0.7 or higher suggests good reliability, and reliability between 0.6 and 0.7 may be acceptable provided that other indicators of a model’s construct validity are good. The AVE for ‘attitude toward the extension’ and ‘consumer innovativeness’ are greater than squared inter-construct correlation, thus showing evidence of discriminant validity. As seen in Table 4, the χ2 is not significant for GS brand extension but significant for GG brand extension. Other indices of model fit, that is, Normed χ2 (χ2/Degree of freedom), Goodness of Fit index (GFI), Comparative Fit Index (CFI) and Root Mean Square Error of Approximation (RMSEA) are consistent with the range of acceptable values ( < 3 for Normed χ2, > 0.9 for GFI and CFI, < 0.10 for RMSEA) discussed by Hair et al (2006). In addition, Incremental Fit Index (IFI), Tucker Lewis Index (TLI) and Standardized Root Mean Square Residual (SRMR) are also reported in Table 4.
792
Overall, the CFA model was accepted and multi-item measures were used in SEM for ‘attitude toward the extension’ and ‘consumer innovativeness’. The measures used for all constructs in SEM are given in Table 5.
SEM and findings The SEM is given in Figure 4. Correlations are assumed among all exogenous factors except those between parent brand factors (that is, attitude toward the parent brand – APB, number of products in brand portfolio – NPBP, quality variance across brand portfolio – QVBP, prototypicality of the brand – PTB, brand breadth – BB) and extension product category factors (that is, extension product category involvement – EPCI, quality variance across brands in the extension product category – QVBEPC, consumer knowledge of the extension product category – CKEPC). The model was analyzed using AMOS 20 with maximum likelihood estimation method. In each context, review of Mahalanobis 2 D values showed minimal evidence of serious multivariate outliers. Examination of Mardia’s coefficient suggested the presence of multivariate kurtosis (Bentler, 2005 as cited by Byrne, 2010). However, problems arising from issues in achieving multivariate normality in SEM were mitigated by the use of maximum likelihood estimation (Joreskog and Sorbom, 1996). Results regarding model fit are given in Table 6. The χ2 is significant for both brand
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
Consumer evaluation of brand extensions
Table 5: Measures used in SEM Concept
Concept labels used in SEM
Item labels/variable Formation in SEM
Attitude toward parent brand Prototypicality of the brand Number of products in brand portfolio Quality variance across brand portfolio Brand breadth Product category fit Brand image fit Attitude toward the extension
APB PTB NPBP QVBP BB PCF BIF AE
Revision of attitude toward the parent brand Extension product category involvement Quality variance across brands in the extension product category Consumer knowledge of extension product category Consumer innovativeness
RAPB EPCI QVBEPC
APB1 PTB4 NPBP2 QVBP1 BB_F1+BB_F2+BB_F3 PCF_F3 BIF_F1+BIF_F2 AE1 AE2 AE3 RAPB2 EPCI1 QVBEPC1
CKEPC CI
Consumer brand consciousness
CBC
Figure 4:
CKEPC1 CI1 CI2 CI3 CI4 CBC3
Structural equation model.
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
793
Ramanathan and Velayudhan
Table 6: Model fit for goods to goods and goods to services brand extensions Brand extension Goods to goods Goods to services
χ2
DF
χ2/DF
IFI
GFI
TLI
CFI
RMSEA
SRMR
236.489*** 356.106***
88 88
2.687 4.047
0.945 0.915
0.962 0.946
0.890 0.830
0.943 0.913
0.052 0.070
0.0521 0.0732
***P < 0.001.
Table 7: Path coefficients for goods to goods and goods to services brand extensions Standardized path coefficients APB→AE NPBP→AE QVBP→AE PCF→AE AE→RAPB PCF→RAPB
Goods to goods
Goods to services
0.078** −0.004 −0.077** 0.179*** 0.568*** 0.007
0.082** 0.003 −0.081* 0.028 0.770*** 0.092**
*P < 0.05; **P < 0.10; ***P < 0.001.
extensions but other indices of model fit, that is, Normed χ2 (χ2/Degree of freedom), GFI, CFI and RMSEA are consistent with the range of acceptable values given by Hair et al (2006). In addition, IFI, TLI and SRMR are also reported in Table 6. Further, an examination of correlations, for all contexts, revealed that all values are well less than 0.5 and there is little evidence of multicollinearity (cf. Hair et al, 2006). The path coefficients of the structural model (that are the focus of this study) are reported in Table 7.
DISCUSSION OF RESULTS Hypothesis 1 predicted that the positive relationship between attitude toward the parent brand and attitude toward the extension is stronger for goods to goods but weaker for goods to services. As seen in Table 7, there is no effect of attitude toward parent brand (APB) on attitude toward extension (AE) either for GG or GS brand extension at 5 per cent significance level.
794
Therefore, Hypothesis 1 is not supported. Aaker and Keller (1990) also found no effect of attitude toward parent brand on attitude toward extension. As seen in Table 7, at 10 per cent significance level, there is a weak positive effect in GG and GS brand extensions. This suggests that not all respondents may use attitude toward the parent brand as a regular criterion to form their attitude toward the extension. As Feldman and Lynch (1988) posited, response to an earlier item in the questionnaire may be used as an input to a subsequent response if the former is accessible and if it is perceived to be more diagnostic than other accessible inputs. In a study done with several variables to reduce omitted variable bias, such order effects could be reduced, but not eliminated, by rotating the listing of items as was done in this study (through the use of two different versions of the questionnaire). Hypothesis 2 predicted that the positive relationship between number of products in brand portfolio and attitude toward extension is stronger for goods to goods but weaker for goods to services. As seen in Table 7, there is no effect of number of products in brand portfolio (NPBP) on attitude toward extension (AE) either in GG or GS brand extension. Therefore, Hypothesis 2 is not supported. Volckner and Sattler (2007) and Yeu et al (2013) also found no relationship between number of products in a brand portfolio and attitude toward extension. Further, the path coefficient of the relationship between number of products in brand portfolio (NPBP) and attitude toward extension (AE) is negative for goods to goods and positive for goods to
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
Consumer evaluation of brand extensions
services. However, the path coefficients are almost close to zero and this could be because on the one hand number of products in brand portfolio may act as collateral for the extension, but on the other hand it may blur the meaning of the brand for consumer (Smith and Park, 1992). Hypothesis 3 predicted that the negative relationship between quality variance across brand portfolio and attitude toward the extension is weaker for goods to goods but stronger for goods to services. As seen in Table 7, the effect of quality variance across brand portfolio (QVBP) on attitude toward extension (AE) is negative and significant for GS brand extension, and negative and not significant for GG brand extension. Thus, Hypothesis 3 is supported. Hypothesis 4 predicted that the positive relationship between product category fit and attitude toward extension is stronger for goods to goods but weaker for goods to services. As seen in Table 7, the effect of product category fit (PCF) on attitude toward extension (AE) is positive and significant for GG brand extension, and positive and not significant for GS brand extension. Thus, Hypothesis 4 is supported. Hypothesis 5 predicted that the positive relationship between attitude toward the extension and revision of attitude toward the parent brand is stronger for goods to goods but weaker for goods to services. As seen in Table 7, there is a positive effect of attitude toward the extension (AE) on revision of attitude toward the parent brand (RAPB) for both GG and GS brand extensions. However, χ2 difference test does not show any difference in size of effect between GG and GS brand extensions. Therefore, Hypothesis 5 is not supported. Since this is a survey study, respondents’ recognition of goods–services distinction was assumed and not confirmed through manipulation check. Hence, a lack of difference in strength of effect could owe likely to moderate diagnosticity of goods–
services distinction. As Feldman and Lynch (1988) posited, inputs that are perceived to be moderately diagnostic may be ignored when more diagnostic ones can be retrieved, but used when they cannot. Hypothesis 6 predicted that the positive relationship between product category fit and revision of attitude toward the parent brand is stronger for goods to goods but weaker for goods to services. As seen in Table 7, there is no effect of product category fit (PCF) on revision of attitude toward the parent brand (RAPB) either for GG or GS brand extension at 5 per cent significance level. Therefore, Hypothesis 6 is not supported. However, at 10 per cent significance level, there is a weak positive effect in GS brand extension. Martinez and Pina (2010) also found no effect and observed that lack of direct effect could owe likely to strong mediation through attitude toward the extension. Evidence for mediation was assessed through the Kim et al (2001) approach as follows. In SEM, the path from product category fit to attitude toward the extension, and the path from attitude toward the extension to revision of attitude toward the parent brand were constrained to zero. On estimating the parameters, a statistically significant direct relationship was found between product category fit and revision of attitude toward the parent brand. On freeing the paths earlier constrained, the direct relationship between product category fit and revision of attitude toward the parent brand was found to be statistically insignificant. Thus, lack of direct effect is owing to strong mediation through attitude toward the extension. The results on hypotheses are summed up in Table 8.
THEORETICAL CONTRIBUTIONS In this research, the observation on paucity of published studies in brand extension from goods-intensive offerings to services-intensive
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
795
Ramanathan and Velayudhan
Table 8: Results on hypotheses H1: APB→AE GG > GS
H2: NPBP→AE GG > GS
H3: QVBP→AE GG < GS
H4: PCF→AE GG > GS
H5: AE→RAPB GG > GS
H6: PCF→RAPB GG > GS
Not supported
Not supported
Supported
Supported
Not supported
Not supported
offerings by Lei et al (2004) is sought to be addressed. The following are some of the resultant contributions to literature. Study by Lei et al (2004) compared ‘brand extensions from moderately services-intensive offerings to low servicesintensive offerings’ with ‘brand extensions from moderately services-intensive offerings to high services-intensive offerings’. However, Lei et al (2004) noted that the selected parent product used in their study was perceived by respondents as ‘moderate’ in terms of services-intensiveness level and proposed that it is also interesting to see whether the parent product’s initial services-intensiveness level would influence the research findings in the future study. This study considered offerings that are highly goods-intensive and highly services-intensive as opposed to moderately goods-intensive or moderately servicesintensive. This study compared ‘brand extensions from low services-intensive offerings to low services-intensive offerings’ with ‘brand extensions from low services-intensive offerings to high servicesintensive offerings’. Whereas the study by Lei et al (2004) considered two contexts, this study has considered two different contexts thereby providing better resolution and comparison across underexplored contexts. Study by Lei et al (2004) used a parent brand that was associated only with one product category. Given that a study by Dacin and Smith (1994) revealed the effects of brand portfolio characteristics on consumers’ extension evaluations, Lei et al (2004) proposed for future research that it might be valuable to investigate the
796
services-intensiveness effect in a brand portfolio condition. This study used brands that were already extended and thereby considered Dacin and Smith’s (1994) effects of brand portfolio characteristics such as number of products in the brand portfolio and quality variance across brand portfolio on consumer evaluation of brand extensions. Lei et al (2004) observed that various consumers might react to risk differently depending on their personal traits. The effect of consumer traits was not examined in their study, and Lei et al (2004) proposed it for future study. This study controlled for the effects of consumer traits such as consumer innovativeness and consumer brand consciousness on consumer evaluation of brand extensions especially in the underexplored contexts, thereby reducing omitted variable bias. Lei et al (2004) recommended the use of real brand names and products, rather than fictitious ones, for future studies. Real brands were used in this study and also participants were recruited from the general public as opposed to merely students, thus providing better external validity of effects reported in brand extensions literature. The specific contribution of this research is that of explaining the moderating influence of context (GG vis-à-vis GS) on consumer evaluation of brand extensions. Drawing from categorization theory, this study sought to examine the extent to which the perception of difference between GG and GS contexts influences consumer evaluations. Following are findings from this research that provide further support to categorization theory, and that are specific
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
Consumer evaluation of brand extensions
contributions to the brand extensions literature. The negative relationship between quality variance across brand portfolio and attitude toward the extension is found to be stronger for goods to services but weaker for goods to goods. The positive relationship between product category fit and attitude toward extension is found to be stronger for goods to goods but weaker for goods to services. Whereas prior literature is scarce with studies on brand extensions from goods to services, the above findings suggest that brand extensions from goods to goods is preferable to brand extensions from goods to services under some conditions such as when quality variance across brand portfolio is high and when product category fit is present.
MANAGERIAL IMPLICATIONS This study helps addressing the managerial decision: When should a brand manager extend a brand across goods and services? Such a decision from the marketing perspective may also become quite pertinent when a firm strategizes to diversify across different offerings. The results from the study suggest that consumers’ attitude toward the parent brand may not play a role in their acceptance of GG and GS brand extensions. When consumers have favorable attitude toward a firm’s ‘consumer goods’ brand, it should be equally advantageous for the firm to extend the brand into ‘consumer goods’ or ‘consumer services’. Another managerially relevant observation from this study is that number of prior extensions of a brand by a firm does not matter for stretching brand across goods and services. When the number of products in brand portfolio is high, the firm should consider this equally advantageous for extending the ‘consumer goods brand’ into ‘consumer goods’ or ‘consumer services’. Although the number of products in brand portfolio may not matter, the findings
of this study suggest that quality variance across brand portfolio may play a stronger negative role for GS than GG brand extension. When quality variance across brand portfolio is high, marketers may need to exercise caution in extending ‘consumer goods’ brand to consumer service offering. In addition, the results of the study indicate that although product category fit may be assumed to be important for a manager to extend brands within GG and GS contexts, product category fit may play a stronger positive role for GG than GS brand extension. When product category fit is seen to be present both for a potential GG extension and GS extension, it may be more advantageous for a firm to extend the ‘consumer goods brand’ into ‘consumer goods’ than ‘consumer services’.
LIMITATIONS AND FUTURE RESEARCH DIRECTIONS One limitation of this study is a lack of use of multi-item measures. There is a general lack of multi-item measures for constructs in the brand extensions literature and scale development could be undertaken in future research to address the scarcity of multiitem measures. In addition, a qualitative study using focus groups could be conducted in order to further examine the proposed constructs of the study in more depth and may provide very useful supplementary information to further support the findings/assertions of this article. Another limitation of this study is that only one brand and a few product categories were studied allowing scope for replication in the future studies with different brands and product categories. In this study, moderating influence of contexts such as GG and GS was examined. Goods categorized as durable and nondurable, and services categorized into people processing, possession processing, mental stimulus processing and information
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
797
Ramanathan and Velayudhan
processing can possibly influence the effect of brand extension because of consumerrelated variables of involvement and brand consciousness, the latter being controlled in this study. In this study, ‘non-durable to non-durable’ extension is compared with ‘non-durable to people processing’ extension. Therefore generalization of results to other types of comparison has to be made with caution. Further, moderating influence of subcontexts within each context (for example, non-durable to non-durable, non-durable to durable and so on within GG) could be explored in future studies. In this study, survey design was used resulting in better external validity of results. However, laboratory experimental designs could be used in future research to assess the internal validity of findings reported in this study. Finally, since this study was conducted in a single nation, that is, India, the findings reported here may not be strictly generalizable beyond the Indian context (cf. Henseler et al, 2010).
ACKNOWLEDGEMENTS The funding received from Reckitt & Benckiser India Limited toward the field work for this study is gratefully acknowledged. The first author would like to gratefully acknowledge the following academics for their inputs during various stages of this study: Keyoor Purani, Unnikrishnan K. Nair, Shubhasis Dey, Kausik Gangopadhyay, Suram Balasubrahmanyam, Sridhar Guda, Ashok Pratap Arora and one anonymous dissertation examiner.
NOTES 1 Apple: 118.863; Google: 107.439; Coca-Cola: 81.563; IBM: 72.244; Microsoft: 61.154; GE: 45.480; Samsung: 45.462; Toyota: 42.392; McDonald’s: 42.254; MercedesBenz: 34.338. All numbers are in $ billions. 2 Aaker and Keller (1990) mention that if the fit is incongruous, the extension may be regarded as humorous or ridiculous.
798
REFERENCES Aaker, D.A. and Keller, K.L. (1990) Consumer evaluations of brand extensions. Journal of Marketing 54(1): 27–41. Albrecht, C., Backhaus, C., Gurzki, H. and Woisetschlager, D.M. (2013) Drivers of brand extension success: What really matters for luxury brands. Psychology and Marketing 30(8): 647–659. Arslan, F.M. and Altuna, O.K. (2011) Which category to extend to – Product or service? Journal of Brand Management 19(5): 359–376. Bartlett, J.E., Kotrlik, J.W. and Higgins, C. (2001) Organizational research: Determining appropriate sample size in survey research. Information Technology, Learning, and Performance Journal 19(1): 43–50. Bass, A. (2004) Licensed extensions – Stretching to communicate. Journal of Brand Management 12(1): 31–38. Bentler, P.M. (2005) EQS 6 Structural Equations Program Manual. Encino, CA: Multivariate Software. Bhat, S. and Reddy, S.K. (2001) The Impact of parent brand attribute associations and affect on brand extension evaluation. Journal of Business Research 53(3): 111–122. Bottomley, P.A. and Doyle, J.R. (1996) The formation of attitudes towards brand extensions: Testing and generalizing Aaker and Keller’s model. International Journal of Research in Marketing 13(4): 365–377. Bottomley, P.A. and Holden, S.J.S. (2001) Do we really know how consumers evaluate brand extensions: Empirical generalizations based on secondary analysis of eight studies. Journal of Marketing Research 38(4): 494–500. Boush, D. et al. (1987) Affect generalization to similar and dissimilar brand extensions. Psychology and Marketing 4(3): 225–237. Boush, D.M. and Loken, B. (1991) A process-tracing study of brand extension evaluation. Journal of Marketing Research 28(1): 16–28. Broniarczyk, S.M. and Alba, J.W. (1994) The importance of the brand in brand extension. Journal of Marketing Research 31(2): 214–228. Brown, B., Sichtmann, C. and Musante, M. (2011) A model of product-to-service brand extension success factors in B2B buying contexts. Journal of Business and Industrial Marketing 26(3): 202–210. Byrne, B.M. (2010) Structural Equation Modeling with AMOS: Basic Concepts, Applications, and Programming, 2nd edn. New York: Taylor and Francis Group. Chapman, K.J. and Aylesworth, A. (1999) Riding the coat-tails of a positive review: Rave reviews and attitude transfer. Journal of Consumer Marketing 16(5): 418–440. Chen, A.C. and Chen, S.K. (2000) Brand dilution effect of extension failure – A Taiwan study. Journal of Product and Brand Management 9(4): 243–254. Choi, T., Liu, N., Liu, S., Mak, J. and To, Y. (2010) Fast fashion brand extension: An empirical study of consumer preferences. Journal of Brand Management 17(7): 472–487.
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
Consumer evaluation of brand extensions
Cochran, W.G. (1977) Sampling Techniques, 3rd edn. New York: John Wiley and Sons. Czellar, S. (2003) Consumer attitude toward brand extensions: An integrative model and research propositions. International Journal of Research in Marketing 20(1): 97–115. Dacin, P.A. and Smith, D.C. (1994) The effect of brand portfolio characteristics on consumer evaluations of brand extensions. Journal of Marketing Research 31(2): 229–242. Dens, N. and Pelsmacker, P.D. (2010a) Attitudes toward the extension and parent brand in response to extension advertising. Journal of Business Research 63(11): 1237–1244. Dens, N. and Pelsmacker, P.D. (2010b) Advertising for extensions: Moderating effects of extension type, advertising strategy, and product category involvement on extension evaluation. Marketing Letters 21(2): 175–189. Dwivedi, A. and Merrilees, B. (2013) Brand extension feedback effects: An Asian branding perspective. Asia Pacific Journal of Marketing and Logistics 25(2): 321–340. Dwivedi, A., Merrilees, B. and Sweeney, A. (2010) Brand extension feedback effects: A holistic framework. Journal of Brand Management 17(5): 328–342. Echambadi, R., Arroniz, I., Reinartz, W. and Lee, J. (2006) Empirical generalizations from brand extension research: How sure are we? International Journal of Research in Marketing 23(3): 253–261. Estelle, B. (2010) Consumer evaluation of brand extensions from products to services. Master’s thesis, School of Business and Economics, Maastricht University, The Netherlands, http://arno.unimaas. nl/show.cgi?fid=19160, accessed 10 November 2011. Feldman, J.M. and Lynch, J.G. (1988) Self-generated validity and other effects of measurement on belief, attitude, intention, and behavior. Journal of Applied Psychology 73(3): 421–435. Fiske, S.T. and Pavelchak, M.A. (1986) Category-based versus piecemeal-based affective responses: Development in schema-triggered affect. In: R.M. Sorrentino and E.T. Higgins (eds.) Handbook of Motivation and Cognition: Foundations of Social Behavior. New York: Guilford Press, Vol. 10, pp. 167–203. Fu, G., Ding, J. and Qu, R. (2009) Ownership effects in consumers’ brand extension evaluations. Journal of Brand Management 16(4): 221–233. Goh, S.Y., Chattaraman, V. and Forsythe, S. (2013) Brand and category design consistency in brand extensions. Journal of Product and Brand Management 22(4): 272–285. Gurhan-Canli, Z. and Maheswaran, D. (1998) The effects of extensions on brand name dilution and enhancement. Journal of Marketing Research 35(4): 464–473. Hair, J.F., Black., W.C., Babin, B.J., Anderson, R.E. and Tatham, R.L. (2006) Multivariate Data Analysis, 6th edn. New Delhi, India: Pearson Education.
Hem, L.E., Iversen, N.M. and Olsen, L.E. (2013) Category characteristics’ effects on brand extension attitudes: A research note. Journal of Business Research 67(8): 1589–1594. Henseler, J., Horvath, C., Sarstedt, M. and Zimmermann, L. (2010) A cross-cultural comparison of brand extension success factors: A meta-study. Journal of Brand Management 18(1): 5–20. Hinkin, T.R. (1995) A review of scale development practices in the study of organizations. Journal of Management 21(5): 967–988. Iacobucci, D. (1992) An empirical examination of some basic tenets in services: Goods-services continua. In: T.A. Swartz, D.E. Bowen and S.W. Brown (eds.) Advances in Services Marketing and Management. Greenwich, CT: JAI Press, Vol. 1, pp. 23–52. Iacobucci, D. (1998) Services: What do we know and where shall we go? A view from marketing. In: T.A. Swartz, D.E. Bowen and S.W. Brown (eds.) Advances in Services Marketing and Management. Greenwich, CT: JAI Press, Vol. 7, pp. 1–96. Interbrand (2014) Rankings, http://www.bestglobal brands.com/2014/ranking/, accessed 9 January 2015. Iyer, S.G., Banerjee, B. and Garber, L.L. (2011) Determinants of consumer attitudes toward brand extensions: An experimental study. International Journal of Management 28(3): 809–823. Joreskog, K. and Sorbom, D. (1996) LISREL 8: User’s Reference Guide. Lincolnwood, IL: Scientific Software. Kalamas, M., Cleveland, M., Laroche, M. and Laufer, R. (2015) The critical role of congruency in prototypical brand extensions. In: H.E. Spotts (ed.) Assessing the Different Roles of Marketing Theory and Practice in the Jaws of Economic Uncertainty. Proceedings of the 2004 Academy of Marketing Science (AMS) Annual Conference; New York, USA: Springer International Publishing, Vol. 27, pp. 241. Kardes, F.R. and Allen, C.T. (1991) Perceived variability and inferences about brand extensions. In: R.H. Holman and M.R. Solomon (eds.) Advances in Consumer Research. Provo, UT: Association for Consumer Research, Vol. 18, pp. 392–398. Keaveney, S.M., Herrmann, A., Befurt, R. and Landwehr, J.R. (2012) The eyes have it: How a car’s face influences consumer categorization and evaluation of product line extensions. Psychology and Marketing 29(1): 36–51. Keller, K.L. (1993) Conceptualizing, measuring, and managing customer-based brand equity. Journal of Marketing 57(1): 1–22. Keller, K.L. and Aaker, D.A. (1992) The effects of sequential introduction of brand extensions. Journal of Marketing Research 29(1): 35–50. Kim, J.S., Kaye, J. and Wright, L.K. (2001) Moderating and mediating effects in causal models. Issues in Mental Health Nursing 22(1): 63–75. Kline, R.B. (2011) Principles and Practice of Structural Equation Modeling, 3rd edn. New York: The Guilford Press. Klink, R.R. and Athaide, G.A. (2010) Consumer innovativeness and the use of new versus extended
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
799
Ramanathan and Velayudhan
brand names for new products. Journal of Product Innovation Management 27(1): 23–32. Klink, R.R. and Smith, D.C. (2001) Threats to the external validity of brand extension research. Journal of Marketing Research 38(3): 326–335. Lei, J., Pruppers, R., Ouwersloot, H. and Lemmink, J. (2004) Service intensiveness and brand extension evaluations. Journal of Service Research 6(3): 243–255. Loken, B. and John, D.R. (1993) Diluting brand beliefs: When do brand extensions have a negative impact? Journal of Marketing 57(3): 71–84. Lovelock, C. and Gummesson, E. (2004) Whither services marketing? In search of a new paradigm and fresh perspectives. Journal of Service Research 7(1): 20–41. Mann, B.J.S. and Kohli, R. (2012) Do brand acquisitions create wealth for acquiring company shareholders? Evidence from India. Journal of Product and Brand Management 21(4): 265–274. Martinez, E. and Pina, J.M. (2010) Consumer responses to brand extensions: A comprehensive model. European Journal of Marketing 44(7–8): 1182–1205. Mervis, C.B. and Rosch, E. (1981) Categorization of natural objects. In: M.R. Rosenzweig and L.W. Porter (eds.) Annual Review of Psychology. Palo Alto, CA: Annual Reviews, Vol, 32, pp. 89–115. Milberg, S.J., Goodstein, R.C., Sinn, F., Cuneo, A. and Epstein, L.D. (2013) Call back the jury: Reinvestigating the effects of fit and parent brand quality in determining brand extension success. Journal of Marketing Management 29(3–4): 374–390. Milberg, S.J., Park, C.W. and McCarthy, M.S. (1997) Managing negative feedback effects associated with brand extensions: The impact of alternative branding strategies. Journal of Consumer Psychology 6(2): 119–140. Milberg, S.J., Sinn, F. and Goodstein, R.C. (2010) Consumer reactions to brand extensions in a competitive context: Does fit still matter? Journal of Consumer Research 37(3): 543–553. Mitchell, A.A. (eds.) (1982) Models of memory: Implications for measuring knowledge structures In: Advances in Consumer Research. Ann Abor, MI: Association for Consumer Research, Vol. 9, pp. 45–51. Mittal, B. (1995) A comparative assessment of four scales of consumer involvement. Psychology and Marketing 12(7): 663–682. Nan, X. and Heo, K. (2007) Consumer responses to corporate social responsibility (CSR) initiatives: Examining the role of brand-cause fit in causerelated marketing. Journal of Advertising 36(2): 63–74. Nijssen, E.J. and Agustin, C. (2005) Brand extensions: A manager’s perspective. Journal of Brand Management 13(1): 33–49. Nijssen, E.J. and Bucklin, P.L. (1998) The Effect of Involvement Upon Brand Extensions. Nijmegen, The Netherlands: Radboud University. Working Paper. Nijssen, E.J., Bucklin, P.L. and Ujji, R. (1995) The effect of involvement upon brand extensions. In: M. Bergera (ed.) Proceedings of the 24th EMAC Annual
800
Conference. Cergy-Pontoise, France: ESSEC, pp. 1020–1024. Nkowcha, I., Bao, Y., Johnson, W.C. and Brotspies, H.V. (2005) Product fit and consumer attitude toward brand extensions: The moderating role of product involvement. Journal of Marketing Theory and Practice 13(3): 49–61. Park, C.W., Milberg, S. and Lawson, R. (1991) Evaluation of brand extensions: The role of product level similarity and brand concept consistency. Journal of Consumer Research 18(2): 185–193. Ramanathan, J. (2013) Consumer evaluation of brand extensions: Good to service and service to good. Vikalpa: The Journal for Decision Makers 38(2): 81–96. Romeo, J.B. (1991) The effect of negative information on the evaluations of brand extensions and the family brand. In: R.H. Holman and M.R. Solomon (eds.) Advances in Consumer Research. Provo, UT: Association for Consumer Research, Vol. 18, pp. 399–406. Sanjay, P.R. and Rebello, J. (2014) Creditors put Kingfisher Airlines brand up for sale. Live Mint, 7 April, http://www.livemint.com/Companies/ T84PBavRojG7zZadN1T3WO/Kingfisher-brandput-up-for-sale-by-banks-to-recover-dues.html, accessed 25 March 2015. Sattler, H., Volckner, F., Riediger, C. and Ringle, C.M. (2010) The impact of brand extension success drivers on brand extension price premiums. International Journal of Research in Marketing 27(4): 319–328. Sichtmann, C. and Diamantopoulos, A. (2013) The impact of perceived brand globalness, brand origin image, and brand origin – Extension fit on brand extension success. Journal of the Academy of Marketing Science 41(5): 567–585. Smith, D.C. and Park, C.W. (1992) The effects of brand extensions on market share and advertising efficiency. Journal of Marketing Research 29(3): 296–313. Song, P., Zhang, C., Xu, Y. and Huang, L. (2010) Brand extension of online technology products: Evidence from search engine to virtual communities and online news. Decision Support Systems 49(1): 91–99. Song, P., Zhang, C. and Zhang, P. (2013) Online information product design: The influence of product integration on brand extension. Decision Support Systems 54(2): 826–837. Sood, S. and Keller, K.L. (2012) The effects of brand name structure on brand extension evaluations and parent brand dilution. Journal of Marketing Research 49 (3): 373–382. Spiggle, S., Nguyen, H.T. and Caravella, M. (2012) More than fit: Brand extension authenticity. Journal of Marketing Research 49(6): 967–983. Sujan, M. (1985) Consumer knowledge: Effects of evaluation strategies mediating consumer judgments. Journal of Consumer Research 12(1): 31–46. Sujan, M. and Bettman, J.R. (1989) The effects of brand positioning strategies on consumers’ brand and category perceptions: Some insights from schema research. Journal of Marketing Research 26(4): 454–467.
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
Consumer evaluation of brand extensions
Sullivan, M.W. (1990) Measuring image spillovers in umbrella-branded products. Journal of Business 63(3): 309–329. Sunde, L. and Brodie, R.J. (1993) Consumer evaluations of brand extensions: Further empirical results. International Journal of Research in Marketing 10(1): 47–53. Tauber, E.M. (1988) Brand leverage: Strategy for growth in a cost-control world. Journal of Advertising Research 28(4): 26–30. Van Riel, A.C.R. and Ouwersloot, H. (2005) Extending electronic portals with new services: Exploring the usefulness of brand extension models. Journal of Retailing and Consumer Services 12(4): 245–254. Van Riel, A.C.R., Lemmink, J. and Ouwersloot, H. (2001) Consumer evaluations of service brand extensions. Journal of Service Research 3(3): 220–231. Viot, C. (2011) Can brand identity predict brand extensions’ success or failure? Journal of Product and Brand Management 20(3): 216–227. Volckner, F. and Sattler, H. (2006) Drivers of brand extension success. Journal of Marketing 70(2): 18–34. Volckner, F. and Sattler, H. (2007) Empirical generalizability of consumer evaluations of brand extensions. International Journal of Research in Marketing 24(2): 149–162.
Volckner, F., Sattler, H., Hennig-Thurau, T. and Ringle, C.M. (2010) The role of parent brand quality for service brand extension success. Journal of Service Research 13(4): 379–396. Weber, R. and Crocker, J. (1983) Cognitive processes in the revision of stereotypic beliefs. Journal of Personality and Social Psychology 45(5): 961–977. Wernerfelt, B. (1988) Umbrella branding as a signal of new product quality: An example of signaling by posting a bond. Rand Journal of Economics 19(3): 458–466. Winer, R.S. (1999) Experimentation in the 21st century: The importance of external validity. Journal of the Academy of Marketing Science 27(3): 349–358. Woolridge, J.M. (2007) Introductory Economics, 3rd edn. New Delhi, India: Thomson South-Western. Yeu, M. et al. (2013) Effects of product number and brand breadth on the evaluations of an extended product. Asia Marketing Journal 15(3): 97–115. Zeithaml, V.A., Parasuraman, A. and Berry, L.L. (1985) Problems and strategies in services marketing. Journal of Marketing 49(2): 33–46. Zimmer, M.R. and Bhat, S. (2004) The reciprocal effects of extension quality and fit on parent brand attitude. Journal of Product and Brand Management 13(1): 37–46.
© 2015 Macmillan Publishers Ltd. 1350-231X Journal of Brand Management Vol. 22, 9, 778–801
801