Creating value through discount : how store image influences loyalty by altering the perception of price.

Creating value through discount.

How store image influences loyalty by altering the perception of price.

S.M.E. de Jong

6095070

Master Thesis

24 March 2017 – final draft

MSc. in Business Administration

Marketing Track

Amsterdam Business School, UvA

Supervisor: A.C.J. Meulemans

Statement of originality

This document is written by Student Sandra M. E. de Jong

who declares to take full responsibility for the contents of this

document.

I declare that the text and the work presented in this document

is original and that no sources other than those mentioned in

the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely

for the supervision of completion of the work, not for the

contents.

Abstract

Nowadays discounts are part of the daily reality of the retail business. It is no longer a choice of the retailer to give discounts, it is a demand from the customer: buying a product at the full price is an exception for them. Over time, discount has changed the way people shop, shifting the focus of the customer to the value and the price of a product. For stores that practically act as intermediates who sell same branded products this focus makes adding value to a product and creating loyalty to the store brand a difficult task. Still stores are focusing on these two strategies. Is it still worth the effort to give discount?

When the same products are sold in different stores (e.g. National brands products) the store has little room to add value. With an equal evaluation of the value of a product how is it possible discount is evaluated differently in different stores? Why does discount feel less or more attractive in different stores? How do customers decide for which percentage of discount to purchase the product? There is a gap between the actual price and the perceived price.

The research focuses on the two main drugstores of the Netherlands: Etos and Kruidvat. This industry is the most suitable to do the research in due to the similarity of the stores (e.g. same products, pricing strategy) but also on the differences (e.g. brand image). The question that this research will try to answer is the following: how does store image influence the store brand loyalty through discount?

An experiment was conducted with a pre-test to determine the level of the customer equity of the two drugstores and a second collection of data has happened for a two (Level of customer equity: Etos and Kruidvat) × two (Condition: discount and no discount) x two (Product price perception) between-subject design. A total of respectively 60 and 112 valid questionnaires were completed with a minimum of 28 participants per cell. Ultimately the effect of these variables on the store brand loyalty

The results of the research indicate support for only hypothesis 1. The other hypotheses are not supported. This non-significant results give space for an interesting discussion and conclusion part of the thesis.

Content

2.2 Value made tangible ... 10

2.3 Changing the perspective on value ... 12

2.4 Customer equity ... 14 2.5 Loyalty ... 16 3. Conceptual framework ... 19 4. Methodology ... 21 4.1 Data ... 22 4.2 Hypothesis testing ... 28 Results ... 28 5. Discussion ... 39 6. Conclusion ... 43 7. Bibliography ... 45 8. Appendix ... 50

1. Introduction

In the past discount was used by the retailers as a tool to increase sales ad hoc and in a substantial way (Ailawadi, Beauchamp, Donthu, Gauri, & Shankar, 2009) by giving the customer an opportunity to try new products (Kotler 1988; Webster 1971), making them buy more products at the same time and by shortening the time in between different purchases moments (Laroche, Pons, Zgolli & Kim, 2001; Shoemaker,1979). Due to various factors, including changed legislation, discounts are no longer contained to a certain period, making the promotions more common and making them available more often. (Yi & Yoo, 2011). The earlier found effects of discount are becoming less effective. Nowadays the customer underestimates the real reference price of the products because of the frequency of discounts (Liefeld & Heslop, 1985) making the usage of discounts necessary to meet the customers with the by them expected price of the product.

According to a recent news article (Wanders, Volkskrant, 2016) one possible solution to this inflation of product value is to actually stop discounting, but this is only possible when stores own their own brand products, since competitors offer the exact same national brand products making it hard to stop with discounts. If a store cannot play the same discount game as the competitor the store will not survive in the current environment. However, customers are willing to pay more for quality, expert advice and service, which creates opportunities for competitive advantages.

The paper offers another managerial application for customer equity by combining it with discount. Because of the frequency of discounts in the retail industry there is a need to know more about discounts and how these are valued by the customer. In particular the question if the thoughts of the customer about the store are of any influences on the appreciation of the discount needs to be answered. The value of the discount could differ if presented in different stores.

While there has been done a fair amount of research on the product level this research will expend from the product level to the store level and give useful insights whether discount has influence on the customer store loyalty or not. The retailer could get useful insights on discounts and the usage of discount from this research.

Due to the nature of the problem that the retail business is facing and based on the characteristics of the stores there has been chosen to analyze the Dutch drugstores Etos and Kruidvat in the research. First the thoughts of the customer on the store will be measured leading to the concept of store image. Store image is expressed in terms of customer equity. Customer equity consists of three drivers namely value equity, brand equity and relationship equity. It is used to determine the thoughts of the customer about the company, and ultimately customer equity can be used to determine how much the customer is worth to the company. There is much resemblance between store image and customer equity, but because of the usability of the latter concept and the extensive knowledge available about it, customer equity has priority in this research. Next, the effect of discounts on product price perception will be measured, followed by the effect of product price perception on the store brand loyalty. This research will ultimately analyze how the level of customer equity will moderate the effect of discounts on the product price perception and how this ultimately influences the store loyalty of the customer. This leads to the following research question:

To what extend does the store image, measured in terms of Customer Equity, influence the product price perception and store loyalty?

This research question will be answered using the following sub questions: - Does discount positively influence product price perception?

- Can product price perception be predicted by the level of customer equity? - Can the store loyalty be predicted by the level of product price perception?

The main articles in this research are the article of Rust, Lemon, and Zeithaml of 2000, where the concept customer equity is explored, and the article of Diallo, Coutelle-Brillet and Rivière of 2015, where the research investigates the relationship among price perception of different brand types, shopping value dimensions and store loyalty.

The practical relevance of this research is offering an innovative way for the calculation of the concept customer equity. Furthermore, the research will give insight on the importance of price for the store loyalty of the customer. The theoretical contribution will try to fill the gap between price and perceived price. Also the leap between product valuation and store valuation will be attempted.

The research will need an analysis of the promotion, the retailer, the product and the customer and its loyalty. In the next section, a theoretical framework is proposed and the research hypotheses are developed, based on a literature review of the central concepts. After presenting the research methodology used to test the proposed model, the results of the measurement are exposed. Lastly, the theoretical and managerial implications of the findings are highlighted and some limitations and orientations for further studies are underlined.

2. Literature review

The following concepts and the relationships between them will be explained: discounts, value, product price perception, store experience, store image, customer equity and store loyalty. The concepts will be placed within the retail industry context. From these the conceptual framework will follow where the relationships will be explained between the variables with the hypotheses that will be researched in this study. The structure of the literature review is based on the hypotheses of this study: (H1) discount has a positive effect on product price perception. (H2) Store image measured by the concept of customer equity strengthens the effect of discount on product price perception. (H3) The level of product price perception is a predictor for the level of customer loyalty. Because of the choice of making store image measurable with customer equity the leap from product level to store brand is easily made. Firstly the concept discount will be explained and how it is evaluated in the mind of the customer.

2.1 Position of discount

To understand how the customer evaluates discount, it is necessary to understand the use and effects of discounts more generally. Discounts come in different ways and these differences can be seen as different strategies. Examples of discount strategies are everyday low pricing, seasonal price of, special price discount, free gifts and buy one, get one free (Yi & Yoo, 2011). Discount strategies differ mainly in time, length and depth. These differences result in different effects on the outcomes for the customer and for the retailer and researchers suggest that some strategies work better than others. This research will focus on one group where the previous mentioned examples belong to: the Regular Price Discounts, being for the customer certain discount that brings a downshift in the internal reference price (Alavi, Bornemann & Wieseke, 2015). Due to the downshift in the internal reference price the gathered information about discount until now might not be current anymore. Discounts nowadays are more common, repeated more often and used for a longer period than before (Yi & Yoo, 2011). Next to the discount effects that are beneficial to the retailers, the academic world is more aware of the negative effects, short or long term, that the retailers endure.

For the retailers these strategies have the positive side of discount in common: the reason of the use of discount and the goal that the retailer is trying to achieve with them. At the end the different types of discounts are used by the retailers in order to improve sales and/or profit

(by adding a strong purchase incentive (DelVecchio, Krishnan & Smith, 2007)), market share and “sales surrogates” such as brand awareness, purchase intention and brand image. (Berger & Bechwati, 2001). The negative effects mainly have to do with the way the customer reacts to discount. These reactions can be observed in the way customers shop and how customers think of the product, the product brand and the store brand. Promotions can reduce the post-promotion choice for the brand. Price post-promotions lowers the consumers' perceptions of brand quality (e.g. (perceived) value), by training consumers to wait for promotions, or by lowering consumers' price expectations for the brand (e.g. lowering the customers’ internal reference price). Also the depth and/or the framing of the promotion has a negative effect on the customer post-promotion brand preference (DelVecchio, Krishnan & Smith, 2007).

It is questionable whether the benefits of having discounts as a retailers are worth the costs. It is important to find out how discount is being connected to the product. There is a need for the retailer to know how the customer evaluates discount and how discount is taken into consideration by the customer while shopping.

How the customer evaluates discount starts with how the customer evaluates a buy. What happens in the mind of the customer before the purchase and how is discount added in this equation? Whether a customer buys a product or not is “decided” after an evaluation of the benefits and the costs of the purchase (Keller, 1993). Using different theories about value and looking at three different types of value models (e.g. value dynamics model, value buildup model and value exchange model) where the area of what value is becomes broader, Khalifa (2004) came up with a value exchange model uniting different components that usually were discussed separately.

The model gives a simplistic view on value and is basically a benefits-costs model that the customer evaluates when buying a product. The difference between the benefits and the sacrifices needs to be zero or above zero will it lead to a purchase decision. The total benefits consist of utility value and psychic value. The total customer sacrifices, which we call here the total customer cost, consists of financial and non-financial customer costs, i.e. the total customer ownership cost (pre-use, at-use, and post-use costs) (Khalifa, 2004). In the following figure the model is showed. The configuration shows that customer value goes beyond products and services and as such, the emphasis on products or services per se should give way to a broader and more comprehensive basis of satisfying and delighting customers

by nesting those products and services in the total meaningful customer experience (Khalifa, 2004).

Figure 1 – Khalifa, 2004

Looking at the value exchange model the values that matter most in this context are the utility value and the psychic value. When discount comes into play the utility value of a product stays the same but the psychic value changes. The psychic value consists of meeting the psychic needs of the customer. What falls within the psychic needs is not easy to define due to the intangible nature of the benefit for the customer and the influences of different factors (e.g. how the customer is being treated in the store) (Khalifa, 2004).

The effects of discount should first be seen on a product level. For the customer the discount is an extra incentive to buy the product in question (Schultz & Robinson, 1982). It changes the perceived value of the product, being one of the factor that meets the psychological needs, often making the product worth high enough for the customer to buy it (Neslin et. al., 1984). Discount is included into the benefits and costs equation, influencing the assessment of the customer on the benefits and the costs of buying the product and ultimately the value of the product. The discount becomes a benefit that is provided by the product (Strang, 1983).

2.2 Value made tangible

As mentioned before value can be determined in different ways, one more tangible than the other. Sirohi, McLaughlin & Wittink (1998) define value as "what you get for what you pay." This is a simplified version of the following definition of value: perceived value is the consumer’s overall assessment of the utility of a product based on perceptions of what is

received and what is given. The definition given by Sirohi et al. (1998) is based on four definitions of value given by customers itself:

- 1. value is low price

- 2. value is whatever I want in a product - 3. value is the quality I get for the price I pay - 4. value is what I get for what I give

The tradeoff still stands of benefits and costs. Variation of the thoughts of the customers about value occurs when defining what is received (i.e. volume, quality, convenience) and what is given (i.e. money, time, effort). Using the word “pay” Sirohi et al. go towards a definition of value as value for money where the given is money.

Discount does not necessary change the value that the customer gives to the product: it changes how the customer thinks about the price. It is a mental process that ends with an evaluation of worthiness and price. Customers already have an idea of what a product that meets their benefits must cost and this is called the price reference. Consumers form expectations of prices and use them in formulating their response to retail pricing. The consumer reaction depends not only on retail price, but also on the comparison they make with the reference price. In other words, consumers use the price they expect to pay for a brand on a given purchase occasion as reference in forming price judgments (Kalwani & Yim, 1992). When discounts come into play the equation between the benefits and the costs changes. Price promotions significantly affect consumers' price perception (Folkes & Wheat, 1995). There is a gap between the actual and the perceived price, making it important to know how the customer encodes the prices of the products. The gap between actual and perceived price could be explained by the thoughts of the customer about the store. This is why product price perception will be introduced next.

At the end the price of a product represents the monetary value that the customer must incur in order to make a purchase (Keller, 1993). Price is thus the sum of the value that the customer gives to the product in order to benefit from the product (Kotler & Armstrong, 2008). Definition used in this research for the concept product price perception is how the price of the product is perceived by the customer of the store. By looking at this definition the subjectivity of the concept has been made clear. Conditioned by the customers’ experiences in the past, the expectations of the products price and fairness of the buy are crucial for the

perception of the actual price (Calvo Porral & Levy-Mangin, 2015). How customers perceive a price is as important as the price itself. This is assessed by the image of the store itself and the image of the competitors (Heda, Mewborn & Caine, 2017). Product price perception is positively influence by discount (Calvo Porral & Levy-Mangin, 2015). Discount can be assessed as a separate part of the price by using the store image. Within different stores discount can on its own be evaluated differently affecting the relationship between discount and product price perception. How discount is perceived, and ultimately the product price perception, depends on the store image.

It has been observed that discount ultimately influences the store image (Heda, Mewborn & Caine, 2017). In this research the opposite will be considered: the influence of store image on the discounts and indirect on the perceived product price. Next the concept store image will be explained.

2.3 Changing the perspective on value

Keller (1993) looks into a more complete idea of the concept and defines the benefits of the promotion as an extra value attached to the experience of the promotion. Customers respond to discount, according to Keller and in line with the previous mentioned research about value, as a characteristic that increases value but not from the product itself but from other clues, namely clues that alter experience. Promotion is seen as part of the experience a customer has in a store (Khalifa, 2004). By connecting the two concepts it needs to be included into the overall value of the product, meaning in the overall perception of the price of the product. When discount is being evaluated with the particular store in mind, this has influence on the relationship between discount and product price perception. It means that the influence of discount on the product price perception is different when the discount is evaluated with the store brand in mind. The thoughts of the customer need to be measurable and this process starts with the assessment of the concept store experience. This will ultimately lead to the customer equity concept.

Store experience has gotten a bigger role within retail, meaning that a specific collection of clues in the store have an effect on the customer that most certain is observed in the purchase behavior of the customer. Cues are formed from different touch points that the customer encounters during the customer journey. Focusing on the cues that are encountered in the store clues can be arranged by the retailer into the following groups: brand-owned,

partner-owned, customer-partner-owned, and social/external/independent. This division is more apt for the retailer it helps with knowing on what they can have influence on. Brand-owned and partner-owned touch points are the two groups of cues where the retailer can have influence on. Sometimes these groups overlap. Cues that fall within these definitions are advertising, websites, loyalty programs, attributes of product, packaging, service price convenience and sales force (Lemon & Verhoef, 2016). For the customer another segregation is more relevant. The clues just mentioned can be divided into three groups: sense, feel and think (Sung-Joon Yoon, 2013). Kotler (1973) stressed the role of the store atmosphere regarding the buying decision. Store atmospherics include visual, auditory, olfactory, and tactile clues. The feel group includes the emotions that are evoked for the customer by the store. The intangible atmosphere and the contact with the employees influence this. When the customer is focusing on the thinking part of the experience it will focus on price, quality and features of the product that the store sells. (Sung-Joon Yoon, 2013) Discount falls in the last group and is thus part of the experience that the customer has in the store. Based on the need of the customer their type of experience will change (Sung-Joon Yoon, 2013). Even though this shifting between the sense, feel and think takes place it does not mean that they substitute each other. It is then prudent to include as much store clues as possible. When including other factors next to the discount in the concept of customers’ experience they could be summed up with the term store image. Store experience could just be one incident, one shopping trip with these specific circumstances ant that specific time, while store image could be seen as the memory and recollection of several shopping experiences that overtime will be connected to the store brand name, and thus made into a generalization of the retailer (Imran, Ghani & Kashif, 2013).

In retail the experience of the customer while shopping has become more and more relevant for business management. It is a factor that needs to be taken into consideration when evaluating how the customer valuates their purchase. All the experiences of the customer with a store brand come together in the image that the customer has of the store. Store image can thus be defined as the collective thought of the customer about the store and the brand. Store image can be defined as a composition of attributes that consumers attach to a particular retail entity (Beneke & Zimmerman, 2014). Store Image is specific to a store or retail chain and its ownership (Delgado-Ballester, Hernandez-Espallardo & Rodriguez-Orejuela, 2014).

Martineau (1958) identified layout of the store, communication and personnel as important image dimension of a retailer. Swoboda, Weindel and Halsig (2016) try to come up with attributes for store image that are valid in different retail sectors. They include the following retail attributions in their research: assortment, price, layout, communication and service. The sectors they have considered are grocery, fashion, electronics and DIY. In the context of drugstores the same attributes will be used in this research. Their final hypothesis is that the importance of the attributes are different within every retail sector. Therefore, the attributes where the customer pays attention to, grades and stores in their mind differ depending on the context. This needs to be taken into consideration.

The combination of these factors do not completely fit within the variable store image. The thoughts of the customer about the store, which are collected within the concept of store image, could explain a big part of the value proposition. The appreciation (or not) of all the factors that are connected to the store and the brand flows into the valuation of the product. Store image needs to be quantifiable and this will happen by using the concept of customer equity.

2.4 Customer equity

Customer equity is defined as the value of the customer to the brand (Rust, Lemon & Zeithaml, 2004). Customers are an asset of a retail business and with the help of the concept of customer equity this is made tangible. The customer equity can also be used the other way around, in order for the retailer to know how the customer valuates the store. It consists of three on their own standing concepts: value equity, brand equity and relationship equity. The three equities together form the customer equity (Lee, Ko, Tikkanen, Phan, Aiello, Donvito & Raithel, 2014). Customer equity is used within business as a predictor of the customer loyalty. Improving one of all the equities will lead to a better customer equity, which will induce a better customer loyalty. Looking at the theory of customer equity, the higher the equity the more a customer is willing to pay. In a retail setting this means the higher a store is evaluated the more a customer is willing to pay for a product that is available at the store.

Value equity is used in order to evaluate products for their credibility, quality and marketability. The equity reflects objective consumer opinions about products or services and it says a lot about the competence of a company's product or service (Rust et al., 2004). Consumers recognize value by comparing the rewards they get with the efforts they spend by

acquiring products or services. Value can be determined by looking at quality, price and convenience (Lee et al., 2014).

Brand equity refers to consumers' evaluations of a personal preference for brand images, which then grant value to products or services (Aaker, 1991). It is an intangible monetary equity that involves the recognition of the brand by the customer. (Rust et al., 2004). The equity will be determined by using two key antecedents of store brand equity: store image and price (Collins-Dodd & Lindley, 2003; Semeijn, Van Riel, & Ambrosini, 2004; De Wulf, Odekerken-Schroder, Van Ossel, 2005; Calvo et al., 2015).

Relationship equity evaluates consumer perceptions of value in their relationships with the company regarding corporate provision of products or services (Rust et al., 2004). Consumers evaluate the quality of face-to-face encounters to assess the benefits of company offerings in relationships and special treatments. For the retail store it is important to get emotional attachment (Dwyer, 1997).

The industry determines which of the aspects within the levers need to be taken into account. The retail industry is characterized by the great amount of choice and the low switching costs (Allaway, Huddleston, Whipple & Ellinger, 2011). Due to this value equity seems to have no weight within the construct. Within retail the focus very often lies on the brand equity. Customer-based brand equity can be seen as a further elaboration of the brand equity concept and focuses on the thoughts of the customer about the brand. The other driver that seems to be making the difference when it comes down to make the customer loyal is the relationship equity. Relationship building strategies have become the way to impress the customer and win them over to be loyal to your brand. By measuring the different components of the customer equity conclusions can be made about the influence of the different equities on the store brand loyalty. By using the concept of customer equity this research tries to connect the product with the store. This concept also gives the opportunity to connect product discount with the customer loyalty of the store brand. Next the concept of customer loyalty will be explained.

2.5 Loyalty

Customer satisfaction leads to customer loyalty (Flint, Blocker & Boutin, 2011). When a store brand gets a high valuation on the three component of customer equity the assumption is that the customer is satisfied and thus loyal (Rust et al., 2001). Customer loyalty is positively related to profitability and market share of the retailer (Flint, Blocker & Boutin, 2011). Customer loyalty is needed to keep up store traffic (Heda, Mewborn & Caine, 2017) and to be able to make business decisions with less risk (Szczepańska & Gawron, 2011). Also retaining the customers that the store already has is five to ten times more convenient than acquiring new ones (Slater & Narver, 2000). These are excellent reasons to create more loyal customers. Companies can choose among tactics to reach this goal in four categories: offering lower prices, shouting out those prices, giving great deals, and tailoring the experience (Heda, Mewborn & Caine, 2017). Discount is part of the experience a customer has in a store, and by connecting the discount to the store image of the customer (e.g. by terms of customer equity) we can speak of an attempt of the store to create loyalty through all of the four categories mentioned.

Different researches have made an effort to collect all the definitions of customer loyalty to find the common denominator, a factor which the company can use when looking for a way to increase loyalty. Looking at the research of Szczepańska and Gawron (2011) most

definitions of loyalty (see table 1) are based on the assumption that there are two main sets of loyalty drivers (i.e. behavioral and psychological drivers). By looking at these drivers two new loyalty drivers are identified (i.e.: action-based and follow-up drivers).

Table 1 - Definitions of customer loyalty by Szczepańska & Gawron, 2011.

Author Definition

The Global Loyalty Agency all the feelings or experiences that would incline a customer to consider the repurchase of a particular product, service or brand or re-visit particular company, shop or website

Newnan J.W., Werbel R.A. repeat purchase of a particular brand, without considering purchase of any other available brand Jacoby J., Chestnut R.W., Day G.S. customer’s predisposition towards the brand as a

function of psychological processes

Storbacka K., Lehtinen J.R. intention to act and willingness to interact with others Bloemer J., de Ruyter K. customers’ non-incidental and intentional actions

displayed over a long period of time towards a particular service/product supplier which operates among numerous and similar service/product suppliers Olivier R.L. deeply-term engagement and product/service/brand re-purchase intention displayed toward a particular product, service or brand

Reichheld F.F. willingness to invest in further product/service/brand relationship development

Dick A.S., Basu K. function of attitude manifested in behavior

Jacoby J., Kyner B.D the biased (i.e. non-random), behavioral response (i.e. purchase), expressed over time, by some decision making unit, with respect to one or more alternative brands out of a set of such brands, and is a function of psychological (i.e. decision making, evaluation) processes

Zawadzka A.M. the result of rational-functional motivation teamed up with emotional-symbolic motivation

In the study of Diallo et. al. (2015) the behavioral and psychological drivers are also highlighted as common dividers of the definition. The definition of customer loyalty used for this research is in line with the research of Diallo et al. (2015). The focus lies there on the tangible effect of loyalty, the measurable effects. Customer loyalty is then composed of three

different ways: intent to continue shopping, intent to increase purchases and intent to recommend the store (Topcu & Uzundumlu, 2009). The focus will lie on the store retention and the word of mouth where a longitudinal study will not be needed. The two concepts will be the drivers of the variable customer store loyalty. The study of Diallo et al. (2015) confirms that shopping value positively affects store retention loyalty, such that quality value, price value and emotional value have positive influences on store retention loyalty. Shopping value also positively affects store WOM, such that price value, emotional value and social value have positive influences on store WOM.

Pricing is thus a factor of importance for the loyalty towards a store. According to Noyam and Simsek (2013) loyalty depends on comparative price perception, discount perceptions and customer satisfaction. In the study of Calvo Porral and Levy-Mangin (2015) the perception of an adequate and affordable price has been proven to have a positive effect on the loyalty of the customer towards the store brand. It has been established that there is a positive relationship between perceived value and customer loyalty in the retail context (Diallo et al., 2015). The link between product price perception and customer loyalty is hereby explained.

Discount is used by stores to create loyalty but it is an asset that is easy to copy. What happens when the same product is on sale at different stores? How will store image influence the reaction on the customer on the discount? Can discounts on products affect the loyalty of the customer to the store? If this is the case it would be unprofitable for the retailer to continue to give discounts while this action is damaging the store brand. If the main goals of discount are not achieved, the usage of discount needs to be revised. Goal of this research is to know what the relationship between discount, customer equity and the product price perception is, and how interaction between these factors effects the customer loyalty to the store brand. With these concepts the following research question will be answered:

• To what extend does the store image, measured in terms of Customer Equity, influence the product price perception and store loyalty?

The next part of the thesis outlines how this question will be answered and made visual. The conceptual framework of the theory will follow next and the relationships between the variables will be depicted in this part of the thesis.

3. Conceptual framework

The following illustration is a schematic view of the literature that has been reviewed previously. The relationships that will be tested within this research are illustrated in the following conceptual framework.

Figure 3 – Conceptual framework

The corresponding hypotheses are the following ones:

H1: Discount has a positive effect on the product price perception.

H2: The positive relationship between discount (IV) and product Price Perception (DV) is positively moderated by the factor customer equity (Moderator) (e.g. the relationship between discount and product price perception is weaker in a low customer equity situation rather than a high customer equity situation).

H3: Product price perception has a positive effect on store brand loyalty.

The variable customer equity will serve in this research as a measurement of store image. The relationship between customer equity and product price perception is a positive one. Meaning that if the customer equity is high the product price perception will be more positive compared to the product price perception number that is coupled with lower values of the customer equity. When the customer equity is low there will follow a lower product price perception. The independent factor customer equity has a positive moderating effect on this relationship. So the effect of discount on product price perception will be stronger. With a high level of customer equity the discount will have a stronger positive effect on product price perception then when there is a low level of customer equity.

Customer

Equity

Discount

Product Price

Perception

Store Brand

Loyalty

When the customer equity is high the customer has positive thoughts about the store. In this condition the price of the product is positively perceived. Meaning that the price is perceived as a fair price for the product. When the discount condition is present the price of the product will be perceived more positively. When the customer equity is low the customer has negative thoughts about the store. In this condition the price of the product is negatively perceived. Meaning that the price that the store wants for the product almost always is perceived as too high for the product. When the discount condition is present the price of the product will be perceived more positively.

Product price perception has a positive effect on the store brand loyalty. Inducing the right product price perception in the mind of the customer is thus vital for the creation of loyalty, even more so when the customer is focusing more and more onto getting the right price, getting the most out of their buy. The influence of moderator on the relationship between customer equity and product price perception will follow through and have an effect on the loyalty created by the product price perception variable. When the product price perception is negative the loyalty is negative. When the product price perception is positive the loyalty will also be positive. It is crucial to know if discount has an effect on this relationship.

When putting the conceptual framework within context the expectations are that the customer equity of the Etos will be higher than the customer equity of the Kruidvat. A high customer equity will have a positive effect on the product price perception. So the product price perception will be more positive in the Etos than in the Kruidvat. Discount will affect the product price perception as follow: the price of the product will be perceived better in the Etos and worse in the Kruidvat. The loyalty at the Etos will be higher than the loyalty at the Kruidvat based on logical reasoning of the positive and negative product price perception and the relationship of this variable with the loyalty variable.

4. Methodology

In the previous section the concepts leading to the conceptual framework have been explained. In the following section there will follow an explanation on how the variables of the conceptual framework will be measured and how the experiment will be carried out. An explanation of the nature of the experiment will follow now.

Nature of research

The subject will be approached with the interpretivist philosophy. A model will be put on to the reality making this research a deductive one. Because the field in which the research will take place is the Dutch retail we will consider it a case study by mean of an experimental survey. It is an explanatory study because there are hypothesizes that will be tested in the research (Saunders, Lewis & Thornhill, 2012).

Research design

To collect the data a survey of experimental nature will be sent out using Facebook as distribution channel. For homogeneity reasons two similar stores within the same industry have been chosen for this case study: Etos and Kruidvat. The survey consists of six questionnaires and only one of the six needs to be answered by the participant, this in order to prevent bias. Two questionnaires were designed to determine the customer equity of the Etos and of the Kruidvat. It can be seen as a pre-test. Four other questionnaires were designed to collect data on the Product Price Perception and Store Brand Loyalty with a manipulation of the condition discount or no discount (low percentage) and a manipulation of store (Etos or Kruidvat) within groups. Here the choice has been made to use a Dutch brand shampoo as the product that the participant will buy. Since it is a product that is available at both stores, with the highest probability to be known by the participants and that can be bought by both genders. For the discount condition a discount percentage of 20 is used. A higher percentage of discount would most certainly have led to an always buy and an always satisfied about the buy situation in the discount condition. Two of these questionnaires have been sent out on a second timeframe in order to measure the discount manipulation on the variable loyalty. The scale that is use in the research is the 5-point Likert scale. Thoughts and intentions are best measured with this scale: it forces the respondent to choose between answers with more confidence. In the survey the words store and drugstore (winkel/drogist) have not been used. Instead the name of the drugstore Kruidvat and/or Etos has been used. The complete

questionnaire is made available in the appendix. Also, see the appendix for an overview of the constructs and scale items used in the questionnaires.

Sample

Due to the social media chosen to distribute the survey, it can be stated that the respondents are part of a convenience sample. Two surveys have been sent out. The first survey had 4 randomly distributed questionnaires resulting into 4 groups. The second survey had 2 randomly distributed questionnaires resulting into 2 groups. The outcomes of the two surveys have been collected into two datasets: the first dataset focuses on the customer equity, while the other dataset focuses on the product price perception and loyalty get. The second survey complemented the second dataset. Both datasets put together consist of 172 respondents. Specific information about the sample can be found in the appendix. No abnormalities have been found. Because of the focus on the Dutch retail the survey has been made and distributed in Dutch: only Dutch speaking people have been able to fill out the survey. The survey is included in the appendix.

4.1 Data

The data is separated into two datasets. The first one is used to determine the customer equity of both stores. When this variable is determined it will be put into the second datasets where further computation with the variables product price perception and loyalty, will be made. The results are thus merged into one dataset to test the final hypothesis. Next, the different constructs need to be calculated by merging the items. The aim is to have a Cronbach’s alpha bigger than 0.7 per variable. The variables that need to be measured are the following: the different equities for the stores, product price perception, loyalty and product attitude.

The conceptual framework variables Product price perception

The product price perception variable is highly reliable with 5 items (α= 0.852, N=112). All items are kept for further tests. Deleting the items wouldn’t increase the Cronbach’s Alpha enough (>0.1). The inter-item correlation matrix demonstrates no correlations (All numbers are <0.8). Furthermore, a factor analysis has been run to further explore the variable product price perception (KMO=0.852, sign.=0.001). The total variance explained by the items of the variable happens with two components. They have respectively the Eigenvalues of 3.251 and 0.622. Even though the second component has an EV<1 it has been extracted because of the

percentage of variance explained (>10%). The cumulative percentage of variance explained is 77.5%.

Store brand loyalty

The loyalty variable is highly reliable with 10 items (α= 0.924, N=112). All items are kept for further tests. Deleting the items wouldn’t increase the Cronbach’s Alpha enough (>0.1). The inter-item correlation matrix demonstrates one correlation (item 9 with item 10 corr.= 0.928). This means that the following tests will have to take into consideration the correlation. Next, a factor analysis of the variable loyalty has been run with an oblimin rotation. (KMO=0.878, sign.=0.001). The total variance explained by the items of the variable happens with two components. They have respectively the Eigenvalues of 6.017 and 1.38. The cumulative percentage of variance explained by the items is 74%.

Other variables Product attitude

The variable Product attitude is not sufficient reliable with a Cronbach’s alpha lower than the norm if constructed by 3 items (α= 0.65, N=112). If item 1 is removed the Cronbach’s Alpha would increase to 0.801. Due to the increase of the Cronbach’s alpha of 0.151 making the variable meet the reliability norm of α > 0.7 it has been chosen to remove one item of the construct. Also, a factor analysis has been run for this variable (KMO=0.5, sign.=0.001). The total variance explained by the items of the variable happens with one component: the Eigenvalue is 1.67. The cumulative percentage of variance explained by the items is 83.45%.

The equities Value equity Etos

A reliability analysis is performed with the questions that measure the different variables. The value equity variable consists of 10 items. The Cronbach’s alpha is 0.366 (n=31). Looking at the inter-items correlations matrix it results that the items are not correlated with each other (all numbers < than 0.8). This means that no rotation is needed when performing the factor analysis. The Cronbach’s number is very low. But it has been decides not to alter the construct of the variable too much in order to continue to perform analysis and to keep the variables that will be compared as equal as possible. A factor analysis has been run (KMO=0,508; sign.=0.001) and there are three components with a Eigenvalue above the 1. Component 4 (EV=0.979) is also taken into consideration because it explains 10 percent

more of the total variance. Cumulative the four components explain 74 percent of the total variance. This way no information needs to be excluded. For all the equities the percentage of variance explain per component is important. The weight needs to be taken into account in order to possibly make a conclusion about the hypothesis differences in customer equity between the stores. In the appendix an overview of the weights is made available.

The formula that is used to calculate the variable value equity of the Etos is the following: (((Value1+Value5+Value6+Value7)/4)*0.364)+(((Value2+Value8+Value9)/3)*0.262)+(((Va lue3+Value4)/2)*0.242)+(Value10*0.132).

Value equity Kruidvat

A reliability analysis is performed with the items that measures the variable. The value equity variable consists of 10 items. The Cronbach’s alpha is 0.654 (n=29). Even though the Cronbach’s Alpha is not optimal, it has been chosen to not remove any questions in order to increase this number. The main reasons are the low increase of the number ( <0.1) and the opportunity to keep the amount of questions that measure the variable between the two stores equal. Looking at the inter-items correlations matrix it results that the items are not correlated with each other (all numbers < than 0.8). This means that no rotation is needed when performing the factor analysis. After running a factor analysis with the 10 items (KMO=0,493; sign.=0.003), it can be said that there are four components with a Eigenvalue above the 1. The cumulative percentage of the total variance that the variable explains is 70.9 %.

The formula that has been used to calculate the Value equity of the Kruidvat is the following: (((Value4+Value3+Value6+Value7)/4)*0.4)+((Value5)*0.227)+(((Value1+Value8+Value10) /3)*0.195)+(((Value2+Value9)/2)*0.178).

Brand equity Etos

A reliability analysis is performed with the questions that measure the different variables. The brand equity variable consists of 7 items. The Cronbach’s alpha is 0.723 (n=31). Looking at the inter-items correlations matrix it results that the items are not correlated with each other (all numbers < than 0.8). This means that no rotation is needed when performing the factor analysis. Next, a factor analysis has been run (KMO=0,737; sign.=0.001). There are 2 components with a Eigenvalue bigger than one, that cumulative explains 62.9 percent of

the total variance. Component 3 (EV=0.740) is not taken into consideration. Even though it explains 10 percent more of the total variance of the variable all the questions load onto one of the two components with a Eigenvalue above the one, so no information needs to be excluded.

The formula that has been used to calculate the Brand equity of the Etos is the following: (((Brand1+Brand2+Brand3+Brand6+Brand7)/5)*0.649)+(((Brand4+Brand5)/2)*0.351).

Brand equity Kruidvat

The brand equity variable consists of 7 items. The Cronbach’s alpha is 0.862 (n=29). No items are removed due to the high Cronbach’s Alpha. Looking at the inter-items correlations matrix it results that the items are not correlated with each other (all numbers < than 0.8). This means that no rotation is needed when performing the factor analysis. Next follows the results of the performed factor analysis (KMO=0,785; sign.=0.001). There are two components with a Eigenvalue above the 1. The cumulative percentage of the explained variance of the variable is 71.8 percent.

The formula that has been used to calculate the Brand equity of the Kruidvat is the following: (((Brand2+Brand3+Brand4+Brand5+Brand6)/5)*0.777)+(((Brand1+Brand7)/2)*0.223).

Relationship equity Etos

A reliability analysis is performed with the questions that measure the variable. The relationship equity variable consists of 7 items. The Cronbach’s alpha is 0.769 (n=31). Looking at the inter-items correlations matrix it results that the items are not correlated with each other (all numbers < than 0.8). This means that no rotation is needed when performing the factor analysis. A factor analysis has been performed (KMO=0,617; sign.=0.001) and there are two components with a Eigenvalue above the 1. Component 3 (EV=0.92) is taken into consideration. The extra component explains more than 10 percent of the total variance of the variable Relationship equity. In this manner no information needs to be excluded. In total the two components explain cumulative 76,7 percent.

The formula that has been used to calculate the Relationship equity of the Etos is the following:

Relationship equity Kruidvat

The relationship equity variable consists of 7 items. The Cronbach’s alpha is 0.795 (n=29). No items were removed from the analyses due to the sufficient level of the Cronbach’s Alpha and the lack of benefit if removing one items (increase CA < 0.1). Looking at the inter-items correlations matrix it results that the items are not correlated with each other (all numbers < than 0.8). This means that no rotation is needed when performing the factor analysis. A factor analysis has been run (KMO=0,749; sign.=0.001) and there are 2 components with a Eigenvalue above the 1. Cumulative they explain 75 percent of the total variance of the variable.

The formula that has been used to calculate the Relationship equity of the Kruidvat is the following:

(((Rela3+Rela4+Rela5+Rela6+Rela7)/5)*0.668)+(((Rela1+Rela2)/2)*0.332).

Customer equity Etos

A reliability analysis is performed with all three drivers of customer equity. The customer equity variable consists of 3 items. The Cronbach’s alpha is 0.63 (n=31). Even though the Cronbach’s Alpha is not optimal it has been chosen to not remove any of the variables in order to keep the construct of customer equity as complete as possible and as close to the theory as possible. Looking at the inter-items correlations matrix it results that the items are not correlated with each other (all numbers < than 0.8). This means that no rotation is needed when performing the factor analysis.

After running a factor analysis with the 3 items (KMO=0,492; sign.=0.000), there have been extracted three component with respectively the following EV: 1.78, 0.88 and 0.34. Component 1 explains 59.3%, component 2 explains 29.4% and component 3 explains 11.3% of the total variance. The items load as follows: Component 1 represents Brand equity, component 2 represents value equity and component 3 represent relationship equity.

The following formula is used to calculate the customer equity of Etos:

Customer equity Kruidvat

A reliability analysis is performed with all three drivers of customer equity. The customer equity variable consists of 3 items. The Cronbach’s alpha is 0.666 (n=29). Even though the Cronbach’s Alpha is not optimal it has been chosen to not remove any of the variables in order to keep the construct of customer equity as complete as possible and as close to the theory as possible. Looking at the inter-items correlations matrix it results that the items are not correlated with each other (all numbers < than 0.8). This means that no rotation is needed when performing the factor analysis.

After running a factor analysis with the 3 items (KMO=0,541; sign.=0.001), there have been extracted three component with respectively the following EV: 1.8, 0.87 and 0.33. Component 1 explains 60%, component 2 explains 29% and component 3 explains 11% of the total variance. The items load as follows: Component 1 represents Brand equity, component 2 represents Value equity and component 3 represent Relationship equity.

The following formula is used to calculate the customer equity of Kruidvat:

(Value equity Kruidvat*0.29)+(Brand equity Kruidvat*0.6)+(Relationship equity Kruidvat* 0.11)

An overview of the weight of the different items within the drivers of the customer equity and also of the weight of the three drivers within the customer equity itself is made available in the appendix.

A table with an overview of the means, standard deviations and correlation numbers of the variables is made available in the appendix. There are no abnormalities found here. Moreover, the sample of every group is not large enough to assume normal distribution. To proceed with other calculations the skewness and kurtosis number have been checked of every outcome variable to see if the sample is normal distributed. The norms for skewness are -0.8 and 0.8. The norms for kurtosis are -3 and 3. There are two exceptions: the skewness of the value equity of Etos is 0.84 and the skewness of PPPTot in condition 4 is -1.516. When looking at the plots of the abnormalities, that are available in the appendix, these do not give reason to use test for abnormal distribution. There will follow no actions to use test for an abnormal distributed sample for further calculations: the assumption is normal distribution for all variables.

4.2 Hypothesis testing

The hypotheses that will be tested in this research are the following ones:

H1: Discount has a positive effect on the product price perception.

H2: The positive relationship between discount (IV) and product price perception (DV) is negatively moderated by the factor customer equity (Moderator) (e.g. the relationship between discount and product price perception is weaker with higher values of customer equity).

H3: Product price perception has a positive effect on store brand loyalty.

In order to test the first hypothesis the variable product price perception needs to be compared of both stores in the discount and in the no discount condition. If there is a significant difference this will give an incentive to look for a reason for this difference. A t- test will be performed.

To test the second hypothesis the first hypothesis needs to be accepted. If there is significant difference the next test will be a regression. Due to the assumption of normal distribution with this test you find out if customer equity can predict the product price perception in the different conditions.

For the third hypothesis there is a need for a regression to test if product price perception predicts loyalty. A previous test that needs to be performed is an ANOVA to see if there is a significant difference between the groups.

Results

The first test is a t-test to see if there is a difference between the product price perception in the discount and no discount condition. A comparison has been made between the means of the product price perception variable in both the conditions with and without discount. Next to the equal variance (F=0.526, p>0.05) there is a significant difference between the two groups. The mean of the product price perception without discount is lower (M=2.86, SD=.71) than the product price perception with discount (M=3.53, SD=.70, t(110)= 5.07, p=0.001. Hypothesis 1 is accepted.

Not only there has been taken a look if there is a difference in product price perception between the conditions discount and no discount, but also in both the condition of store. A comparison with all four groups has been made with an ANOVA. The main effect of product price perception was significant , F(3,108)=8.611, p=0.000. The four different condition differ on the level of product price perception. These results have been checked by looking at results of the Welch and Brown-Forsythe test and are correct. Also the post hoc test verifies the results of the performed ANOVA. There are significant differences between the four conditions in the product price perception. The most interesting difference would be the difference between the differences of the condition discount of both stores. A graph has been made of the means of all 4 conditions.

Figure 4 – Means four conditions product price perception

The product price perception is in general higher for the Kruidvat than the Etos. The graph furthermore shows the positive effect that discount has on the dependent variable product price perception. The difference between the means of the variable product price perception without and with discount for the Etos is 0.7072. The difference between the means of the variable product price perception without and with discount for the Kruidvat is 0.6428. The slope of the Etos is more steep than the slope of Kruidvat. Discount alters the product price perception more at the Etos than at the Kruidvat.

For the second hypothesis a regression needs to be run in order to find out if customer equity is a moderator of the relationship between discount and product price perception. There has been run a PROCESS using model 1. The used variables were the following ones: Y = PPPTot, X = Discount and M = MeanCE. The sample size is 112. The model summary shows that there is a predicting variable of the dependable variable. The variables discount and customer equity explain 43.9% of the variance (R2=0.19, p=0.000). The model show that only discount significantly predicts product price perception (t(112)=4.94, p=0.000). The level of customer equity and the interaction of the variables discount and customer equity are not significant and are thus not predictors of product price perception. The whole PROCESS analysis is included in the appendix. It can be concluded that the second hypothesis is not supported.

Looking for more insight into that the non-significance result of the last test different t-tests has been run on all the equities. A comparison has been made between the different equities (e.g. value equity, brand equity, relationship equity and customer equity).

The mean for the value equity of Etos (M=3.62, SD=0.29) is higher than the mean for the value equity of Kruidvat (M=3.56, SD=0.36). The difference in means between these groups of equal variance is insignificant (t(58)=0.765, p=0.45). The mean of brand equity for the Etos (M=2.95, SD=0.43) is higher than the mean of brand equity for the Kruidvat (M=2.77, SD=0.65). The difference in means between these groups of unequal variances (F=4.52, p<0.05) is insignificant (t(58)=1.30,p=0.195). Because no other calculations will be made with this variable the unequal variance has consequences for any test. The relationship equity mean for Etos (M=2.62, SD=0.58) is higher than the mean of the relationship equity of Kruidvat (M=2.47, SD=0.63). The difference between the means of these groups of equal variance (F=0.019, p>0.05) is insignificant (t(58)=0.718, p>0.05). The mean of the customer equity of Etos (M=3.11, SD=0.34) is higher than the customer equity of Kruidvat (M=2.97, SD=0.48). The difference between the means of these groups with equal variance (F=3.43, p=0.07) is insignificant (t(58)=1.39, p=0.171). All the means have been collected and made visual in the following graph.

Figure 5 – Means equities

There have not been found significant differences between the two stores in any equity but when interpreting the graph the following tendencies are visual: all the equities are found higher at the Etos rather than the Kruidvat.

For the third hypothesis a regression has been run to find out if product price perception can significantly predict loyalty. The result of the regression indicates that the into four conditions divided predictor product price perception explained cumulative 45.3% (R2=0.18) All the F values (1,26) and the Beta values are not significant. For an overview the results see the appendix. The conclusion can be made that the variable product price perception does not predict loyalty. Hypothesis 3 is not supported.

An ANOVA has been run to see if the different condition were different. The main effect of loyalty was not significant, F(3,108)=0.494, p=0.687. The four different conditions do not differ on the level of loyalty. These results have been checked by looking at results of the Welch and Brown-Forsythe test and are correct. Also the post hoc test verify the results of the performed ANOVA. Apparently the condition discount and the condition store do not have any influence in the variable loyalty. Two t-tests confirm this assumption. A comparison between loyalty in the condition with and without discount has been made. Because of the Levene’s test (F=0.152, p>0.05) we assume equal variance. Looking at the significance of the difference between the two groups the p>0.005 indicates an insignificant difference between loyalty without discount (M=2.83, SD=0.82) and loyalty with discount (M=2.9, SD=0.78). A

comparison has been made between the means of the loyalty of the Etos (M=2.95, SD=0.83) and the Kruidvat (M=2.78, SD=0.76). The groups are of equal of variance (F=0.691, p>0.05) and the difference between the means of the two groups is not significant (t(110)=1.105, p=0.272). A graph has been made to make the means of loyalty of the different groups visual.

Figure 6 – Means four conditions loyalty

Even though the differences are not significant there are certain tendencies present. It has been made visual that the Etos scores higher then the Kruidvat. Further it can be said that discount has a positive relationship with loyalty. The difference between the means of the variable loyalty without and with discount for the Etos is 0.0536. The difference between the means of the variable loyalty without and with discount for the Kruidvat is 0.1. The slope of Kruidvat is steeper than the slope of Etos, meaning that the discount condition has more effect on the loyalty of the customer at the Kruidvat than the discount condition has on the loyalty towards the Etos.

Other results

Some questions of the questionnaire give some extra inside in the thoughts of the participants about the stores. In this section of the results these questions have been analyzed.

The participants were asked how much discount they would except in the drugstore that they were confronted with. For the Kruidvat the most chosen answer is 50% discount (7x), 0% discount (6x) en 30% discount (5x). For Etos the respondent have most chosen for 25%

discount (7x), 20% discount (5x) en 10% discount (4x). From this overview it can be concluded that the participants of the survey except higher discount at the Kruidvat then the Etos. Peculiar is the 0 % discount that has been mentioned 6 times for the Kruidvat. An explanation for this odd percentage could be the low price image that Kruidvat has (product price perception Kruidvat slope between discount and no discount condition less steep then product price perception Etos slope between discount and no discount condition). Product price perception does not change much at the Kruidvat in the two different conditions.

Another question was proposed in the survey to the see how the customer valued the different equities. The respondents needed to distribute 10 points between the three equities (Value equity, brand equity and relationship equity).

Table 2 – Means answers distribution question

Mean weight equities Etos (n=31) Mean weight equities Kruidvat (n=29)

With this responses there is insight on the importance of each equity according to the customer. These results can be compared with the actual measurement of the distribution of the equities that are made visual in the following table.

Table 3 – Weight equities

A first observation is the valuation of brand equity which has been measured to be more important than the Value equity for both the stores. Second the differences between the equities are bigger in the thoughts of the customer then actual measured. Also the differences measured between the equities of the stores are much smaller than the differences that are visual in the responses to the distribution question with the exception of relationship equity.

Value equity 5.7 Brand equity 2.7 Relationship equity 1.6 Value equity 5.2 Brand equity 3.1 Relationship equity 1.7 Etos (%) Kruidvat (%) Brand equity 59.3 60 Value equity 29.4 29 Relationship equity 11.3 11

A question was proposed in the survey to see if the participant would feel about the 20 percentage discount and if this amount of discount fits with the image of the store. 56 people had filled out this question. For Etos the following number are applicable: n=28 (M=3.68, SD=0.67). For the Kruidvat the following numbers are applicable: n=28 (M=3.89, SD=0.685).

The following chart shows the distribution of the answers for both conditions (Etos & Kruidvat) according to the 5-point Likert scale.

Figure 6 – Visualization answers fit discount

An independent t-test shows that the means of the answers to this questions, assuming equal variances (F=1.067, p=0.306) the groups do not significantly differ from each other (t(54)= -1.183, p=0.242). The tendency that can be seen from the chart that the 20 percent discount is more likely to fit with the store Kruidvat then the Etos. The higher mean of Kruidvat confirms this tendency.

The question “Which drugstore do you preference go to shopping?” has been asked to all the 112 persons that have taken the survey. 39,3% preferred the Etos, 33% preferred the Kruidvat and 27,7 % did have any preference.

Figure 7 – Visualization answers preference question

Because there has not been a pretest to see where the preference lied these could have had influence on the customer equity and the weight that every question had on it. Almost one third of the respondents had no preference for one of the drugstores, probably flattening out the results.

The main effect of product attitude was not significant, F(3,108)=0.932, p=0.428. The four different conditions do not differ on the level of product attitude. See table in the appendix for an overview of the means. In the following graph all the means of the four groups are made visual.