1
The effect of cooperative differentiation: what
is the actual payoff of exclusive promotions on
consumer purchase behaviour?
2
The effect of cooperative differentiation: what
is the actual payoff of exclusive promotions on
consumer purchase behaviour?
University of Groningen
Faculty of Economics and Business MSc Marketing Management Master thesis
29 July 2013
Supervisor: Prof. Dr. L.M. Sloot 2nd supervisor: dr. J.E.M. van Nierop
Author:
A.H. (Sanne) de Vries Oosterkade 9d11 9711RS Groningen Telephone: 06-52466324
3
Management summary
Supermarkets find it increasingly difficult to differentiate themselves from the competition. As a result, a new phenomenon that is increasingly observed is that of cooperative differentiation. Cooperative differentiation is an agreement between a manufacturer and a retailer that a specific brand, promotion, format or variety is exclusively supplied to the retailer. In this way the retailer differentiates its assortment from competitors, which could lead to a stronger positioning of the retailer. The goal of this research is to find out whether or not retailers can benefit from cooperative differentiation, specifically exclusive promotions, and whether or not the consumer‟s brand equity and deal proneness affect this. This leads to the following research question:
What is the effect of exclusive product promotions on the consumer’s buying intention, perceived deal value, store differentiation and store switching intention and to what extent do brand equity and deal proneness affect this?
4 Table of contents Management summary p. 3 1. Introduction p. 6 2. Theoretical framework p. 9 2.1 Product differentiation p. 9 2.2 Exclusivity p. 11 2.3 Cooperative differentiation p. 12 2.4 Brand equity p. 15 2.5 Deal proneness p. 16
3. Hypotheses and conceptual model p. 18
3.1 Promotions p. 18
5 5. Results p. 36 5.1 Cronbach‟s alpha p. 40 5.2 Normality test p. 41 5.3 Basic analysis p. 42 5.4 Multicollinearity check p. 43 5.5 Multiple regression p. 45 5.5.1 Independent variables and buying intentions p. 45 5.5.2 Independent variables and store differentiation p. 46 5.5.3 Independent variables and store switching intention p. 48 5.5.4 Independent variables and perceived deal value p. 48 5.5.5 Moderator effects of brand equity p. 50 5.5.6 Moderator effect of deal proneness p. 52 5.6 Other findings p. 54 5.7 Overview of the results p. 55
6. Conclusions and recommendations p. 56
6.1 Discussion p. 56
6.1.1 Effect of exclusive promotions p. 56 6.1.2. Effect of retail related promotions p. 57 6.1.3 Effect of other independent variables p. 58 6.2 Managerial implications p. 59 6.2.1 Implications for retailers p. 59 6.2.2 Implications for manufacturers p. 60 6.3 Limitations and further research p. 60
References p. 62
6
1. Introduction
In the last decade retailers are becoming bigger and bigger. Where in the past every little town had its own unique local grocery store, nowadays they have been replaced by big supermarkets belonging to a couple of big chains. In the Netherlands there are currently through mergers and acquisitions two major supermarket chains who dominate the market (Albert Heijn and Jumbo). As a result supermarkets are becoming highly transparent and find it more difficult to differentiate themselves from competitors. A new phenomenon that as a response increasingly is observed is cooperative differentiation. Cooperative differentiation is an agreement between the manufacturer and the retailer that a specific brand, promotion, format or variety is exclusively supplied to the retailer. In this way the retailer differentiates its assortment from competitors, which could lead to a stronger positioning of the retailer. A common example of cooperative differentiation is Blue Band Bread that is exclusively available at Albert Heijn.
In today‟s competitive environment, companies spend a lot of time and money on customer acquisition and retention (Senthil, Chandrasekar and Selvabaskar, 2012). A condition to acquire and keep customers is that the customer should perceive the retailer and the retailer‟s assortment as being different to that of the competitors. Differentiated products are defined as product offerings that consumers perceive to differ from their competition on any physical or non-physical characteristic, including price (Dickson and Ginter, 1987). According to Dickson and Ginter (1987) the perceptual differences are created by usage experience, word of mouth and communication, where the actual differences are a matter of product characteristics. So by selling more exclusively differentiated products retailers are trying to acquire more customers. But why would a supplier produce and sell their differentiated products exclusively to a specific retailer?
7
of their retail partners (Hultink et al., 1999). Differentiated products in a retailer‟s assortment can create increased customer traffic and points of differentiation but also indicates that the retailer is supplying the latest and best products (Kaufman, 2002; Andritsos and Tang, 2010). Co-operation between a retailer and a supplier could thus benefit both parties. The retailer has an exclusive product in his assortment which leads to more differentiation and the supplier has guaranteed shelf space to sell their products.
Cooperative differentiation is not only observed in the supermarket sector. Examples of industries where this exclusive arrangement also appears are video games and wireless communications. In the video games industry game producers sometimes agree to make their games available exclusively for particular brands of console (Hermanlin and Katz, 2013). This phenomenon is also observed in the wireless communications industry. In this industry, networks often enter into exclusive arrangements with handset manufacturers. For example, Apple‟s agreement with AT&T, under which the iPhone is available in the U.S. exclusively for use on AT&T‟s wireless network (Hermanlin and Katz, 2013). When the retailer has exclusive products in his assortment, the retailer can enjoy various benefits including higher retailer price (because the product is not available elsewhere) and higher store traffic (due to exclusive sales of the product) that may generate a spill-over effect for consumers to buy other regular products also in the store (Cheng, 2008)
8
value, buying intentions, store differentiation and store switching intentions of the consumers. The effect of exclusive promotions on consumer purchase behavior might be influenced by the brand equity and the deal proneness of consumers. It is proposed in this research that these two variables might moderate the relationship between exclusive promotions and the consumers‟ perceived deal value, buying intention, store differentiation and store switching intention. This leads to the following research question.
What is the effect of exclusive product promotions on the consumer’s buying intention, perceived deal value, store differentiation and store switching intention and to what extent do brand equity and deal proneness of the consumer influence this?
The goal of this research is to find out whether or not retailers can benefit from cooperative differentiation, specifically exclusive promotions and whether or not the consumer‟s brand equity and deal proneness affect this. The contribution of this research to academic literature lies in the exploration of the actual payoff of a relatively new and unexplored field of differentiation for retailers. Furthermore, through the two variables added (brand equity and deal proneness) a better understanding is gained of the specific situations in which exclusive promotions have a positive effect on the consumer‟s purchase behavior.
9
2. Theoretical framework
As mentioned in the introduction, this study will describe the effect of cooperative differentiation on consumer purchase behaviour. The concept of cooperative differentiation in a retail setting is to our knowledge not yet described in literature. In order to obtain information about this concept three interviews were held. The first person who was interviewed was a formula manager at Jumbo/C1000. The other interviews were conducted with a brand manager at Unox and a brand manager at Nestlé. Through these interviews an attempt is made to describe the concept of cooperative differentiation as fully as possible. Before describing the concept in practice a literary review of the most important characteristics of the concept of cooperative differentiation, differentiation and exclusivity, will be discussed. After that we will review the concepts of brand equity and deal proneness.
2.1 Product differentiation
When selecting their assortment nowadays retailers have an enormous choice of products. They typically have to choose from hundreds, possibly thousands, of products offered by dozens of different manufacturers (Pan and Honhon, 2012). The average number of stock-keeping units in a supermarket increased heavily from 16.500 to 25.153 over the 1990-2004 period (Progressive Grocer, 2005). However the unlimited growth of the assortment size has recently come to a halt because retailers increasingly understand that bigger is not always better and that assortment composition has a significant effect on store patronage and sales (Koelemeijer, 2002). Selecting products to offer in their assortment is a challenging problem for retailers because customer choice depends not only on the customer‟s preferences but also on the available assortment (Pan and Honhon, 2012). And because a retailer‟s profit depends on the sales and relative profitability of the products in his assortment, the selection of products is very important (Pan and Honhon, 2012).
10
purchases (Blattberg and Neslin, 1990). This is another way to differentiate their product offering and to use one product to promote another. However, in order to be effective and to increase sales, it is important that consumers find the offered promotions attractive. Simonson, Carman and O‟Curry (1994) show that adding a product feature, a gift or a premium that many consumers perceive as having limited or no value, whether or not it is free, can significantly decrease sales of the core product.
Consumers are not all equal in their taste and preferences for products; they face interpersonal differences (Kahn, 1995). Manufacturers and retailers have been trying to satisfy this heterogeneity in consumer tastes through increased product variety (Rajagopolan and Xia, 2012). These product variations help differentiate the retailer‟s products and may help decrease retail price competition (Rajagopolan & Xia, 2012). According to one of the pioneers of marketing thought, Shaw, the strategy of product differentiation can be described as meeting human wants more accurately than the competition (Shaw, 1912). According to Chamberlin (1965) product differentiation could be defined as distinguishing the products of one seller from those of another on any basis that is important to the buyer and leads to a preference. The basis of product differentiation could be real or imagined, arising from product distinction, packaging or distribution differences, or the prestige value of a trademark and trade name (e.g. Coca Cola) (Chamberlin, 1965). Porter (1976) also viewed product differentiation as depending on both physical product characteristics and other elements of the marketing mix. The creation of imaginary differences when no real differences exist through such devices as product names has been labelled „pseudo differentiation‟ (Lancaster, 1979). A more recent definition of product differentiation according to the Business & Management Dictionary of Bloomsbury Business Library (2007) is that product differentiation refers to a marketing method that promotes and emphasizes a product‟s difference from other products of a similar type.
11
challenging as the importance of location continues to decline and the share of internet-based retailing increases (Alba et al., 1997;Simonson, 1999). Nonetheless, Simonson (1999) suggests that product assortment can still play a key role not only in satisfying wants, but also in influencing buyer wants and preferences. A potential way for retailers to further distinguish themselves from competitors with their assortment is through exclusivity (Clark, 2004).
2.2 Exclusivity
Exclusivity can be defined as the situation in which vertical contracts between a buyer and a seller contain exclusivity provisions that restrict the ability of either of them to transact with third parties (Matouschek and Ramezzana, 2007). According to Chang (1992), exclusive dealing could be considered in a more general sense and he differentiates between „downward‟ and „upward‟ exclusive dealing. Downward exclusive dealing refers to the situation in which a manufacturer prohibits a partnering retailer (or distributor) from carrying the products of the manufacturers competitors (Bernheim and Whinston, 1998). Upward exclusive dealing occurs when the upstream manufacturer is restricted from distributing his product through any of the downstream retailers‟ direct competitors (Andritsos and Tang, 2010). In this paper exclusivity means that a specific product is exclusively available at a retailer initiated by the retailer and/or the manufacturer.
12
manufacturer essentially trades market coverage with investment in the partnership made with the downstream retailer (Frazier, 1999).
In the current market there is intense competition among manufacturers and there is increased concentration at the retail level (Kadiyali et al., 2000). The power seems to be shifting from the manufacturers to the retailers (Kadiyali et al., 2000). If the manufacturer is the dominant channel partner in the context of product exclusivity, we assume that it is the uniqueness of the manufacturer’s product and its ability to generate store traffic and act as a differentiator from the competition which allows the manufacturer to act as the leader (Andritsos and Tang, 2010). Through an exclusive sales channel the manufacturer can better control the customer experience and interaction with the new product because the products get more attention from the retailers (Yu, Chavan and Dowden, 2008). However, a potential disadvantage could be that the manufacturer obtains a smaller fraction from the market than would be possible under a non-exclusive product launch (Andritsos and Tang, 2010). In the case of non-exclusive products the manufacturers expose their products to consumers through the extended reach of different channels and provide convenience for consumers to make purchases through alternatives channels (Andritsos and Tang, 2010). In the case of non-exclusive products the retailer faces a dilemma if they want to add the product to their assortment. If they carry the new product their profit margins will be low due to retail competition, if they do not carry the new product they may potentially lose store traffic. By establishing exclusive sales arrangements retailers are trying to reduce competition (Andritsos and Tang, 2010). The retailer can enjoy various benefits such as asking a higher retail price because the product is not available elsewhere and higher store traffic due to the exclusive sales of the new product that may generate a spill-over effect for consumers (Cheng, 2008). However, although exclusivity and cooperative differentiation have some similarities they are not the same.
2.3 Cooperative differentiation
13
C1000; the exclusive availability of XXL knaks (sausages) at Super Unie and the exclusive promotion of free Unox dishes when you buy three Unox products at Albert Heijn.
Cooperative differentiation can be initiated by both parties. The retailer can approach the manufacturer if they would like to do something with a particular product, category or event. The manufacturer can also provide suggestions for a product or event though the retailer may only want to include it in its assortment if it is exclusively supplied to them. This also depends on the power of the retailer. Large supermarkets can exert more pressure on manufacturers to enforce exclusivity than smaller supermarkets. In the case of small supermarkets the initiative for cooperative differentiation therefore usually comes from the manufacturer. An example of an idea initiated by the manufacturer is the free DVD giveaway of the „Vrienden van Amstel’ concert when you buy Amstel beer at Jumbo. In this case the manufacturer was a big sponsor of the event and he initiated the exclusive promotion to the retailer. In the end, it does not matter which party initiates cooperative differentiation, in all instances the idea will be elaborated upon together.
A major benefit of cooperative differentiation from a retail perspective is that the retailer can differentiate its store with an assortment distinct from its competitors. Another advantage is that the exclusively available products can profit from the brand strength of the manufacturer. Furthermore, because the product or promotions are solely available through a specific retailer the probability of price erosion is smaller. A potential risk for retailers is that the exclusive products do not satisfy real consumer needs, but are only developed in order to differentiate. Another disadvantage is that the manufacturer may not invest much in the product because of the limited availability.
14
through one retail chain. Manufacturers always prefer the situation in which their products are available in all retail stores. An exclusive contract with one retailer may also damage the relationship with another retailer who is thereby put to a disadvantage.
Cooperative differentiation can occur in different forms. It can appear in the form of an exclusive product, an exclusive promotion, an exclusive format or an exclusive product variety. An example of an exclusive product is the luxury tea from Douwe Egberts only available at Albert Heijn. An example of an exclusive promotion is the free Christmas books from Douwe Egberts available at C1000. And the exclusive availability of the ketchup variant of Suze & Thomas knaks (sausages) from Unox at Albert Heijn is a good example of an exclusive product variety. The decision to choose a particular form depends on the situation. Exclusive promotions are the most practical and easy form to use. Something is usually given away when you buy a certain amount of products. This is also interesting for manufacturers because the sales of their products will increase through the promotion. With the other forms of cooperative differentiation the biggest advantage lies with the retailer.
The duration of cooperative differentiation depends on the situation. The duration on product level is dependent on the agreements made by both parties. In some cases the products have about a year‟s exclusivity, in others only a few months‟. It is often also dependent on the product performance. Promotions always have a limited duration agreement. They are unique and appear only once, so the duration is mostly a couple of weeks. Some promotions, however, are repeated every year during the same period (e.g. the Douwe Egberts Christmas books).
15
2.4 Brand equity
Building brand equity is considered to be an important part of brand building (Keller, 1998). Branding is a powerful means of differentiation (Pappu, Quester and Cooksey, 2005) and as said before differentiation is one of the key competitive positioning strategies suggested by Porter (1990). If consumers perceive a particular brand favourably, branding might develop a competitive advantage for firms (Aaker, 1989). The concept of brand equity has been defined in a number of different ways and no common viewpoint has emerged about how to conceptualize brand equity (Keller, 2013). However, the broad meaning of the term brand equity could be described as the value endowed by the brand to the product (Farquhar, 1989). That means that brand equity explains why different outcomes result from the marketing of a branded product or service than if it were not branded (Keller, 2013). The definitions of brand equity can be classified into two categories, definitions based on the financial perspective that include the value of a brand to the firm (Brasco, 1988; Mahajan et al., 1990; Shocker and Weitz, 1988; Simon and Sullivan, 1993) and definitions based on the consumer perspective, which define brand equity as the value of brand to the consumer (Aaker, 1991; Kamakura and Russell, 1993; Keller, 1993; Kim and Lehmann, 1990; Rangaswamy et al., 1993). In this paper we will make use of customer-based brand equity. Customer-based brand equity occurs when the consumer has a high level of awareness and familiarity with the brand and holds some strong, favourable and unique brand associations in memory (Keller, 2013). Brand awareness consists of brand recognition and brand recall performance (Keller, 2013). Customer-based retail brand equity involves a “shortcut” in the minds of consumers that recalls from memory the most salient positive elements of satisfaction with past shopping experiences and goods purchased, which in turn influences future patronage and minimizes the potential influence of competitor efforts (Ailawadi and Keller, 2004).
16
enhance brand equity by providing additional points of differentiation for the manufacturer‟s brand. The advertising of new products can, however, either increase or decrease the parent brand equity (Sinapuelas and Sisadiya, 2010).
The level of brand equity of a product may influence the effect of exclusive promotions on the consumer‟s purchase behaviour. If consumers have strong brand equity for a particular brand of product in a promotion, we expect their purchase behaviour to be more favourable. The same applies conversely: if consumers have a strong negative brand equity towards the product in a promotion, we expect that their purchase behaviour is negatively affected. However, there is another variable that could possibly moderate the relationship between exclusive promotions and consumer purchase behaviour, namely the consumer‟s sensitivity for promotions, also called deal proneness.
2.5 Deal proneness
17
18
3. Hypotheses and conceptual model
In this chapter the research of this study will be further clarified. First the independent variable promotions will be described in more detail. After that the hypotheses will be formed using literature and at the end a visual representation of the research will be displayed in the form of a conceptual model.
3.1 Promotions
Promotions are the most common form of cooperative differentiation. According to an early definition of Cooke (1985) promotions can be defined as that it communicates information about the products and persuades potential customers to buy them. A more recent definition stated that promotions refer to activities, usually short-term, designed to attract attention to a particular product and to increase its sales, using advertising and publicity (Bloomsburry business library, 2007). A lot of research has been done about consumer response to non-exclusive promotional activities on packaged goods in a supermarket setting (Grewal et al., 2011). This research has shown that short-term sales are positively affected by offering promotions (Blattberg and Neslin, 1990). There are a lot of different types of product promotions. Products can be promoted by e.g. bundled products and premiums, price reductions or gifts (Blattberg and Neslin, 1990). A common type of product promotion within cooperative differentiation is free gift offers. “Free gift with purchase” offers appear to be enticing the consumer to buy products through the offer of a free gift with their purchase (Raghubir, 2004). Through the free gift the value of the transaction increases, and so the offer may increase sales of the promoted product (Raghubir, 2004).
19
3.2 Buying intentions
Consumer purchase behavior can often be better understood based on the assumption that people select the option that is easiest to justify to oneself and to others, as opposed to the traditional assumption of utility maximization (Shafir et al., 1993). Retailing firms build the most profitable strategies through differentiation and competitive advantage, offering customers something new and which they value that other retail outlets do not (Rajagopal, 2008). According to Rajagopol (2008), supermarkets can be potential outlets where customers can experience innovative (exclusive) promotions on a variety of products which drive purchase decisions.
Customers purchase behavior can usually be predicted by their intentions (Hsu et al., 2011). Purchase intentions can be defined as the likelihood of buying the brand or of switching to another brand (Keller, 2013) or as the decision to act, or psychological states which represent the individual‟s perception to engage in a particular behavior (Fishbein and Ajzen, 1975). The purchase intentions can be influenced by the knowledge, attitude and personality of the consumer as well as by the characteristics of the product, the price and the relevance of the product promotion (Barber et al., 2012). Product promotions offer value to customers and by offering exclusive retail related product promotions the retailer increases this value even more. We therefore expect that the exclusive retail related promotion will lead, in comparison with a widely available non retail related promotion, to an increase in customer buying intention. In other words, we expect that offering a customer an exclusive product promotion, compared to offering a non-exclusive promotion, will have a positive effect on their purchase intention. We also expect that the degree of retail relatedness of the promotion will be positively related to the buying intention. This leads to the following hypotheses:
H1a: The exclusive availability of a promotion at a retailer is, compared to a non-exclusive promotion , positively related to the customer’s buying intention.
20
3.3 Store differentiation
Repeated buying behavior of customers is largely determined by the value acquired on the products (Rajagopol, 2008). Value has become an important concept among consumers, retailers, and manufactures (Albrecht, 1995) because „„creating outstanding customer value is the only secure route to achieving sustainable financial and market success‟‟ (Sweeney et al., 1999; Coopers and Lybrand, 1998). Consumers prefer to shop where they believe they will receive the most satisfaction and value from the store and the merchandise acquired (May, 1989). Exclusive product promotions are intended to offer consumers more value than non-exclusive product promotions even as retail related promotions. It is crucial for the retailing firms to apprehend the mechanism involved at the consumer level regarding these product promotions (Rajagopol, 2008). Variables such as variety seeking, perceived financial benefit, brand loyalty and store loyalty towards product promotions have specific influences on the buying behavior and volume of retail sales (Laroche et al., 2003). So the exclusive product promotions offered by supermarkets are not only aimed at boosting the sales (purchase intention) but also at augmenting the store value (Rajagopol, 2008). Through exclusive product promotions consumers can obtain extra value on specific products from a certain retailer that is not available on the same products from other retailers. So in this way the retailer can further distinguish itself from competitors by offering extra value on specific products that cannot be gained anywhere else. Sharpe and Dawes (2001) defined store differentiation as the situation in which a store outperforms rivals stores in the provision of features such that it faces reduced sensitivity for other features. So store differentiation exists when a store‟s offering is preferred, on some buying occasions (or by some customers all of the time), over rival store‟s offerings (Sharpe and Dawes, 2001). We thus expect that exclusive product promotions have a positive influence on the perceived store distinctiveness because exclusive promotions offer more value and customers prefer more value over less value. We also expect that the degree to which the promotion fits with the retailer may also have a positive effect on the customer‟s perceived store differentiation. This lead to the following hypotheses:
21
H2b: The degree of retail relatedness of the promotion is positively related to the customer’s perceived store differentiation.
3.4 Store switching intention
The abundance of choice in today‟s market place provides a great opportunity for consumers to switch between stores (Shaklu and Babun, 2013). Consumers nearly always have the opportunity to change their shopping behaviour because there are other shopping opportunities available to them (Findlay and Sparks, 2008). So it is possible for them to change their behaviour by switching between stores or retailers (Findlay and Sparks, 2008). The consumer‟s store switching intention can be defined as the intention to change the main shop for a main shopping trip (Findlay and Sparks, 2008). Store switching intention is an important consideration for retailers, because if consumers visit another store to buy products, the retailer will suffer negative consequences in terms of store sales and store profits (Borle et al. 2005). Also, when consumers switch to other stores following a promotion, they probably buy other items in other categories as well in that store, generating spill over effects (Sloot and Verhoef, 2008; Cheng, 2008). Shopping value has according to Shaklu and Bubin (2013) a significant influence on reducing the store switching intention. The retailer offers value to the consumers and by offering exclusive, retail related promotions we expect that the value offered to the consumers is increased. So through offering exclusive retail related promotions we expect that the ability to generate change in behaviour and then to retain the „switched‟ customer will increase (Findlay and Sparks, 2008). This leads to the following hypotheses:
H3a: The exclusive availability of a promotion at a retailer is, compared to a non-exclusive promotion, positively related to the customer’s store switching intention.
H3b: The degree of retail relatedness of the promotion is positively related to the customer’s store switching intention.
3.5 Perceived deal value
22
(1997) investigated whether restrictions on promotion availability and promotion type had an influence on the consumer‟s perceived deal value. It was stated that coupon promotions were found to have significantly higher perceptions of deal value compared to reduced price promotions available to consumers because of the more perceived limited availability (Chatterjee and McGinnes, 2010). With coupon promotions you can join the promotion if you are in the possession of a coupon. So consumers perceive this as more valuable than promotions which are offered universally. Exclusive promotions also are limited so we expect that this gives consumers more value than promotions that are widely available and they perceive a higher deal value, especially when the promotions fit well with the retailer. This leads to the following hypotheses:
H4a: The exclusive availability of a promotion at a retailer is, compared to a non-exclusive promotion, positively related to the customer’s perceived deal value.
H4b: The degree of retail relatedness of the promotion is positively related to the customer’s perceived deal value.
3.6 Brand equity – moderator effect
23
2013). Brand equity will thus be influenced by the strength of the brand associations, the favorability of the brand associations and the uniqueness of the brand associations. These factors could be different for every customer but also for every brand or product. We therefore expect that brand equity has a positively moderated influence on the effect of (exclusive retail related) promotions on the perceived deal value, buying intentions, store differentiation and store switching intention. The following hypotheses are proposed as a result:
H5a: The expected positive effect between exclusive product promotions, compared to non-exclusive promotions and the consumer’s buying intention is positively affected by the brand equity of the products. H5b: The expected positive effect between the retail relatedness of promotions and the consumer’s buying intention is positively affected by the brand equity of the products.
H5c: The expected positive effect between exclusive product promotions, compared to non-exclusive promotions and the consumer’s perceived store differentiation is positively affected by the brand equity of the products.
H5d: The expected positive effect between the retail relatedness of promotions and the consumer’s perceived store differentiation is positively affected by the brand equity of the products.
H5e: The expected positive effect between exclusive product promotions, compared to non-exclusive promotions and the consumer’s store switching intention is positively affected by the brand equity of the products.
H5f: The expected positive effect between the retail relatedness of promotions and the consumer’s store switching intention is positively affected by the brand equity of the products.
H5g: The expected positive effect between exclusive product promotions, compared to non-exclusive promotions and the consumer’s perceived deal value is affected by the brand equity of the products.
H5h: The expected positive effect between the retail relatedness of promotions and the consumer’s perceived deal value is affected by the brand equity of the products.
3.7 Deal proneness – moderator effect
24
consumer‟s internal reference price. This in turn can generate increased value within the consumer‟s mind (Shukla and Babin, 2013). In presence of the various forms of deals, the consumer‟s overall deal proneness can increase the consumer‟s excitement and participation in the purchase process and in turn affect the overall perception of shopping value (Carpenter and Moore, 2008). So the deal proneness of the consumer influences the sensitivity to promotions and the likelihood of the consumer‟s purchase behaviour as a result of these promotions. Therefore we expect that deal proneness has a positively moderated influence on the effect of (exclusive, retail related) promotions on the perceived deal value, buying intentions, store differentiation and store switching intention. This leads to the following hypotheses:
H6a: The expected positive effect between exclusive product promotions, compared to non-exclusive promotions, and the consumer’s buying intention is positively affected by the deal proneness of the customer.
H6b: The expected positive effect between the retail relatedness of promotions and the consumer’s buying intentions is positively affected by the deal proneness of the customer.
H6c: The expected positive effect between exclusive product promotions, compared to non-exclusive promotions, and the consumer’s perceived store differentiation is positively affected by the deal proneness of the customer.
H6d: The expected positive effect between the retail relatedness of promotions and the consumer’s perceived store differentiation is positively affected by the deal proneness of the customer.
H6e: The expected positive effect between exclusive product promotions, compared to non-exclusive promotions and the consumer’s store switching intention is positively affected by the deal proneness of the customer.
H6f: The expected positive effect between the retail relatedness of promotions and the consumer’s store switching intention is positively affected by the deal proneness of the customer.
25
3.8 Conceptual model
From the previously described hypotheses the following conceptual model is derived.
26
4. Methodology
In the following chapter we will discuss the research methodology. First the experimental design of the study will be explained. After that we will explain the data collection. We will then discuss the survey design of the research and finally we will describe our plan of analysis.
4.1 Experimental design
In order to gain insight into how consumer purchase behaviour is enhanced through exclusive promotions, a quantitative research in the form of an experiment is conducted. More specifically, a 2 (level of exclusivity) x 2 (level of retail relatedness) x 2 (type of gift) factorial design with a questionnaire is used. The goal of this experiment is to detect differences in consumer purchase behaviour between the situation with exclusive promotions and the situations without exclusive promotions. According to Xie and Boush (2011) an experimental study is an appropriate method because experimental studies explore the potential causal relationship between claims and consumer psychological or behavioural responses.
In order to analyse the effect of exclusive promotions on consumer purchase behaviour, a mixed model design is adopted. The respondents receive all retail related and non-retail related promotions that are exclusively available at the retailer or not. The promotions also differ in the type of gifts, a free service or a free product, that are given away. In figure 2 an overview of the experimental conditions are given. The retail relatedness of the promotions refers to the fit of the promotion with the retailer and its customers. To test whether or not the promotions are retail related an expert study with five experts is conducted. In our survey we used two promotions that fit with the retailer (a product and a service) and two promotions that do not fit with the retailer(also a product and a service). The promotions that fit with the retailer give away a free glass or tickets to a museum. The promotions that are not retail related gave away a classical CD or tickets to the theatre. The promotions that we use in our survey are all validated by experts.
27
Testing the hypotheses is done by first asking the respondents about their socio-demographics and shopping behaviour (control variables). The respondents are then asked questions about the different brands in the promotions (Pickwick, Optimel, Ariel and Tuc) to measure their brand equity regarding these brands. Four different types of products were used to ensure that the product type itself does not have too much influence. After that the scenario, in which the respondent is asked to imagine themselves, will be described. Subsequently, the respondent is shown an advertisement with a promotion whereby buying several products of a specific brand leads to them receiving a free gift. The advertisements contain different promotions (retail related and non-retail related) and different gifts (service or a product) and differ according to where the promotion is obtainable (exclusive at C1000 or non-exclusive at the C1000, Albert Heijn, Jumbo and Plus). After being shown the advertisements, the respondent is asked to answer several questions related to their buying intention, perceived store differentiation, store switching intention and their perceived deal value. The last part of the questionnaire consists of questions regarding the deal proneness of the consumers.
4.2 Data collection
The survey questions are asked by means of an online survey. The reason for choosing this method is because it enables a large amount of information to be gathered in a relatively short time span (De Leeuw, 1996). An online survey for a descriptive research also enables us to identify and describe the variability in different phenomena (Saunders et al., 2000). According to Malhotra (2004) another benefit of an online survey is that the data is already digital and easily processed. For respondents it also has benefits because the survey can be completed at any time, which is convenient and is completely voluntary.
28
respondents in a store other than C1000, which is why we in this case recruited at Albert Heijn. For the purposes of this study, it was however intended that there be a C1000 near these supermarkets so that the distance to the supermarket plays no role in the decision to switch from store. It was also important to recruit only at Albert Heijns located in the area of Groningen because some of the promotions relate to activities in Groningen. By only recruiting people at a supermarket in Groningen we attempted to avoid the risk of the promotion being less attractive due to the distance to the activities.
The visitors of the supermarkets were approached at the entrance of the store as to whether the investigator could ask them something. After shortly introducing and explaining the basic procedure, the customers were asked if they would be willing to participate in the research. Following a positive response they received a card with some brief information about the research and the link to the survey. The respondents were also offered a Merci chocolate as a thank you. The chocolate was in line with the rule of reciprocation, which stated that upon receiving a gift or a favour from a stranger, people usually feel obliged to repay in kind what they were provided with (Cialdini, 2009). The cards were handed out over a period of twee weeks.
29
Promotion Retail relatedness promotion Type of free gift Manipulation
1: Optimel Retail related Service Exclusive / non exclusive 2: Ariel Non retail related Service Exclusive / non exclusive 3: Pickwick Retail related Product Exclusive/ non exclusive 4:Tuc Non retail related Product Exclusive / non exclusive Table 1: Overview promotions
As mentioned before there were eight different scenarios. According to Hair et al (2010) a minimum of 20 respondents per condition is recommended, to ensure statistical power. However the researcher aimed at collecting at least 200 completely finished questionnaires to ensure statistical power (so 100 per condition).
4.3 Survey design
The survey consists of several questions about the variables mentioned in the conceptual model. The questions consist of the following parts: socio-demographics, brand equity, perceived deal value, buying intentions, store differentiation, store switching intention and deal proneness. Mainly all the questions are measured on a Likert scale and are based on academic literature. Scale questions are often used to collect data on attitudes and beliefs of the respondents (Saunders et al., 2000). This research mainly uses a 5-point Likert scale, which creates enough possibilities for the respondent to express their opinion and it is measurable (Malhotra, 2004). The odd number of the scales gives respondents the opportunity to give a neutral answer.
30
4.3.1 Dependent variables
The first questions the respondents have to answer after they have seen the promotions are related to their perceived deal value. Prior research in marketing has firmly established that consumers differ in their perceptions of promotional offers (Suri, Swaminathan, and Monroe, 2004). Multiple scales have been used since to measure the perceived deal value. Chatterjee and McGinnis (2010) measured the concept by using a three-item scale where respondents have to assign values to these items on a 7-point Likert scale. Another scale which measured perceived deal values based on a 9-item scale comes from Grewal et al. (1998). The scale from Dodds et al. (1991) also measured the perceived deal value and reported a high internal consistency with Cronbach‟s alpha of 0,93. Based on these scales from literature, the following scale is used in this research to measure the perceived deal value. These items should be rated on a 5-point Likert scale with a score of 1 totally disagreeing and a score of 5 totally agreeing.
1. I think this is an attractive offer
2. I think this promotion is a better deal than most promotions where a discount is given 3. This offer gives customers great value for their money
The questions thereafter measure the buying intention of the respondents. Buying intention is a variable that is frequently researched in literature. There are consequently a lot of different scales in the available literature used to measure this. For example, the scale of Baker and Churchill (1997) that measures the behavioural intention towards a product. Another scale to measure buying intention is the scale of Dodds, Monroe and Grewal (1991). This scale has a high internal consistency with Cronbach‟s alpha of 0,96 and the scale used in this research to measure buying intention is mainly based on this scale. However, instead of using a 5-item scale we use a 3 item scale in this research. The items are rated on a 5-point Likert scale where 1 indicates total disagreement and 5 total agreement.
1. The likelihood of purchasing this product is high
31
The concept of store differentiation will be measured next on a 3-item scale. Customers perceive the store as being different as it offers unique value to the customers (Rajagopol, 2008) According to Sweeney et al. (1999) and Coopers and Lybrand (1998) creating outstanding customer value is the only secure route to achieve outstanding financial and market success. Good promotions and a clearly distinctive shop are variables that add value to a store. Based on this literature a 3-item measurement scale has been developed. The items are measured on a 5-point Likert scale where 1 represents total disagreement and 5 total agreement.
1. C1000 is a distinctive supermarket
2. There are always surprising offers at C1000
3. C1000 is clearly different than most other supermarkets
Finally the last questions will measure on a 3-item scale the store switching intention of the customers. Sloot and Verhoef (2008) and Shaklu and Babin (2013) both developed a scale to measure the construct of store switching intention. Sloot and Verhoef (2008) defined the concept as „the degree to which a consumer is likely to switch to another store in the case of a brand delisting‟. In this research we measure the degree to which a consumer is likely to switch to C1000 in the case of an exclusive retail related promotion. Based on these investigations we developed the following measurement scale. This scale is a 5-point Likert scale, where a score of 1 indicates total disagreement and 5 indicates total agreement.
1. It is likely that I would do my weekly shopping at C1000 instead of my regular supermarket if the nearest C1000 had this offer.
2. I do not find the promotions interesting enough to change my regular supermarket.
3. This promotion is so interesting that I would not mind spending extra effort on my grocery shopping.
4.3.2 Moderators
32
towards the brands Optimel, Ariel, Pickwick and Tuc. After some sociodemographic questions, questions related to the consumer‟s brand equity are asked. Based on the literature from Verhoef, Langerak and Donkers (2007) a 4-item scale was used and measured on a 5-point Likert scale.
1. Brand X is a strong brand 2. Brand X is a well-known brand 3. Brand X is an attractive brand 4. Brand X is a unique brand
Questions about the deal proneness of the customers are asked at the end of the questionnaire. Based on the literature of Bailey et al. (2008) and Rani & Velaydudhan (2008) a 5-item scale is used. The questions are measured on a 5-point Likert scale with 1 totally disagreeing and 5 totally agreeing.
1. A free gift with the purchase of a product can be reason enough for me to actually buy the product
2. I find it hard to resist promotions
3. If a product is on sale, it can be a reason for me to buy it
4. Although I have some favorite brands, I usually buy the brand that is on sale 5. If a product is on sale, I buy more than I normally would
4.3.3 Control variables
33
Control questions Measurement scale
1. What is your gender? Male or Female Nominal variable 2. What is your age? Open question Interval data 3. What is the number of people in your
household?(including yourself)
Open question Interval data 4. What is the composition of your household? Single, partner no
children, family with children, other
Nominal
5. Are you responsible for the shopping? Yes or No Nominal variable 6. What is your highest level of education? Primary education,
Mavo/VMBO, Havo, Vwo, MBO, HBO, WO, Don‟t tell, other
Nominal variable
7. How many hours per week do you spend on paid work?
Open question Interval data 8. Which work situation is applicable to you? Self employed,
Full-time worker, part-time worker,
unemployed, Student, Housewife/man, Retired, Unfit for work
Nominal variable
9.What is your monthly net income? <€500, €501-€1000,
€1001-€1500, €1501-€2000, €2001-€2500, €2501-€3000, €3001-€3500, >€3001-€3500, prefer not to tell Nominal variable
10. Which store (s) are in your area, i.e. are accessible for you?
C1000, Jumbo, Albert Heijn, Plus,
Discounter (Aldi / Lidl), Coop/Boni/Dirk van de broek, Other
34
11. In which store(s) do you do your shopping? C1000, Jumbo, Albert Heijn, Plus,
Discounter (Aldi / Lidl), Coop/Boni/Dirk van de broek, Other
Nominal variable
12. In which supermarket do you do most of your shopping?
C1000, Jumbo, Albert Heijn, Plus,
Discounter (Aldi / Lidl), Coop/Boni/Dirk van de broek, Other
Nominal variable
13. Which means of transport do you usually use to go to the supermarket?
Car, bicycle, scooter, walking
Nominal variable Table 2: overview controll variables
4.3.4 Survey results
The questionnaire described above will lead to the variables shown in table 3. Table 3 also gives an overview of the way the variables are measured. The complete survey with all the questions can be found in appendix 1.
Construct Way of measuring Literature
Perceived deal value Three 5-item scales (1 = totally
disagree, 5 = totally agree). measuring the customer‟s perceived deal value of the promotion.
Grewal et al,1998: Dodds et al, 1991; Chatterjee and McGinnis ,2010 Buying intention Three 5-item scales (1 = totally
disagree, 5 = totally agree). measuring the extent of buying intention a respondent feels for the product in the promotion.
Dodds, Monroe and Grewal (1991).
Store differentiation
Three 5-item scales (1 = totally
disagree, 5 = totally agree). measuring the customer‟s perceived store differentiation Rajagopol, 2008; Sweeney et al., 1999 ;Coopers and Lybrand, 1998; Researcher Store switching intention Three 5-item scales (1 = totally
disagree, 5 = totally agree). measuring the customer‟s store switching intention
35
Exclusive promotions Dummy variable, equal to 1 if
promotion is exclusive Researcher Retail related promotions Dummy variable, equal to 1 if
promotion is retail related
Researcher Brand equity Four 5-item scales (1 = totally
disagree, 5 = totally agree). measuring the customer‟s brand equity
Verhoef, Langerak and Donkers (2007) Deal proneness Five 5-item scales (1 = totally disagree,
5 = totally agree). measuring the customer‟s deal proneness
Bailey et al,2008; Rani and
Velaydudhan, 2008 Table 3: Overview
4.4 Plan of analysis
In order to answer the research question and the formulated hypotheses, several analyses are performed on the obtained data. To gain knowledge about our sample, descriptive statistics are used. The dependent variables and the moderators are all measured on a multi-item scale. It is therefore important to determine if there is internal consistency between the items of the construct (Malhotra, 2007). Through Cronbach‟s Alpha we will check the reliability of the scales and the underlying questions.
After that we will check the nature of the distribution with a normality test. Subsequently we will perform some basic analysis to get some insight into the data. Before conducting a multiple regression analysis it is important to check for multicollinearity between the independent variables. Multicollinearity is a state of high intercorrelation among independent variables (Malhotra, 2007). If the correlation coefficient is too big (>4 moderatore multicolliniearity, >10 strong multicolliniearity) it is difficult to distinguish which of the independent variables influences the dependent variables (buying intention, store differentiation, store switching intention and perceived deal value).
36
5. Results
5 days after distributing the last card the survey was taken offline. In total, 236 persons had begun to fill in the questionnaire but only 201 of these 236 questionnaires could be used. There were several reasons for not including certain respondents. Four respondents indicated that they were not responsible for the grocery shopping, so these respondents did not fit in the target group and were excluded. Twelve respondents already indicated that their regular supermarket was C1000, although they were recruited at the Albert Heijn. These respondents were expected to react differently to the promotions and the related questions because they already had a preference for C1000, so they too were excluded from the dataset. The other 19 respondents did not complete the whole questionnaire and were therefore also excluded. All in all, a total of 35 respondents were removed from the dataset. This leads to a usable response rate of 85,1%.
Socio-demographic characteristics
Demographic variable Sloot et al. Regular Dutch shopper CBS 2012/2013 Our sample
(2005) (CBL, 2012) Sample size (n) 749 2682 201 Sex Female (%) 77 71 51 51 Male (%) 23 29 49 49 Age 34 or below (%) 32 23 26 63,5 35 to 54 (%) 40 42 37 24.5 55 or older (%) 28 35 37 12 Household size 1 person (%) 59 36 37 19.7 2 persons (%) 33 33 34.1 3 or more persons (%) 41 31 30 46.2
Education (based on Dutch system) N/A N/A
Lower (%) 27 14,9 Middle (%) 42 15,9 Higher (%) 30 68,2 Doesn't say (%) 2 1
Income N/A N/A N/A
€500 or below (%) 12,5 €501 - €1000 (%) 13,9 €1001 - €1500 (%) 4,9 €1501 - €2000 (%) 12 €2000 - €2501 (%) 10,5 €2501 - €3000 (%) 9 €3001 - €3500 (%) 9,6 €3501 or more (%) 15,4 Doesn't say (%) 13
37
In order to determine whether the sample is representative of the regular Dutch shopper the sample was compared to other studies (Sloot, Verhoef & Franses, 2005; ConsumentenTrends 2012). In table 4 the sociodemographics of the sample are presented in comparison with other studies conducted in 2005 and 2012/2013.
The distribution in age and education of the sample is not the same compared to the other three studies. This can partly be explained by the location of the sampling because one of the supermarkets was located in an area where many students live. Although more elderly people were present at the other location, a lot of them indicated that they were not in possession of a computer and/or internet or had difficulties in using a computer, which resulted in them not participating in the study. Another possible explanation could be that the respondents in the other studies were interviewed in the supermarket itself, whereas our respondents were asked if they would participate in the study at home. A comment frequently heard from people was that they were too busy, especially people aged 35 to 54. The permissiveness of the cards may also have lead to a lower participation of this group. The average age and education of our sample is thus lower and higher compared to the regular Dutch shopper. The distribution of the sex of the participants is also not comparable to the regular shopper in the first two studies, though it is representative of the distribution of gender in the Netherlands (Consumententrends, 2012).
Because the gap in age and education between the regular Dutch shopper and our sample is quite large, the sample is weighted on age. We weighted the sample on age and not on education because we expect that younger people are generally more highly educated and by doing this we expect the education to automatically be corrected. The variable age is recoded in three groups where 0 refers to the group aged 34 years old or below, 1 refers to the group between 35 years and 54 years of age and 2 refers to the group aged 55 or older. Compared to the regular Dutch shopper there are too many respondents in the first group, so this group is weighted 0.36 (group 0: 23/63.5). In the other two groups there are too few respondents so these groups are weighted 1.71 (group 1: 42/24.5) and 2.92 (group 2: 35/12). Taking these weights into account, the sociodemographic characteristics will change (see table 5).
38
supermarket was located in a wealthy neighbourhood where generally more highly educated people live. The demographics of the weighted sample are, however, all more in line with the regular Dutch shopper than the original sample, so from now on the weighted data will be used in the analysis.
Sociodemographic characteristics
Demographic variable Our sample Sample size (n) 201 Sex Female (%) 50 Male (%) 50 Age 34 or below (%) 23 35 to 54 (%) 42 55 or older (%) 35 Household size 1 person (%) 21.1 2 persons (%) 33.3 3 or more persons (%) 45.6 Education (based on Dutch system)
Lower (%) 19.8
Middle (%) 19.9
Higher (%) 59.3
Doesn't say (%) 1
Table 5. Socio-demographic characteristics weighted on age
39
Graph 1: work situation respondents.
Figure 3 shows where the other respondents do most of their shopping. Of the respondents 17% indicated that they do most of their shopping at a discounter and 8% of the respondents shop mostly at Jumbo. Most of the respondents indicated that they do their shopping by car (53.4%). The other respondents indicated that they do their groceries by bike (29,3%), scooter (0,05%) or by walking (16.8%).
40
5.1 Cronbach’s Alpha
To measure the perceived deal value, buying intention, store differentiation, store switching intention, brand equity and deal proneness multiple items are used. In the regression model we used the mean of the answers of the related variables as the overall answer to the construct. However, before being able to combine these variables it is vital to be certain that the questions measure the same construct. Because we want to know how reliable the constructs are we used Cronbach‟s Alpha to measure the internal consistency. The reliability of a scale refers to the extent to which it produces consistent results if repeated measurements are made (Malhotra, 2007). With the internal consistency we measure the strength to proceed with these variables instead of the original variables. Cronbach‟s Alpha should be higher than 0,6 for internal consistency. With a Cronbach‟s Alpha of 0,6 or higher we can be confident that the items measure the same construct. In table 6 an overview of the Cronbach‟s Alpha is given for all constructs with multiple items. However, before the items could be checked by Cronbach‟s Alpha, the reversed item of store switching intention should be reformulated.
Cronbach’s Alpha Number of items
Buying intention 0.879 3 Perceived deal value 0.861 3 Store differentiation 0.970 3 Store switching intention 0.858 2 Brand equity 0.872 4 Deal proneness 0.567 4 Table 6: Overview of Cronbach‟s Alpha
41
the same construct so we take the mean of the answers of the questions into the regression model as the overall answer to the variable.
The Cronbach‟s Alpha of deal proneness, however, is slightly below the critical level of 0.6. The cronbach alpha of 0.567 is based on a four item construct instead of five because item 5 decreased the value. This item „If a product is on sale, I buy more than I normally would‟ was therefore deleted. Although the value is still below the critical threshold of 0,6, we nevertheless take, based on expert opinion, the mean of the four items as the overall answer to the variable. We should take this in mind and be aware of it when we look critically at the outcome of this variable in the analysis.
5.2 Normality test
The importance of normal distribution for fitting continuous data is well known. However, in many practical situations data distribution departs from normality (Paul and Zhang, 2010). According to Pallant (2007), particularly in social sciences research dependent variables are often not nicely normally distributed. To test if our sample is normally distributed we make use of the Skewness and Kurtosis tests. Skewness is the measure of the symmetry of distribution and Kurtosis refers to the peakedness or flatness of the distribution. In table 7 the results of the tests for the dependent variables can be found.
The scores of the Skewness test for buying intention and store switching intention are between 1 and 1/2 which indicates that the distribution is moderately skewed. The values of the scores of perceived deal value and store differentiation are between (-)1/2 and +1/2 which indicates that the distribution is approximately symmetric. The kurtosis scores are <1 for all variables which indicates that all the distributions are a bit more flatter than a normal distribution.
Skewness Std. error Skewness
42
The scores of the tests indicated that the distributions are not perfectly normal distributions. However, because the Skewness scores are all values lower than |1| and also the kurtosis scores are not extremely flat, we don‟t have to indicate any violation of the normality assumption. Due to the large sample size of 201 respondents we can also assume that even a violation of the assumption does not cause major problems (Gravetter and Wallnau, 2000).
5.3 Basic analysis
43 Buying intentions Service Product NR RR Total NR RR Total Non-exclusive (0) 2,14 2,23 2,18 1,84 2,76 2,30 Exclusive (1) 2,09 2,15 2,12 2,02 2,74 2,38 Total 2,12 2,19 2,15 1,93 2,75 2,34 P-value exclusivity 0,567 0,437
P-value retail related 0,499 0,000*** P-value EX*RR 0,885 0,372 Store differentiation Non-exclusive (0) 2,50 2,47 2,48 2,44 2,54 2,49 Exclusive (1) 2,38 2,65 2,52 2,43 2,57 2,50 Total 2,44 2,56 2,50 2,43 2,56 2,49 P-value exclusivity 0,740 0,891
P-value retail related 0,153 0,149 P-value EX*RR 0,083 0,770
Store switching intention
Non-exclusive (0) 2,06 1,93 2,00 1,75 2,08 1,91 Exclusive (1) 2,05 2,00 2,02 1,85 2,23 2,04 Total 2,06 1,96 2,01 1,80 2,15 1,98 P-value exclusivity 0,754 0,139
P-value retail related 0,278 0,000*** P-value EX*RR 0,646 0,771
Perceived deal value
Non-exclusive (0) 2,62 2,68 2,65 2,02 3,04 2,53 Exclusive (1) 2,66 2,73 2,70 2,37 2,99 2,68 Total 2,64 2,71 2,67 2,20 3,01 2,61 P-value exclusivity 0,687 0,140
P-value retail related 0,512 0,000*** P-value EX*RR 0,954 0,045** Table 8: Two-way ANOVA
5.4 Multicollinearity check
44
Our main independent variables (the manipulated conditions) are constructed by dummy variables where the variable promotion exclusivity is coded as „‟0=non-exclusive‟‟ and „‟1=exclusive‟‟ and the variable retail related promotion is coded as „‟ 0=non retail related‟‟ and „‟1= retail related‟‟. Because the two moderator variables brand equity and deal proneness are measured on a Likert scale (interpreted as continuous variables) they are mean centered to reduce possible multicollinearity which can occur because the moderators are included. The interaction variables are created by multiplying the mean centred brand equity (BE) and deal proneness with the variables advertising exclusivity and retail related promotions as these form our independent variables.
Independent variables VIF
Main independent variables
Advertising exclusivity (EX) 1,024 (0=non-exclusive 1=exclusive)
Retail related advertising (RR) 1,063 (0=non-retail related 1= retail related)
Moderators
Interaction EX*BE 1,555
Interaction EX*DP 1,424
Interaction RR*BE 1,501
Interaction RR*DP 1,416
Consumer related variables
Gender 1,261 Age 1,420 Income 1,615 Education level 1,063 Household size 1,828 Household composition 1,775
Weekly working hours 3,449
Profession 2,843
Buys product 1,706
Bought products 1,635
Transport for groceries 1,297 Most visited supermarket 1,082
Table 9: Multicollinearity between independent variables
45
5.5 Multiple regression
In this section we will perform the regression analyses. By performing a multiple regression analysis we are able to examine how much of the variance in consumer buying intention, perceived store differentiation, store switching intention and perceived deal value can be explained by our proposed independent and moderating variables. We will also test our previously formulated hypotheses. By using the results of the regression analysis the hypotheses can either be accepted or rejected.
5.5.1 Independent variables and buying intentions
We will first test the relationship between the independent variables and buying intention. The results of the multiple regression analysis can be found in table 10. The overall model is significant with an F-value of 10.448. The model has an R square of 0,221, which indicates that the regression model explains 22.1% of the original variation (Hair et al., 2010). We find that retail related promotions have a significant effect on the consumer‟s buying intention. Nonetheless, the exclusive availability of a promotion in a supermarket does not have a significantly positive effect. Therefore we can‟t support hypothesis 1a but we can accept hypothesis 1b.
H1a: The exclusive availability of a promotion at a retailer is, compared to a non-exclusive promotion, positively related to the customer’s buying intention.
H1b: The degree of retail relatedness of the promotion is positively related to the customer’s buying intention.
