The effect of pricing on the perceived quality and willingness to buy
of a national brand, compared to its private label competitor.
-The consumer’s perspective-
The effect of pricing on the perceived quality and willingness to buy of a
national brand, compared to its private label competitor.
-The consumer’s perspective-
Willemijn Sanders
August 2010
University of Groningen Faculty of Economics & Business Master Thesis Business Administration
Marketing Management
Supervisor RUG: Prof. Dr. Tammo H.A. Bijmolt Co-assessor RUG: Dr. Laurens M. Sloot Supervisor GfK Panel Services: Joop Holla
Van Kinsbergenstraat 24-3 1057 PR Amsterdam willemijnsanders@hotmail.com
MANAGEMENT SUMMARY
This thesis focuses on the phenomenon of private labels (PL) in retail and how manufacturers of national brands (NB) should cope with the threat of private label, in terms of the price setting of their products. The goal of this study is to come up with guidelines regarding the price gap between national brands and private label copycats in order to surpress the market share of the private label product. The central research question of this study is: “What is the effect of pricing on the perceived
quality and willingness to buy of a national brand, compared to its private label competitor?” -The consumer’s perspective- The ‘consumer’s perspective’ is added to emphasize the point of view that
product prices have an effect on the perceived quality of consumers towards those products.
The first step in answering the central question is a literature research of articles regarding private label and the effect of price on perceived quality. Based on this literature two main effects are extracted; a hypothesized positive relationship between the price gap between private label and the national brand and perceived quality of the national brand and a hypothesized positive relationship between the price gap and market share of the national brand, through perceived quality. The hypotheses are then empirically tested with data from the GfK Autumn Report. At first, a factor analysis is conducted on thirteen items, to construct the dimension ‘perceived quality’. In order to test the hypothesized main effects two General Linear Models (GLM) are constructed.
The first GLM tests the effect of the price gap between the national brand and its private label competitor on the perceived quality of the national brand. This model shows a significant and positive effect of price gap on perceived quality. Additional analysis shows a positive relationship between price gap and perceived quality for products that fall within a small (0-20%) or medium (20-40%) price gap and a negative relationship for products in the 40≥% price gap.
Store name does not have a significant effect in the GLM, which could stem from the fact that store name has only a small effect on perceived quality, a finding that was already described in earlier published literature.
The second GLM tested the effect of the price gap on market share, direct as well as indirect, through perceived quality. This model led to the conclusion that there is a negative direct effect of price gap on market share. Since the interaction effect was found to be positive, it can be concluded that the indirect effect of price on market share, through perceived quality, is indeed positive. It was also found that the decrease in market share between the medium and large price gap was more gradual then the decrease between the small and medium price gap. Also, as hypothesized, the effect of perceived quality on market share was positive.
The results lead to the overall conclusion: “When the price gap between private label and national
brand products does not exceed the 40%, higher price gaps have a positive effect on the perceived quality. Although a direct effect of price gap on market share is negative, price gap can lead to higher market share of the NB through enhancing perceived quality.”
Based on the conclusions the following recommendations are formed. In order to achieve the highest perceived quality, manufacturers should strive for a 20-40% price gap between their national brand product and the private label competitor. In addition, in categories with high perceived quality variance, price should be used as a tool to emphasize the variance between NB and PL. In product categories that are perceived as hedonic, price could also be used as a tool to enhance perceived quality.
PREFACE
This thesis is the final step in finishing the Master Marketing Management. After my bachelor Business Administration I decided to continue my study with the Master Marketing Management. For me this turned out to be the perfect master, with courses that fitted all my interests, dedicated teachers and an inspiring learning environment. The course ‘Retailer Marketing’ drew my intention to the interesting development of private label products. Therefore, I decided to dedicate my final assignment of the master to this subject.
I would like to thank my supervisor Tammo Bijmolt and my co-assessor Laurens Sloot for their feedback and constructive comments. Also, I would like to thank Joop Holla from GfK Panel Services for his cooperation and his approval that lead to the use of very interesting GfK data for this research.
At last, I would like to thank my parents. Through every phase of my study they supported me and enabled me to develop myself in numerous ways, for which I am very grateful. In addition, I would like to thank my boyfriend Pim for his support and patience, especially during the last months.
TABLE OF CONTENT
MANAGEMENT SUMMARY ... 3
PREFACE ... 5
1. INTRODUCTION ... 10
1.1. Private label ... 10
1.2. Strategies for manufacturers ... 11
1.3. Research goal ... 11 1.4. Sub questions ... 12 1.5. Scope ... 12 1.6. Relevance ... 13 1.6.1. Scientific Relevance ... 13 1.6.2. Managerial Relevance ... 13 1.7. Data collection ... 13 1.8. Study outline ... 14 2. LITERATURE REVIEW ... 15 2.1. Introduction ... 15
2.2. The concept of perceived quality ... 15
2.3. Informational cues ... 17
2.3.1. Price as an external information cue... 18
2.3.2 The influence of price on perceived quality ... 18
2.3.3. Price premium as a strategy ... 19
2.3.4. The price-perceived quality relationship in fast moving consumer goods ... 19
2.3.5. The price-perceived quality relationship across product categories ... 20
2.3.6. Non price external cues ... 21
2.3.7. The influence of non price external cues on perceived quality ... 21
2.4. The influence of price on purchase ... 23
2.4.1. Direct effect of price on purchase intention ... 23
2.4.3. Influence of price on purchase intention of store vs. national brands ... 24
2.5. The price gap between store brands and national brands ... 25
2.6. The influence of perceived quality on purchase ... 26
2.6.1. The influence of perceived quality on purchase intentions ... 26
2.6.2. The influence of perceived quality on the purchase of store brands ... 27
2.7. Hypotheses ... 29 2.8. Conceptual Model ... 29 3. RESEARCH DESIGN ... 31 3.1. Data ... 31 3.1.1. Autumn Report ... 31 3.1.2. Sample ... 31 3.1.3. Aspects ... 33 3.1.4. Perceived Quality ... 34 3.1.5. Chosen brands ... 35 3.1.6. Chosen retailers ... 35
3.1.7. Price Gap between National Brand products and Private Label products ... 36
3.1.8. Perceived quality gap ... 37
3.1.9. Utilitarian vs. Hedonic products ... 37
3.1.10. Market Share ... 38
3.2. Analysis ... 38
3.2.1. General Linear Model ... 38
3.2.2. Compare Means ... 39
3.2.3. Single Regression Analysis ... 39
3.2.4. Multiple Regression Analysis ... 39
4. RESULTS ... 43
4.1. The price gap perceived quality relationship ... 43
4.1.2. Hypothesis testing ... 46
4.2. The price gap market share relationship ... 52
4.2.1. General Linear Model ... 53
4.2.2. Hypothesis testing ... 55
4.3. Overview of results ... 57
5. CONCLUSIONS & RECOMMENDATIONS ... 58
5.1. Conclusions ... 58 5.2. Recommendations ... 60 5.3. Limitations ... 61 5.4. Future research ... 63 6. REFERENCES ... 64 6.1. Articles ... 64 7.2. Websites ... 68 7. APPENDIX ... 69
7.1. Questionnaire GfK Autumn Report ... 69
7.2. Thirteen aspects from the questionnaire ... 73
7.3. Chosen Brands ... 74
7.3.1. Albert Heijn... 74
7.3.2. C1000 ... 75
7.4. Factor Analysis ... 76
7.4.1. KMO and Bartlett’s Test ... 76
7.4.2. Reliability ... 76
7.4.3. Importance ... 77
7.4.4. Component Matrix ... 77
7.4.5. Communalities ... 79
7.5. AH and C1000 store brand ... 79
7.5.2. ANOVA ... 80
7.6. Price gap perceived quality relationship ... 80
7.6.1. GLM according to the conceptual model ... 80
7.6.2. Regression on price gap and perceived quality for each price gap ... 80
7.6.3. Regression analyses on perceived quality for PL and NB ... 81
7.6.4. Means perceived quality ... 82
7.6.5. Regression analyses on four product categories... 82
7.7. Relationship between price gap and market share ... 83
7.7.1. Descriptive dummy variable perceived quality ... 83
7.7.2. GLM according to the conceptual model ... 84
7.7.3. ANOVA market share per price gap ... 84
1. INTRODUCTION
At first, it seemed a rather innocent introduction of a small assortment with inferior, low priced goods. But nowadays this development has evolved to a worldwide battle of the brands. The battle with private label is one of the major issues that national brands worry about. Private labels or store brands are products sold under complete ownership, control and exclusively sold by the retailer (Sethuraman & Cole, 2009). The continuous rise of private label share in food products poses serious challenges for manufacturers. As the threat of private label expands to numerous product categories, the importance to compete with private label increases (Abe, 1995; Ashley, 1998; Cotterill, Putsis & Dhar, 2000).
The Dutch market is no exception in this competition between branded products and store brands; with a turnover growth of 7% compared to 2.2%, private label still grows at the expense of national brands (IRI, 2010).
1.1. Private label
Where private label started as cheap inferior goods, it developed to several types of products that serve different segments of the market. Kumar and Steenkamp (2007) divide private label in four groups including copycat brands. The group of copycat brands makes up for 50% of the total private label products (Kumar & Steenkamp, 2007). These products are characterized by a quality that is very close to that of the national brand, but with a lower price. Hoch (1993) recognizes the copycat products as a specific threat to manufacturers. Figure 1.1. shows two examples of copycat private label products.
1.2. Strategies for manufacturers
Battling private label has several differences compared to competing with other national brands (Hoch, 1996). The fact that the retailer is both customer and rival makes the competition with private label very complex (Hoch, 1996, Kumar & Steenkamp, 2007). Ashley (1998) states that there are two ways in which manufacturers can compete against private label; with innovation and with pricing. Hoch (1996) elaborates on the two dimensions quality and price and distinguishes six options for leading brands to compete, where reducing the price gap is one of them and the only one that directly influences the pricing of the products. Changing the price gap has a statistically significant effect on sales and profits of both national and private label across all the eighteen categories tested. With a small price gap between the national brand and the private label the national brand is in favor, while the private label benefits from a large price gap (Hoch, 1996). Changing the price gap can be achieved by permanently setting lower prices or by executing price promotions. However, while reducing the price gap seems a strong strategy for the manufacturer, it does not seem logic that the retailer will reduce the price gap voluntary, in favor of the national brand (Hoch, 1996). So, when developing a strategy for the national brand to compete with its private label competitor with the price dimension
as a tool, this limitation should be kept in mind.
The research of Hoch and Banerji (1993) shows that the price gap between national and private label brands in different categories is not related to private label market share in the category. They conclude that quality, which is significantly related to private label market share, is more important than price. Kumar & Steenkamp (2007) also mention that the price gap with the national brand may not be of particular influence on private label. On the contrary, the research of Sinha & Batra (1999) shows that price consciousness, and thereby price, is in fact a highly significant predictor of private label purchase. Abe (1995) also acknowledges pricing as an important tool in fighting against private label, although the outcomes of this tool are mixed.
1.3. Research goal
research goal of this thesis is, to provide theoretical guidelines for manufacturers to cope with the threat of a private label competitor, using pricing as a tool. This leads to the central research question:
“What is the effect of pricing on the perceived quality and willingness to buy of a national brand, compared to its private label competitor?”
-The consumer’s perspective-
1.4. Sub questions
In order to answer this central question, several sub-questions need to be answered. This research starts with a review of literature on the topic of private label and price setting, and price setting and perceived quality. This literature review will find answers on the following sub-questions:
SQ I. What is the definition of the concept of perceived quality?
SQ II. What characteristics make a category more conducive for price as an indicator for product quality?
SQ III. What is the influence of product price on purchase intention?
SQ IV. What is the influence of perceived product quality on purchase intention?
SQ V. What is the influence of the price gap between the national brand and its private label competitor on the market share of the national brand compared to the private label variant?
1.5. Scope
In this research several relationships between price, perceived quality and willingness to buy (reflected in market share) are investigated in order to answer the central research question. However, the relationships between these concepts are complex and influenced by numerous factors. To investigate these relationships in detail is way beyond the scope of this research. In order to keep this research manageable, strict boundaries should be added. The direct relationship of price on perceived quality will be investigated, moderated by store name, perceived quality differences within the category, and whether the product is hedonic or functional. For the relationship perceived quality and market share, no moderating variables are included.
1.6. Relevance
1.6.1. Scientific Relevance
Although the development of private label is generally considered as a very serious threat to national brands, it seems that empirical work on how to compete with private label is rather scarce and more aspects of strategies to compete with private labels should be studied (Cotterill, et al. 2000; Verhoef, Nijssen & Sloot, 2002). Abe (1995) also notes the lack of theoretical work in the literature regarding private label. In order to compete private label effectively, theoretical guidance, among other things, on pricing is necessary. This theoretical work should consist of research specifically focused on private label situations, and not, as is the currently the case, of theories ‘borrowed’ from the established fields of marketing and/or economics (Abe, 1995). This thesis researches the possibilities to battle private label with the use of price setting, in the specific case of copycat private label products.
1.6.2. Managerial Relevance
Previously, national brand manufacturers underestimated private label as an equal competitor. In order to compete with private label in a constructive way, the first thing managers should do is acknowledge the threat of private label, and treat them as an equal competitor, compared to the national brand competition. Subsequently, they should tailor their strategy to this competition (Hoch, 1996). The emphasis in possible strategies for managers is mainly on advertisement and innovation (Ashley, 1998; Hoch, 1996; Verhoef, Nijssen & Sloot, 2002). Pricing strategies to fight private label mainly focus on managing the price gap (Hoch, 1996, Verhoef, Nijssen & Sloot, 2002) and not on the different consequences of price setting on willingness to buy. In addition, strategies that do focus on the price gap between national and store brands deliver mix results (Abe, 1995). This thesis aims to give manager additional guidelines regarding the price gap between national brands and private label. These additional guidelines are relevant because they include also the indirect effect of price setting, through perceived quality, and not only the direct effect price has on willingness to buy. However, despite the managerial relevance, it should be kept in mind that the ability of national brands to influence consumer product prices is limited.
1.7. Data collection
2007). Chapter 2 consists of a literature review with data from leading marketing journals, such as Journal of Marketing, Journal of Advertising Research Journal of Marketing Research and Journal of
Marketing Theory.
The research of literature will lead to the formulation of several hypotheses. These hypotheses are tested with the use of primary data. This method of using two types of data is according to the general rule of Malhotra (2007). The primary data is the results of a survey of 5927 Dutch households. This data was previously used in the Autumn Report 2009 of GfK Panel Services. The questionnaires were constructed in such a way that they reflected the brand equity of 50 leading brands in the Netherlands, according to the framework of Steenkamp & Dekimpe (see appendix 7.8.) and Keller (2008). Although the data was not primarly collected in order to answer the research question of this thesis, it is still suitable for solving the specific problem statement of this thesis.
1.8. Study outline
2. LITERATURE REVIEW 2.1. Introduction
As pointed out previously in the introduction, product price can be used as a powerful tool for manufacturers in their battle against private label. This thesis focuses primarily on two effects of the price gap between a national brand and its private label competitor. These two effects of the price gap are represented in figure 2.1.; a direct effect of the price gap on market share, and an indirect effect, through perceived quality. In this chapter, the concept of perceived quality is discussed, including factors that influence this concept. After that, the influence of price on both perceived quality and market share is pointed out, followed by a section that focuses specifically on the price gap between national and store brands. The secondary data on these different subject leads to the formulation of several hypotheses, which are summarized at the end of this chapter, complemented with a conceptual model.
Figure 2.1. The effects of the price gap between NB and its PL competitor
2.2. The concept of perceived quality
The concept of perceived quality plays a central role in this research. This concept is complex and hard to define. In this section an overview of literature on the concepts of quality and perceived quality will be given in order to come up with a definition that is applicable to this research.
The article of Garvin (1984) gives five approaches in defining quality. Because of the consumer perspective of this research, the user-based definition of quality will be used throughout this thesis. This definition is characterized by the premise, that individual consumers have different preferences and quality is the degree in which the product fulfills the wants or needs of the consumer (Garvin, 1984). This definition of quality is, like perceived quality, already highly personal and consequently subjective. Taking all five perspectives into account, a framework with eight basis elements of product quality was constructed. The other dimensions of product quality are: performance, features, reliability, conformance, durability, serviceability and aesthetics. In this framework, perceived quality is the eighth element. The user-based definition is believed to focus on aesthetics and perceived quality.
Perceived Quality NB Price Gap
NB-PL Competitor
The article of Brucks, Zeithaml & Naylor (2000) also divides quality in six dimensions which are comparable to those of Garvin (1984). Instead of perceived quality the term ‘prestige’ is used, based on included references to brand image and appearance. These articles compare perceived quality with prestige and the aesthetics of the product. Without actually using the product, the consumer uses the ‘outside’ of the product to judge the actual performance.
One leading article on the concept of perceived quality is the research of Zeithaml (1988). The definition of perceived quality given in this article is as follows: “The consumer’s judgment about a product’s
overall excellence or superiority”. Monroe and Krishnan (1985) suggest a slightly different definition: “Perceived product quality is the perceived ability of a product to provide satisfaction relative to the available alternatives?” A third, widely used, definition of perceived quality is simply ‘fitness for use’
(Steenkamp, 1989). Several variations on these definitions are available. To provide some clarity, Trinkle (1984) divides all the definitions of perceived quality in three divisions. First, the neutral concept is a distinction between ‘much quality’ and ‘not much quality’, and quality is treated as the nature of the product. The second concept is evaluative and distinct between good and bad quality, and focuses on the fitness for use of the product. The term ‘quality product’ fits the third type of perceived quality definitions and represents the overall superiority or excellence of a product with respect to all the attributes of that product.
As discussed in the previous part, quality as well as perceived quality are complex concepts and hard to capture in a single definition. However, this research focuses primarily on the influence of extrinsic attributes (price and store name) on perceived quality. With this perspective in mind, the definition of Zeithaml (1988) seems to fit the scope of this research and will thus be used as the correct definition for perceived quality. The findings of Garvin (1984, 1987) will be used as additional information on the concept of perceived quality. In the following part of this thesis, the factors influencing perceived quality according to Zeithaml (1988) will be further discussed.
Figure 2.2. Factors influencing perceived quality
Perceived quality is not the same as objective or actual quality nor is it a specific product attribute. However, the term objective or actual quality could be debated, because quality is always perceived by someone (Zeithaml, 1988). According to Garvin (1987) perceived quality is fed mostly by reputation and the assumption, that if a certain brand delivered good products in the past, the product quality will also be high in the future.
The lack of congruity and clearness of the concept of quality is the motive in the research of Ghylin et al. (2008), where empirical support for the dimension of product quality is given. The article includes a representation of several clusters of words associated with product quality. Words that are associated with product quality are, among others, good value and conformance, which represents the ‘degree that a product’s design matched established standards’ (Garvin, 1984, 1987).
2.3. Informational cues
As can be seen in figure 2.2., in the process of product evaluations, consumers use internal and external cues. According to Monroe & Krishnan (1985) a cue can be defined as “any informational stimulus about
or relating to the product”. Internal cues are those attributes, that are derived from the physical product
(such as nutritional content), whereas external cues are not directly related to the performance of the product itself (Olson, 1973 in: Rao & Monroe, 1989). Although intrinsic cues should have a direct relationship with perceived quality, the capabilities of the consumer to judge the quality of a product on the bases of intrinsic cues may be limited (Chang & Wildt, 1996). Because of the easiness to understand and process extrinsic cues, consumers will often rely on extrinsic cues to judge the quality of a product instead limited (Chang & Wildt, 1996). Intrinsic informational cues can vary widely across product or service categories. On the contrary, external cues such as price, brand name and the level of advertising are not specific for one product and are thus used as quality indicators for a variety of products or services (Zeithaml, 1988).
Intrinsic Attributes
Extrinsic Attributes
Perceived Monetary Price
2.3.1. Price as an external information cue
Price is believed to be an indicator for quality; this follows from the reasoning that a higher price reflects the higher costs of producing the product. Assuming efficient production of products, the price per product is higher when better and thus more expensive ones, materials etcetera are used (Steenkamp, 1989). In a situation with rational and perfectly informed consumers, it seems logic that a more expensive product than a competing product, but not of higher quality, would simply not be bought. However, the paradox is that in the situation of perfectly informed consumers, the need for price as an informational cue would also disappear (Steenkamp, 1989). In a more realistic situation, where the consumer is not perfectly informed, there is simply assumed that there is a positive relationship between price and quality. This premise is called the signal theory (Cataluña et al. 2006).
2.3.2 The influence of price on perceived quality
Several studies investigated the influence of price on perceived quality. The article of McConnell (1968) is an important study, where respondents had to rate a non durable product without actual quality differences, but with three different prices, on quality. This resulted in a positive, non linear relationship between price and quality and important evidence that consumers use price in order to judge the quality of a product. Because this research includes non-durable consumer goods, it could function as a basis for this research. Unfortunately, after years, an error in the computation of the results was discovered (Steenkamp, 1989), so it would be academically unjustifiable to use the research of McConnell as a guideline.
change (Shapiro, 1968). Another notable result of this research was an increased strength of price as a cue for quality for consumers that had trust in the capability and honesty in the price setting.
2.3.3. Price premium as a strategy
In the previous part, the relationship between price and perceived quality is described. Assuming a rational and continuous calculating consumer, that will only pay a price premium if this is justified by the actual quality, defined in this case as fitness for use (Rao, 2005). This means that influencing perceived quality of the product could be a strategy, but only in the situation of product trial. In establishing long term sales, this pricing strategy would fail. However, the article of (Shiv, Carmon & Ariely, 2005) describes a continuous and nonconscious effect of price on the quality perception after actually using the product. This is another argument supporting the idea that actual quality does not really exists and an important finding for the managerial implications of this thesis.
2.3.4. The price-perceived quality relationship in fast moving consumer goods
Because this research focuses on packaged consumer goods, at first it is necessary to figure out if there is actually a price-quality relationship for these products. According to Zeithaml (1988) the use of intrinsic or extrinsic informational cues differs across product categories. Consumers will rely more on extrinsic cues instead of intrinsic attributes when the effort of evaluating intrinsic information (for example in the case of a packaged good, the nutritional information of the product) exceeds the value that can be extracted from this information (Zeithaml, 1988). In the case of fast moving consumer goods, the main intrinsic cues available are nutritional information (in the case of food and drinks) or other information about the ingredients of the product. Assuming that for the average consumer the time to judge this intrinsic information properly and that the time to do this, will exceed the value that could be extracted from this information, packaged consumer goods will be mostly judged on the basis of external attributes.
HI: In categories with a wide price gap between NB and PL, the perceived quality of the NB is also
higher.
2.3.5. The price-perceived quality relationship across product categories
However, the strength of the price-perceived quality relationship is not equal across product categories, and this indicator should therefore be used selective in order to lead to correct implications (Shapiro, 1983). According to Shapiro (1983) products of which the performance is hard to judge due to technical complexity or products for which the future performance is hard to abstract from observable aspects, will experience a stronger price-perceived quality relationship.
Another requirement for price as an indicator for perceived quality is large differences in perceived quality within the category (Shapiro, 1983). This is also called category quality variance and reflects the perception of a consumer of the extent, to which brands in a certain product category differ in quality (Bettman, 1974). This finding is also supported in the empirical research of Sethuraman & Cole (1999) which takes national brand and store brands into account. In addition this article advises to use price as a specific tool for increasing quality perceptions in these categories. High category quality variance leads to a stronger price perceived quality relationship, because consumers judge the quality of the private label brand compared to the national brand. Bases on the premise that private label products are of a
less quality than national brands (Huang, 2009) a high category quality variance leads to the idea that
there is also a higher variance between the national brand and its private label competitor, which means ‘automatically’ a more negative perception of the quality of the private label (Delvecchio, 2001). This leads to hypothesis II(a):
HII(a): In categories with higher perceived quality differences between products, the effect of the price
gap between NB and PL on the perceived quality of the NB will be stronger.
yield a higher consumption pleasure experience a stronger price-perceived quality effect (Sethuraman & Cole, 1999). Although a classification of products in hedonic and function will be suggestive rather than mutually exclusive, a rough definition of a functional product would be a ‘fundamental necessity’ (Lee & Hyman, 2008). An overview of hedonic versus functional products is not included in the research, but in the market of packaged consumer goods this will imply, that for example cookies will experience a stronger effect of price on perceived quality compared to toilet paper. Inclusion of this moderating variable leads to hypothesis HII(b):
HII(b): In categories perceived as hedonic, the relationship between the price gap between NB and PL
and the perceived quality of the NB will be stronger.
2.3.6. Non price external cues
Several other non price cues have been reported to influence perceived quality. In single-cue studies, scent, country of origin and brand name where found to have significant effects on perceived quality. (Steenkamp, 1989). The downside of these single cue studies is that they oversimplify the reality and have limited internal and external validity (Olson, 1977 from: Dodds & Monroe, 1985; Steenkamp, 1989). In order to reach a more realistic model, Dodds et al. (1991) add external cues to their concept. Although the complexity of the relationship between price, perceived quality and product purchase is too high to be captured in this thesis, an attempt is made to create some guidelines regarding these concepts and the relationships between them.
2.3.7. The influence of non price external cues on perceived quality
A number of articles mention the positive effect of price on perceived quality. This effect is particularly present when price is the only extrinsic cue available (Dodds & Monroe, 1985). Adding extrinsic cues diminishes the effect of price on quality (Dodds et al. 1991; Chang & Wildt, 1996), although this finding contradicts with the results of Monroe & Krishnan (1985 from: Dodds, et al. 1991). Another notable result of this research is the effect of store name in brand low and brand absent situations only. In order to create a more realistic model, at least one other informational cue should be included in the study. As mentioned before, brand name has the strongest direct effect on perceived quality.
Figure 2.3.Extended conceptualization to include brand name and store name by Dodds et al. (1991)
Previous research of the effect of external cues on quality perceptions reveals inconsistent results. According to Dodds et al. (1991) the importance of price in relation to extrinsic cues, such as brand and store name, may differ across products with a different nature, price range, and the research method used. In order to come up with theoretical guidelines it is necessary to research a narrow framework in which product type, private label type, price range and research method are constructed in a precise and realistic method for the specific situation (Abe, 1995).
However, as a result from a combination of three previous studies Dodds et al. (1991) obtain an average weighted effect for price, brand name and store name. Price has a strong effect on willingness to buy, brand name a moderate and store name a small effect. The effect of brand name on perceived quality is also supported in the study of Gardner (1971), which states that the effect of brand name in a multiple-cue study even exceeds the effect of price. Store name on the other hand is also found to have no significant direct effect on perceived quality (Stafford & Enis, 1969). In this context, the influence of store name on perceived quality is in fact a carry-over effect from the retailer to the branded product. In this research store brands are compared to national brands, hence the store name is also the name of the brand. Thus, the effect of the store name could be expected to be more or less the same as the effect of brand name on the relationship between price and perceived quality. This leads to the following hypothesis:
HII(c): Store name has a moderating effect which diminishes the price-perceived quality relationship of
the national brand.
for hedonic products such as cookies presumingly, there will be a stronger effect of price on perceived quality.
2.4. The influence of price on purchase
The influence of price on purchase is not as simple to capture as may seem. Although the concept of price elasticity seems rather straightforward, there are several dimensions that have to been taken into account. Price has a dual effect on purchase, which is represented in figure 2.4.
Figure 2.4. Conceptual relationship of price effect by Dodds et al. (1991)
The direct effect of price on purchase is straightforward: when the costs of a product increase, the willingness to buy will decrease. The indirect effect of price is through the perceived quality. The theory holds that this relationship is positive; when the product price increase, perceived quality will be higher and thus will the purchase intention increase. However, in practice a lot of other factors should be taken into account. The following section will go deeper into the direct and indirect effect of price on purchase.
2.4.1. Direct effect of price on purchase intention
2.4.2. Indirect effect of price on purchase
One relationship that has been under particular interest, and causes an indirect relationship between price and purchase intention, is the price-perceived quality relationship. Whereas many studies have tried to capture this relationship, its exact boundaries and conditions remain unclear (Dodds et al. 1991). On the one hand, the willingness to pay a certain price for a product depends on the utility that consumer derives from the product, where perceived quality is one major source of this utility (Kumar & Steenkamp, 2007). The perceived quality of a product reflects the perception of the consumer and is not the same as the actual quality of the product. On the other hand, price is also used as an indicator for the quality of the product.
The article of Dodds et al. (1991) elaborates on the price-quality relationship and describes the relationship between price and willingness to buy (figure 2.4.), based on the conceptualization of Monroe and Krishnan (1985), using a model of Monroe (1979). In this model the concepts of perceived value, sacrifice and quality are used. Perceived value is considered as a tradeoff between quality and sacrifice. Price has a positive relationship with perceived quality as well as perceived sacrifice. Because sacrifice has a negative and quality a positive effect on perceived value, price plays a dual role in the tradeoff between price and value.
In order to understand the relationship between price and quality, the concept of the acceptable price range should also be taken into account (Dodds et al. 1991). The consumer will refrain from purchasing the product if the price exceeds the acceptable price range. However, when the product price will fall below the range, the consumer will doubt the quality of the product, which will also decrease its purchase intention (Cooper 1969a from: Monroe & Grewal, 1991).
2.4.3. Influence of price on purchase intention of store vs. national brands
differences across categories, it appears that for national brands the purchase intention increases more after a price discount compared to the intention for store brands. This outcome does demand a sufficient quality gap between the store and the national brand (Bronnenberg & Wathieu, 1996). Aggarwal & Cha (1998) go even further by stating that the purchase intention between national brand and store brand is solely based on the price of the national brand, and that the price of the store brand is not taken into account.
After analyzing literature on the effect of product prices, it can be concluded that price has a direct effect on purchase intention as well as an indirect effect, because it influences the perception a consumer has of the quality of the product. National brands can use price discounts to steal consumers who would otherwise buy store brands, but the influence of price discounts on the purchase intention of store brands has much less influence.
2.5. The price gap between store brands and national brands
Although there is no consensus about the effect of price on private label share, the existence of a price gap between national brands and copycat private labels is evident (Kumar & Steenkamp, 2006). Contrary to economic reasoning, a large price gap between private label and the national brand will not necessarily drive private label success. Dhar & Hoch (1997) argue that this is due to price sensitivity, where price insensitive categories experience big gaps, whereas price sensitive categories have small gaps. Adding the fact that demand for private label is higher in the latter category, the link between a high share of private label and a small price gap is clear.
The influence of the price gap on store brand performance is not a straight line. Dhar & Hoch (1997) mention several factors that affect the relation between the price gap and store brand performance. In categories where a lower quality store brand is offered, the price gap and other factors that influence consumer price sensitivity, such as demographics, may be more influential in explaining retailer performance. Dhar and Hoch (1997) conclude that the average price gap across categories between national brands and private label is 40%, within a range of 60%. They also notice a positive effect between the price gap and store brand performance: a 10% change in price gap leads to a 0.8% change
in store brand share (average=8.2). Reflecting on the conceptual relationship by Dodds et al. (1991) as
shown in figure 2.4., the conclusion of Dhar and Hoch (1997) only takes perceived sacrifice into account, instead of the trade off between perceived quality and perceived sacrifice. Although the existence of a ‘turning point’ where perceived sacrifice exceeds perceived quality seems logic, the interpretation of the model of Dodds et al. (1991) leads to the third hypothesis:
HIII: A wide price gap between NB and PL has a positive direct effect on the market share of the NB.
Summarizing this section of the literature review, it can be concluded that a wide price gap between private label products and national brands is appointed as one of the drivers of private label share (Quelch & Harding, 1996). As a consequence of this diminishing the price gap is advised as a strategy in order to control the threat of private label. Nevertheless, empirical research also shows that a small price gap would not necessarily lead to a high market share of the store brand.
2.6. The influence of perceived quality on purchase
2.6.1. The influence of perceived quality on purchase intentions
are mainly the result of using different definitions of these concepts. They also conclude that, although the strength and exact effect of perceived quality on purchase intentions remains somewhat unclear, there is at least a positive relationship between perceived quality and purchase intention.
2.6.2. The influence of perceived quality on the purchase of store brands
As discussed previously, perceived quality does have an effect on the purchase of products. However, the strength of private label products compared to national brands was traditionally inferior product quality, compensated by very low prices. This is no longer the case, as a substantial amount of store brands actually emphasizes their high quality (Hoch & Banerji, 1993). Consequently, product quality is found to be an important driver of store brand success (Hoch, 1996, Steenkamp & Dekimpe, 1997).
Although improving, the quality of store brands is still perceived to be lower than the quality of national brands (Huang, 2009), even if the ingredients of the store brand product equals the national brand (Richardson, Dick & Jain , 1994). It is important to note that perceived quality is the most important factor for purchase intention. Even for store brand buyers the effect of value for money is not as strong as the effect of perceived quality (Richardson et al. 1994). This could be the effect of a stronger reliance on extrinsic cues as indicators for perceived quality compared to the reliance on price as a direct reason for purchase. A store brand strategy that is purely focused on price is unlikely to become successful (Richardson et al. 1994).
High perceived quality is especially important in convincing non store brand buyers to buy private label instead of national brands. This concept is named the conquesting power of the store brand. This should not be confused with the intrinsic loyalty of the brand, which refers to the amount of consumers which are already dedicated private label buyers. However, both factors could increase the market share. The correlation between perceived quality and conquesting power is high, while the distinction in perceived quality between high and low intrinsic loyalty is only weak, but positive (Steenkamp & Dekimpe, 1997). Private label products that score high on conquesting power (such as potato chips and coffee) have a significantly higher perceived quality then products that score low on this dimension (beer and margarine). It is interesting to note that this goes together with a small price gap between private label and national brands (+/- 12% compared to +/- 25% in categories that score low on conquesting power) (Steenkamp & Dekimpe, 1997).
strategy (Huang, 2009). However, it appears that the intended carry-over of perceived quality from the national brand to its private label competitor does not necessarily exist (Sayman, Hoch & Raju, 2002).
Because perceived quality is to be found to have a direct and an indirect positive effect on purchase intention and perceived quality is found to be especially important in convincing national brand buyers to switch to store brand products, the fourth hypothesis will be:
HIV: There is a positive relationship between perceived quality of the NB and the marketshare of the NB.
2.7. Hypotheses
After answering the sub questions, some main guidelines will appear, regarding the effect of pricing on the perceived quality and willingness to buy. However, in order to answer the central research question, empirical research, focused on the specific situation of a national brand and its private label competitor, has to be conducted. Based on the outcomes of the literature review, several hypotheses are constructed and empirically tested. The hypotheses tested in this research are:
HI: In consumer goods product categories with a wide price gap between NB and PL, the perceived
quality of the NB is higher.
HII(a): In consumer goods product categories with higher perceived quality differences between
products, the effect of the price gap between NB and PL on the perceived quality of the NB will be stronger.
HII(b): In consumer goods product categories perceived as hedonic, the relationship between the price
gap and the perceived quality of the NB is stronger.
HII(c): Store name has a moderating effect which diminishes the price-perceived quality relationship of
the national brand.
HIII: A wide price gap between NB and PL has a positive direct effect on the market share of the NB. HIV: There is a positive relationship between perceived quality of the NB and the marketshare of the NB. 2.8. Conceptual Model
The conceptual model is a simplified version of the ‘Conceptual relationship of price effect’ by Dodds, Monroe & Grewal (1991) and adapted to the situation of a NB and its PL competitor. Because of the narrow scope of this thesis and the availability of data, some variables had to be deleted from this model.
The conceptual model reflects two main effects, appointed by the dotted squares:
The effect of price the price gap on perceived quality of the NB.
The effect of price on market share, as well indirect through perceived quality, as the direct
effect.
HII(c): - HII(b): + HII(a): + HI: + HIV: + HIII: + HI: + HIV: + HIII:
perceived quality, where store name (HII(c)) is expected to have a diminishing effect on this relationship. Including all the hypotheses stated in the previous section, the conceptual model then becomes:
Figure 2.5. Conceptual Model
3. RESEARCH DESIGN
The research design in this thesis is aimed at obtaining managerial implications. It does so by testing specific hypotheses with a large and representitive sample. This research is therefore defined as conclusive, opposed to exploratory research (Malhotra, 2007). The hypotheses that are stated in the literature review will be tested by using data from the Autumn Report of GfK Panel Services. In this section the method with which the data is used to get relevant results, is discussed.
3.1. Data
3.1.1. Autumn Report
GfK Panel Services is an organization, specialized in market and consumer research. GfK publishes an annual summer and winter report, which represents an image research of Dutch retailers. In addition, the autumn report is published. This research focuses on the image of leading brands in the Netherlands (source: www.gfk.nl & www.gfk.com). The autumn report uses 2 types of data:
Continuous data from online registration of grocery product purchase
Additional data, collected with a questionnaire (appendix 7.1.)
3.1.2. Sample
The GfK autumn report uses a a sample called the ConsumerJury. In this report the ConsumerJury consists of 6000 respondents. This sample size matches the guidelines for conclusive research (Malhotra, 2007). Respondents of the ConsumerJury register their groceries online, on a continuous base. In order to create the autumn report, GfK combines this data with additional research regarding consumer attitudes to several leading brands.
Figure 3.1. Frequency of ratings for each brand
When all the duplicate respondents are excluded, this leads to a sample with N=5749. The sample used in this thesis has the same distribution as the total sample of the autumn report. An overview of the demographic variables of the sample is given in tabel 3.1. and figure 3.2.:
Table 3.1. Sample divided in age & gender Figure 3.2. Sample divided in districts
This is an example of a nonprobability sampling technique, where different quotas on relevant characteristics, such as demographics, are developed before the sample is drawn (Malhotra, 2007). In addition, the consumer types are measured, which is represented in table 3.2.
In order to create a representative sample, several selection methods are used. Respondents that were used in previous research are contacted repeatedly. New respondents are contacted online, with a written request or at markets. Hard to reach target groups are contacted with alternative selection methods, which means written as well as requests by telephone. There were no respondents deleted due to missing data (source: www.gfk.nl & www.gfk.com).
3.1.3. Aspects
The data used in this thesis comes from the Autumn Report of GfK Panel Services. In the original dataset the respondents were questioned about 50 leading brands and each respondent had to rate 5 brands; the sample is thus divided in 10 groups of approximately 600 respondents. The sum of all the data leads to an N of 29635, hence the N of the sample is 5927. However, the dataset used for this thesis consists of a selection of 22 brands, and consequently of fewer respondents. The total data has an N of 13036 and the sample consists of 5749 respondents. The respondents had to rate the importance of 13 aspects regarding, among other things, quality, price and social responsibility. Subsequently, 5 brands had to be rated on each of the components. The aspects that had to be rated are shown in appendix 7.2.
The respondents had to answer on a 10-item Likert Scale where 1=very bad and 10=very good. There
was also an 11th option: ‘I don’t know’. Because this 11th option is not part of the Likert Scale, this value
is treated like a missing value.
The 13 items are constructed in such a way that they measure the brand equity of the 50 brands. Brand equity is the consequence of three types of activities; the marketing activities of a company, the marketing activities of the competition and private label activities (Steenkamp & Dekimpe, see appendix 7.8.). Private label activities damage the brand equity of national brands by introducing copycat brands and attack national brands from different sides by introducing store brands in different price segments (Kumar & Steenkamp, 2007).
Steenkamp & Dekimpe (see appendix 7.8.) propose a conceptual model for measuring brand equity that consists of four aspects: differentiation, relevance, esteem and intimacy. The 13 items used in the Autumn Report are constructed in such a way that it measures two of these aspects, namely esteem (45% and intimacy (30%). The third aspect is brand innovation (30%), which can be compared to differentiation. The items divided per aspect can be found in appendix 7.2. In addition, the definition of
brand equity from Keller (2008) is used. Keller (2008) identifies three key drivers of brand equity;
Figure 3.2. Overview of the 13 aspects
3.1.4. Perceived Quality
As can be seen in figure 2.1. & 2.5., perceived quality is an essential concept in the conceptual model. However, this concept could not be reflected with one question. In order to discover whether the dimension of perceived quality is present in the 13 aspects, a factor analysis is conducted. The KMO and Bartlett’s test that shows whether factor analysis is appropriate can be found in appendix 7.4.1. Factor analysis is conducted on 13 items. When all the aspects are extracted based on eigenvalue 1, the solution includes only one factor appendix, which can be seen in appendix 7.4.4. In order to improve the interpretation of the factors, there was chosen to work with a fixed number of factors, also represented in appendix 7.4.4. Based on the factor solution of Altenburg (2008) using the GFK Autumn Report 2008, the number was set at 3 factors. As can be seen in appendix 7.4.2., where 0.70 is the lower limit, for non-exploratory research (Hair, 2006), the Cronbach’s Alpha of these aspects is still very high at .916. To improve the accuracy of the dimension ‘perceived quality’, the 4 aspects of perceived quality are then corrected with their importance. This is reflected in appendix 7.4.3.
Based on the conducted factor analysis, the dimension perceived quality is bases on the aspects good quality, reliability, value for money and positive differentiation. This slightly differs from the dimension used in the thesis of Altenburg (2008) where reliability, quality, positive feeling, value for money, fits me, and appeals to me, were included. The term ‘perceived quality’ is defined by Zeithaml (1988) as: ‘the consumer's judgment about a product's overall excellence or superiority.' However, this definition could lead to including all the 13 aspects. For example, Yoo et al. (2000) states that ‘high perceived quality
The amount in which ...
1. ... <brand name> offers good quality 2. ... <brand name> is reliable
3. ... <brand name> offers value for money 4. ... <brand name> recalls a positive feeling 5. ... <brand name> fits you
6. ... <brand name> is innovative (offers new products regularly) 7. ... the logo of <brand name> appeals to you
8. ... <brand name> appeals to you 9. ... <brand name> contributes to society 10. ... is clear where <brand name> stands for
11. ... <brand name> responds to the needs of the consumer
means the consumer’s recognition of the differentiation and superiority of the brand’. Based on this literature and reasoning, the construction of the new dimension with 4 aspects seems correct.
3.1.5. Chosen brands
The brands included in the autumn report are the 50 brands with the highest turnover in the Netherlands, chosen from the IRI brand top 100 (Koomen, 2009). Tobacco brands are excluded from the report and replaced by brands just outside the top 50 turnover brands. For this research, data should be at product level. However, the data available is at brand level. To overcome this limitation, brands are chosen that represent a clear and more or less homogenous product. This means that the brand Honig, which includes products in several categories like soup and bakery items, is excluded from this research, while Milner, a brand that represents the product ‘cheese’, is included in the selection of brands. The chosen brands are represented in appendix 7.3.
3.1.6. Chosen retailers
In order to test HI the size of the price gap between the different national brand products and the private label products should be made clear. For this purpose, product data of two retailers are compared. The chosen retailers are Albert Heijn and C1000. Both retailers have a well developed private label assortment, ‘Albert Heijn Huismerk’ for Albert Heijn and ‘Selekt’ for C1000. The price gap between A-brands and private label differs between these retailers. For 35 tested products, Albert Heijn has an average price gap of 37.9%, where C1000 has an even larger price gap of 48.2% (Smit, 2009, online). Congruent with the literature on price gaps between national brands and private label (Kumar & Steenkamp, 2007), the price gap of the chosen brands, differs significantly. The price gap can be divided in low (0≥<20%), medium (20≥<40%) and high (40%≥). Table 3.2. & 3.3. represents the overview of the different price gaps at Albert Heijn & C1000:
Table 3.2. Overview price gap between national brands and private label competitors at Albert Heijn
Small (0≥<20%) Medium (20≥<40%) Large (40%≥)
Toilet Paper (-6%) 45+ Cheese (23%) Tea (58%)
30+ Cheese (6%) Peas (24%) Beer (59-100%)
Chilled Orange Juice (9%) Diapers (25%) Fruit Mix Syrup (66%) Potato Chips (14%) Laundry Detergent Special (25%) Coke (104%)
Chocolate Milk (17%) Filter Coffee (26%) Laundry Detergent Basic (116%) Potato Chips Special (17%) Gold Melange Coffee (27%)
Table 3.3. Overview price gap between national brands and private label competitors at C1000
In addition, to test the moderator effect of store name on the relationship between price and perceived quality (HII(c)), the equity of both store names is compared. For this purpose, the variable brand power
consumer is used, which is further explained in section 3.2.4.
3.1.7. Price Gap between National Brand products and Private Label products
As can be seen in table 3.3. & 3.4., the price gap between national brand products and their private label competitor differs across retailers. In order to proof HI an evenly distributed selection of products with a different price gap is made as can be seen in table 3.4.
Table 3.4. Overview price gap Albert Heijn & C1000 combined
In order to test HI, the perceived quality across these three different price gaps is compared. To prevent bias, the price gap of Heineken was the only beer brand included, because the PQ was influenced significantly by including all the beer brands. Two variables were constructed that reflect the price gap between PL and NB products. One is the continuous variable price gap NB-PL that reflects the mean of price gaps between NB and PL at AH and C1000. In addition, the variable ‘price gap’ reflects the price gap between PL and NB as depicted in table 3.3., were 1=small, 2=medium and 3=large.
Small (0≥<20%) Medium (20≥<40%) Large (40%≥)
Chilled Orange Juice (8%) Margarine (20%) Chocolate Milk (51%)
Toilet Paper (14%) Cheese 30+ (20%) Beer (57-100%)
Potato Chips (14%) Peas (24% Fruit Mix Syrup (60%)
Breakfast Cake (19%) Multigrain Biscuit (25%) Tea (82%)
Margarine (18%) Filter Coffee (26%) Laundry Detergent Special (92%) Potato Chips Special (26%) Coke (162%)
Diapers (33%) Laundry Detergent Basic (233%) Cheese 45+ (36%)
Orange Juice Basic (39% Breakfast Cake (42%)
Small (0≥<20%) Medium (20≥<40%) Large (40%≥)
Toilet Paper Peas Beer
Potato Chips Filter Coffee Fruit Mix Syrup
Margarine Diapers Tea
Orange Juice Cheese Laundry Detergent
Breakfast Cake Coke
3.1.8. Perceived quality gap
In order to test HII(a) the products that are used in this research have to be divided in categories with a large quality gap between the products and categories with a small quality gap. The overview of these differences in the perceived quality gap is given in table 3.5.:
Table 3.5. Overview perceived quality variance within categories
3.1.9. Utilitarian vs. Hedonic products
In order to proof HII(b), the products included in this research should be divided in either a hedonic or functional good. Although some guidelines regarding the division of products in these two categories are given in the literature review, this classification is not objective (Lee & Hyman, 2008). When using the literature on this topic, as described previously, the classification of the products would be as follows:
Table 3.6. Classification of categories
Brand Mean Perceived Quality Perceived Quality Variance
Milner 4.58 Maaslander 4.55 Robijn 4.83 Ariel 4.63 Douwe Egberts 5.12 Nescafé 4.66 Grolsch 4.69 Bavaria 4.10 .031 .588 .455 .204 Functional Hedonic
Margarine Chocolate milk
Cheese Fruit Mix Syrup
Peas Beer
Washing Detergent Orange Juice
Diapers Potato chips
Potato Chips Special Breakfast cake Whole grain biscuit
Coke Filter Coffee
3.1.10. Market Share
In HIII & HIV the effect of the price gap between NB and PL and the effect of perceived quality of the NB on the market share of NB is investigated. Market share is a concept which consists of three different parts (source: GfK). The formula for computing market share is:
purchase share X consumer’s loyalty X purchase intensity = market share
However, for the 50 brands, only the purchase share is available. Purchase share will thus be treated as market share. Hence, the hypotheses test the effect of perceived quality and price gap on the percentage of households that purchase the product, not their loyalty to that product and the amount of money they spend on the product per purchase. Especially purchase share (penetration) and purchase intensity differ across categories (Dhar, Hoch & Kumar, 2001). These different category roles are represented in figure 3.4. Consequently, in interpreting the results, this limitation should be kept in mind.
Figure 3.3. Category Roles (Dhar et al. 2001)
3.2. Analysis
3.2.1. General Linear Model
As mentioned in chapter 2.8., the conceptual model focuses on two main effects. The model that represents the effect of the price gap between PL and NB on the perceived quality of the NB is complemented with three moderator variables. To analyse the effect of the independent variable on the dependent as well as the effect of the moderator variables, a General Linear Model, or GLM, is used. GLM is capable of analyzing several statistical models within one model estimation and provides the researcher with a flexible and simplistic model (Hair, 2006). The GLM is an example of a multiple regression analysis, which is discussed in detail in section 3.2.4.
High Penetration Low Penetration
Frequency of Purchase Percent of Households Buying
3.2.2. Compare Means
In order to test if several groups differ significantly from each other, the means of several variables are compared. The analyses used for this purpose are the Independent-Samples T Test and ANOVA. The Test compares the means of two variables, if it is interval or ratio data (Huizingh, 2006). Comparing means can lead to conclusions whether means do not or do differ significantly from each other, but does not give conclusions about the causes of these differences. In order to explain relationships between variables, regression analyses are conducted.
3.2.3. Single Regression Analysis
In this research several single regression analysis are conducted. With a single regression analysis the aim is to fit a linear relationship. In answering HI, HIII, and HIV the effect of the price gap on perceived quality, the effect of the price gap on market share and the effect perceived quality on market share will be fitted according this linear relationship. Table 3.7. represents an overview of dependent and independent variables, according to the conceptual model.
Table 3.7. Variables in regression analyses
3.2.4. Multiple Regression Analysis
In addition to the single regression analysis, HII(a), HII(b) and HII(c) focus on a moderator effect. A moderator effect occurs when a second independent variable, changes the shape of the relationship between the other independent and dependent variable (Hair, 2006). Including a moderator effect leads to conducting a multiple regression analysis (de Vries & Huisman, 2008). Moderator effects included in this research are perceived quality differences within a category, whether a category is perceived as hedonic or functional and store name.
HII(a) focuses on the influence of perceived quality variance within the category on the price-perceived quality relationship. This hypothesis is based on literature of Bettman (1974) and Sethuraman & Cole (1999) that states that the price-perceived quality increases when consumers perceive more quality differences within a category.
Hypothesis Independent Variable Dependent Variable
HI Price Gap Perceived Quality
HIII Perceived Quality Market Share
The data used in this research contains only a couple of products that fall within a category. In order to prove HII(a) four categories, which contains multiple brands, are used: cheese, chips, beer and washing detergents. The perceived quality of these brands is then compared, which is represented in table 3.5. Note that this method is based on the assumption that perceived quality variance between two national brands is representative for the perceived quality variance in the entire category, including private label. According to this figure, consumers perceive the strongest quality variance within beer and the least variance in the cheese category. These data are used to conduct analyses in two different ways. First, the dummy variable ‘perceived quality variance’ is constructed, where 0 = low perceived quality variance (cheese and detergent) and 1 = high perceived quality variance (coffee and beer).
For Nescafé, only the price gap between National Brands a Private Label at Albert Heijn is available, so the price at Albert Heijn is also treated as the price for Albert Heijn and C1000 combined. Therefore, the
regression analyses are performed with ‘Price Gap’ at Albert Heijn as the independent variable. In addition, four regression analyses are conducted on the quality differences as depicted in table 3.5.,
and the standardized coefficients are compared.
HII(b) includes the classification whether a product is perceived as hedonic or utilitarian as a moderator variable between price and perceived quality. Based on the classification as depicted in table 3.6., a dummy variable was constructed, with utilitarian = 0 and hedonic = 1. The multiple regression than includes the Price Gap of AH and C1000 combined, the dummy variable and the interaction effect between these two as independent variables and perceived quality as the dependent.
The inclusion of store name as a moderator variable and the associated hypothesis HII(c) is based on the article of Chang and Wildt (1996) that states that the price-perceived quality relationship is diminished when other intrinsic cues beside price are available. To test the moderator effect, the variable brand power consumer two retailers is compared. Brand power consumer is the sum of the weighted variables brand equity (75%) and brand innovation (25%) and represents the rating of the retailer on all the 13 items. The impact of the store or brand name is assumingly not caused by the perceived quality of the retailer alone, but is the result of all the 13 items from the autumn report. Therefore, the decision is made to include brand power consumer, so all the 13 items are indirectly represented in the judgment of the retailer. The included store brands are Select (C1000 store brand) and AH store brand. These store brands differ significantly on brand power consumer, as can be seen in the SPSS output in appendix 7.5.
