C
ONSUMERS’
PURCHASE INTENTION OF PRIVATE LABEL PRODUCTS IN RETAIL:
THE EFFECTS OF PRIVATE LABEL BRANDINGby
C
ONSUMERS’
PURCHASE INTENTION OF PRIVATE LABEL PRODUCTS IN RETAIL:
THE EFFECTS OF PRIVATE LABEL BRANDINGDate: June 28th, 2012
Author: Robert van Ginneken
Studentnumber: s1532391
Adress: Zaagmuldersweg 582
9713 LZ Groningen Mobile number: (+31)(0)651867177
Email: robert_vanginneken@hotmail.com
University: RijksUniversiteit Groningen Department: Business Administration Specialization: Marketing Management Qualification: Masters’ Thesis
MANAGEMENT SUMMARY
In today’s retailing environment, private label brands are becoming more and more important. Manufacturers’ brands are struggling to keep their shelf space in the supermarkets and retailers are introducing private label products in more than just the basic categories. Since private label brands are getting more attention from retailers, aspects of brand management are being applied to these products as well. This paper looks into the branding of the private label products. Most private label brands carry the name of the retailer on the packaging, but sometimes retailers choose to brand the private labels with a name that is meaningful for the category or the segment, but has no link to the retailer. Current practice seems to be one of trial and error. The goal of this thesis is to give insights for retailers into the branding of private label products, whether the private label is already existing or if it is being newly introduced.
This thesis is based on three types of branding that can be applied to private label products: own-name branding, which means that the private label product carries the name of the retailer in the name, endorsed name branding, which means using a new name for the private label product but with the endorsement of the retailers’ name, and lastly, other-name branding, which means using a new name for the private label product, in this case meaningful for the category in which the product is introduced. The effect of these types of brand naming on the product evaluation and the purchase intention is researched. Extraneous factors that are taken into account within this study are the retailer brand equity (service-oriented retailer or price-oriented retailer), product category (category with high hedonic value or category with high utilitarian value), and product segment (discount segment or premium segment). The purchase intention and the product evaluation data were collected using an online survey where participants were shown a number of assortments with differing branding of the private label within these assortments.
The results show that service-oriented retailers have an advantage over price-oriented retailers for branding the private label products with their own name in terms of purchase intention. When looking at the product category, private label products in product categories with high hedonic value, such as wine or chocolate, have significantly higher purchase intention when using an new (other) name for the private label than when using the retailers’ name in case of a price-oriented retailer. Service-oriented retailer however, are known for better quality and can use their own name in this case. Although consumers are less involved when buying products with high utilitarian value, since shopping behavior then is more task-related and rational and brand names play a less important role, using an own-name branding by price-oriented retailers still affects the purchase intention negatively. For the discount segment, price-oriented retailers best use an own-name branding, but service-oriented retailers should use other name branding in order to reduce the effect of brand dilution by linking the inferior discount private label to the retailers’ brand. For the premium segment this is the reverse. Purchase intention is higher when price-oriented retailers make use of another name instead of their own name for the branding. Consumers see this proposition as not credible for this type of retailer. Service-oriented retailers should use their own-name here. Overall, endorsed naming seems not to have significant benefits in relation to the own-name branding strategy.
This paper offers strong support that decisions about branding the private label products by retailers have a large impact on purchase intention and evaluation and that this is an useful marketing tool in order to increase these indicators.
PREFACE
This paper had been written as the final chapter of my master in Marketing Management, specialization of the master program Business Administration at the University of Groningen. With the completion of this thesis I will finish my master study after having finished the bachelor program of Business Administration earlier during my career as a student in Groningen.
The orientation towards retail marketing and my interest in this area of the marketing profession has started with the introductory classes in marketing by prof. dr. P.S.H. Leeflang during the bachelor phase of my study. During my most recent study in Marketing Management, I chose the facultative course of Retail Marketing. This course has instigated my interest for the world of retail even further. After finishing all courses for the master program, I preferred, although it was not part of the master program Marketing Management, to gain some experience in the field of retail marketing by doing an internship. Therefore I solicited at FrieslandCampina (a FMCG company) and worked there for six months at the marketing division where I was given a large number of responsibilities and projects to work on. During this period, I decided that my master thesis would be about retail marketing as well and in September of 2011 I therefore started with my thesis with prof. dr. L.M. Sloot as my supervisor.
The subject of this master thesis has become a mixture of the field of branding, retail marketing and category management. The initial plan was to write this master thesis partly during my internship, but this turned out to be too demanding since the internship was practically a full-time job. In the end though, although this thesis is finalized somewhat later than expected, the result is nonetheless very satisfying.
I would like to thank my supervisor prof. dr. L.M. (Laurens) Sloot for his constructive feedback, help and suggestions during the process. Furthermore I would like to thank my friends and family for their support throughout all the years of my studies.
Groningen, 28/06/2012
TABLE OF CONTENTS
page
1. INTRODUCTION 5
1.1 The emergence of private labels 5
1.2 Problem statement 5
1.3 Research relevance 6
1.4 Structure 6
2 THEORETICAL BACKGROUND 7
2.1 Private labels 7
2.1.1 Reason behind private labels 9
2.1.2 Quality perception of private labels 9
2.1.3 Extending the private label brand 10
2.2 Branding the private label 11
2.2.1 Brand naming strategies 11
2.2.2 Own-name vs. other-name brand naming 12
2.2.3 Endorsed naming strategy 13
2.3 Categories and private labels 13
2.3.1 Hedonic vs. utilitarian value 13
2.3.2 Social influence on shopping behavior 14
2.4 Brand equity 14
2.4.1 Defining brand equity 14
2.4.2 Measuring brand equity 15
2.4.3 Brand equity of retailers & private labels 16
3 CONCEPTUAL MODEL 17
3.1 Hypotheses 17
3.1.1 Retailer characteristics 17
3.1.2 Product category characteristics 17
3.1.3 Product segment characteristics 18
3.1.4 Product evaluation vs. likeliness to buy 19
3.2 Conceptual model 20
4 RESEARCH METHODOLOGY 21
4.1 Research method 21
4.2 Variables & measurement 21
4.3 Subjects & stimuli 23
4.3.1 Retailers 23 4.3.2 Product categories 24 4.3.3 Product names 24 4.4 Research design 25 4.5 Data collection 25 4.5.1 Questionnaire 25 4.5.2 Respondents 26 4.6 Method of analysis 26 5 RESULTS 28 5.1 Demographics 28 5.1.1 Descriptive statistics 28
5.1.2 Demographic difference in customers 29
5.2 Construct reliability & classification 30
5.3 Likeliness to buy private label & retailer brand equity 31
5.4 Likeliness to buy private label & hedonic value 32
5.5 Likeliness to buy private label & product segment 34
5.6 Likeliness to buy private label & product evaluation 36
6 CONCLUSION 39
6.1 Conclusions 39
6.2 Managerial implications 41
6.3 Limitations 42
6.4 Further research 42
1. INTRODUCTION
The first chapter of this paper will introduce the subject of the research. After giving a background about private label brands, the problem statement of the research at hand is explained. Next, the relevance of the research for both science and practice are discussed. This part ends with the structure of the rest of the paper.
1.1 The emergence of private label
Today’s retailing world is hard to imagine without the existence of private label products. Whereas two decades ago the retailers’ shelves were completely dominated by the manufacturer brands, now these manufacturers have to turn in space on the shelves year on year. In 2008, 36 percent of the shelf space in supermarkets was occupied by private label brands in the Netherlands (EFMI, 2008). This is more than the 25,2 percent market share of these private label brands (EFMI, 2008). Europe is leader in private label share in 2009 with Switzerland, the UK and Germany leading the field with 46, 43 and 32 percent value share respectively (NielsenWire, 2010).
In order for these manufacturers to keep level, they have to be continuously more creative in the field of product innovations, packaging and price differentiation. One reason why retailers can be more demanding towards manufacturers is that the margins on the private label products is generally higher than on manufacturer products (Hoch & Banerji, 1993; Ailawadi & Harlam, 2004). The reasons why retailers increasingly choose for private label brands are further discussed in the chapter about the theoretical background. The retailers however are still depending on the manufacturers as well. That is to say, they produce the brands which attract a large number of consumers to the stores of the retailer. This battle between the interests of manufacturers and retailers will continue and may even get more rigid.
1.2 Problem statement
This continuing battle described above means that retailers will have to behave more and more like manufacturers. Branding and brand management principles can and should be applied to retail brands as well (Ailawadi & Keller, 2004). Within certain categories retailers are introducing ranges of private label products from the discount segment to the premium segment. The marketing of all these private label products raises questions about the branding of these products. Should the retailer use its own name for branding these products? How will it affect the image of the retailer and is there a fit between this and the private label brand? Does the branding of the private label depend on the category and/or segment in which it is introduced? These questions lead to the following problem statement: How can retailers use branding of private label products most effectively?
1.3 Research relevance
The Marketing Science Institute (MSI) has stated a number of high priority research topics for the period 2010-2012. One of them is managing brands in a transformed marketplace and the perception of consumers that private label brands are now comparable in quality to national (manufacturer) brands (MSI, 2011). This research can help understand the mechanisms behind the choice of consumers for private label brands. Furthermore, Ailawadi & Keller (2004) state that, although a lot of work has been done on retailer actions and consumer perceptions of retailer image that has direct relevance to branding, there is a gap in academic research on retail branding per se. Next to the scientific relevance, this paper can give insight for managers working with private label brands as well. The conclusions can give insight into the strategic options they have available when introducing private label products and in which occasions certain strategies should be preferred in order to get the optimal return from the introduction. This is relevant especially in the current economic downturn since private label brands seem to have the advantage in that case. Cotterill, Putsis & Dhar (2000) expect private labels to suffer during stronger economic times. During bad economic times however, consumers switch more extensively to store brands than they switch back in a subsequent recovery (Lamey et al., 2007).
1.4 Structure
2. THEORETICAL BACKGROUND
In this chapter the scientific literature regarding private labels and subjects that correlate with private labels and the research at hand is discussed. First is a paragraph about private labels in general. Next the branding of private labels, the categories and brand equity are discussed.
2.1 Private labels
As stated in the introduction of this paper, the retailing environment has changed greatly in the last decades. Not only the emergence of private label brands in itself, but the concentration of retailers has largely increased as well. The three largest buying associations in the Netherlands (AH, Bijéén and Superunie) had a market share of 85,5 percent in 2009 (Distrifood, 2011). With this increasing competition amongst retailers, it has become more important for every retailer to differentiate from the competitors. Since the manufacturer brands are offered by all the retailers (except for the hard discounters), it is hard to differentiate on assortment. Therefore many retailers introduced the private label brands. Starting with the standard private labels, which are positioned against mainstream-quality manufacturer brands, retailers are now introducing a discount (economy) and a premium private label line (Ailawadi & Keller, 2004; EFMI, 2008). Not only the number and type of private label brands have changed over the past decade, the position of the private label brand on the shelf has improved as well (e.g. more private label brands at eye level) (EFMI, 2008).
This paper will focus on the branding strategies of private label brands for two segments, namely the discount segment and the premium segment within a category. In order to show the development of new private label brands in the past years, we will take a view of the different segments of private label brands next. Picture 1 shows a number of private label brands who operate at the discount segment of their respective categories. Although most private label buyers tend to be value conscious and inclined to make use of price deals (Burton et al., 1998), private label brands in the discount segment are specifically designed for the price-sensitive shoppers. These shoppers are essentially concerned with buying their products at the lowest price or getting the best value-for-money (Brown, Pope & Voges, 2001; Jayawardhena, Wright & Dennis, 2007). Arnold & Reynolds (2003) describe these consumers as the value shoppers who look for bargains and discounts. Picture 1 shows a number of discount private label brands of retailers in the UK and the Netherlands. Picture 2 shows an entire range of discount private label products at C1000 (Dutch retailer) across categories.
The first segment for which retailers introduced private label brands is the value segment. This range of private label brands is positioned against the mainstream-quality manufacturer (national) brands (Ailawadi & Keller, 2004). These private label brands generally sell for a slightly lower price than their (manufacturers’) competitors. Picture 3 depicts a range of private label products, from discount to the premium segment, at Tesco’s (UK) within one category.
Currently, retailers are offering private label products that are of a higher quality than the regular private label brands and which improve the overall quality image of private label brands. This will be discussed later. The premium private label brands are positioned at the top end of the market and deliver quality equal to that of premium-quality manufacturers’ brands (Geyskens, Gielens & Gijsbrechts, 2010). Picture 4 gives examples of premium positioned private label brands; Sainsbury’s Taste the Difference, Tesco’s Finest and AH Excellent.
Following the success of the different private label brands, retailers are starting to offer private label brands for specific target groups and seasonal products. Examples of these are organic and fair trade products and seasonal products for holidays. Picture 5 depicts a number of these private label brands.
Picture 4: Premium private label brands at Sainsbury’s (UK), Tesco (UK) and AH (NL)
Picture 3: Example of private label range within category soup (Tesco, UK) (NL)
2.1.1 Reason behind private labels
In literature, there can be found four reasons for retailers to offer private label products. First, the private label products give a higher percent margin for the retailer than manufacturer brands do (Hoch & Banerji, 1993). Ailawadi & Harlam (2004) also show that percentage gross margins and percentage net margins are substantially higher for store brands than for manufacturer brands.
A second reason is that retailers can, through private labels, gain higher negotiating leverage over manufacturers (Ailawadi & Keller, 2004; Geylani, Dukes & Srinivasan, 2007; Narasimhan & Wilcox, 1998). Private labels provide additional market power to retailers since retailers are less dependent on specific upstream suppliers and so can strengthen their bargaining position and extract more profits (Bontemps, Orozco & Réquillart, 2008; Pauwels & Srinivasan, 2004). Kadiyali, Chintagunta & Vilcassim (2000) state that the high quality of private labels in some categories gives retailers a leverage when dealing with manufacturers. Also, the increasing number of private labels makes the shelf space even scarcer and gives retailers even more market power (Kadiyali, Chintagunta & Vilcassim, 2000; Walsh & Mitchell, 2010).
Third, it is assumed that private label brands create loyalty to the retailer. Steenkamp & Dekimpe (1997) state that private label brands can be used to increase store loyalty and that retailer chains have improved the quality of their private label products in order to encourage consumer loyalty to the chain rather than to manufacturer (national) brands. Ngobo (2011) shows that an increase in the number of private labels a retailer carries can increase the share of wallet the consumer devotes to the store. There is however a point where adding more private label products or replacing manufacturer brands with private label brands starts to decrease the store loyalty (Ngobo, 2011). This result is supported by Ailawadi, Pauwels & Steenkamp (2008) who found that the share of wallet initially increases with private label share but begins to decrease beyond a private label share of approximately 40 percent. Store brand usage is also associated with store loyalty (Ailawadi, Neslin & Gedenk, 2001). Beristain & Zorrilla (2011) say that private label brands provide an opportunity to build store image and retailer equity and, ultimately, generate store loyalty.
Lastly, private labels are being used by retailers as a differentiation tool which can reduce price competition amongst the retailers (Bontemps, Orozco & Réquillart, 2008; Bergès-Sennou, Bontems & Réquillart, 2004). Private labels can be a critical component of competitive advantage, especially in an economic downturn (Walsh & Mitchell, 2010).
2.1.2 Quality perception of private labels
the private label brands as well and a growing number of consumers now view private label brands as being comparable to manufacturer brands in terms of quality (Walsh & Mitchell, 2010; Quelch & Harding, 1996).
2.1.3 Extending the private label brand
The subject of brand extensions is thoroughly researched in the past years. We can see the expansion of private label brands by retailers into new categories and segments as a kind of brand extension and therefore we will discuss brand extension literature in relation to private label brands in this next paragraph.
Well-known retailers can use their retailer brand name for the branding of a large number of private label products. They can take advantage of their reputation for quality by using this brand and transfer these associations to the private label brand (Erdem, 1998). In order to leverage the existing brand equity, a retailer can use the same brand name and use a brand-extension strategy (Pina, Iversen & Martinez, 2010; Völckner & Sattler, 2006). The idea behind this brand-extension strategy is that the brand imagery of the parent brand (in this case the retailer brand) is transferred to the brand extension through the reciprocal transfer of associations and emotions (Pina, Iversen & Martinez, 2010; Czellar, 2003). This reciprocal effect can enhance or dilute the original retail brand (Czellar, 2003). These enhancement and dilution effects can be named the forward and the backward feedback (or spillover) effects (Pina, Iversen & Martinez, 2010; Balachander & Ghose, 2003). These theories could work as well for retailers and their private label brands. When we see the retailer brand as the parent brand and the private label brand as brand extensions, we can expect the same spillover effects to take place. We will discuss the impact of the brand image and brand equity of the parent brand on the branding of the private label brands in the next chapter.
feedback effect. They found that high-quality extensions improve the parent brand evaluation far more than lower-quality extensions dilute it (Heath, DelVecchio & McCarthy, 2011). Lei, De Ruyter & Wetzels (2008) find that consumers have lower risk perceptions and more positive evaluations in case of a lower-quality extension. They argue that this can explain why lower-quality extensions are much more often sighted in the marketplace and that it has become a competitive necessity (Lei, De Ruyter & Wetzels, 2008). High-quality extensions however, can be introduced in order to gain higher profits in this segment and most importantly reinforce the parent brand image and attract new consumers to the market segment (Aaker, 1997; Lei, De Ruyter & Wetzels, 2008). It can be noted though that through this strategy the quality variation in private label products increases through either downscale or upscale line extensions and thereby can cause the consumers to become less confident in the private label brand name as a signal of a given quality level (Geyskens, Gielens & Gijsbrechts, 2010).
2.2 Branding the private label
Brands are important tools that influence the role of a product. Fischer, Völckner & Sattler (2010) define two major functions of brands based on a literature research, namely risk reduction and social demonstrance. The consumer uses what he/she knows about the brand in terms of overall quality and specific characteristics. With this consumers can form reasonable expectations about the functional and other benefits of the brand (Fischer, Völckner & Sattler, 2010). Through this, brands can reduce risks in product decisions (Keller, 2008; Smith & Park, 1992; Hoeffler & Keller, 2003). The social function of brands means that brands can serve as symbolic devices that allow consumers to project their self-image (Matzler et al., 2011; Keller, 2008). Brands are also used to communicate and show to others the type of person he or she is or would like to be seen as (Matzler et al., 2011; Escalas & Bettman, 2005; Keller, 2008). An example is that sometimes consumers buy the more expensive brand or manufacturers’ brand in case they have visitors whereas they would normally buy the (discount) private label brand. Since retailers introduced the premium brands and since the quality of the private label brands is regarded as higher than when they appeared on the market (Walsh & Mitchell, 2010; Quelch & Harding, 1996), the social function of private label brands has improved as well. In this paper we will examine one of the most important extrinsic cues of a product, namely the brand name. When the consumer has limited prior knowledge of a product (which is the case when introducing new private label brands, especially in new segments), the brand name may be the most accessible and diagnostic cue available (Hoeffler & Keller, 2003). The brand name is important since it captures the central theme or key associations of a product in a very compact and economical fashion (Keller, 2008).
2.2.1 Brand naming strategies
In this paper, we distinguish three different brand naming strategies based on literature. The first is a direct naming strategy (own-name branding strategy), were the parent brand, in this case the retailer brand, becomes the brand name of the product (Vanhonacker, 2007; Ngobo, 2011; Rao, Agarwall & Dahlhoff, 2004). An example of this can be Tesco Tomato Soup (See picture 3). A second strategy makes use of the parent brand but only by endorsement (Rao, Agarwall & Dahlhoff, 2004). Vanhonacker (2007) names this a brand-bridging strategy. An example of this naming strategy is Bounty by P&G. The advantage of this strategy is that the
Brand naming strategy Sub-strategy Examples Own-name branding Direct naming strategy
Endorsed naming strategy
Tesco Tomato soup; Sainsbury’s Cola; AH Orange Juice.
Perla by Albert Heijn; Taste the Difference by Sainsbury’s
Other-name branding Delicata; Euroshopper; Djoezz
Table 1: Brand naming strategies and examples
2.2.2 Own-name vs. other-name brand naming
Using the retailer brand name for the private label products (own-name branding strategy) has certainly advantages. Collins-Dodd & Lindley (2003) state that private label brands, when included with the retailer name or logo, can be viewed as an extension of the brand name of the retailer. Retailers can leverage the positive attitudes consumers hold to increase awareness and build positive retailer brand attitudes by using packaging, merchandising and advertising strategies to reinforce the linkage between the private label products and the retailer brand (Collins-Dodd & Lindley, 2003; Wernerfelt, 1988). The retailers’ name provides a signal about the quality of the stores’ products (Erdem & Swait, 1998). Another advantage is that search costs for consumers are lower in case of a high number of private labels with the retailer brand name. These private label brands provide easily recognizable cues for simplifying the buying process (Ailawadi, Neslin & Gedenk, 2001). Having the same retailer brand name on products in a wide array of categories strengthens awareness and recall of the retailer brand and may facilitate consumer’s decision making (Ailawadi & Keller, 2004). Dhar & Hoch (1997) justify the use of the retailer brand name for private label products since it enhances the private label performance in all studied categories.
As said earlier however, the introduction of discount and premium private labels can cannibalize sales of the regular private label products. Geyskens, Gielens & Gijsbrechts (2010) suggest that in order to counter this effect, retailers can create stand-alone brands instead of subbrands under the retailer brand name (an other-name branding strategy). Consumer perceptions of a private label product which is branded with the retailer brand name will affect the impression of the retailer as a whole (Ailawadi & Keller, 2004; Beristain & Zorrilla, 2011; Collins-Dodd & Lindley, 2003). Therefore it can be argued that a high service oriented retailer should not use the own retailer brand name on its discount private label brand in order not to harm the retailer brand image. On the other hand, a premium quality private label under the retailer brand name of a price-oriented (or discount) retailer, would not be credible in the eyes of the consumer. These effects can therefore affect the private label brand performance as well as affect the brand image of the retailer. Ailawadi & Keller (2004) suggest that if the retailer chooses not to use the retailer brand name for private label products, these feedback effects would be less strong. In addition to this reasoning, Ngobo (2011) showed that the strategy of multiple private label brand names is more effective than having all the private labels under the retailer brand name.
brand name can have the capability of linking new brand associations at a later time. Some studies show that the overall liking among respondents was higher for meaningful brand names than for unrelated brand names (Kohli & Suri, 2000; Kohli, Harich & Leuthesser, 2005). Meaningful brand names however, are less distinctive and accordingly have less protection from trademark infringement (Klink, 2003).
2.2.3. Endorsed naming strategy
In literature however, it is suggested that an endorsement strategy (brand-bridging strategy) is a best-of-both-worlds strategy. Keller & Lehmann (2006) state that when an extension is given another name in addition to the parent brand name, this can effectively shield a parent brand from dilution effects. In case of a product failure, the strength of the endorsement by the parent brand can moderate the perceived responsibility (Einwiller, 2006). As said, with this strategy, the extension can benefit from the consumers’ familiarity with the parent brand name as well (Vanhonacker, 2007). This middle-of-the-road option can be a solution to the findings that new brands can work as good as brand extensions depending on the situation at hand (McCarthy, Heath & Milberg, 2001; Klink & Athaide, 2010). Instead of using a new brand name as suggested above, the endorsement strategy can create reduced risk for the consumer, since a retailer with an established reputation is backing the product (Keller, 2008).
2.3 Categories and private labels
The branding of the private label can depend on multiple factors such as the category in which it will be introduced or the image of the retailer. The latter we will define in the next chapter. Here we will take a look at branding the private label depending on the different categories in which the private label can be introduced and its effect on the proposed branding strategy.
2.3.1 Hedonic vs. utilitarian value
Consumers differ in their shopping behavior regarding to different product categories. Product categories can be differentiated based on offering high hedonic value or high utilitarian value (Dhar & Wertenbroch, 2000; Batra & Ahtola, 1991). Product categories with high hedonic value provide more experiential consumption, fun, pleasure and excitement (e.g. wine, chocolate), whereas product categories with high utilitarian value are primarily instrumental and functional (e.g. detergent, toilet paper) (Dhar & Wertenbroch, 2000). Products that aim at hedonic consumption evoke those facets of behavior that relate to the multisensory, fantasy, and emotive aspects of consumption (Hirschman & Holbrook, 1982). This distinction is closely related to the subject of category involvement. Involvement generally refers to a person’s perceived relevance of the stimulus object based on inherent needs, values and interests (Zaichkowsky, 1994). We can imagine that a shopper in a retail setting will be more involved for a product category such as wine (with high hedonic value) than for a product category such as kitchen towels.
2.3.2 Social influence on shopping behavior
Furthermore, we can look at the social influence when consumers buy private label brands. Earlier, we stated that consumers could be more likely to buy products that are more luxury (such as premium private label brands or manufacturer brands) if they shop for a social event. The overall importance of brands is greater in categories when consumption is more visible to others (Fischer, Völckner & Sattler, 2010). Bearden & Etzel (1982) found that the consumption of products should be visible to others in order to capitalize on the social demonstrance effect of the brand. This brings us to the issue that brands can signal what type of person a consumer is or would like to be. Consumers select brands that represent their personal attitudes and feelings (Jones & Kim, 2011) Consumers look for a fit between the consumers’ self and the brand personality or image (the self-congruence of the brand) (Malär et al. 2011).
With this in mind, we can argue that branding the private label can depend on the role of the category. Consumers will pay more attention to the brand and the brand name for products that will be used in a social context or that will express the person he or she is or would like to be. This situation will be more often the case when shopping in a category with high involvement and with high hedonic value. Therefore, we can state that an other-name branding strategy for these categories can be more relevant in that the brand name then can expresses a category-specific benefit.
An example of the fact that brands and brand names are less relevant in low-involvement utilitarian categories is that Proctor & Gamble had quit most of the paper towels and other categories with their higher value brands in the late 1990’s. This was caused by more intense price competition and most of all the rise in the market share of private label brands in these categories (the market share was higher than 70 percent in many markets) (Fischer, Völckner & Sattler, 2010).
2.4 Brand equity
Next we will discuss the brand equity and brand image of the retailer and its effect on the branding strategies that could be used by these retailers. First, an overview of the brand equity concept is given. Hereafter we will apply it to the branding strategies as mentioned earlier in this paper.
2.4.1 Defining brand equity
Brand awareness consists of brand recognition and brand recall. Brand recognition is the ability of the consumer to recognize prior exposure when prompted with the brand. Brand recall is their ability to retrieve the brand from memory in case a category or a purchase situation is given (Keller, 2008). The types of brand associations that form the brand image can be classified into three categories: attributes, benefits and attitudes (Keller, 1993). Attributes are the descriptive features that characterize a product or service. These can be divided in product-related attributes (the attributes necessary for performing the function) and non-product-related attributes (external aspects of the product or service that relate to its purchase or consumption). Benefits are the personal value consumers attach to the product or service. Lastly, brand attitudes are defined as consumers’ overall evaluations of a brand (Keller, 1993). See figure 1 for an overview of the brand equity concept according to Keller (1993).
Next to the types of brand associations, the brand image also consists of the strength, the favorability and the uniqueness of these brand associations. The strength of the associations is based on how deep a consumer thinks about product information and relates it to existing brand knowledge. Favorable brand associations are those associations that are desirable to consumers, successfully delivered by the product and conveyed by the supporting marketing program (Keller, 2008). The uniqueness of brand associations is based on the essence that a brand should have unique selling proposition that gives consumers a compelling reason why they should buy it (Keller, 2008).
2.4.2 Measuring brand equity
Keller (2008) says that brand equity can be measured in two ways, namely directly and indirectly. Direct measurement means measuring its potential sources such as the aspects of brand awareness and brand image. Measuring indirectly means measuring outcomes and requires to estimate the benefits that are created from the sources of brand equity (Keller, 2008). Fischer, Völckner & Sattler (2010) say that the willingness of consumers to pay a price premium and to build a loyal brand relationship (both potential sources) translates into a higher overall level of brand equity.
Sources of brand equity Brand awareness Brand image Brand recall Brand recognition Strength of brand associations Uniqueness of brand associations Favorability of brand associations Types of brand associations Attitudes Benefits Attributes Product-related Non-product-related Symbolic Experiential Functional
This is based on the assumption that when brands are more relevant to the consumers, they should be more willing to pay a higher price for the branded products and should be more loyal to their preferred brand. Ailawadi, Lehmann & Neslin (2003) also underline this theory and propose a revenue premium measure. This measure is based on the revenue a brand achieves in the market less the revenue it would achieve if it had no brand name (Ailawadi, Lehmann & Neslin, 2003). Aaker (1996) also states that the price premium a consumer is willing to pay may be the best single measure of brand equity since any driver of driver equity should eventually affect the price premium.
Based on a large number of frameworks in the literature about brand equity and customer-based brand equity (used interchangeably in literature) Netemeyer et al. (2004) propose and test four facets of customer-based brand equity. These are perceived quality (PQ), perceived value for the cost (PVC), uniqueness and the willingness to pay a price premium. Perceived quality is here defined as the customer’s judgment of the overall excellence, esteem, or superiority of a brand relative to alternative brands (Netemeyer et al., 2004). Perceived value for the cost is defined as the consumer’s overall assessment of the utility of the brand based on perceptions of what is received and what is given (e.g. price and nonmonetary costs) relative to other brands. Uniqueness is the degree to which customers feel the brand is different from competitors. Lastly, the willingness to pay a price premium is defined as the amount a customer is willing to pay for their preferred brand over a comparable or lesser brand (Netemeyer et al., 2004).
2.4.3 Brand equity of retailers & private labels
As discussed earlier, consumers can be less involved in some categories than others. In these categories, brand awareness alone can be enough to create favorable responses (Keller, 2008). This can be applied to the branding strategies in that a retailer could best use an own-name strategy in these situations since consumers base their choice merely on familiarity. If the imagery of the brand is taken into account as well, we can argue that a retailer with a high brand equity (and thereby has a high perceived quality and consumers are willing to pay a price premium) can more easily extend their private label brands into new premium segments. This can be based on the idea that consumers take into account the credibility of the private label brand. This can be defined as the extent to which consumers believe a product can deliver the benefits that it promises and satisfy the consumers’ needs and wants. The credibility of a brand as a signal of a products position may increase the perceived quality and decrease information costs and the risk perceived by consumers (Erdem & Swait, 1998).
If we apply the brand naming strategies to this, we can say that retailers with high brand equity should probably choose an own-name branding strategy when introducing a premium private label brand. If the private label brand is highly credible, the risk for buying a private label brand is lower (Ngobo, 2011). On the other hand, a retailer with low brand equity should probably choose an other-name branding strategy when introducing a premium private label brand since consumers may use brand names as a signal of the credibility of the product claims (Hoeffler & Keller, 2003).
3. CONCEPTUAL MODEL
Now that we have given an overview of the theoretical background of this paper, we can explain the conceptual model of the study and its hypotheses that are derived from that model. We will start with a description of the hypotheses that make up the conceptual model. These are formed in order to answer the following problem statement given in the introduction: Which brand naming strategy is the best option for a retailer introducing a private label brand (either discount or premium) given the circumstances? This will depend on three variables, namely the retailer characteristics, the product category characteristics and the product segment characteristics. Using this classification we will form the hypotheses in the next paragraph.
3.1 Hypotheses
First, we have to explain the different brand naming strategies a retailer can choose since the hypotheses that are being introduced use these to explain the effect of the variables. As described in the chapter about the theoretical background (see chapter 2), we state two main brand naming strategies: Own-name brand naming strategy and other-name brand naming strategy. The former can be divided in a direct naming strategy (using the retailers name only) and an endorsed naming strategy (using the retailers name as endorser of the brand).
3.1.1 Retailer characteristics
The choice of brand naming strategy can depend first of all on the characteristics of the retailer and then most importantly the retailers brand image and equity. We can argue that when a retailer which is known for its price-oriented offering introduces a private label branded with the retailers’ name, this private label can be seen as less credible by consumers compared to when a retailer which is known for service and quality would do the same. Although the product would be the same, the name of the price-oriented retailer would make it less credible in the eyes of the consumer and therefore the likeliness to buy would also be lower. Consumers would have a higher likeliness to buy private label products of a service-oriented retailer since the quality of these private label products is regarded higher. When using an endorsed name branding strategy, the consumer will be less able to link the private label product to the retailer and this effect will be lower. Following this reasoning the first hypothesis would be:
H1a: The likeliness to buy private label products is higher for a service-oriented retailer than for a price-oriented retailer in case of an own-name branding strategy.
H1b: The likeliness to buy private label products is equal for a service-oriented retailer or a price-oriented retailer in case of an endorsed name branding strategy.
3.1.2 Product category characteristics
utilitarian value is described as task-related and rational (Babin, Darden & Griffin, 1994). Here the emotional value of a brand name will play a much smaller role. Also, the hedonic product categories are more often subject of choice in case of shopping for social events. There is a greater importance for brands when consumption is more visible to others (Fischer, Völckner & Sattler, 2010). Furthermore, consumer select brand that represent their personal attitudes and feelings (Jones & Kim, 2011). Following this reasoning we can state the next hypothesis:
H2a: The likeliness to buy private label products is higher when using an other-name branding strategy than when using an own-name branding strategy in case of a product category with high hedonic value. H2b: This effect will be smaller in case of a service-oriented retailer than a price-oriented retailer.
H2c: The likeliness to buy private label products is equal when using an other-name branding strategy or when using an own-name branding strategy in case of a product category with high utilitarian value.
3.1.3 Product segment characteristics
The branding of the private label can also depend on the segment in which the private label is being introduced. For discount private label brands for instance, using an own-name branding strategy would be highly credible in case of a price-oriented retailer. For the same retailer, an own-name branding strategy for a premium private label brand can be far less credible. This is based on the theory that consumers may use brand names as a signal of the credibility of the product claims (Hoeffler & Keller, 2003). We expect that a service oriented retailer can use its own name for discount private labels as well and the likeliness to buy would not be lower. In this last situation however, the association of the retailers brand with discount (lower quality and price) products can deteriorate the brand name since consumer perceptions of a private label product which is branded with the retailer brand name will affect the impression of the retailer as a whole (Ailawadi & Keller, 2004; Beristain & Zorrilla, 2011; Collins-Dodd & Lindley, 2003). In the same way, a premium private label brand, using an own-name strategy, can be highly credible for a service-oriented retailer. This leads to the following hypothesis:
H3a: The likeliness to buy discount private label products is higher when using an own-name branding strategy than when using an other-name branding strategy in case of a price-oriented retailer.
H3b: The likeliness to buy discount private label products is equal when using an own-name branding strategy or when using an other-name branding strategy in case of a service-oriented retailer.
H3c: The likeliness to buy premium private label products is lower when using an own-name branding strategy than when using an other-name branding strategy in case of a price-oriented retailer.
H3d: The likeliness to buy premium private label products is higher when using an own-name branding strategy than when using an other-name branding strategy in case a service-oriented retailer.
3.1.4 Product evaluation vs. likeliness to buy
Consumers can evaluate a product offering as very good and still would not buy the private label product. Consumers who buy mainly national brands for instance, are less inclined to buy a product from a private label brand although they evaluate the offering positively. Consumers shopping for low prices however, can show the opposite behavior. They might evaluate a proposition lower than they say they are likely to buy the same proposition. This is because consumers are willing to sacrifice quality of a product when they receive a price benefit in return. Expected is that this effect is mainly noticeable for discount private label products, since the price benefit for these products are the largest.
Second, there might be a difference in the evaluation of private label propositions when using an endorsed naming strategy or when using an own-name branding strategy. We expect premium private label products from service-oriented retailers to be evaluated as good. When using an endorsed naming strategy however, consumers can link the product less easily to the retailer and therefore rate the product lower if the cannot link the product to the retailer. The opposite effect is true for the price-oriented retailers. If the link is less easily made by using an endorsed naming strategy, the consumers will evaluate the propositions as being better than they would do if the link to the retailer is very clear, which is the case with an own-name branding strategy. This leads us to the following hypotheses:
H4a: In case of a discount private label product, the likeliness to buy is equal to the evaluation of the proposition in case of a service-oriented retailer.
H4b: In case of a discount private label product, the likeliness to buy is inversely proportional to the evaluation of the proposition in case of a price-oriented retailer.
H4c: The evaluation of the proposition of a premium private label product is lower when using an endorsed naming strategy than when using an own-name branding strategy in case of a service-oriented retailer. H4d: The evaluation of the proposition of a premium private label product is higher when using an endorsed
naming strategy than when using an own-name branding strategy in case of a price-oriented retailer.
High High
Discount PL Premium PL Own-name branding strategy Other-name branding strategy
Discount PL Premium PL Own-name branding strategy Other-name branding strategy
Low Low
Likeliness to buy / Price-oriented retailer Likeliness to buy / Service-oriented retailer
3.2 Conceptual model
The above formulated hypotheses lead to the following conceptual model (see figure 3). In this conceptual model there is one independent variable (predictor variable), namely the type of brand naming strategy of the private label. In combination with three extraneous variables, namely retailer characteristics, product category characteristics and product segment characteristics, this leads to the different treatments that are used in this research. There are two dependent variables which are the likeliness to buy the private label brand and the evaluation of the private label offering. The effect of the extraneous variables in combination with the brand naming strategies on the likeliness to buy and the product evaluation are explained in the earlier paragraphs and in the theoretical background. The method of research and the definition of the variables and concepts is the subject of the next chapter.
Product category characteristics
- Hedonic value vs. utilitarian value
Product segment characteristics
- Discount segment vs. premium segment
Retailer characteristics
- Service oriented vs. price-oriented
Brand naming strategy
- Own-name brand naming strategy - direct naming strategy - endorsed naming strategy - Other-name brand naming strategy
Likeliness to buy (consumers
preference for the PL)
Figure 3: Conceptual model
4. RESEARCH METHODOLOGY
This chapter gives a description in detail of the methodology of the research in the paper. First we will describe the research method. Hereafter, an explanation of the different dependent, independent and extraneous variables and their measurements is provided. Next, a description of the stimuli and the data collection is given and the chapter concludes with a description of the analysis methods used.
4.1 Research method
With investigating the link between the brand naming strategy of private label brands and the likeliness to buy, as previously stated in the conceptual model, we look at a cause-and-effect relationship. The three controlled extraneous variables (see conceptual model) only either dilute or strengthen this relationship. In short, we want to find which brand naming strategy is best under what circumstances in order to reach the highest likeliness to buy of the private label brand. For this reason, we will make use of an experiment. An experiment is formed when the researcher manipulates one or more independent variables systematically and measures their effect on one or more dependent variables, while controlling for the effect of extraneous variables (Malhotra, 2007; Cooper & Schindler, 2008). We choose for an experimental research since Malhotra (2007) also states that the main method of causal research is experimentation. In this research the variable which is to be manipulated is the independent variable type of brand naming strategy and we will look at its effect on the likeliness to buy and the product evaluation (the dependent variables).
4.2 Variables & Measurement
The independent and the extraneous variables will be explained next, together with the dependent variables mentioned earlier in the conceptual model.
The independent variable of brand naming strategy is based on theory and consists of three levels: Direct naming, endorsed naming and other-naming. This variable will be manipulated in this study. There are two dependent variables in this study. The first is the likeliness to buy, which is here simply defined as the likeliness that the respondent would buy the product if they would have the choice as given in the stimuli of the research. The measurement used for the research is based on the measurements of the willingness to buy as used by Dodds, Monroe & Grewall (1991) and the likelihood of purchasing as used by Steenkamp, Batra & Alden (2003). The second is the product evaluation. This is here defined as the assessment of the strengths and weaknesses of a specific product offering. This dependent variable is measured using a number of indicators based on the measurement of perceived quality and perceived value as used by Dodds, Monroe & Grewall (1991). Both dependent variables are to be measured using statements and 7-point Likert scales. The internal and external consistency of these will be assessed as well.
a scale based on the HED/UT scale developed by Voss, Spangenberg & Grohmann (2003). See table 2 for the scale and its items.
The second extraneous variable is the product segment characteristic. This variable has two levels and is based on the quality and pricing of the product segment at hand. This can thus either be a discount segment, which is a segment with a lower quality than the ‘standard’ private label brands and sells for a lower price as well. The second level is that the segment is a premium segment, were the product within this segment are generally of higher quality and sell for a higher average price than the standard private label brands. A more in-depth explanation of these levels and the segments within private label brands, see the theoretical background.
The third and last extraneous variable is the retailer characteristic. This again has two levels and is either service-oriented or price-oriented. The definition of a service-oriented retailer is that it is based on providing excellent service and a wide product assortment and availability. This is generally paired with slightly higher prices. The price-oriented retailer is focused at offering products for low prices and thereby often offering a smaller assortment and less service within the store. As with the product category variable, this variable is not one or the other. There is a line along which the retailers can be placed according to their image (service- vs. price-oriented). For this variable we will take the view of the consumer of the retailer as the starting point since he or she will base their opinion on their own experiences with the retailer and not on the image the retailer says it would like to be viewed as. Aaker (1991) defines the brand equity as a set of brand assets and liabilities linked to a brand, its name and symbol, that add to or subtract from the value provided by a product or a service to a firm and/or to that firm’s customers. In this research we focus at the retailers’ brand equity. We will base our measurement and scale on the work of Arnett, Laverie & Meiers (2003) and Jinfeng & Zhilong (2009) who both developed a scale for measuring retailers’ brand equity. For an overview of these variables and the previous research on which the measurement and scales are based, see table 3.
Construct Items
Hedonic value (HED/UT) Utilitarian
UT1 Effective/ineffective
UT2 Functional/not functional
UT3 Necessary/unnecessary
UT4 Practical/impractical
Hedonic
HED1 Not fun/fun
HED2 Dull/exciting
HED3 Not delightful/delightful
HED4 Enjoyable/unenjoyable
Retailer brand equity (RBE) RBE1 This retailer is very well known
Next to the variables mentioned in the conceptual model and the above paragraph, we will take into account some demographic variables and the respondents preference in order to see if there is a correlation between these and the outcomes of the likeliness to buy and evaluation of the private label products.
4.3 Subjects & Stimuli
Next, the stimuli and the subject will be discussed. First, we will start with the chosen retailers and the product categories. Then the stimuli regarding the product names and the assortments are discussed. These stimuli form the basis of the questionnaire which can be found in exhibit 1 and will be further explained in paragraph 4.5.1.
4.3.1 Retailers
The choice of retailers that are subject in this research is important since it has to distinguish between the service-oriented versus the price-oriented retailers. Table 4 shows the retailers chosen for this research, using face validity checks, and their respective orientation. This selection is based on the price policy of these retailers and the service and assortment they offer to consumers. Since it is important that the chosen retailers belong to the category of orientation they are assigned to, the orientation of the retailers is measured in the questionnaire. This measurement used is based on the literature mentioned earlier and shown in table 3. The items used for this measurement can be found in exhibit 1. The construct of brand equity is measured using a multi-item scale with a 7-point Likert scale (strongly agree vs. strongly disagree) with a neutral midpoint.
Retailers Service-oriented Price-oriented
Albert Heijn Dirk
Plus C1000
Table 4: Retailers and their orientation
Variable Previous research
Likeliness to buy
(dependent variable)
- Willingness to buy
Dodds, Monroe & Grewall (1991); Gamliel & Herstein (2007) - Purchase likelihood
Steenkamp, Batra & Alden (2003) Product evaluation
(dependent variable)
- Product evaluation (product quality & product value) Dodds, Monroe & Grewall (1991);
Richardson, Jain & Dick (1996) Hedonic value - HED/UT scale
Voss, Spangenberg & Grohmann (2003) Retailer brand equity - Retailer equity
Arnett, Laverie & Meiers (2003); Jinfeng & Zhilong (2009)
- Brand equity
Aaker (1991); Keller (2008)
4.3.2 Product categories
The product categories chosen for the research are based on the associations consumers have with these categories. Table 5 shows the chosen categories based on their face validity. In order to measure if the perception of these product categories by the respondents is the same, a multi-item scale is used to measure the hedonic versus the utilitarian value of these product categories. This multi-item scale is based on the HED/UT scale by Voss, Spangenberg & Grohmann (2003) as mentioned before. Using 7-point scales, respondents are asked to state their associations with the product categories. If respondents think neither word fits the product category, there is a neutral midpoint in the scales. In total 8 scales are presented, whereby 4 are of utilitarian nature and 4 of them are of hedonic nature.
Product category High hedonic value Low hedonic value
Chocolate Paper towels
Ice cream Detergent
Table 5: Product categories and the hedonic value
4.3.3 Product names
In order to see whether different names for the private label products evoke different responses from respondents when the product itself remains the same, we give each product three names based on the brand naming strategies discussed in the literature paragraph. Every stimulus will have a version with the name of the retailer (own-name strategy), a version with a name which is relevant for the category and endorsed by the name of the retailer (own-name strategy with endorsement) and a version with a fictitious name (other-name strategy). Table 6 shows the chosen brand (other-names for each condition presented to the respondents. For the fictitious names the choice was made to use meaningful brand names that are relevant for the category the product belongs to. Chosen is for suggestive brand names, as termed by Keller, Heckler & Houston (1998), whereby a particular attribute or benefit making up the main selling point of the brand is reinforced semantically (Keller, Heckler & Houston, 1998). Meaningful brand names are limited to the products and categories for which they have a meaning, whereas non-meaningful brand names can be applied to any other product or product category (Kohli, Harich & Leuthesser, 2005). Since application to other categories is not important for this research, we choose the meaningful brand names as stimuli. The two other names, for the own-name strategies, are simply based on the retailer brand name and the segment in which the product is positioned within the category.
Brand names Discount segment Premium segment
Chocolate Direct naming Endorsed naming Other naming
[retailer] Voordeel Delicacao van [retailer]
Delicacao -
Ice cream Direct naming Endorsed naming
Other naming -
[retailer] Premium Gelato van [retailer]
Gelato Paper towels Direct naming Endorsed naming
Other naming
[retailer] Voordeel Wipe van [retailer]
Wipe -
4.4 Research design
The research at hand is using different assortments (scenarios) that are presented to the respondents. These scenarios are based on the variables explained in the previous paragraph. Since there are four different retailers, four product categories, two product segments and three different brand naming strategies, this would lead to a total of 96 assortments that have to be presented. This, however, would mean a very extensive questionnaire for the respondents and make the chance of respondents to quit during filling out the questionnaire very high. Therefore, we chose to reduce the number of scenarios presented to respondents. Since the four product categories are grouped binominal (either being high or low on offering hedonic value), we can reduce the assortments to a total of 48. However, the scenario were the other name branding is presented, does not different between retailers since no retailer branding is used in this scenario. Therefore we can reduce the number of assortments that have to be presented to 36. Since this is still a large number for one single respondent, we split the respondents into three randomly selected groups who are presented the twelve assortments. An example of the scenarios with assortments can be found in exhibit 1 at the end of this paper.
The assortments (scenarios) consist of four products each of which can be found in the corresponding product categories of the fictitious product. Chosen is here for the products of the two main market leaders in that segment added with a discount option without branding. The assortments only give the name, type, volume and price of the product and show no actual picture of the packaging. This is because the packaging can interfere in the evaluation of the product by the respondent and with that the conclusions of the evaluation can be attributed to either the name of the product ór to the packaging and visual cues the respondent has evaluated. The same can be said about the pricing of the products within each assortment. The price ranges across all categories are kept the same in order to rule out the possibility of interference of the price in the evaluation by the respondents.
4.5 Data collection
This paragraph will describe the method used for the collection of the data, in this case a questionnaire, and will describe the methods for acquiring respondents.
4.5.1 Questionnaire
As said before, the measurements in the questionnaire are mainly balanced multi-item scales. In order to reduce the effort for the respondents, no unstructured questions are used but there is given the possibility of a midpoint answer in case the respondent neither agrees nor disagrees.
4.5.2 Respondents
Because of the need to reach many respondents in a short time period, the survey was held online based on a nonprobability sampling method With this method, those who respond may not represent a true cross section of those who receive the questionnaire but there is, however, some of this self-selection in almost all surveys because every respondent chooses whether to be interviewed (Cooper & Schindler, 2006). The online survey has been facilitated by Thesistools, a website dedicated to providing students a way to create and spread their online surveys. An online survey is also used because the response quality is comparable to that of offline surveys, whereas their response time is faster and their costs are lower (Ilieva, Baron & Healy, 2002; Deutskens, de Ruyter & Wetzels, 2006).
In order to reach a larger representativeness of the sample, the respondents are not only sought in the acquaintances of the researcher, since the sample would then have a selection bias, which occurs when some population units are sampled at a different rate than intended by the investigator (Keller, 2005; Lohr, 2010). The respondents are acquired through a number of Dutch internet forums (such as vrouwenpower.nl) and through mailings within companies with a diversely aged workforce.
In order to minimalize the effort of the respondents and to reduce the chance of unfinished questionnaires, there are formed three groups of respondents. These do not vary in their composition. The respondents are randomly assigned to one of the three versions of the questionnaire. The three versions only differ in the assortments that are shown to the respondents and therefore the total number of respondents for the assortment questions is 1/3 of the number of respondents of the other questions in the questionnaire.
4.6 Method of analysis
5. RESULTS
In this chapter, the results and outcomes will be analyzed and the hypotheses will be discussed. First the demographics of the group of respondents are discussed in order to gain insight into the representativeness of this group for a population. Next, internal consistency tests are performed to check whether the variables of retailer brand equity and hedonic value of the segment are measured in the correct manner. After that, the hypotheses and their judgment are presented in the order they were stated in the previous chapters.
5.1 Demographics
The demographics of the respondents group are discussed using descriptive statistics. In total 189 respondents filled out the questionnaire. Within this group however, some respondents answered only a small number of questions. After data cleaning, these respondents were excluded from the analyses performed. A small number of respondents skipped a few questions randomly. These respondents are taken into account for the analyses and the unanswered questions are indicated as missing values. Furthermore, none of the respondents indicate they never buy one of the products chosen in the research. After this data cleaning, the sample consists of a total of 171 respondents. The analyses that will follow in this chapter are mainly based on multiple regression techniques. The sample size we use here is sufficient in order to be able to generalize the results and to reach sufficient power of regression tests (Hair et al., 2010).
5.1.1 Descriptive statistics
The final sample (n=171) consists for 57.9 percent of female and for 42.1 percent of male respondents. So the gender statistic is slightly skewed towards female respondents. The average age of the sample is 38.3 years, which is close to the average age of the Dutch population of 40.3 (CBS Statline, 2011). Figure 4 shows the distribution over the age groups. The level of education of the respondents is also evenly distributed with LBO (6.4%), MAVO (5.8%), MBO (27.5%), HAVO/VWO (8.8%), HBO (24.6%) and the university educated (26.9%). In the sample, the university educated respondents are overrepresented compared to the Dutch population were this is only 9.5% (CBS Statline, 2011). The monthly income of the household of respondents is, as expected, a sensitive question in that a large number of respondents preferred not to answer this question. The respondents that did however are also evenly distributed over the income groups (see figure 5). The average household size is 2.9. Most respondents frequently visit the Albert Heijn (27.5%), C1000 (29.8%), Plus (17.5%) and Dirk (12.9%) or another retailer (12.3%). See exhibit 2 for all graphs of the descriptive statistics.
Gender CBS 2010/2011 ConsumentenTrends 2011* Present sample (n = 171)
Male
Female 49 % 51 % 29 % 71 % 42 % 58 %
Age
18 – 34 years 35 – 54 years 55 years and older
26 % 38 % 36 % 23 % 42 % 35 % 42 % 47 % 11 % Household size 1 person 2 persons 3 or more persons 36 % 33 % 31 % 36 % 33 % 31 % 9 % 35 % 56 %
