MASTER THESIS
The Effect of the Use of M-shopping Applications on Attitudinal Brand
Loyalty, and the factors influencing this relationship
University of Amsterdam Faculty of Economics and Business
MSc in Business Administration Track: Marketing
First supervisor: Dr. Umut Konuç Second supervisor: Dr. Frederik Situmeang
By:
Student: Griselda Morales Moreno UvAID: 10490736
Statement of Originality
This document is written by Student Griselda Morales Moreno who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
ABSTRACT
The fast paced environment being experienced nowadays within the Internet dimension is leading towards a more hyper connected world, which is encouraging mobile commerce (m-commerce) to become a profound economic and social phenomenon (Kourouthanassis et al, 2012). Only in 2014, the usage in shopping applications increased by 174% (mobilecommercedaily.com, 2015). Furthermore, the acquisition and retention of customers posits a challenge that needs to be fulfilled within a market full of competitors, so as to win market share and develop a sustainable competitive advantage (Lin et al., 2006). Therefore, the research question emerged out of these two dimensions, focusing the research on the effect of the use of mobile shopping (m-shopping) applications on attitudinal brand loyalty, and the effect of some specific moderators (innovativeness, impulsiveness, and trustfulness). The online survey data from 156 Dutch and Spanish mobile shoppers was used in order to test the hypotheses. After conducting a multiple linear regression analysis, the results demonstrated that neither the usage nor the moderators have a significant effect on attitudinal brand loyalty. Several arguments are given in the discussion in order to explain the possible reasoning for the hypotheses not being supported. Furthermore, this study concludes with several managerial implications of the results, as well as some suggestions and directions for further research.
Keywords: m-shopping applications, attitudinal brand loyalty, innovativeness, impulsiveness, and trustfulness, m-commerce, mobile applications
Acknowledgements
I would like to thank my supervisor Dr. Umut Konuç for providing me with periodical feedback and recommendations about my work. Without his help this Master Thesis could not have been developed.
I am also thankful for the support from my colleagues and especially to Sergi Albiol who encouraged me during the whole process and helped me with the final check of the document. Finally, I would like to thank my second supervisor Dr. Frederik Situmeang for reading and grading this thesis.
TABLE OF CONTENT
1. INTRODUCTION ... 7
2. LITERATURE REVIEW ... 10
2.1.M-COMMERCE AND E-COMMERCE DEFINITIONS, AND PREVIOUS RESEARCH ... 11
2.1.1. M-shopping ... 12
2.2.CUSTOMER LOYALTY DEFINITION AND PREVIOUS RESEARCH ... 15
2.3.M-COMMERCE AND E-COMMERCE LINKED WITH CUSTOMER LOYALTY PREVIOUS RESEARCH . 18 2.4.GAPS IDENTIFIED AND RESEARCH QUESTION ... 20
2.4.1. Industries Analysed ... 22
3. CONCEPTUAL FRAMEWORK ... 23
4. METHODOLOGY ... 28
4.1 THE SAMPLE ... 28
4.2 DATA COLLECTION PROCEDURE ... 29
4.3MEASURES ... 29
4.4STRENGTHS AND LIMITATIONS ... 31
5. RESULTS ... 31
5.1RELIABILITY AND VALIDITY ... 31
5.2CORRELATION CHECK ... 32
5.3MODEL TESTING ... 34
5.3.1. Results Consumer Electronics Database ... 35
5.3.2. Results Flight Tickets Database ... 37
5.3.3. Summary of Results ... 38
6. DISCUSSION ... 41
6.1.MANAGERIALIMPLICATIONS ... 44
6.2.LIMITATIONSANDFURTHERRESEARCH ... 45
APPENDIX ... 54
I.QUESTIONNAIRE ... 54
II.SPSSOUTPUT ... 57
Demographic statistics for Consumer Electronic data ... 57
Q-Q plots for Consumer Electronic data ... 58
Demographic statistics for Flight Tickets data ... 60
Q-Q plots for Flight Tickets data ... 61
INDEX OF TABLES Table 1: Prior customer loyalty research overview 17
Table 2: Review of relevant papers in the field of online loyalty 20
Table 3: Scales and items used in the questionnaire 30 Table 4: Reliability of scales 32 Table 5: Correlations for electronic products data 33
Table 6: Correlations for flight tickets data 33
Table 7: Explanation of the items within the regression 35 Table 8: Linear regression for consumer electronics respondents 39
Table 9: Linear regression for flight tickets respondents 40
Table 10: Summary of the results 38 INDEX OF FIGURES Figure 1: Schema of the concepts related with online commerce 14
Figure 2: Schema of the different types of loyalty 18
1.
INTRODUCTION
Nowadays, smartphones represent one of the leading creations from the technological revolution, allowing people to play games, check e-mails, surf and even check prices on the stock market, by only using a mobile device (Chen et al., 2011). One of the main features that smartphones have is the opportunity of buying products and services through it; this is referred to as mobile commerce (henceforth, m-commerce).
The following research is focused on the use of m-shopping applications and its effect on attitudinal brand loyalty, as well as how specific variables moderate this relationship. The selected psychographic moderators are: impulsiveness, innovativeness, and trustfulness; and the demographic ones are: age, gender, and education. The study was conducted within the consumer electronics and flight tickets industries, in order to acquire two different points of view and make the results more generalizable. This research is important due to different reasons. On the one hand, the fast paced environment being experienced nowadays within the Internet dimension is leading towards a more hyper connected world, which is encouraging m-commerce to become a profound economic and social phenomenon (Kourouthanassis et al, 2012). Specifically in the second quarter of 2014, m-commerce experienced a worldwide increase in spending of 47% (comScore.com, 2014). Moreover, statistics from March 2014 show that, 17% of smartphones users in the UE purchased goods or services using their smartphones at least once per month (statista.com, 2014), and this percentage is expected to heavily increase in the upcoming years. Only in 2014, the usage in shopping applications increased by 174% (mobilecommercedaily.com, 2015). Moreover, the research specifically focuses on mobile applications as m-shopping tool, because as stated in a recent study, 86% of the time spent on a mobile device is using applications, and only 14% spent on surfing the Internet browser (forbes.com, 2014). On the other hand, companies’ main goal is to sell their products or services; therefore, organisations aim to be visible and reachable in those channels
where their target customers are present. Furthermore, the acquisition and retention of customers posits a challenge that needs to be fulfilled within a market full of competitors, so as to win market share and develop a sustainable competitive advantage (Lin et al., 2006). As a consequence, global managers fear on how this new shopping phenomenon will impact customers’ loyalty towards the brand. Thus, the research field lacks a thorough investigation about the relationship between m-shopping applications use and attitudinal brand loyalty.
The research question emerged after analysing the current research gaps in the relevant literature. On the one hand, the research about m-shopping remains scarce, as well as brand loyalty related research (Shankar et al., 2003). And on the other hand, there is a latent need of further investigating the role of specific characteristics of consumers that influence their attitude towards loyalty (Liao et al., 2014). This research will be relevant due to several reasons. First, there is an exponential increase of the m-shopping channel, so that may become the leading shopping form of the near future. This fact converts the channel into one of the most interesting ones in order to study. Second, considering that brand loyalty is a key determinant of the firms in order to succeed and keep growing, the relationship between this concept and m-shopping applications use will provide some interesting insights in how this new form of shopping affects loyalty. Moreover, it will provide actionable insights contributing to managers’ decisions in marketing investment. And third, m-shopping applications apply to a wide range of businesses of different industries. On the one hand, the companies that are already present in the mobile shopping channel will better understand how some psychographic moderators affect the relationship between their mobile application and brand loyalty. Thus, being able to target customers in different categories depending on their personal characteristics, so that they can adapt and make more efficient brand loyalty strategies by better using their mobile applications. And on the other hand, firms that do not have a
mobile application yet, will be able to receive further information on how brand loyalty could be improved with the existence of a mobile application with specific characteristics.
Contributions
The topic to be studied posits a relevant starting point for the related literature. There is currently a lack of understanding of consumer behaviours’ in the mobile technology-mediated shopping environment, so that retailers’ ability remains limited when developing appropriate shopping services in the mobile shopping stage (Yang and Kim, 2012). This study aims to provide further knowledge into the m-shopping and brand loyalty research area, as well as offer some managerial contributions so as to help decision makers to efficiently allocate marketing investments.
Managerial relevance
Some managerial contributions will result from the study so as to provide mobile professionals with an interesting insight on how to improve their efficiency. First, the increase in the development of specific mobile applications for companies posits a new challenge in retaining customers, thus, making them stay loyal to the firm. Organizations are spending a substantial amount of resources on m-commerce technologies; so delivering added value to customers posits a differentiation strategy in order to achieve a competitive advantage (Lin et al., 2006). The outcomes from the research will help mobile professionals to better understand how loyal customers feel towards the applications they use, depending on the different uses they undertake. Second, the research contributes to the understanding of how innovative, impulsive, or trustful customers identify themselves loyal to the brand. Third, once known to what extent customers identify themselves loyal to the brand, managers could use that information for developing and implementing more accurate loyalty strategies. They will be able to better target their mobile customers by using the relevant data from the analysed moderators. And fourth, it may help organisations that are not willing to implement a mobile
strategy to realize about how their loyalty strategies could benefit from introducing a mobile application.
Theoretical relevance
Regarding the theoretical contributions, some improvements will be added to the existing literature. First, it will provide a better insight on m-shopping literature, since it is an area under study and its contributions remain limited. Second, brand loyalty research will gain awareness about its relationship with this relatively new way of shopping. Third, the combination of the selected moderators and the study of their impact on the previous mentioned relationship would deliver an added value to the previous research. And fourth, it will provide a starting point for researchers in this area, in order to develop an extended version of the research by including more moderators; expand the scope by analysing other industries; and increase the sample of respondents.
Thesis overview
In the following paragraphs, the literature relevant to the topic and the research gaps are reviewed, it follows the conceptual framework and methodology used. In the last stage the results are offered, as well as the discussion and the managerial implications. Moreover, some further research guidelines are presented.
2.
LITERATURE REVIEW
First, the concepts of m-commerce, e-commerce, and specifically m-shopping are introduced. Moreover, their previous literature is analysed. Second, the concepts of customer loyalty and brand loyalty are widely examined. Third, the literature regarding m-commerce and customer loyalty is studied. And fourth, after examining the mentioned literature, the research gaps and question are identified.
2.1. M-commerce and E-commerce Definitions, and Previous Research
Mobile commerce (m-commerce) is a relatively new concept of shopping. It can be defined as online transactions conducted through mobile devices (e.g. smartphones and tablets), using wireless telecommunication networks (Siau et al., 2001; Chong, 2013). There are two different tools in order to buy through a mobile device: mobile websites, which consist of an adapted version of the regular website, so as to be more user-friendly in palm-size screens, and mobile apps, which are specific software downloadable to a mobile device (Bellman et al., 2011). The mobile Internet usage has increased since the introduction of the smartphones. Nowadays, its use is becoming increasingly convenient for users, since they can have nearly the same utilities as in a laptop but available in a palm size device 24 hours per day, 7 days per week. One of the determinants of the channel success is the capability that its providers have in understanding the different roles of customers in their busy lifestyles (Dholakia et al., 2004). So, that they are able to recognize their customers’ needs and adapt consequently its business models. Some of the advantages perceived from this channel, that are shared with electronic commerce (e-commerce), are: providing great flexibility, low cost structures, fast transactions, broad product lines, convenience, customization, and one of the most important features overall is the fact that it reduces information asymmetries between sellers and buyers (Srinivasan et al., 2002).
Since m-commerce research remains scarce, it is important to include in the study the e-commerce concept, which shares some commonalities with m-e-commerce. Both m-e-commerce and commerce represent a form of online commerce. Moreover, in several cases, the e-commerce research serves as a starting point in order to investigate about similar concepts in the m-commerce context, and to set up some basis for the future research of m-commerce. E-commerce concept emerged some years before the m-E-commerce. E-E-commerce can be defined
as computer-mediated shopping transactions, conducted through computers and laptops (Dholakia et al., 2004).
Regarding the adoption of the m-commerce channel, it is influenced by several factors such as performance expectancy, effort expectancy, social influence, facilitating conditions, perceived value, trust, perceived enjoyment, and personal innovativeness (Chong, 2013). However, these factors are not the only ones that might affect the channel adoption. It is thought that several opportunities position the shopping channel with some competitive advantages in comparison to the e-commerce, so that can lead as well to influence the channel adoption. As Nassuora (2013) stated, some of these opportunities are: faster access, more powerful, more effective and accessible almost anytime, anywhere for its users. All these factors explain accurately which elements motivate customers in order to adopt this new channel. However, m-commerce’s main advantage is the ability to offer greater ubiquity and accessibility to its users in comparison to e-commerce (Nassuora, 2013). The perceived advantages of the channel help converting it into an appealing way of shopping which keeps attracting more new users. Moreover, regarding the decision-making stage of consumers, several variables affect them when interacting with m-commerce applications. These variables are: the digital environment, the physical environment, the social environment, the mobile device, and the mobile connection (Benou et al., 2012).
2.1.1. M-shopping
When considering the m-commerce idea, it seems to be too broad; therefore, a more specific concept needs to be considered for this research. This concept is defined as mobile shopping (m-shopping). M-shopping can be defined as the activities undertaken by consumers who use wireless Internet services when shopping using mobile phones or tablets (Ko, Kim and Lee, 2009). These activities are three in total: information search, purchase, and after-sales service (Neslin et al., 2006). Whereas information search can be defined as the stage of the
decision-making process in which the customers actively compile information from different sources (Köksal, 2011). The purchase stage is the actual purchase of a specific product or service. And the after-sales service is formed by the requirements and enquiries of the customers, which engage firms in providing information and support after the purchase.
The mobile environment, as well as the online environment, offer the companies the possibility to present massive amounts of information to their consumers when searching for products or services through the app (Grant et al., 2007). When consumers face this vast amount of information, they spend different amounts of time searching through it. It all depends on the product category they are looking for (Bhatnagar and Ghose, 2004). During the pre-purchase or information search stage, customers look for information about product quality, features and services, possible prices, and availability of the product (Umit Kucuk and Maddux, 2010). The consumers’ behaviour during this research phase is influenced by three personal skills, namely technological skills, search skill, and information processing skills (Grant et al., 2007). In that stage it can be identified the research-shopper phenomenon, which is the tendency of customers to use different channels for searching information and for purchasing (Verhoef et al., 2007). There can be defined two types of research-shoppers: the competitive research shopper who searches at one firm and purchases from another one, and the loyal research shopper who searches and purchases from the same firm (Neslin and Shankar, 2009).
Some studies have demonstrated that customers are able to make better quality decisions when shopping online, than when shopping on a brick-and-mortar (Punj, 2012). This is due to the benefits that customers perceive from shopping online, that can be described as: saving time, saving money, and helping finding out the products that best match customers’ needs (Punj, 2011). Apart from the mentioned beliefs, there are some motives and impediments that determine the engagement of customers with m-shopping usage. The motives comprise
convenience, company performance, and marketing-mix; the impediments are composed by the lack of interpersonal relations, problems using the mobile device, transaction costs, and barriers to the products (San-Martín et al., 2013). Moreover, some authors have researched several variables that seem to influence the intention to buy through m-shopping applications. On the one hand, personality variables such as affinity to mobile phones, compatibility, and innovativeness (Aldás-Manzano et al., 2009), as well as enjoyment (Su and Lu, 2009), influence positively the intention to engage in m-shopping. While on the other hand, customers seem to be negative about several aspects such as trust, fun, excitement, clarity, and friendliness, so they perceive multiple variables which negatively influence the intention of using their mobile phones for shopping (Holmes et al., 2014).
M-shopping applications concept is restricted to specialized applications that can be found in mobile devices (Bellman et al., 2011). Internet mobile browsers are not considered applications, as mentioned in the first section of the literature review. Moreover, the present research refers to the concept of m-shopping applications use. What is meant by use is the actual and intentional usage that consumers exhibit towards the selected brand. Thus, the current behaviour showed when shopping their preferred brand, and the intention of shopping in the near future. As an example, a customer that searches information or buys a determined number of times from a specific brand is exhibiting actual usage, whereas a customer that declares that he or she will buy from a specific brand is showing intentional usage.
The next section introduces the concept of customer loyalty and discusses the existing research about the topic.
2.2. Customer Loyalty Definition and Previous Research
Customer loyalty is a topic that has been extensively researched about in the last decades, especially in the offline commerce context, as it has existed for several more years than the online commerce. One of the most recognized definitions states that, customer loyalty is “a deeply held commitment to rebuy or repatronize a preferred product/service consistently in the future, thereby causing repetitive same-brand or same brand-set purchasing, despite
situational influence and marketing efforts having the potential to cause switching behaviour”
(Oliver, 1999, p.34). More precisely, it can also be defined as the proportion of times that a customer chooses the same product or service in the same category compared to the total amount of purchases made in that category (Neal, 1999). When analysing customer loyalty conceptualization, previous research has established two different types of loyalty; attitudinal loyalty, refers to the customers’ psychological environment and the declared intention to rebuy a product or a service from a concrete brand, and behavioural loyalty, refers to the tendency to repeat and continue past purchases, it is the observed behaviour of a concrete customer (Kim et al. 2008, as cited in San-Martín et al. 2013; Valvi et al., 2012; and Hawkins et al., 2013). For example, when a customer declares that he or she will buy from a specific brand, or that he or she feels loyal towards a brand, they are exhibiting attitudinal loyalty. Whereas in the other hand, if a customer’s behaviour expresses loyalty towards a brand, for example, buying a product category in the same brand 4 times out of 5, then those customers are exhibiting behavioural loyalty. These two definitions seem accurate enough and accepted by authors to describe the types of customer loyalty. They provide a clear understanding of how a loyal customer can react towards a business, either voluntarily and conscious or behaviour driven without much thinking involved. Odin et al. (2001) defined two different types of customer
loyalty, based on the situation that the customer encounters when repurchasing: reflective loyalty, which is the result of brand commitment or a positive attitude towards a brand, or an inertia to purchase, which is the repeated purchase of the same brand without any specific motive for choosing it. The current research studies specifically whether the m-shopping applications usage has an effect on the attitudinal loyalty dimension from customers, so as to assess how loyal they feel towards the brand featuring an m-shopping application.
Several authors have analysed which factors, moderators and relationships existed between customers’ experiences and customer loyalty. The results up to now highlight several variables that lead the increase in customer loyalty. Some of these variables can be identified as: perceived firm innovativeness (Kunz et al. 2011), service customization (Coelho et al., 2012), habit strengthening (Olsen et al., 2013), and perceived multichannel service quality (Hsieh et al., 2012). Among the research conducted it can be concluded that, one of the main outcomes, that several of them share, is the fact that customer satisfaction is one of the key mediators in order to achieve customer loyalty (Coelho et al., 2012; Kunz et al., 2011; Olsen et al., 2013; Wang et al., 2007). However, there is still the need to find more variables so as to help decision makers and researchers to better understand and predict customer loyalty.
As previously mentioned, the research in the field of customer loyalty has been vast throughout the last decades. The Table 1 presents a collection of papers conducted within the field of customer loyalty, as well as some papers related to brand loyalty, which summarize the general concepts that have been researched until now. After reviewing the past research, several general topics are highlighted and found in different papers. First, some of them centre their focus on analysing the different concepts of customer loyalty, as well as brand loyalty (Wernerfelt, 1991; and Hawkins, 2013). Second, another topic is the development of new conceptual models in order to explain customer loyalty (Dick et al., 1994; and Kim et al. 2008, as cited in San-Martín et al. 2013). Third, some researchers focused their efforts towards
understanding the effect of different moderators on customer loyalty (Yang et al., 2004; Liu et al., 2012; and Yi et al., 2003). Fourth, the notorious relationship between customer satisfaction and customer loyalty has received special attention by several authors (Hallowell, 1996; Oliver, 1999; Nam et al., 2011; and Hsieh et al., 2012). And fifth, the interest of building up a measurement model for brand loyalty derived in the research of Odin et al. (2001).
After the review of relevant past research, a clarification must be done. Within the different expressions of customer loyalty, this study specifically centres its research on brand loyalty. While the term customer loyalty may seem too broad, and customers can exhibit loyalty to several aspects, as can be a concrete firm, product, brand, channel, or a retailer, this research focuses on the exhibited loyalty by customers towards the brand. Brand loyalty could be defined as the degree of attachment that customers exhibit for a specific brand, and this attachment is strongly linked to use experience (Liu et al., 2012).
Table 1. Prior customer loyalty research overview Concepts of customer/brand loyalty New conceptual models to explain customer/brand loyalty Effect of different moderators on customer loyalty Relation customer satisfaction and customer/brand loyalty Measurement procedure for brand loyalty Guest, 1944 ✓ Wernerfelt, 1991 ✓
Dick and Basu, 1994 ✓
Hallowell, 1996 ✓
Oliver, 1999 ✓
Odin et al., 2001 ✓
Yi and Jeon, 2003 ✓
Yang and Peterson, 2004 ✓ ✓
Kim et al., 2008 ✓
Nam et al., 2011 ✓
Hsieh et al., 2012 ✓
Liu et al., 2012 ✓
Figure 2. Schema of the different types of loyalty
The following part analyses the literature published until now about the relationship between e-commerce and m-commerce with customer loyalty.
2.3. M-commerce and E-commerce linked with Customer Loyalty Previous Research
Some research has been conducted about customer loyalty in the context of m-commerce, although it is limited to few topics and the concept has exponentially evolved since its inception. It should be mentioned again, that m-commerce is an extension from commerce. Therefore, as a starting point it should be considered the research about e-commerce and customer loyalty, as it will serve as a proxy for m-e-commerce. As a general overview of loyalty applied to e-commerce there should be considered the critical review from Valvi et al. (2012), which conducted a review of multiple empirical studies on e-loyalty. Several definitions for e-loyalty were identified such as continuance intention, re-purchase intention, repatronize intention, commitment, stickiness, and word-of-mouth (WOM). Moreover, it highlighted the main factors influencing customer loyalty in each shopping stage; pre-purchase e-competitors attitudes, e-reputation, customer characteristics, and PC
knowledge, during purchase web servQual, and customer e-pleasure, after purchase e-satisfaction, e-trust, perceived value, convenience motivation. (Valvi et al., 2012) It is noteworthy understanding these factors, because they provide companies with a clearer approach towards what their customers value when shopping, in order to increase their repurchase intentions.
Authors may be assuming that online and offline loyalty are driven and enhanced by different factors and moderators. Thus, different research is being conducted in order to better understand and prove so. Following this argument, as Shankar et al. (2003) stated in their study, loyalty appears to be higher when customers buy online rather than offline. And as previously mentioned, customer satisfaction proved to be the leading factor towards loyalty. The explanation relies on the fact that, the online environment offers more information and helps customers accessing and processing the information available. (Shankar et al., 2003) Nevertheless, the results were achieved from a research conducted in the early years of e-commerce; therefore, one cannot assume that these findings still apply nowadays. Following Shankar et al. (2003) argumentation, Huang (2011) argued that customers generally exhibit high excess of behavioural loyalty in online purchases. Moreover, this e-loyalty can be increased with the 8 Cs developed by Srinivasan et al. (2002), such as customization, contact interactivity, care, community, convenience, cultivation, choice, and character. Following the same line of argument, but focusing on one of the few researches that has been conducted about loyalty in m-commerce, Lin et al. (2006) reported that in this case, loyalty might be affected by perceived value, trust, habit, and customer satisfaction. Since the m-commerce channel has emerged relatively few years ago, and its increase and development have nothing to do with the early days of its inception, one may conclude that its findings could be obsolete. The main reason being the exponential increase in the transaction volume achieved through mobile devices. In 2010, the global mobile payment transactions totalled a value of 52.9
bn/$US, while in 2013, the sum was 235.4 bn/$US (Statista.com, 2013). These data reflect the fast-growth that mobile transactions are acquiring in a relatively small period of time. Additionally, the research from Lin et al. (2006) was the first one conducted to examine the factors affecting loyalty in m-commerce and was limited to Taiwan. Then, one might think that the factors researched by Lin et al. (2006) may have varied, or that different moderators are influencing the relationship between m-commerce and customer loyalty nowadays.
In the Table 2, one may identify a summary of the research that has been conducted until now in the field of online loyalty. The next section provides a closer approach towards analysing the existing research gaps, and finding the relevant gaps so as to be researched within this Master thesis.
Customer loyalty in e-commerce/ m-commerce Behavioural loyalty and online retailers Satisfaction in multichannel customers Review studies of e-loyalty Types of loyalty and influence of social media
Hawkins and Vel, 2013 ✓
Huang, 2011 ✓
Hsieh et al., 2012 ✓
Liao, Wang, and Yeh, 2014 ✓
Lin and Wang, 2006 ✓ Shankar et al., 2003 ✓ Srinivasan, Anderson, and Ponnavolu, 2002
✓
Valvi and Fragkos, 2012 ✓
Table 2. Review of relevant papers in the field of online loyalty.
2.4. Gaps Identified and Research Question
This section provides the analysis of the research gaps, providing its relevance and importance for the research. Moreover, there is the identification of the specific research gap, and the subsequent research question.
The review of the literature in Table 2 serves as an approach in order to better recognize the research gaps. Once reviewed the mentioned past research, several gaps should be emphasized. First, the research has been focused on customer loyalty but not specifically on brand loyalty (Shankar et al., 2003). Second, there have been analysed the influence of some specific moderators in the relationship between e-commerce and customer loyalty, but there is
still space for further investigation of other moderators (Chang et al., 2008; Liao et al., 2014). Third, there is the latent need to investigate the role of individual characteristics of consumers that influence their attitude towards loyalty (Liao et al., 2014). And fourth, the research conducted up to now has been mainly focused on the banking and travel industry, so there is a gap for further studying different industries (Lin et al., 2006; Shankar et al., 2003).
The aforementioned gaps are important to research due to several reasons. First, the increasing use of mobile shopping technologies posits a challenge for companies in general. They need to make sure whether they are attracting and retaining customers in a proper way through this new channel. Considering that m-shopping applications provide accessibility anytime and anywhere as long as one has a wireless connection, it is important for companies to efficiently attract customers and make them stay loyal to the brand. Otherwise, the easiness of finding different brands where to buy from, posits a threat for firms selling products or services through mobile applications. Second, there is the opportunity to investigate how different psychographic and demographic moderators influence the relationship between m-shopping applications use and brand loyalty, which will provide with new insights on the mobile shopping field. These results will provide managers with specific information on the factors that most affect the level of attitudinal brand loyalty exerted by m-shopping users, so as to check if their mobile applications succeed in those specific moderators, or if they are targeting the proper audience. Consequently, mobile applications developers will gain knowledge when understanding mobile shoppers’ behaviours. So as to better know how to loyalize them to the brand, by engaging them in the use of mobile shopping applications. Third, studying different industries allows researchers to prove the generalization of the results. As well as to provide managers with an extended view of the research, not only limited to a specific industry. Fourth, there has been little research in order to investigate the relationship between m-commerce with customer loyalty. Nevertheless, the existing research becomes less
useful as years pass by; this is due to the fact that online commerce suffers from a fast-paced environment with an on-going development happening. As an example, the research from Lin et al. (2006) was conducted 10 years ago, and the m-commerce context has exponentially increased and changed dramatically. Therefore, managers find themselves with out-dated information that may not apply in today’s context, and there is the latent need to acquire new and updated data about consumer behaviours towards brand loyalty when using m-shopping applications.
Once identified the possible research gaps, limitations, and its relevance, it follows the research gap and question for the current thesis. It has been recognized a need to further understand the consumer behaviours related with m-shopping, and how its use may affect attitudinal brand loyalty, since there has not been conducted specific research about this topic. The aim of this research will be to determine if m-shopping applications have an effect on attitudinal brand loyalty, and if some moderators are influencing this relationship. The psychographic moderators are: impulsiveness, innovativeness, and trustfulness; and the demographic moderators: age, gender, and education. After all the considerations, the research questions follow:
- Does m-shopping application use affect attitudinal brand loyalty?
- Whether and to what extent do the selected moderators impact the relationship between the use of m-shopping applications and attitudinal brand loyalty?
2.4.1. Industries Analysed
The proposed research questions need to be analysed within specific constrained industries so as to determine the research boundaries. In order to have different points of view, and not be constrained to only one industry, two industries were selected for the study: consumer electronics and airlines industry. First, consumer electronics industry was selected due to its product tangibility nature. Moreover, this industry showed to be the most purchased
category in online retail in 2013, with 69% of the respondents assuming having bought products in this category (Walker Sands’ 2014, Future of retail study). Second, the airlines industry was chosen because it is a service purchased online, with low purchasing risk because customers have almost completely shifted their travel buying behaviour towards the online industry. Moreover, these two categories were chosen following the next criteria: consumer electronics and the flight tickets purchase represent nowadays two of the activities that are mainly conducted online. These are not only two different industries, but also two different categories with different consumer behaviours and expectations, which will bring specific insights to generalize the findings to diverse industries.
In the following part, the conceptual framework is introduced so as to establish the basis of the research that will be conducted.
3. CONCEPTUAL FRAMEWORK
Once identified the research question, the conceptual framework designed for the study is explained. First, the framework is presented so as to map the general concepts. And second, the hypothesis formulation follows in order to establish the basis of the research.
Figure 3: Conceptual Framework
The figure 3 draws the framework that will be the core guideline of this research. The goal of the study is to find out the effect of the use of m-shopping applications on brand loyalty, and how the psychological (innovativeness, impulsiveness, and trust) and demographic
(age, gender, education) moderators stated in figure 3 moderate this effect. The next paragraphs develop the hypotheses that are to be analysed through an online survey.
M-shopping applications use: Information search and Purchase
Regarding the different usage situations that consumers face when shopping through a mobile application, this research will be only focused on the information search and purchase dimensions. The after-sales phase is excluded because in the researched industries (consumer electronics and airlines industries) the m-shopping applications are rarely used for after-sales services. In the information search phase, two types of shoppers can be identified: the competitive research shoppers, who search in one firm but purchase from a competitor, and the loyal research shoppers, who search and purchase from the same firm (Neslin et al., 2009). Therefore, considering the case of the loyal research shopper, when researching information through an m-shopping application, a consumer may feel loyal towards the brand and willing to buy from it either in the same mobile application or through another shopping channel. Consumers who research information in different channels but select the m-shopping application for the final purchase represent another option of customer loyalty. Specifically, highly involved shoppers are more likely to use the Internet for finalizing the shopping process (Seock and Bailey, 2008). So, one may imply that those customers who engage in purchasing through a mobile application, would engage in loyal attitude towards the firm because of its convenience and accessibility (Holmes et al., 2014)
Therefore, after the explained reasoning, the first hypotheses state as follows:
H1a: The use of m-shopping applications for information search purposes has a positive impact on attitudinal brand loyalty.
H1b: The use of m-shopping applications for purchase purposes has a positive impact on attitudinal brand loyalty.
H2: The positive effect of the use of m-shopping applications for purchase purposes on attitudinal brand loyalty is stronger than the positive effect of the use of m-shopping applications for information search purposes.
Innovativeness
All consumers are at some point innovators; there are some points in time when people adopt some objects or ideas that are new for them (Hirschman, 1980). Rogers and Shoemaker 1971 (as cited in Hirschman 1980) argued that innovativeness can be defined as the degree to which an individual is relatively earlier in adopting an innovation than other members of his social system. Moreover, to catalogue a product as an innovation, there are five characteristics in order to assess its innovativeness: relative advantage, compatibility, complexity, trialability, and observability (Rogers and Shoemaker 1971, as cited in Hirschman 1980). When shopping, consumers make adoption decisions regarding their perceptions of the relative advantage of the innovation (Choudhoury and Karahanna, 2008). Following this reasoning, one may think that perceived innovativeness of an m-shopping application can lead to achieve higher rates of brand loyalty. Furthermore, this argument could be possible if the statement by Hirschman (1980) still applies nowadays. She said that taking into account the individual propensity of consumers to adopt innovations might have an impact in the theories of brand loyalty (Hirschman, 1980). So, by considering m-shopping applications as innovations constantly changing, one may think that the innovative customers will be more willing to shop through them and become more loyal to these firms. Therefore, the third hypotheses state that:
H3: The use of m-shopping applications has a positive impact on attitudinal brand loyalty, and this effect is stronger for innovative people.
Impulsiveness
Impulsiveness when shopping online can be defined as, a consumer internal trait of quick response towards a given stimulus without deliberating about the action outcomes (Chih
et al., 2012). Impulsive buying is characterised by being unplanned, resulting from an exposure to stimulus, decided on the spot, and involving an emotional and cognitive reaction (Piron 1991, as cited in Lin and Chuan 2013). Customers find high levels of convenience and anonymity when engaging in online shopping, to the extent that the opportunities to buy impulsively increase (Chih et al., 2012). Moreover, the impulsiveness tendency will increase if the customer finds the online shopping an enjoyable activity (Floh and Madlberger, 2013), or if they find sufficient stimuli in the Internet environment (Kaisheng, 2010). In a survey from 2011 to British consumers, the impulsive behaviours of consumers in the bricks-and-mortars was deeply analysed in order to empirically support the previously mentioned statements. From the study it was found that, 76% of the respondents admitted to purchase groceries on impulse, 57% of the respondents purchased mid-cost products on impulse, and 28% of the respondents purchased high cost products on impulse (such as electronics or furniture) (Shoppercent, 2011, as cited in Floh and Madlberger 2013)). Following this logic, the same could happen in the m-shopping context, and even reporting higher rates, since it is a more convenient way of shopping that allows the consumer to shop impulsively whenever and wherever they want to. And as Donthu and Garcia (1999) stated, Internet shoppers are more convenience seekers, innovative, impulsive and less risk averse than traditional shoppers are. Therefore, since customers who buy impulsively would not pay much attention to the retailer where they buy from, their loyalty levels will turn out to be lower. Their impatience and compulsive reaction, altogether with the continuous availability of the mobile device, is expected to report buying patterns that will exhibit no clear loyalty to a specific firm. So, the fourth hypothesis states as follows:
H4: The use of m-shopping applications has a positive impact on attitudinal brand loyalty, and this effect is less strong for impulsive people.
Trustfulness
One of the biggest challenges that the online shopping has brought is the uncertainty of not seeing, touching or feeling the products before buying them and the fact that the vendor is always unobserved (Salo and Karjaluoto, 2007; Yeh et al., 2009). Therefore, trust in the online context has become extremely important, so as to deal with this uncertainty (McKnight et al. 2002, as cited in Salo and Karjaluoto 2007), and as stated by Salo and Karjaluoto (2007), trust is the most significant factor for success in an online environment. In the m-commerce context, trusting beliefs can be defined as consumers’ perceptions of particular characteristics of m-vendors when dealing with consumers’ orders (Kim and Benbasat 2003, as cited in Lin and Wang 2006). There are some additional reasons that enhance customers’ uncertainty when buying online, such as disclosure of credit card details (McKnight et al. 2002, as cited in Salo and Karjaluoto 2007), the risk of not receiving the product at all (Pavlou 2003, as cited in Salo and Karjaluoto 2007), the presence of unscrupulous or ineffective merchants, and the doubtful security of the Internet (Lee and Turban 2001, as cited in Salo and Karjaluoto 2007). Companies aim to achieve consumers’ trust with the expectation that it will translate into sales and profits (Rodgers and Harris, 2003). However, since e-commerce channels provoke higher perceived risk for many consumers (Liebermann and Stashevksy, 2002), trust is likely to play a major role when determining the future intentions and behaviours of consumers (Weisberg and Arman, 2011). So companies, in order to achieve higher trust perceptions from their customers, need to work in developing trusting beliefs towards the firm. Moreover, older people perceive greater risks when buying online (Hernandez et al., 2011), so it seems probable that they are the ones to distrust more online companies. And regarding gender characteristics, an exploratory study from Rodgers and Harris (2003) reported that, women trust e-commerce less than men, because they are more sceptical and do not find it as convenient as males. This logic relates to brand loyalty in the sense that, if customers are sceptical about the trustworthiness of
a specific firm, they would rarely engage in loyalty behaviours with the firm. And as found in Lin et al. (2006) study, trust has a positive effect on customer loyalty. Therefore, it could be implied that consumers buying through m-shopping applications would report in higher brand loyalty if they trust the providers from where they buy. Then, the fifth hypothesis states that: H5: The use of m-shopping applications has a positive impact on attitudinal brand loyalty, and this effect is stronger for trustful people.
Once the hypotheses have been formulated, the following section describes the methodology that is applied within this research.
4. METHODOLOGY
In order to be able to investigate the proposed research about how does the use m-shopping applications affect attitudinal brand loyalty, and to what extent the moderators influence the mentioned relationship, the study was carried out through an Internet mediated questionnaire. Specifically, the survey was related to the consumer electronics and airlines industries, in order to focus in the behaviour of customers on those specific industries. Although the questionnaire included both questions for behavioural and attitudinal loyalty, the study mainly considered the attitudinal brand loyalty data. The behavioural loyalty data was compiled as a control measure.
This chapter describes the research design. In the first part, the sample is described. In the second part, the data collection procedure is presented. In the third part, there is a description of the measures used in the research. And in the fourth part, the strengths and weaknesses of the research design are described.
4.1 The sample
The population for this study consists of online shoppers residents in the Netherlands and Spain, which use their mobile devices applications (smartphone or tablet) to shop online.
These countries were chosen because of having one of the highest smartphones penetration rate in Europe. Spain accounts the 72% of penetration rate of smartphones, while the Netherlands has 65% (Consumer Barometer, 2014). This population is estimated to be around 8 million people in total for both countries (Arayoweb.com, 2014; Ecommercenews.eu, 2014). Since the population is too broad, and makes it difficult to retrieve a probability sample, a convenience (non-probability) sample was chosen in this case. The main goal was trying to achieve a sample as large as possible, in order to have a more representative sample and generalizable outcome.
4.2Data collection procedure
The data was collected through a cross-sectional study by using an online survey. The online questionnaire represents the most adequate format of data collection, in order to obtain the required psychological and behavioural information. First, a pilot study with five respondents was conducted in order to assess the suitability of the questionnaire. After this pilot questionnaire, some of the questions were modified to make them more understandable. Second, the main study was launched online on the 19th of April. The expected response rate for an online survey of relative short length is 25% of the total amount of distributed questionnaires (Deutskens et al., 2004). A total of 156 respondents formed the research sample.
4.3 Measures
The questionnaire consisted of three parts. The first part, presented questions related to the consumer’s behaviour when shopping through mobile devices (smartphone or tablet). The measures of the questions were self-reported. First, the consumer was asked to think about which online brand does he/she use when shopping through a mobile device. Then they had to answer the following questions thinking about this brand. Some questions demanded information about whether or not the consumer used the mobile applications for different
purposes, then the answer was nominal by responding yes or no. Some attitudinal loyalty questions required a five point Likert scale answer, using an adapted scale of Arnold and Reynolds (2003). Moreover, respondents were asked to respond some questions about shopping frequency by using discrete ratio answers. This first part was first answered about mobile shopping of consumer electronics, and then repeated again for mobile shopping of flight tickets. The second part required answering several psychographic scales by using a five point Likert scale. The scales were related to the hypothesized moderators, namely: innovativeness, impulsiveness, and trustfulness. In order to measure the moderators, specific items from some scales were adopted. Innovativeness was measured using items from the scale developed by Manning, Bearden, and Madden (1995), impulsiveness through the scale of Valence, d’Astous, and Fortier (1988), and trustfulness with the scale of Yamagishi and Yamagishi (1994). Finally, respondents were asked to answer questions about their age, gender, and level of education. These 3 variables were measured using multiple-choice questions.
Scales Items Type of scale Source/adapted
from Loyalty I feel myself loyal to this firm
I plan to stay with this firm
5-point Likert scale
Arnold and Reynolds (2003)
Innovativeness I often seek out information about new products and brands
I frequently look for new products and services
5-point Likert scale
Manning, Bearden, and Madden (1995)
Impulsiveness There are times when I have a strong urge to buy (clothes, books, …) I have often bought a product that I did not need, while knowing that I have very little money
5-point Likert scale
Valence, d’Astous, and Fortier (1988)
Trustfulness Most people are trustworthy
Most people are basically good and kind
5-point Likert scale
Yamagishi and Yamagishi (1994) Table 3. Scales and items used in the questionnaire
4.4 Strengths and limitations
This research design faces some strengths and limitations. On the one hand, the strengths are associated with the ability to obtain large amount of data through the Internet, and the opportunity to obtain data from two countries so as to have a wider overview of the situation. On the other hand, the limitations are related to the limitation of the study by doing it cross-sectional and not being able to see the evolution over time, and the fact that the convenience sample could lead towards a biased sample of the population. Moreover, the use of some self-reported measures presents a limitation in the study.
5. RESULTS
In this chapter, the reliability of the scales and hypotheses was tested by performing several analyses in SPSS. The survey was distributed online for a period of 2 weeks. The result was a database of 177 respondents who fully completed the survey. 21 surveys were deleted because they answered that they did not use mobile applications for information search or purchase purposes in neither of the two industries. Therefore, there were 156 respondents who at least used an app in one of the two industries or in both of them. The database was duplicated in order to separate the respondents for the consumer electronics industry and the flight tickets industry; therefore, the missing data for either one or the other industry could be eliminated. Moreover, the analysis could be run separately for both industries. The final database consisted of: 117 respondents for consumer electronics and 137 for flight tickets.
5.1 Reliability and Validity
In order to achieve a meaningful final answer, the Cronbach’s alpha of the muli-item scale constructs was computed. The Cronbach’s alpha was tested in order to ensure if all items in one scale were measuring the same, or if some questions should be omitted for the analysis. The Cronbach’s alpha of the five scales analysed in the survey resulted to be >0.70, which is
the minimum value acceptable. Therefore, all the items were kept for further analysis, because they demonstrated to be internally consistent. Moreover, there were not any counter-indicative items.
Construct Cronbach’s alpha
Innovativeness 0.861
Impulsiveness 0.715
Trust 0.807
Loyalty towards electronic products’ companies 0.845 Loyalty towards flight tickets’ companies 0.851 Table 4. Reliability of scales
The validity of the scales is satisfied due to the fact that the scales were used from different studies that developed them. Innovativeness was developed in the study by Manning, Bearden, and Madden (1995), impulsiveness in the research of Valence, d’Astous, and Fortier (1988), and trust in Yamagishi and Yamagishi (1994).
5.2 Correlation check
The Pearson correlation coefficient was calculated in order to establish if there was any case of multicollinearity (Table 5 and 6).
The correlation coefficients that exceed 0.7 or -0.7 display a case of multicollinearity (Field, 2013). In this study, only the variable of the total usage showed multicollinearity, this was due to the fact that it represented the sum of the times used for information search and the times for purchase. Therefore, the variable total usage was kept within the model.
Mean SD 1 2 3 4 5 6 7 8 9 10 1.Times_IS 5.66 4.711 1 2.Times_PU 2.30 1.922 0.345** 1 3. Total_Usage 7.96 5.668 0.948** 0.630** 1 4. HowLong 0.40 0.492 -0.160* 0.018 -0.127 1 5. Loyal 3.38 0.847 0.139 0.109 0.152 0.360** 1 6. AppIS 3.94 0.903 0.021 -0.079 -0.009 0.016 0.092 1 7. AppPU 3.39 0.965 0.161* 0.220** 0.208* -0.027 0.211* -0.020 1 8. Innov 3.90 0.750 0.234** 0.183* 0.250** 0.007 0.220** 0.175* 0.080 1 9. Impul 3.29 1.081 0.156* 0.220** 0.205* -0.024 0.073 0.088 0.172* 0.084 1 10. Trust 3.05 0.786 -0.117 -0.173* -0.150* -0.120 -0.068 0.156* -0.055 -0.082 -0.071 1
Table 5. Correlations for electronic products data ** Correlation is significant at the 0.01 level (1-tailed) * Correlation is significant at the 0.05 level (1-tailed)
Mean SD 1 2 3 4 5 6 7 8 9 10 1.Times_IS 7.07 5.158 1 2.Times_PU 2.36 2.648 0.252** 1 3. Total_Usage 9.42 6.365 0.915** 0.620** 1 4. HowLong 0.36 0.483 0.026 0.070 0.050 1 5. Loyal 3.32 0.901 0.274** 0.159* 0.290** 0.163* 1 6. AppIS 4.18 0.688 0.260** -0.040 0.194* -0.003 0.220** 1 7. AppPU 3.36 1.069 0.126 0.131 0.157* 0.200** 0.350** 0.300** 1 8. Innov 3.77 0.851 0.062 -0.010 0.046 0.093 0.220** 0.210** 0.129 1 9. Impul 3.27 1.048 0.193* -0.052 0.134 0.119 0.048 0.090 0.019 0.175* 1 10. Trust 3.11 0.796 0.051 -0.018 0.034* -0.120 -0.101 -0.042 0.016 -0.023 0.028 1
Table 6. Correlations for flight tickets data ** Correlation is significant at the 0.01 level (1-tailed) * Correlation is significant at the 0.05 level (1-tailed)
Regarding the normality of the data some measures were taken. After analysing the skewness coefficients, several variables were normalized so as to have a normal distribution of the data. The skewness levels were assessed as adequate if they were compressed between 0.5 and -0.5, otherwise, the distribution of the data was not considered to follow a normal distribution (Field, 2013).
5.3 Model Testing
In order to test the hypothesis, a linear regression analysis was used, so as to determine the relationships between the predictor variables and the outcome variable. A hierarchical multiple regression was performed in order to analyse how 8 different independent variables (use for information search, use for purchase, appropriateness of the app for information search, appropriateness of the app for purchase, duration of the relationship with the company, level of innovativeness of the customer, level of impulsiveness of the customer, level of trust of the customer) and 6 moderated interactions (innovativeness and use for information search, innovativeness and use for purchase, impulsiveness and use for information search, impulsiveness and use for purchase, trustfulness and use for information search, trustfulness and use for purchase) were able to predict levels of attitudinal brand loyalty, after controlling for the variable gender. As mentioned, the moderators were included within the regression so as to have an integrated equation considering all the variables present in the model. The moderators are present in the equation through variables composed by the multiplication of the usage variables and the specific moderator. The multiple regression was first conducted for consumer electronics data and then duplicated the model but using flight tickets data.
The multiple linear regression model developed was constructed as follows:
ABLi=(b0+b1UseIS+b2UsePU+b3AppIS+b4AppPU+b5HowLong+b6Innov+b7Impul+b8Trust+
b9ModInnovIS+b10ModInnovPU+b11ModImpulIS+b12ModImpulPU+b13ModTrustIS+b14ModTrustPU+
Table 7. Explanation of the items within the regression
The following two subsections test the hypotheses for both industries analysed and elaborate on the acceptance or rejection of the hypotheses. Thus, the discussion of the results is now presented.
5.3.1. Results Consumer Electronics Database
Within the consumer electronics database, the multiple regression analysis was calculated as appears in Table 8. The model including the 14 variables and gender as a control variable explained a total variance of 33,1% F(14, 101) = .331; p<.001.
The first hypothesis states that the use of m-shopping applications for information search purposes has a positive impact on attitudinal brand loyalty. Thus, the coefficient from the regression for the information search use variable should be positive, in order to increase the level of loyalty for higher levels of usage. In this case, the coefficient of the use of m-shopping applications for information search purposes has resulted to be negative and non-significant (b=-2.19, p=.08). Therefore, H1a is rejected. Regarding H1b, the coefficient studying the use of m-shopping applications for purchase purposes should also be positive, so as to increase attitudinal brand loyalty for higher levels of purchase usage. However, in this study, the coefficient presented a negative and non-significant coefficient within the linear Abbreviation Variable
ABL Attitudinal Brand Loyalty
UseIS Times the m-shopping application was used for information search purposes
UsePU Times the m-shopping application was used for purchase purposes
AppIS The appropriateness of the m-shopping application for searching information
AppPU The appropriateness of the m-shopping application for purchasing
HowLong The duration of the relationship with the selected company
Innov Degree of innovativeness of the respondent
Impul Degree of impulsiveness of the respondent
Trust Degree of trustfulness of the respondent
ModInnovIS Innovativeness * Use for information search
ModInnovPU Innovativeness * Use for purchase
ModImpulIS Impulsiveness * Use for information search
ModImpulPU Impulsiveness * Use for purchase
ModTrustIS Trustfulness * Use for information search
ModTrustPU Trustfulness * Use for purchase
regression (b=-2.17, p=.18). Therefore, H1b is rejected.
H2 hypothesized that the positive effect of the use of m-shopping applications for purchase purposes on attitudinal brand loyalty was stronger than the positive effect of the use of m-shopping applications for information search. The regression shows that the coefficient for information search usage is negative and non-significant (b=-2.19, p=.08) while the purchase usage coefficient is negative related with loyalty and non-significant as well (b=-2.17, p=.18). Thus, neither of the two coefficients is positive, so since the first premise of having a positive relationship is not fulfilled, the hypothesis 2 is rejected. Apart from this reasoning, none of the coefficients resulted in a significant result.
Hypothesis 3 stated that the use of m-shopping applications, including information search and purchase purposes, had a positive impact on attitudinal brand loyalty, and that effect was supposed to be moderated by innovativeness. Thus, the effect should be stronger for innovative people. The interaction coefficient between the use for information search and innovative degree shows a positive but non-significant interaction effect, b=.27, 95% CI [-.44, .98], t=.76, p=.45. Moreover, the coefficient of interaction between purchase use and innovative degree was positive but non-significant as well b=.25, 95% CI [-.57, 1.08], t=.60, p=.55. Therefore, hypothesis 3 is rejected.
Hypothesis 4 considered that the use of m-shopping applications, including information search and purchase purposes, had a positive impact on attitudinal brand loyalty, and that effect was moderated by impulsiveness. Thus, the effect should be less strong for impulsive people. The interaction coefficient between the use for information search and impulsiveness resulted to be positive but non-significant, b=.25, 95% CI [-.18, .68], t=1.15, p=.25. Moreover, the coefficient of interaction between purchase use and impulsiveness was negative and non-significant as well, b=-.13, 95% CI [-.62, .35], t=-.55, p=.58. Therefore, hypothesis 4 is rejected.
Hypothesis 5 considered that the use of m-shopping applications, including information search and purchase purposes, had a positive impact on attitudinal brand loyalty, and that effect was moderated by trustfulness. Thus, the effect should be stronger for trustful people. The interaction coefficient between the use for information search and trustfulness resulted to be positive but non-significant, b=.37, 95% CI [-.26, 1.00], t=1.17, p=.24. Moreover, the coefficient of interaction between purchase use and trustfulness was positive but non-significant as well, b=.63, 95% CI [-.03, 1.30], t=1.89, p=.06. Therefore, hypothesis 5 is rejected.
5.3.2. Results Flight Tickets Database
Within the flight tickets database, the multiple regression analysis was calculated as appears in Table 9. The model explained a total variance of 28,2% F(14, 121) = .28; p<.001.
In order to test the hypothesis 1a, the coefficient of the use of m-shopping applications for information search purposes has resulted to be positive but non-significant (b=.07, p=.51). Therefore, H1a is rejected. Regarding H1b, the variable studying the use of m-shopping applications for purchase purposes presented a negative and non-significant coefficient within the linear regression (b=-1.21, p=.42). Therefore, H1b is rejected.
The multiple linear regression displays a coefficient for information search usage positive but non-significant (b=.07, p=.51) while the purchase usage coefficient is negative related with loyalty but non-significant as well (b=-1.21, p=.42). Therefore, hypothesis 2 is rejected.
For testing hypothesis 3, the interaction coefficient between the use for information search and innovative degree shows a negative and non-significant interaction effect, b=-.006, 95% CI [-.04, .03], t=-.35, p=.73. Moreover, the coefficient of interaction between purchase use and innovative degree was negative and non-significant as well b=-.04, 95% CI [-.60, .52], t=-.152, p=.88. Thus, not only the coefficients report the contrary to what was hypothesized but also they are not significant statistically. Therefore, hypothesis 3 is rejected.
