The impact of brand equity on the possession and usage of
mobile shopping apps;
and the behavioral & attitudinal factors affecting this
relationship
Nikki Tsiftis (10591370) Supervisor: Prof. Umut Konus Master Thesis Business Administration – Digital BusinessStatement of Originality
This document is written by Nikki Tsiftis who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are 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.Acknowledgement
Writing this master thesis was the This master thesis is the final step towards achieving my master’s degree in Business Administration. I would like to thank a number of supporters, who made this thesis possible in its final form. First of all, I would like to thank my supervisor dr. Umut Konuș for all his support, his very useful feedback and his time to discuss all my ideas. It was a great pleasure working together and therefore I really enjoyed the whole process of writing my master thesis. Furthermore, I would like to thank my family and friends, and in particular my mother and sister, for always believing in me and providng me support.Abstract
Mobile shopping has become an increasingly important topic among marketers and scholar as mobile commerce is rapidly growing. However, still more knowledge is needed on the potential drivers of mobile shopping app adoption, i.e. possession and use. The adoption of a firm’s mobile shopping app leads not only to extra mobile sales, it also benefits consumer’s spending behavior and it inreases in-store visits. The importance of brand equity has been widely recognized, though the impact of brand equity on the adoption of mobile shopping apps is not known. lThe relative impact of brand equity on possession and use of mobile shopping apps is investigated by comparing it to other attitudinal and behavioral factors, e.g. demographics, psychographics, privacy/security concerns and digital experience. This study also examines the moderating effects of product category, digital experience and privacy/security concerns. In this study, 247 respondents were surveyed by an online questionnaire. Binary logistic regression was used to test the hypotheses. The results show that among all factors, brand equity is the most important predictor of possession within the groceries category and use within the clothing category and groceries category. Age has the biggest impact on use wthin the groceries category and shopping enjoyment on possession within the clothing category. Findings of this study enables the manager to target its marketing communication better by focusing on the most influential factors of mobile shopping app adoption within their product category.Table of Contents
1. Introduction ... 1 2. Literature review ... 5 2.1 The rise of mobile commerce ... 5 2.2 Mobile Shopping Applications ... 7 2.3 Possible Factors Affecting Mobile Shopping Application Usage ... 8 2.4 Brand Equity and Its Impact On Mobile Shopping App Possession/Usage ... 13 2.4.1. Customer Based Brand Equity ... 13 2.5 Contributions ... 20 2.6 Research Gap and Research Question ... 23 3. Conceptual Framework ... 24 3.1 Brand equity ... 25 3.2 Moderators and Control variables ... 26 4. Research design ... 32 4.1 Population Sample ... 32 4.2 Measures ... 33 4.3 Analyses ... 37 5. Results ... 41 5.1 Reliability ... 42 5.2 Clothing ... 43 5.3 Consumer Electronics ... 47 5.4 Flight Tickets ... 49 5.5 Groceries ... 53 5.6 Additional Analysis ... 56 6. Discussion & Conclusion ... 57 6.1 Discussion ... 57 6.2 Conclusion ... 63 6.3 Managerial implications ... 65 6.4 Limitations / Future Research ... 65 5. References ... 67 Appendix A – Tables and Figures ... 72 Appendix B – Survey ... 781. Introduction
The number of smartphone users is growing every year. In 2017, 2.39 billion people used a smartphone, which is an increase of 10.8% and represents 32.3% of the global population (eMarketer, 2017). The worldwide smartphone penetration will continue to increase, and by the end of 2018, it is expected that more than a third of the global population will be using a smartphone. Consequently, Mobile shopping (M-shopping) has become an increasingly important topic that has gained a lot of attention from both practitioners and researchers. From 2013 to 2018, mobile commerce doubled from 7 to 17% and it is the fastest growing shopping channel (PwC, 2018). As the number of smartphone users is increasing continuously, so is the demand for and growth of mobile phone applications (Liu, Zhao, Li, 2017). Despite the tremendous growth of mobile apps, little research has been devoted to understanding drivers of mobile shopping behaviour in apps. Mobile shopping applications (mobile shopping apps) are software programs developed for mobile devices such as smartphones and tablet in which purchases can be made. From practitioner view, it is important to know its drivers, because mobile shopping app usage has a lot of benefits for firms. Prior research shows that people buy more frequently (order rate increases) and spend more (order size increases) when adopting mobile shopping (Wang, Malthouse & Krishnamurthi, 2015). Research also shows that the customer lifetime value is higher for multichannel shoppers (Kumar and Venkatesan, 2005; Neslin and Shankar, 2009). This is the case, because the technology of mobile devices provides convenient access which gives customers the opportunity to incorporate M-shopping into their habitual routines (Wang et al., 2015). Therefore,particularly mobile orders consist of habitual products, which are products that customers are already familiar with and have a history of purchasing (Wang et al., 2015). Kim, Kim, Choi & Trivedi (2017) researched some drivers of mobile app possession and mobile app usage. They found that online experience, i.e. experience accumulated through online shopping, and mobile experience, i.e. experience through smartphone usage, both positively relate to the possession of shopping apps. Digital experience and browsing patterns of shopping apps predicted mobile purchases through shopping apps. Also factors like personal innovativeness and risk perception tend to influence mobile shopping app adoption (Thamaraiselvan et al., 2017). This study extends previous research on mobile app usage by investigating potential consumer-related drivers instead of app-related factors. Brand equity (perceptions) could be an important driver of mobile shopping app adoption. Keller defines customer based brand equity as ‘’the differential effect that brand knowledge has on customer response to the marketing activity of the brand” (1993). High equity brands possess high levels of brand awareness and strong, favourable and unique associations. So it would make sense that a consumer who has favourable perceptions about a brand, is more likely to download that branded app and makes more in-app purchases. Prior research already shows the strong positive relationship between certain dimensions of brand equity and both attitudinal loyalty and behavioural loyalty (Chaudhuri & Holbrook, 2001; (Netemeyer et al. 2004). Since brands that possess high brand equity have more loyal customers, it could be that they show behavioural loyalty in downloading the brand’s app and use it more frequently. This study extends previous studies by introducing brand equity as a potential driver of mobile app possession and usage. It links brand equity to mobile shopping adoption for the first time and therefore bridges the gap. For example, little is known so far about the
effect of certain brand associations on downloading a mobile shopping app and making purchases through the app. From business practice, there are some examples that show the impact of brand equity on customers’ use of mobile apps for shopping. For example, Nike is ranked 18th in the Best Global Brands Rankings 2017 with a brand value of $27,021 million (Interbrand, 2018). Nike’s mobile shopping app ‘Nike SNKRS’ is in the top 10 of most downloaded mobile shopping apps in the United States (SimilarWeb, 2018). The second most downloaded app in the United States is Amazon (SimilarWeb, 2018). Amazon also has well-established brand equity with a total brand value of $64,796 million, being ranked 5th in the Best Global Brands Ranking 2017 (Interbrand, 2018). This could imply that brand equity is a critical factor and certainly has an impact on customers’ use of mobile app for shopping. In order to investigate the relative importance of brand equity as a driver, other (consumer-related) factors as drivers of mobile shopping app adoption are taken into account as well, namely demographic characteristics (age and gender), digital experience (online orientation and mobile orientation), privacy/security concerns and psychographics (shopping enjoyment and technology innovativeness). Furthermore, mobile experience, online experience, privacy/security concerns and product type are investigated as possible moderators affecting the relationship between brand equity and mobile shopping app possession/usage. So the aim of this research is to investigate the impact of brand equity on mobile shopping app adoption. For several reasons this topic is interesting and relevant for managers. First, when people adopt a mobile shopping app, it doesn’t only impact the amount of purchases through shopping apps. In fact, it positively influences the purchase frequencies in all channels, e.g. store, online and mobile. So if managers are able to convince
channels. The reasoning behind this is that the installed mobile shopping apps on the mobile screen functions as a ‘reminder’ for the customer, since the brand will be in their consideration set. After adopting mobile shopping, consumers place a higher number of orders per year (order rate increases) and especially low-spending consumers spend more money per purchase (order size increases) (Wang, 2015). So adopting mobile shopping benefits consumer’s spending behaviour. Furthermore, cross-channel effects could happen (Heerde, 2014), in-store visits could rise as a result mobile shopping. Therefore, it is important for managers to know how to stimulate mobile shopping app adoption. When a relationship between brand equity and mobile app possession and usage exists, managers could benefit from it in several ways. First of all, companies can build brand equity themselves which makes it practically relevant for businesses and managers. For example, they could start with building greater brand awareness; they could target markets strategically to deepen customer connection; they could aim to achieve positive customer feelings and judgements about their brand; they could try to increase loyalty by getting customers more involved via social media, marketing events and remarketing campaigns. Secondly, managers can target the brand’s marketing communication better by focusing on the most influential factors of mobile shopping app adoption. Furthermore, it shows in which situations brand equity influences the adoption of mobile shopping apps. For example, if brand equity does really matter for older or more innovative groups, but not for others, it would help managers in tailoring their marketing and communication strategy in terms of spreading their mobile strategy. Should a manager put more brand focus on its communication with young people when it comes to creating more users for the mobile shopping app? Or should the focus be on people with less previous mobile experience in
order to increase app usage? So managers could use this research as a tool to optimize their mobile app adoption strategy and consequently to be more profitable.
2. Literature review
In this chapter, an overview of the relevant literature is given by elaborating upon the key concepts of this research. First of all, the rise of mobile commerce is briefly discussed to show its relevancy. After that, the concept of mobile shopping applications will be discussed by looking at the different types of shopping apps and how it is relevant for online/mobile marketing. Next, possible factors affecting mobile shopping app usage will be examined like demographic variables, psychographic variables, digital experience and product category. Consequently, attention will be paid to the under-investigated role of brand equity as a driver of mobile shopping app possession and usage. The concept of brand equity will be explained and its dimensions according to different scholar, followed by reasoning why it should be an important driver in mobile app possession and usage. Finally, the chapter ends with the theoretical and managerial contributions based on the theoretical framework, and with the research gap and research question of this study. 2.1 The rise of mobile commerce Since the introduction of third generation (3G) mobile communication technologies, the development of mobile commerce has been triggered. In 2017, 2.39 billion people used a smartphone and the expectation is that around one third of the global population will be using a smartphone by the end of 2018 (eMarketer, 2017). Accordingly, Mobile shopping (M-shopping) has become an increasingly important topic and has gained a lot of attention fromFig. 2 The rise of mobile commerce (PwC, 2018) Figure 2 shows that mobile commerce doubled from 7% to 17% in the last 6 years and it is growing the fastest from all other shopping channels. Tablet buying also increased from 8% to 12%. Purchases by the personal computer (PC) went down from 27% to 20% and is likely to be surpassed by mobile commerce soon (PwC, 2018). In 2018, 52.2% of worldwide online traffic comes from mobile devices, compared to 50.3% in 2017 (Statista, 2018). Surprising is the rise of in-store purchases, which is slightly increasing since 2015 (figure 2). One of the reasons of physical stores’ continued popularity might be that customers can order online and pick it up in the store (PwC, 2018), which shows the importance of online/mobile channels again in stimulating offline channel. Smartphones are also frequently used as a payment channel at bricks-and-mortar stores, via customized orders in advance, in-store apps or a mobile payment platform at checkout (PwC, 2018).
2.2 Mobile Shopping Applications As already mentioned, the mobile channel is increasingly becoming more important for retailers. Not only as a channel to facilitate purchases at the website, but also as a purchase/payment channel via mobile applications (GFK, 2015). This research is focused on a specific part of mobile commerce, namely purchases via mobile shopping applications. So online purchases made on mobile devices are not being considered. Little research has been done yet in the field of mobile shopping applications which makes it extra interesting to consider. A mobile shopping application (mobile shopping app) can be defined as a software program developed for mobile devices such as smartphones and tablets in which purchases can be made. Different types of mobile shopping apps exist which have their own purpose and have different roles/functions. A mobile shopping app could serve the pre-purchase stage, i.e. facilitate search, and give the customer information about different products or services for example. Secondly, a mobile shopping app could serve the purchase phase, i.e. facilitate purchases, which makes it possible for customers to buy products/services through the app. Thirdly, a mobile shopping app could serve the after-purchase phase, i.e. facilitate after-sales service, so it provides the customer with personalized and tailored support, increases engagement and follows up with relevant and tailored offers. The main role of the latter app type is to increase loyalty and build relationships. The advantage of mobile shopping apps is that is allows customers to engage with a retailer through wireless sessions under all types of temporal or spatial situations (Shankar et al. 2010). As mentioned earlier, mobile shopping usage has a lot of benefits for companies. Therefore, it is important to look at the factors predicting mobile app possession and usage in order to enhance the amount of purchases via the mobile channel.
Other research investigating predictors of mobile shopping behavior or adoption can be found in table 1. Some research is focused on app characteristics like Kim & Baek (2018) and Kalinic & Marinkovic (2016), so on the technological/user-context. Yang (2012) found that perceived usefulness and perceived enjoyment predict attitude toward mobile shopping. Attitude towards mobile shopping significantly predicted intention to use mobile shopping. Table 1. Prior research overview of mobile (shopping) app adoption Authors (year) Empirical/
theoretical Drivers Antecedents Mobile adoption
Kim et al. (2017) Empirical Online experience Mobile experience Browsing behaviour (non)-shopping apps Shopping app possession Mobile app purchases Kim, S., & Baek,
T.H. (2018) Empirical Time convenience Interactivity Compatibility Informational and experiential apps Mobile app engagement Thamaraiselvan, N. Natarajan, S., Balasubramanian, D. & Kasilingam, L. (2017) Empirical Perceived risk
Personal innovativeness Gender Experience Frequency of using mobile shopping apps Intention to use mobile shopping applications
Yang (2012) Empirical Perceived usefulness
Perceived enjoyment Attitude towards mobile shopping
This research Empirical Brand equity Age Gender Shopping enjoyment Technology innovativeness Privacy/security concerns Online experience Mobile experience Mobile experience Privacy/security concerns Product type Mobile shopping app possession Mobile shopping app usage
2.3 Possible Factors Affecting Mobile Shopping Application Usage Previous research on mobile commerce, and in particular on mobile applications, found certain determinants of mobile app adoption. Different outcome variables are used such as mobile app engagement, purchases through apps, intention to use the mobile app etc. Also barriers to mobile shopping app adoption are discussed. 2.3.1 Online Experience and Mobile Experience The study of Kim et al. (2017) found certain predictors of mobile shopping app possession and mobile purchases. First of all, online experience (total experience in online shopping) and mobile experience (experience through smartphone usage), both referred to as digital experience, predict the possession of mobile shopping apps. Since experienced online shoppers are more familiar with the online shopping environment, barriers to adopt mobile shopping apps are reduced. That applies because the online environment is quite similar to the app environment (Bang et al. 2013). Experienced mobile shoppers are exposed to a greater number of apps which reduces barriers to download apps (Kim et al., 2017). Secondly, digital experience and browsing information from shopping apps predict purchasing decisions. So they found that more experienced online shoppers and smartphone users use greater number of shopping apps when making mobile purchases. 2.3.2 Psychographics Technology innovativeness Innovative consumers tend to adopt a new technology in a relatively early stage far before average others (Aldas-Manzano, Ruiz-Mafe & Sanz-Blas, 2008; Agarwal & Prasad, 1998). Lu (2014) found that innovative people tend to have more positive perceptions of innovations and more positive intentions to using new technologies. Multichannel shoppers tend to be
more innovative as well and explore new channel initiatives (Konus, Verhoef & Neslin, 2008). Prior research also indicates that consumers with higher technology innovativeness are more likely to adopt mobile shopping (Yang 2012). Therefore, this psychological trait is likely to have a positive impact on the adoption of mobile shopping apps in this study as well. Shopping enjoyment Shopping enjoyment can be defined as the tendency of a consumer to derive pleasure and fun from shopping (Bruner, 2009). Customers who enjoy shopping are more likely to be multichannel shoppers, since they really enjoy shopping and therefore like to try new things in purchase stage and get new ideas anywhere and anytime during the pre-purchase stage (Konus, Verhoef & Neslin, 2008). So it might be that they are more prone to installing new mobile shopping apps on their smartphone and to use it more. 2.3.3 Privacy/Security Concerns Privacy and security concerns is another factor that negatively influences mobile shopping app adoption, so it is considered a barrier. Compared to offline and online commerce, mobile commerce has to deal with more privacy/security concerns. Like for example mobile devices may be infected by viruses and Trojan horses. M-shoppers may also wonder about their payment information, such as credit card accounts and passwords, which are maybe not always safely transmitted and stored (San-Martin & López-Catalán, 2013). Another concern could be location privacy, as the mobile vendors may disclose location information without the consumer knowing it (Mamonov & Benbunan-Fich, 2015).
2.3.4 Demographics Age In general, older people tend to have a lower understanding of new technologies (Williams & Page, 2011), and therefore will be less likely to adopt mobile shopping from apps. Therefore, we assume that when age increases, chances are lower that a mobile shopping app will be installed on a mobile device. However, older people tend to be more loyal to their favourite brand and are less into discovering new brands (McLeod, 2009). So when older people already possess a branded mobile shopping app, it might be that they are more loyal to that app and buy more in comparison to younger people. Gender Men tend to buy more quickly compared to women, in order to not spend a lot of time shopping, so they are quick shoppers (Hansen & Jensen, 2009). Men have less pleasure during shopping than women, whereas women have more patience during the shopping process (Underhill, 2010). Therefore, the assumption is made that male shoppers more strongly hold on to one particular brand, since its more convenient and saves time. So it might be they are on average more likely to download a mobile shopping app than women. When they have an app of a certain brand on their smartphone, for example to buy clothes, it is assumed they will make more purchases in that app than women as well, since they are less researching available other options and spend less time browsing compared to women (Cleaver, 2004). Therefore, we assume that men are more likely to adopt a mobile shopping app than women.
2.3.5. Product category In order to make the study more representative, four different product categories are taken into account: tickets, consumer electronics, clothing and groceries. Prior research shows that different categories can lead to different consumer behaviours (Konuş, Verhoef, & Neslin, 2008). So they all might have a different impact on the adoption of mobile shopping apps. Wang et al. (2015) found that M-shoppers tend to buy products that they have purchased before or already familiair with, e.g. habitual products, rather than items or brands that require research, planning or consideration, e.g. non-habitual products. Due to the small screens of smartphones, searching takes more effort which results in less types and amount of information being retrieved by customers (Shankar et al. 2010; Sweeney & Crestani, 2006), so it negatively affects their cognitive abilities in finding or recalling web information (Wang et al., 2015). The high search costs result in customers buying products that they are already familiar with, rather than products/brands that require more research or consideration (Wang et al., 2015). Therefore, customers prefer to use laptops/PC’s for exploratory search behaviors and use mobile devices for specific tasks (Adipat, Zhang & Zhou, 2011; Ghose, Goldfarb & Han, 2013; Sweeney & Crestani, 2006). Furhermore, the four product categories can be distinguished as offering intangible (tickets) and tangible products (consumer electronics, clothing and groceries). With intangible products or low-touch products, customers don’t need to touch it and can shop faster, they prefer shopping online (Lynch, Kent, & Srinivasan, 2001). Contrary, high-touch products or tangible products consumers want to touch the product before purchasing (Konuş et al., 2008). It could mean that for intangible products, mobile channel is more used and therefore leads to higher possession and usage. However, the focus in this study will be on investigating the difference between habitual and non-habitual products.
2.4 Brand Equity and Its Impact On Mobile Shopping App Possession/Usage 2.4.1. Customer Based Brand Equity Brand equity is studied through various perspectives, from the perspective of the consumer (Aaker, 1991, 1996; Keller, 1993) and from the financial perspective (Shankar, Azar & Fuller, 2008). In this study, the focus is on customer-based brand equity, so theories explaining brand equity from the perspective of the consumer will be discussed. One of those theories is Keller’s well-known Customer Based Brand Equity (CBBE) model (1993). Theories explaining brand equity from the customer perspective will be explained in the following paragraphs. Brand Equity as a construct has many forms and definitions (Yoo & Donthu, 2001). However, some concepts are continuously used by scholar. Table 1 gives an overview of frequently used dimensions of Brand Equity defined by different authors. In the article of Aaker (1991), Brand Equity consists of 5 dimensions: brand awareness, brand associations, perceived quality, brand loyalty and other proprietary brand assets. Keller defined Customer-Based Brand Equity as “the differential effect that brand knowledge has on customer response to the marketing activity of the brand” (1993). Contrary to Aaker’s 5 dimensions, Keller defined Brand Equity in terms of the consumer’s brand knowledge and is simply broken down into two components; brand awareness and brand image/associations. Rust, Zeithaml and Lemon (2004, p. 112) define Brand Equity as “the sum of customers’ subjective assessments of a brand’s intangible qualities”. They used brand awareness, brand image and brand ethics as their dimensions of Brand Equity. Kim & Kim (2004) and Yoo, Donthu & Lee (2000) used the same dimensions as Aaker (1991), namely brand awareness, brand image, brand loyalty and perceived quality. Lastly, Shocker & Weitz (1988) looked at
Table 2: dimensions of Brand Equity Dimension Aaker (1991) Keller (1993) Rust et al. (2004) Kim & Kim (2004) Yoo et al. (2000) Shocker &Weitz (1988) Brand image/associations X X X X X X Brand awareness X X X X X Brand loyalty X X X X Perceived quality X X X Brand ethics X Other brand assets X Brand Equity is considered to be a multidimensional concept in most studies. When combining all dimensions shown in table 1, four dimensions of brand equity are collectively used: brand image/associations, brand awareness, brand loyalty and perceived quality. These 4 dimensions represent Aaker’s model from his 1991 article. So these 4 dimensions will be used in this study as well to measure Brand Equity from the customer perspective. Aaker’s model has been empirically tested in a lot of studies (Yoo et al., 2000; Kim & Kim, 2004; Eagle & Kitchen, 2000). Important to mention is that all the brand equity dimensions have causal relationships (Aaker, 1991). The brand resonance model of Keller (2001) builds upon his highly accredited CBBE-model of his 1993 article. The focus is on the key dimensions of brand knowledge and their effect on relationships between consumers and brands (Keller, 2016). The brand resonance model defines 4 branding stages or steps, each stage having its own branding objective, which figure 1 shows. The model puts emphasis on the duality of brands, the rational route
to brand building on the left and the emotional route to brand building on the right side of the pyramid. Six brand building blocks – salience, imagery, performance, feelings, judgments and resonance - need to be established with consumers in order to pass the 4 steps. The next paragraphs relate the 4 chosen variables of Aaker’s model (1991) to the 4 branding levels as proposed by Keller (2008). The following paragraphs explain the 4 chosen dimensions of brand equity of Aaker’s model (1991) and relate it to Keller’s brand resonance model (2008). Fig. 1 Brand resonance model (Keller, 2016) Brand Awareness The first aspect of brand equity is brand awareness. Brand awareness is “the ability for a buyer to recognize or recall that a brand is a member of a certain product category (Aaker, 1991, p. 61). From the definition follows that brand awareness consists of both brand
in the brand resonance model of Keller (2008), which is the first branding level. Brand salience is defined in terms of breadth and depth of brand awareness. It is about the extent to which consumers think about your brand, often (breath) and easily (depth). And whether it is at the right times, at the right place and in the right ways. Brand salience needs to exist in order to go to the next stage in the pyramid (Keller, 2008). Brand Associations The second aspect of Brand Equity is brand associations. Brand associations can be defined as “anything linked in memory to a brand” and brand image as “a set of associations, usually in some meaningful way” (Aaker, 1991, p. 109). These associations can be identified by the level of strength (Aaker, 1991; Keller, 1993). The link to a brand will be stronger when it is based on many experiences or exposures than when it is based on a few (Aaker, 1991). In the brand resonance model of Keller (2001), brand associations are part of the second branding stage, which are composed of the two brand building blocks Brand performance and Brand imagery. Brand performance is about the intrinsic properties of the brand and refers to the extent to which a product or service performs well in the opinion of the consumer. It can be increased by delivering a product or service that meets consumer needs. Brand imagery means how people think about your brand in terms of values and meaning, which is about the extrinsic properties of the brand (Keller, 2008). Extrinsic properties of the brand tell something about how consumers think about a brand. An increase in brand imagery can be established directly through experiences with the brand or indirectly through advertisements (Keller, 2008).
Perceived Quality The third aspect of Brand Equity is the perceived quality. Perceived quality is “the consumer’s judgment about a product’s overall excellence or superiority” (Zeithaml, 1988, p. 3). Therefore, it can be said that perceived quality is based upon the subjective evaluations of consumers. Perceived quality is seen as one of the components of Brand judgments in Keller’s brand resonance model Keller, 2008). It is part of the third branding stage which is composed of the two brand building blocks Brand judgments and Brand feelings. By building positive and unique brand associations, consumers form opinions about and develop feelings for the brand. The brand is being evaluated and the consumer forms a certain attitude or opinion. Brand judgments are on the rational side of the pyramid and are about the consumer’s opinions about the brand, which are composed of perceived quality, credibility, consideration and superiority. Brand feelings are on the emotional side of the pyramid and are about the consumer’s emotional reaction towards a brand, which are composed of warmth, fun, excitement, security, social approval and self-respect (Keller, 2008). Brand Loyalty The fourth aspect -and most important- of Brand Equity is brand loyalty. Aaker (1991, p. 39) defines brand loyalty as “the attachment that a customer has to a brand.” Consequently, it is about how likely a consumer is to switch to another brand when a change is made in the price or product features (Aaker, 1991). Contrary, other research focuses on the behavioural aspects of brand loyalty (Guadagni & Little, 1983; Gupla, 1988). Often brand loyalty is the core of a brand’s equity that is sufficiently different from the other dimensions (Aaker,
building block Brand resonance, which also consists of emotional attachment, being part of a brand community and active brand engagement (Keller, 2008). In this final branding stage, consumers form (long-term) relationships with the brand. The consumer identifies itself with the values of the brand and is willing to invest in a relationship. This results in for example repurchases, the consumer being less influenced by price reductions of the competitor and consumers being more willing to pay a higher price. Brand Equity – Adoption of mobile shopping apps According to Keller (2001), brand resonance is a combination of attachment/intensity and activity. It is linked to physical activity (consumer behavior) and it is linked to mental activity (consumer mindset). So brand resonance is important to activate consumers and increase purchases. Therefore, it is reasonable to assume that brand resonance will lead to more in-app purchases as well, since it activates purchases. This is strengthened by using behavioural reasoning theory. Brand equity perceptions are important on the adoption of mobile shopping apps for several reasons. Yang (2012) found that perceived usefulness and perceived enjoyment predict attitude toward mobile shopping. Attitude toward mobile shopping significantly predicted intention to use mobile shopping. According to theory of reasoned action, attitude is a combination of communication about attributes and advantages (Jalilvand & Samiei, 2012). Theory of reasoned action assumes that brand attitude is affected by brand awareness and brand image (Jalilvand & Samiei, 2012). In other words, brand equity, which exists of brand awareness and brand image, is related with brand attitude. Both constructs can be positive or negative. So a brand that possesses high brand equity is likely to have consumers with a more positive attitude towards the brand, whereas a brand that possesses
low brand equity is likely to have consumers with a more negative attitude towards the brand. Furthermore, prior research indicated that consumer attitude towards a brand has a strong effect on purchase intention as attitude is a suitable determinant of purchase intention (Abzari, Ghassemi, Vosta, 2014; Laroche, Kim, Zhou; 1996). Behavioral intention often serves as a mediating variable between attitude and real behavior (Abzari, Ghassemi, Vosta, 2014). According to Keller’s definition of brand equity, consumers who possess high brand equity are more likely to react in a favourable manner to the brand. Strong loyalty is one of the important characteristics that belongs to brands with strong brand equity (Keller, 1993). Prior research shows the strong positive relationship between certain dimensions of brand equity and both attitudinal loyalty and behavioural loyalty (Chaudhuri & Holbrook, 2001; (Netemeyer et al. 2004). Since brands that possess high brand equity have more loyal customers, it could be that they show behavioural loyalty in downloading the brand’s app and use it more frequently. Since attitude drives intention to use, and intention to use says something about the actual use of an information system according to TAM-model, brand equity could be the driver of mobile shopping app adoption.
2.5 Contributions 2.5.1 Theoretical Contribution This research contributes to the existing literature and investigates the research gap mentioned in table 1. First of all, it is unique in combining the core brand equity components with different mobile shopping stages (e.g. possession and usage). A few researchers looked already at the relationships between attributes of mobile value-added services and a firm’s brand equity (Wang & Li, 2012). They found that personalization, identifiability and perceived enjoyment positively influence the brand equity factors brand awareness, brand associations, perceived quality and brand loyalty. Our research looks at it the other way around and investigates the potential impact of brand equity on the adoption of mobile shopping apps (e.g. possession and usage). Furthermore, it compares the relative influence of brand equity on the adoption of mobile shoppings to other factors. Also, brand equity as a potential driver of mobile shopping app usage has not been researched yet. Previous research on the adoption of mobile shopping is either focused on app characteristics (Kim & Baek, 2018) or on characteristics of the app users (Kim et al., 2017; Thamaraiselvan et al., 2017. This research looks at the impact of demographic- behavorial and attitudinal factors on the adoption of mobile shopping apps. On top of that, we look at the influence of every factor per product category, so value will be added to previous findings. It also takes into account the moderating effects of digital experience and privacy/security concerns on the relationship between brand equity and mobile shopping app adoption. It will add additional value to previous literature as well.
2.5.2 Managerial Contribution Since mobile shopping app adoption has a lot of benefits for firms, such as increased order rate and order size, enhanced Customer Lifetime Value for M-shoppers and increased store visits, it is important for managers to know its drivers. This study investigates whether brand equity has a significant impact on mobile shopping app possession and usage in comparison to other drivers. It is expected that when a brand has established brand equity, it is more likely that its mobile shopping app is adopted. However, this research offers many other contributions for managers. First of all, we investigate the relative importance of brand equity as a driver of mobile shopping app possession/usage. So the relative importance of the impact of brand equity on mobile shopping app usage is investigated with regard to other factors. This enables the manager to support the more influential factors and make a marketing communication on them. This will lead to an optimized mobile app adoption strategy and therefore increased purchases through the apps, and sales in the long-term. Secondly, we investigate in which situations brand equity influences the adoption of mobile shopping apps. It might not always be the case that this relationship exists. It could be that for people with privacy and security concerns, the impact of brand equity on mobile shopping app adoption is less strong. It could also be that for people with less mobile experience, it might be more important to have established high brand equity before downloading and using the mobile shopping app. They may need additional reasons like established trust before they decide to download and use the app. Furthermore, the study shows for which product categories brand equity is particularly important as a driver of possession or usage.
Thirdly, brand equity can be altered by the company itself which makes it practically relevant. When a brand does not possess high brand equity, then managers could decide to implement a strategy to start building brand equity in order to increase their mobile shopping app usage and experience the benefits. For example, they could start with building greater brand awareness; they could target markets strategically to deepen customer connection; they could aim to achieve positive customer feelings and judgements about their brand; they could try to increase loyalty by getting customers more involved via social media, marketing events and remarketing campaigns. Fourthly, next to brand equity, maangers can act upon the other factors which have a significant influence within their product category. If consumers with mobile experience or online experience are more likely to download or use a brand’s mobile shopping app, managers could focus their marketing communication campaigns on them. Furthermore, apps could be designed in such a way, that they are easy to navigate and use, so those consumers with less mobile- and online experience will not get lost while using the app and will not delete the app. When privacy/security concerns of consumers leads to lower chances of app downloads or lower chances of the app being used at least once, managers could try to reduce those risks by honestly displaying wat happens with the consumer’s data. There are several ways of increasing the transparency for customers which might reduces their concerns and in turn increases app downloads or use. When older people or men are more likely to adopt a mobile shopping app, managers could adjust their assortiment within the app towards their needs or adjust their marketing communication.
2.6 Research Gap and Research Question A lot of research has been done in the field of branding, and in particular brand equity. Contrary, research in the field of mobile applications is more limited. Little is known so far about the drivers of mobile app possession and mobile app usage, especially in the context of Branding. Since we know that brand equity impacts purchase intention, it would be interesting to know whether brand equity has an impact on mobile shopping app adoption as well. It might be that a relationship between Customer Based Brand Equity and mobile app possession and usage exists. It would make sense that a person scoring high on brand equity for a certain brand, is more likely to have installed the shopping app of that brand on its smartphone and to use the app (making purchases in the application). Therefore, the research questions that will be investigated in this study are the following: Main question: What is the impact of brand equity on the possession/usage of mobile shopping apps in comparison to other drivers? Sub question: How (whether) brand equity interact (as moderators) with other factors: demographic, attitudinal and behavioral variables – in driving customers’ possession and use of mobile shopping apps.
(H1)+
(H2c)- (H3b)- (H4)+ (H5a)- (H5b)+ (H6)+ (H7)+ (H8)+ (H2a)+ (H3a)- (H2d)- (H2b)+3. Conceptual Framework
In this chapter, the conceptual framework is presented in which the hypothesized relationships are visualized. Furthermore, the main reasons behind every hypothesis are briefly explained. Fig. 2 Conceptual Framework*Variable not analyzed Demographics Age Gender (male) Psychographics Technology Innovativeness Shopping Enjoyment
Mobile Shopping
App Adoption
Possession Usage Usage Intensity*Brand Equity
Brand Awareness Brand Associations Perceived Quality Brand Loyalty Privacy/Security concerns Online Experience Product Category (non-habitual vs habitual) Mobile ExperienceFigure 2 shows the expected relationships between all the variables of this research in a conceptual framework. The aim of this research is to find out whether there is a difference in mobile shopping app adoption between brands scoring high on brand equity and brands scoring low on brand equity. As discussed in our theoretical framework, it is expected that brand equity has a positive impact on mobile shopping app adoption, i.e. the possession of the mobile shopping app and app usage (whether a purchase has been made once). In order to investigate the relative importance of brand equity as a driver of mobile shopping app adoption, secondary variables are taken into account. Those are demographics (age, gender), psychographics (shopping enjoyment and technology innovativeness), privacy/security concerns and digital experience (online experience and experience). Mobile- and online experience, privacy/security concerns and different product categories are expected to influence the relationship between brand equity and mobile shopping app adoption. 3.1 Brand equity According to Keller’s definition of brand equity, consumers who possess high brand equity are more likely to react in a favourable manner to the brand. Strong loyalty is one of the important characteristics that belongs to brands with strong brand equity (Keller, 1993). Prior research shows the strong positive relationship between certain dimensions of brand equity and both attitudinal loyalty and behavioural loyalty (Chaudhuri & Holbrook, 2001; (Netemeyer et al. 2004). Since brands that possess high brand equity have more loyal customers, it could be that they show behavioural loyalty in downloading the brand’s app and use it more frequently. Since attitude drives intention to use, and intention to use says
something about the actual use of an information system according to TAM-model, brand equity could be the driver of mobile shopping app adoption. H1: Brand equity has a positive impact on mobile shopping app possession and usage. 3.2 Moderators and Control variables Online Orientation and Mobile Orientation Online orientation (total experience in online shopping) positively predicted mobile app possession and mobile purchases in a previous study (Kim et al, 2017). Since experienced online shoppers are more familiar with the online shopping environment, barriers to adopt mobile shopping apps are reduced. That applies because the online environment is quite similar to the app environment (Bang et al. 2013). More experienced online shoppers also used a greater number of shopping apps when making mobile purchases. Therefore, we assume that experienced online shoppers have a higher chance to have downloaded the mobile shopping app and use it. Mobile orientation (experienced smartphone usage) positively predicted mobile app possession and mobile purchases in a previous study (Kim et al., 2017). Experienced mobile shoppers are exposed to a greater number of apps which reduces barriers to download apps (Kim et al., 2017). Kim et al. (2017) found that more experienced online shoppers and smartphone users use greater number of shopping apps when making mobile purchases. This study measures possession of one particular mobile shopping app. Therefore, we assume that chances are higher of someone with a lot of mobile experience to have downloaded the mobile shopping app and use it. H2a: Online experience has a positive impact on mobile shopping app possession and usage. H2b: Mobile experience has a positive impact on mobile shopping app possession and usage.
However, we expect that the positive impact of brand equity on mobile shopping app possession and usage is stronger for people with less mobile experience or online experience. People with less mobile experience or online experience need some additional reasons/clues for trust on why they should download the mobile shopping app. So they need some additional convincing due to their lack of mobile/online experience. For those people, the impact of brand equity on mobile shopping app possession and usage is stronger, because having high brand equity matters more for them. Other people with extensive mobile experience or online experience already go and use mobile apps and therefore need less additional proof. As Kim et al. (2017) found, more experienced online shoppers and more experienced smartphone users make mobile purchases by using a greater number of shopping apps. Therefore, we expect the following: H2c: Mobile experience negatively moderates the positive relationship between brand equity and mobile shopping app adoption and usage - The influence of brand equity will be stronger for people with less mobile experience. H2d: Online experience negatively moderates the positive relationship between brand equity and mobile shopping app adoption and usage - The influence of brand equity will be stronger for people with less online experience. Privacy/Security concerns Compared to offline and online commerce, mobile commerce has to deal with more privacy/security concerns. M-shoppers may wonder about their payment information, such as credit card accounts and passwords, which are maybe not always safely transmitted and stored (San-Martin & López-Catalán, 2013). Another concern could be location privacy, as the mobile vendors may disclose location information without the consumer knowing it (Mamonov & Benbunan-Fich, 2015). Therefore, we propose that privacy and security
concerns negatively influence mobile shopping app adoption (a barrier). We also expect that Privacy and security concerns negatively impact the influence of brand equity on mobile shopping app adoption. H3a: Privacy/security concerns have a negative impact on mobile shopping app possession and usage. H3b: The negative effect of privacy/security concerns negatively moderates the positive impact of brand equity on mobile shopping app possession and usage. Product Type Different product categories will be used to increase the generalizability of the study. Those categories are tickets, consumer electronics, clothing and groceries. It is expected that different product categories influence the effect of brand equity on mobile shopping app adoption. Due to the high search costs of mobile shopping, M-shoppers tend to buy products that they are more familiar with, habitual products. So it is expected that mobile shopping app adoption is higher for apps offering habitual products than for apps offering non-habitual products. We expect that the influence of brand equity on mobile shopping app adoption is more important for brands in the non-habitual product category. Habitual products are defined as products that are used on a daily basis, such as coffee, tea etc. So in this study, the groceries category represents the habitual product category. The other categories, e.g. clothing, consumer electronics and flight tickets, represent the non-habitual product category. H4: Product type moderates the influence of brand equity on mobile shopping app possession and usage – The influence of brand equity will be stronger for non-habitual products than for habitual products.
Demographics Age In general, older people tend to have a lower understanding of new technologies (Williams & Page, 2011), and therefore will be less likely to adopt mobile shopping from apps. Therefore, we assume that when age increases, chances are lower that a mobile shopping app will be installed on a mobile device. However, older people tend to be more loyal to their favorite brand and are less into discovering new brands (McLeod, 2009). So when older people already possess a branded mobile shopping app, it might be that they are more loyal to that app and buy more in comparison to younger people. Therefore, we assume that when they have installed the mobile shopping app on their mobile device, they are purchasing more compared to younger people. H5a: Age has a negative relationship with mobile shopping app possession. H5b: Age has a positive relationship with mbile shopping app usage. Gender Men tend to buy more quickly to not spend a lot of time shopping compared to women, they are more quick shoppers (Hansen & Jensen, 2009). Men have less pleasure during shopping than women, whereas women have more patience during the shopping process (Underhill, 2010). Furthermore, men are less researching available other options and spend less time browsing compared to women (Cleaver, 2004). Therefore, the assumption is made that they more strongly hold on to one particular brand, since its more convenient and saves time, and therefore are more likely to download a mobile shopping app. When they have an app of a certain brand on their smartphone, for example to buy clothes, it is assumed they will make more purchases in that app than women. Therefore, we assume that men are more likely to adopt a mobile shopping app than women. H6: Gender (male) has a positive impact on mobile shopping app possession and usage.
Psychographics Technology Innovativeness Innovative consumers tend to adopt a new technology in a relatively early stage far before average others (Aldas-Manzano, Ruiz-Mafe & Sanz-Blas, 2008; Agarwal & Prasad, 1998). They are also more likely to adopt mobile shopping (Yang, 2012). The reason behind it is that Innovative people have more positive perceptions of innovations and intentions towards using new technologies (Lu, 2014). Multichannel shoppers tend to be more innovative as well and explore new channel initiatives (Konus, Verhoef & Neslin, 2008). Therefore, this psychological trait is likely to have a positive impact on the adoption of mobile shopping apps as well. H7: Technology Innovativeness has a positive impact on mobile shopping app possession and usage. Shopping Enjoyment Shopping enjoyment can be defined as the tendency of a consumer to derive pleasure and fun from shopping (Bruner, 2009). Customers who enjoy shopping are more likely to be multichannel shoppers, since they really enjoy shopping and therefore like to try new things in purchase stage and get new ideas anywhere and anytime during the pre-purchase stage (Konus, Verhoef & Neslin, 2008). So it might be that they are more prone to installing new mobile shopping apps on their smartphone and to use it more. H8: Shopping enjoyment has a positive impact on mobile shopping app possession and usage.
Table 3. Summary table hypothesis
Hypotheses Factor Direction of influence
H1 - Brand Equity positively influences mobile shopping app Possession and Usage Brand Equity + H2a – Online Experience positively influences mobile shopping app Possession and Usage Online Experience + H2b – Mobile Experience positively influences mobile shopping app Possession and Usage Mobile Experience + H2c – Influence of Brand Equity on mobile shopping app Possession and Usage is more pronounced for customers with less Mobile Experience Mobile Experience - H2d – Influence of Brand Equity on mobile shopping app Possession and Usage is more pronounced for customers with less Online Experience Online Experience - H3a – Privacy/Security Concerns negatively influence mobile shopping app Possession and Usage Privacy/Security concerns - H3b – Privacy/Security Concerns negatively influence the impact of Brand Equity on mobile shopping app Possession and Usage Privacy/Security concerns - H4 – Influence of Brand Equity on Mobile shopping app Possession and Usage is more significant for non-habitual products than for habitual products Product Type Positive for non-habitual products; negative for habitual products H5a Age has a negative influence on mobile shopping app Possession Age - H5b Age has a positive influence on mobile shopping app Usage Age + H6 – Gender (male) has a positive influence on Mobile shopping app Possession and Usage Gender (male) + H7 – Technology Innovativeness has a positive influence on mobile shopping Possession and Usage Technology Innovativeness + H8 – Shopping Enjoyment has a positive influence on mobile shopping app Possession and Usage Shopping Enjoyment +
4. Research design
This study aims to investigate the impact of brand equity on the possession and use of mobile shopping apps, and the way some demographics (age, gender, education), psychographics (shopping enjoyment), and digital experience (online orientation, mobile orientation) moderate these relationships. The study will be conducted by using an online-questionnaire based survey with Qualtrics. The advantage of gathering data through an online-questionnaire based survey, is that it is relatively easy to gather a large sample against a low cost (Saunders Lewis, Thornhill, 2009). Since the data needs to be collected in a short period of time, an online survey is the best option. It makes it possible to gather a large group of respondents in a limited amount of time, regardless of the distance. Four different surveys were administered in which the brand-related questions differ per survey, depending on the category (clothing, consumer electronics, flight tickets or groceries). The questionnaire was spread in two different languages, namely Dutch and English. Before the survey was launched, an online and offline pilot-study was done with 10 persons to see whether the questions are understandable and could be answered within the given time slot (Baker, 1994). Risks can be prevented by identifying possible problems and to make sure that the chosen instruments are adequete (Teijlingen & Hundley, 2001). 4.1 Population Sample The population-sample for this study will be people from 18 years and older that use mobile shopping applications in their daily lives. Convenience sampling method will be used due to financial- and time restrictions. The online-questionnaire based survey will be spread via social media, colleagues, students, family and friends.4.2 Measures The first part of the questionnaire started with introduction questions. First mobile experience and online experience were asked which are both ratio variables. Online experience is measured by asking the average amount of hours spent online on a day, and by asking what percentage of someone’s total shopping activity is online. Mobile experience is measured by asking what percentage of someone’s total online activity is spent on a mobile device, and by asking what percentage of someone’s total online shopping activity is through a mobile device. The second part of the questionnaire imposed brand-related questions. Respondents got randomly assigned to a category in Qualtrics, i.e. clothing – H&M, consumer electronics – MediaMarkt, flight tickets – KLM or groceries – Albert Heijn. Respondents were first imposed with the brand equity questions, followed by the mobile shopping app possession/usage questions to not influence their answers on the brand equity questions. Before imposing the respondent with the brand, top-of-mind brand awareness is measured by asking which brand comes to mind first when you either want to buy clothes, consumer electronics, flight tickets or groceries. After that question, the respondent is asked to keep one of the four brands in mind by answering the following questions. Brand equity is measured by an adapted and shortened multi-item scale of Yoo & Donthu with 6 items which reflects all the different dimensions of brand equity. All items have a Cronbach’s a higher than 0.70. An example item to measure brand equity: ‘I can recognize H&M among other competing brands’. The latter corresponds to the dimension ‘brand awareness’. The dimension ‘brand associations’ is measured by the item ‘Some characteristics of X come to my mind quickly’. The dimension ‘perceived quality’ is measured
by the item ‘The likely quality of X is extremely high’. Lastly, the dimension brand loyalty is measured by the items ‘I consider myself to be loyal to X’ and ‘X would be my first choice’. The dependent variables are the mobile shopping app adoption variables, namely possession and usage. Possession is a nominal variable, i.e. dummy variable, and measured by whether the app is installed on a mobile device or not. Use is also a nominal variable, i.e. dummy, and measured by whether someone has ever used the app to buy a product or not. Finally, usage intensity is a ratio variable and measured by asking the number of times the app is used to buy a product, which makes it a count variable. In the third part of the questionnaire, respondents are asked questions about their psychographics and behavorial attitude. Technology innovativeness, shopping enjoyment and security/privacy concerns are all measured by existing and validated 7-points Likert scales (strongly disagree – strongly agree) at the interval level. Shopping enjoyment is measured with 3 items, which are extracted from multiple sources Konus, Verhoef and Neslin (2008). Cronbach’s a is 0.91. An example item to measure shopping enjoyment: ‘I like shopping’. Technology innovativeness is measured by a shortened scale of Konus, Verhoef and Neslin (2008) with 3 items. Cronbach’s a is 0.78. An example item to measure technology innovativeness: ‘I regularly purchase different variants of a product just for a change’. Security/privacy concerns is measured by an adapted and shortened scale of Dinev & Hart (2006) with 3 items. Cronbach’s a is 0.88. An example item to measure security/privacy concerns is: ‘I am concerned about submitting information on the internet, because of what others might do with it’.
The questionnaire ended with asking respondents about their demographics, i.e. age (ratio variable) and gender (nominal – dummy). An overview of all variables, their measures, role and measurement levels can be found in table 4.
Table 4. Summary table measures, roles and measurement levels of variables
Variable Role Measure Level
Digital
Experience Online Orientation Moderator Numeric Ratio
Mobile
Orientation Moderator Numeric Ratio
Brand Equity Brand Equity
Independent
Variable Multi-item 7-points Likert Scale Ordinal Mobile Shopping App Adoption Possession Dependent
Variable No = 0, Yes = 1 Nominal (dummy)
Use Dependent
Variable No = 0, Yes = 1 Nominal (dummy)
Usage Intensity Dependent
Variable Numeric Ratio
Psychographics Technology
Innovativeness Control Variable 7-points Likert Scale Ordinal Shopping
Enjoyment Control Variable 7-points Likert Scale Ordinal
Attitudinal
Factor Security/Privacy Concerns Moderator 7-points Likert Scale Ordinal
Product
Category Tickets Moderator No = 0, Yes = 1 Nominal (dummy)
Consumer
Electronics Moderator No = 0, Yes = 1 Nominal (dummy)
Clothing Moderator No = 0, Yes = 1 Nominal
(dummy)
Groceries Moderator No = 0, Yes = 1 Nominal
(dummy)
Demographics
Age Control Variable Numeric Ratio
Gender Control
Variable Male = 0, Female = 1 Nominal (dummy)
Recoding Variables The variable top-of-mind brand awareness is recoded as a dummy variable in which respondents mentioning either H&M in the clothing category, MediaMarkt in consumer electronics category, KLM in flight tickets category or Albert Heijn in groceries category = 1 and respondents not mentioning H&M, MediaMarkt, KLM or Albert Heijn in one of the categories = 0. The variable mobile shopping app possession is recoded as a dummy variable with respondents having installed the mobile shopping app on a mobile device of either H&M, MediaMarkt, KLM or Albert Heijn =1 and respondents not having installed the mobile shopping app of either H&M, MediaMarkt, KLM or Albert Heijn = 0. The variable mobile shopping app use is recoded as a dummy variable with respondents that have once used the mobile shopping app on a mobile device of either H&M, MediaMarkt, KLM or Albert Heijn = 1 and respondents not having used once the mobile shopping app of either H&M, MediaMarkt, KLM or Albert Heijn = 0. The variable gender is recoded as a dummy variable with males = 0 and females = 1. Creating variables The total score of the variable brand equity is calculated by averaging the score on brand awareness (20%), brand image (20%), brand quality (20%) and brand loyalty (40%). According to Keller’s CBBE pyramid, brand loyalty is the final level and most difficult stage to obtain. Therefore, the total score becomes more realistic when brand loyalty contributes more “weight” compared to the other three elements.
Furthermore, the variable brand equity is multiplied with the moderating variables (I.e. mobile experience, online experience, privacy/security concerns) in order to analyze the moderating effects. The variable mobile experience is calculated by averaging the scores on the two statements regarding the percentage of someone’s total online activity spent on a mobile device and the percentage of someone’s total online shopping activity through a mobile device. The variable online experience is calculated by averaging the scores on the two statements regarding the daily amount of hours spent online as a percentage of 24 hours and the online percentage of someone’s total shopping activity. The predictor variables are all standardized into z-scores, so the relative influence of the different predictor variables could be assessed. This method was chosen, since the unit of measurement is not the same for all predictor variables, which can be seen in table 4. 4.3 Analyses SPSS (Version 22.0.0.0, IBM Corporation) is used to analyze the data. The outcome variables Mobile shopping app possession, usage and intensity are asked separately for all four product categories (i.e. clothing, consumer electronics, flight tickets and groceries), so the data is analyzed separately based on these four product categories. The main effects of brand equity and other behavioral and attitudinal factors on mobile shopping app possession, usage and intensity are also analyzed per category. Econometric analysis, i.e. regression based models, are used in which adoption related variables are the dependent variable (Y) and brand equity and other behavioral and attitudinal factors are explanatory variables (X). Furthermore, interactions between moderator variables and brand equity are
Two different models are used for possession and usage. The regression type depends on the nature of the dependent variable. For possession and usage, binary logistic regression is used. For usage intensity, which is a metric/quantity variable, zero-inflated poisson regression is used. The latter is chosen, since the intensity variable consists of a lot of zero’s in the data. The two different models allow us to check whether drivers differ in different phases of mobile shopping through apps. We investigate two stages in mobile shopping: 1. possession of the app and 2. Usage of the app. For example, is brand equity more important in driving app downloads or is it more important in app usage? Table 5. 16 Analyses Models Possession Use Clothing 1.Main
effect 2.Moderating effects 3.Main effect 4.Moderating effects
Consumer Electronics
5.Main
effect 6.Moderating effects 7.Main effect* 8.Moderating effects*
Flight Tickets
9.Main
effect 10.Moderating effects 11.Main effect 12.Moderating effects
Groceries 13.Main
effect 14.Moderating effects 15.Main effect 16.Moderating effects
*These models are not tested Binary Logistic Regression Analysis Binary logistic regression analysis is used to test the hypothesis. Binary logistic regression is used, since we have a 0/1 binary categorical discrete type of dependent variable in both models. It is not a continuous or numeric quantity. Those dependent variables are influenced by continuous and categorical independent variables. Therefore, binary logistic regression is used, since it is most suitable in case of a binary dependent variable (Osborne, 2008). The following seven assumptions need to be met in order to correctly perform binary logistic regression analysis.
First of all, logistic regression requires the dependent variable to be dichotomous. The models meet this requirement, since all the dependent variables are nominal with two outcomes, i.e. possession and use (two outcomes: “yes” or “no”). Secondly, logistic regression requires one or more independent variables that are measured on either a continuous or nominal scale. In order to meet this assumption, all the independent variables measured at the ordinal level are treated as a continuous variable, i.e. brand equity, shopping enjoyment, technology innovativeness and privacy/security concerns. Third, the observations need to be independent of each other. The categories of the binary dependent variable (i.e. possession and usage) and the nominal independent variables (i.e. gender) should be mutually exclusive and exhaustive. This means that there is no relationship between the observations in each category of the dependent variable or of any nominal independent variables. In addition, there is no relationship between the categories. Fourth, logistic regression requires a bare minimum of 15 cases per independent variable, although some researchers recommend a minimum of 50. Since binomial logistic regression relies on maximum likelihood estimation (MLE), reliability of estimates declines significantly for combinations of cases where there are few cases. In this study, every independent variable has a minimum of 60 cases, so we meet this assumption. Fifth, linearity of the continuous independent variables and logit transformation of the dependent variable is assumed. In this study, the continuous independent variables mobile experience, online experience and age are found to be linearly related to the logic of the dependent variables possession and usage. It is assessed via the Box-Tidwell (1962)
statistical significance being accepted when p<0,00417 (Tabachnick & Fidell, 2014). So we meet this assumption. Sixth, there should be no or little multicollinearity among the independent variables. It implicates that the correlation coefficients of the independent variables should not be too high, so the independent variables should not be too highly correlated with each other. The correlation matrix tables for all four product categories can be found in Appendix A. We meet this assumption as well. Seventh, logistic regression requires no significant outliers, high leverage points or highly influential points. For the variable mobile experience, there is one outlier detected with a standardized value of 10,35(z-score), the value was adjusted from 125% to 1.25hours due to probably a typing error made by the respondent. For the variable technology innovativeness, several cases are found with a standardized value of 2,611(z-score), which were kept in the analysis. No other outliers are detected in the other variables (>2.5 standard deviations from the mean).
