UNIVERSITY OF AMSTERDAM
Customers’ use of and preferences for branded mobile
shopping applications and its antecedents
Master thesis
Kimberley Deen, BSc
10004683
Supervisor: Umut Konus
MSc Business Studies
6/28/2014
Wordcount: 15.731
2
Table of Contents
1. Abstract ... p.5 2. Introduction ... p.6 3. Literature review ... p.9
3.1 Shopping via mobile devices ... p.9 3.2 Branded mobile shopping apps ... p.10 3.3 Effectiveness of branded mobile shopping apps ... p.11 3.4 Different types of branded mobile shopping apps ... p.12 3.5 Technology Acceptance Model ... p.14 3.6 Mobile app shopping antecedents ... p.14
4. Conceptual framework ... p.17
4.1 Four types of branded mobile shopping apps ... p.17 4.2 Online purchase frequency ... p.19 4.3 Differences between smartphone and tablet shopping ... p.20 4.4 Gender differences in shopping app use ... p.20 4.5 Perceived self-efficacy ... p.21 4.6 Consumers’ level of innovativeness ... p.22 4.7 Perceived benefits of branded mobile app shopping ... p.24
4.8 Consumers’ attitude toward branded mobile app shopping ... p.27
3
5.1 Selection ... p.29 5.2 Sample ... p.29 5.3 Research tool ... p.31 5.4 Measurements ... p.32 5.5 Measures perceived self-efficacy ... p.33 5.6 Measures level of innovativeness ... p.33 5.7 Measures benefits of branded mobile shopping apps ... p.34 5.8 Measures attitude toward branded mobile shopping apps ... p.34 5.9 Measures use of different types of branded mobile shopping apps ... p.35 5.10 Measures preferences for different types of branded mobile shopping apps ... p.36 5.11 Measures control variables ... p.37
6. Analyses and results ... p.38
6.1 Effect online purchase frequency ... p.38 6.2 Effect device owner ... p.39 6.3 Gender-effect ... p.40 6.4 Effect control-variables ... p.41 6.5 Effect perceived self-efficacy ... p.43 6.6 Effect level of innovativeness ... p.45 6.7 Effect perceived benefits of branded mobile app shopping ... p.48 6.8 Effect attitude toward branded mobile app shopping ... p.49 6.9 Mediating-effect of attitude toward branded mobile app shopping ... p.51
4 8. Managerial implications ... p.58 9. Limitations ... p.59 10. Further research ... p.60 11. References ... p.61 12. Appendix A: Questionnaire ... p.70
5
1. Abstract
Mobile users increasingly use branded mobile applications (apps) for shopping purposes. Since there has been done limited research to the antecedents of consumers’ use of and preferences for branded mobile shopping apps, this study tries toimprove the explanatory power of the variance in consumers’ use of and preferences for these apps. A distinction was made between four different types of apps: informational, experiential, location-based, and transactional shopping apps. An online survey study was conducted to collect data from 331 Dutch mobile apps users. The findings indicated that, in general, consumers used and preferred informational and transactional shopping apps more than experiential and location-based shopping apps. It appeared that consumers’ self-efficacy, their level of innovativeness, perceived benefits of mobile app shopping and their attitude toward mobile app shopping positively affected the use of branded mobile shopping apps. However, the effect of the perceived self-efficacy and the attitude toward mobile app shopping varied between the four different types of shopping apps. The results showed that the attitude toward mobile app shopping mediated the effect of the perceived self-efficacy, level of innovativeness and benefits of mobile app shopping on the use of shopping apps. The findings of this study may intrigue practitioners to develop more advanced mobile shopping apps that are based on the desires and needs of their customers. It may also help companies to invest in appropriate marketing strategies, communication activities and the ability to allocate their marketing resources more effectively.
6
2. Introduction
Due to the popularity of Internet the constant developments in mobile technology, along with the fast growth and inherent characteristics of mobile devices, i.e. smartphones and tablets, mobile commerce has emerged as a new, potential instrument for marketing activities (Varnali & Toker, 2010). Mobile commerce (M-commerce) can be defined as transactions, either direct or indirect, with a monetary value, carried out via a wireless telecommunication network, e.g. online shopping and banking (Ko, Kim & Lee, 2009). M-commerce can be considered as an extension of e-commerce that delivers omnipresent value by offering ease and accessibility at any given time and moment (Balasubramanian, Peterson, & Jarvenpaa, 2002). According to the IBM Retail Online Index (2013), the online sales percentage of mobile shopping increased more than 60 % in 2013 compared to the percentage in 2012. This has attracted many companies to develop their own consumer apps, i.e. branded mobile apps, as an integral part of their marketing strategy (McKinsey & Company, 2011). Especially the number of branded shopping apps that are available on mobile platforms, such as Apple’s App Store and the Blackberry App World, is growing rapidly since companies saw the business potential of the mobile channel. According to a research from Nielsen (2012a), 47% of the mobile phone owners use branded apps for shopping purposes, e.g. investigating products and reviews, comparing prices, finding store locations, and redeeming coupons. The eBay app appears to be one of the most-used branded app for shopping purposes attracting more than 13 million unique users within one month.
Evidently, mobile devices have become more and more popular as a shopping medium. The penetration of branded mobile shopping apps caused many changes in the possibilities regarding advertising, retailing and online shopping in a marketing context. A limited comprehension of consumer behavior in the mobile technology-mediated shopping environment impedes companies in developing proper shopping services in this early stage of
7
mobile shopping via apps (Yang & Kim, 2012). The unique characteristics of mobile
shopping apps may create diverse consumer wants and needs in comparison to other shopping channels, e.g. in-store, online shopping via a laptop or desktop computer. Companies should therefore stay informed about the latest developments in mobile shopping app technology and simultaneously be aware of the factors that could influence consumers’ shopping use of these apps.
Since there is a lot of business press but very limited academic literature in the area of branded mobile shopping apps (Thomson, Rao Hill & Carrillat, 2013), academic studies that extend the knowledge about mobile shopping apps, consumers’ behavior and the motivators to use shopping apps are highly relevant for practitioners and academics. Insights obtained from earlier studies are still fragmented, indecisive and are often not applicable (Smutkupt, Krairit & Esichaikul, 2010), which is mainly due to the newness of this phenomenon (Ong, 2010), and the lack of experience in developing and using branded mobile apps as a
marketing tool for shopping among practitioners. Considering the fact that there is a large amount of branded apps available on smartphone platforms, it might be helpful to first categorize those apps. Hence, the goal of this article is to find out what the differences in consumers’ preferences for and use of different types of branded mobile apps are, and what main factors are driving these differences. Hereby the focus lays on branded shopping apps that are downloaded by consumers on their mobile phones or tablets. By performing a survey methodology, this research attempts to answer the question of whether and how consumers could differ in terms of their preferences and use of different types of branded mobile shopping apps and to what extent consumers’ perceived self-efficacy, their level of innovativeness, attitude toward mobile shopping, and their perceived benefits of mobile shopping via apps are driving this.
8
and practice. Theoretically, it extends the literature regarding branded mobile apps by focusing on its implications from the customer’s perspective regarding their use of and
preferences for branded shopping apps. The results of this study highlight relevant factors that may have an influence on whether consumers use and prefer certain types of branded mobile apps for shopping purposes. Moreover, it provides basic knowledge about the influence of motivators and the attitude of mobile shoppers to use certain types of shopping apps, which scholars can further elaborate on in future research. This study also enlarges the existing knowledge about the influence of certain characteristics of consumers with regard to their use of and preferences for certain types of apps. Practically, since mobile shopping via branded apps increases among consumers (Nielsen, 2012a), it is important for companies to
understand how consumers may differ in their use of and preferences for different types of branded apps in order to maximize the effectiveness of their marketing strategy. This study offers marketers a conceptual framework of important factors that may explain how
consumers differ in their mobile shopping behavior via apps. This framework may help incorporating apps into existing marketing strategies, which give organizations the possibility to improve their competitive advantage (Smutkupt, Krairit & Esichaikul, 2010). Insights in the antecedents of shopping via branded mobile apps may help companies investing in the right marketing strategy and communication activities and more effectively allocate their marketing resources. The findings of this study may also intrigue practitioners to develop more advanced mobile shopping apps that are based on the desires and needs of their customers. Thus, this study helps companies to develop successful mobile shopping apps, which may increase consumers’ purchase behavior and their engagement with the brand.
This article is structured as follows: the first section contains a literature review of findings from earlier studies that are related to the subject of this study, which will descend in the conceptual framework of this research, based on hypotheses. Then the methodology of
9
this research will be described, followed by the data analyses and results. The paper continues with the discussion of the results, the managerial implications of the findings, and ends with the limitations and suggestions for future research.
3. Literature review
3.1 Shopping via mobile devices
As previously mentioned, shopping via mobile devices is a growing phenomenon among consumers. The versatility of mobile shopping services is increasingly appealing to
consumers (Yang, 2012). Mobile shopping on a smartphone or tablet is different from online shopping via a desktop computer or laptop because it does not contain any temporal and spatial constraints, which enables mobile users to shop any moment and place they want (Heinonen & Pura, 2006; Wu & Wang, 2006). It appears that consumers engage in mobile shopping because it increases the shopping efficiency by saving time and reducing customers’ costs and effort (Yang & Kim, 2012). These authors find that consumers engage in mobile shopping because of idea, adventure, and gratification motivations. In particular, consumers engage in mobile shopping if they want to collect ideas about e.g. new trends and fashion, and product information, look for excitement, stimulation, and new experiences, and experience positive emotions and special treats.
There are differences found in the way consumers use smartphones and tablets for shopping purposes. Generally, tablets are more often used (55 %) for buying products and services than smartphones (28 %), which may be due to the size of the screen (Adobe, 2012), security issues and the quality, i.e. speed, of the Internet connection (Forrester, 2012). Tablets are mainly used in the digital home, whereas smartphones are more consistently distributed and wider used. Thus, it is assumable that tablet owners more often engage in shopping-related activities at home than smartphone owners. Forrester (2012) also finds a general
10
allegiance among smartphone owners to the brand of their mobile phones when buying a tablet, e.g. 83 % of European Iphone owners bought an Ipad. It also appears that smartphone ownership can be linked to tablet ownership.
3.2 Branded mobile shopping apps
Consumers can shop on their mobile device via an Internet browser or via an app, whereby they prefer using online retailers’ apps more than the mobile website version of it (Forrester, 2013). A mobile app is a form of software that works on a mobile phone, also known as a smartphone, and performs particular tasks for a user (Mobile Marketing Association, 2008). Apps use content and data from the Internet, like websites, or download the content in order to provide users access to a certain app without an Internet connection (Compuware, 2013). Mobile app users can download apps via device-specific platforms, such as Apple’s App Store, Android Market or Blackberry App World. Apps can range from tools for chatting and e-mailing, to maps and direction finders, through to online shopping programs, and games (Bellman et al., 2011). Gupta (2013) divides mobile phone apps into five categories: games and entertainment, social networks, utilities (e.g. maps, calendars, e-mail), discovery (e.g. Tripadvisor, Flixster), and brands (e.g. Nike app).
Especially the last category mentioned above, i.e. branded apps, is valuable for companies. Branded mobile apps are conceptually defined as software downloadable to a mobile device, which prominently demonstrates a brand identity, mostly via the name of the app and the presence of a brand logo or icon, throughout the user experience (Bellman et al., 2011). In this study the focus only lays on branded mobile apps that can be used for shopping purposes, such as information searching, and ordering products. Branded shopping apps are a form of ‘pull’ rather than ‘push’ advertising (Bhave, Jain & Roy, 2013). In other words, the consumer approaches the brand instead of the more traditional way whereby the brand
11
approaches their consumers (Shankar, Venkatesh, Hofacker & Naik, 2010). Consumers themselves determine to which apps they are exposed on their mobile phones, and they control how much information they disclose by customizing the app (Bhave, Jain & Roy, 2013).
3.3 Effectiveness of branded mobile shopping apps
According to Bellman et al. (2011), the overall perceptions of engagement with a branded app depend on the number and intensity of the experiences a consumer can have with a certain app. They conclude that the use of branded apps has a positive persuasive impact, increasing interest in the brand and also in the brand’s product category. Additionally, Gupta (2013) argues that these apps may be more effective than traditional ads because consumers do not perceive those apps as advertising. Consumers use the app for its functionality, and therefore do not perceive it as intrusive. Moreover, these apps add convenience in the sense of letting consumers perform certain tasks, e.g. track bank balances and pay bills, check in and monitor flight statuses etc, more quickly and smoothly than traditional ways via desktop computers, laptops or mobile browsers. They also provide functions and benefits traditional desktop computers, laptops and mobile browsers cannot. For example, in South Korea, where the UK-based retailer Tesco offers a grocery delivery business named Home Plus, the
company placed lifelike photos of products on the store shelves on the walls of subway stations with QR-codes that were linked with Home Plus’ mobile app. Customers who scanned the QR-codes could do their groceries and arrange the delivery while waiting for their trains. The app has been downloaded more than a million times since it was launched. It appears that branded apps have a major influence on the favorability of consumers’ attitude toward the brand, but only a relatively small effect on purchase intention (Bellman et al., 2011, Crawford, Gosling, Bagnall & Light, 2014). Neither the familiarity with the brand
12
nor the relevance of the product category made a difference in the effectiveness of the
branded mobile apps (Bellman et al., 2011). Bhave, Jain & Roy (2013) found that particularly Gen Y, i.e. individuals born between 1980 and 2000, has a positive attitude toward branded apps. Especially interactive and engaging mobile apps of their favorite brands and brand story-based games were highly appreciated. The interactive features that these apps provide highlight a symbolic connection to the company through it (Crawford et al., 2014). Gen Y also appears to be interested in service-oriented branded apps, such as shopping apps,
location-based apps, and deals- and discount-based apps (Bhave, Jain & Roy, 2013). Thus, it is assumable that certain types of branded shopping apps may be more effective than other types of branded shopping apps.
3.4 Different types of branded shopping apps
Remarkably, there has not been done a lot of research regarding the existence of different types of branded mobile shopping apps. Therefore it is still not clear whether certain types of shopping apps are more effective than others. Preliminary research focuses more on branded mobile apps in general, while there may exist large differences regarding functions and features between one certain branded app and another. Only Bellman et al. (2011) seem to be aware of the fact that the type of branded shopping apps may influence e.g. consumers’ attitude and purchase behavior. These authors investigate whether using popular branded mobile apps affect consumers’ brand attitude and brand purchase intention, whereby they make a distinction between two different execution styles of branded apps: informational experiences and experiential experiences. The first includes utilitarian branded apps, whereby the emphasis lays on providing informational content to customers. The latter includes
intrinsic branded apps, whereby the focus is on providing enjoyment/entertainment
13
regarding users’ mobile shopping activities, i.e. content delivery, transactions, location-based services, and entertainment. According to Chong, content delivery involves using a mobile device to search for information, such as product descriptions and comparing product prices. Transactional activities include transferring money between consumers and businesses via a mobile device, e.g. paying for products a consumer ordered via a smartphone or tablet.
Location-based services deal with activities that include receiving time- and location sensitive information, such as an app that provides information about the nearest stores. Lastly,
entertainment includes using a mobile phone for hedonic purposes, e.g. watching branded videos of a company. Bellman et al.’s informational experiences are comparable to Chong’s content delivery activities and the same counts for Bellman et al.’s experiential experiences and Chong’s entertainment-related activities. Generally, it appears that consumers mostly engage in content delivery shopping activities via their mobile device, i.e. browsing products, (in store) price comparison, and reading product reviews (Nielsen, 2013; Smith & Sivakumar, 2004). Bellman et al. find that branded apps with an informational style were more effective at shifting purchase intention than experiential apps, probably because this style is very user-centered, and therefore stimulates users to make personal connections with the brand. Additionally, findings of Crawford et al. (2014) indicate that customers prefer an app that simply provides information and an option to purchase products. On the contrary, experiential branded apps were less effective in shifting consumers’ purchase intention, most likely
because they focus attention on the mobile device (Bellman et al., 2011).
Since most of the earlier mentioned studies focus on the effect of branded mobile apps, it is still not clear which factors determine whether consumers engage in mobile shopping and what causes the differences in their mobile shopping behavior. In order to comprehend what makes different types of mobile shopping apps effective in a marketing context, it is firstly crucial to understand the antecedents of mobile app shopping. In the next
14
paragraphs this study discusses the main drivers of shopping via mobile apps.
3.5 Technology Acceptance Model
Whether consumers engage in mobile shopping activities may be explained by their
willingness to adopt it. The adoption of mobile shopping apps may be clarified on the basis of the Technology Acceptance Model (TAM) of Davis (1989), which states that perceived ease of use and perceived usefulness can predict the usage of new technology (Aldás-Manzano, Ruiz-Mafé & Sanz-Blas, 2009; Pagani, 2004). Perceived ease of use can be explained as the extent to which an individual believes that using a particular technology would be effortless (Davis, 1989). Perceived usefulness comprises the degree to which an individual believes that using a certain technology would improve the job performance. Davis, Bagozzi & Warshaw (1989) also find that perceived enjoyment has an impact on the adoption of new technology, whereby a positive interaction between perceived usefulness and perceived enjoyment was found. Perceived enjoyment can be defined as the believing of an individual that the new technology is fun to use. Moreover, these authors suggest that consumers’ perceived ease of use, perceived usefulness, and perceived enjoyment determines their attitude toward the new technology, which in turn determines whether or not they are willing to adopt a certain new technology.
3.6 Mobile app shopping antecedents
According to earlier findings regarding mobile shopping adoption, consumers’ perceived enjoyment and perceived usefulness cause positive attitudes toward mobile shopping, whereby the first was a stronger determinant in affecting a positive attitude (Lu & Su, 2009; Yang, 2012. In particular, consumers may more frequently be involved in mobile shopping activities if they find them enjoyable and easy to use (Chong, 2013). Contrarily,
Aldás-15
Manzano, Ruiz-Mafé and Sanz-Blas (2009) find no evidence for the effect of perceived usefulness on the intention to shop via mobile devices, even though it has a direct effect on consumers’ attitude. The differences in these findings may be explained by the type of shopping activity a consumer performs (Chong, 2013). Moreover, perceived enjoyment appears not to be relevant for users regarding transaction-based activities. In this case, users find the ease of use and perceived usefulness more important than perceived enjoyment. In general, perceived usefulness may positively affect users’ mobile shopping activities, except for location-based activities. Location-based services include e.g. receiving mobile
advertising and finding stores close to consumers’ current geographical location, whereby users may not perceive this as useful.
Based on earlier studies regarding mobile shopping, it appears that, besides the TAM-drivers (i.e. perceived usefulness, perceived ease of use, and perceived enjoyment), many new antecedents were added to improve the explanatory power of differences in consumers’ mobile shopping behavior (Li, Dong & Chen, 2012). These antecedents include user psychographics (e.g. anxiety, attitude, innovativeness, compatibility, affinity, perceived values, social influence, behavioral control), and demographics (e.g. age, education). An overview of these antecedents is provided in Figure 1, whereby only direct effects are displayed. Considering the fact that these findings are found for mobile shopping in general instead of mobile shopping via branded apps, these findings may not be completely applicable to the focus of this study. Since there has been done limited research to the antecedents of consumers’ use of and preferences for branded mobile shopping apps, this study tries to improve the explanatory power of the variance in consumers’ use of and preferences for these apps. Based on preliminary findings, it appears that important predictors to adopt new mobile technologies include the level of consumer’s innovativeness (Cowart, Fox & Wilson, 2008; Pagani, 2004; Yang, 2010; Yang, 2012) and the perceived benefits of the new technology
16
(Eastlick & Lotz, 1999). Furthermore, consumers’ perceived self-efficacy has also been empirically supported as an important antecedent in the adoption of new technologies
(Dabholkar & Bagozzi, 2002; Yang, 2012). Yang (2012) also found a strong relation between the attitude toward mobile shopping and the intention to engage in mobile shopping.
Therefore, this study focuses on the impact of consumers’ level of innovativeness, their self-efficacy, perceived benefits of and attitude toward mobile app shopping on their use of and preferences for branded mobile shopping apps, while controlling for multiple demographical variables, i.e. gender, age and education. Since there appear to be differences in shopping behavior with regard to the online purchase frequency, type of device, i.e. tablet or
smartphone, and the brand of the device consumers own (Adobe, 2012; Forrester, 2012), these factors will also be taken into account as control variables. In the next chapter the expected relationships between the variables of this study will be discussed, whereby hypotheses will be drawn.
17 Notes:
1
Significant effect found by Aldás-Manzano, Ruiz-Mafé and Sanz-Blas (2009)
2
Significant effect found by Ko, Kim and Lee (2009)
3
Significant effect found by Lu & Su (2009)
4
Significant effect found by Chong (2013)
5
Significant effect found by Yang (2010)
6
Significant effect found by Yang (2012)
4. Conceptual framework
4.1 Four types of branded mobile shopping apps
Before discussing the antecedents of this research with regard to the use of branded mobile shopping apps, this study first makes a distinction between four types of shopping apps. The first two types of apps are based on the classification of Bellman et al. (2011), i.e.
informational and experiential branded mobile apps. However, it is noteworthy that they only focus on two different execution styles of branded apps in their research, since Chong (2013) clearly finds that mobile users perform four different mobile shopping activities on their mobile devices, which implies that there also may exist four different types of branded mobile apps. In order to get a better picture of the different conventional branded shopping apps that are available for customers, this article bundles the findings of Bellman et al. and Chong. This research focuses on four main execution styles of branded shopping apps. The first style is an informational app, which is defined as a utilitarian branded shopping app, whereby the emphasis lays on providing informational content to customers. For example, the Dutch company Kieskeurig.nl developed an app (Productchecker) which makes consumers able to e.g. search for information and reviews about different kind of products, scan barcodes, compare products based on their characteristics and price and easily share this comparison with others. The second style is an experiential app, whereby the focus is on providing hedonic, i.e. enjoyable, experiences to customers. According to Higgens (2006), adding sensual elements, i.e. auditory, visual and interactivity, is crucial in order to create positive
18
experiences when a company tries to appeal to consumers’ senses. Therefore, this research distinguishes experiential branded apps from other execution styles by the presence of sensual elements, such as graphics, sounds and animations (Kim, Ling & Sung, 2013). A clear
example of an experiential branded app is the Mercedes-Benz Vehicle app, whereby consumers can explore the different cars through virtual tours, interactive galleries, vehicle customization tools and multiple videos that show the innovations of the cars.
The third and fourth style of branded apps are based on Chong’s (2013) findings regarding mobile users’ mobile shopping activities. The third style this study focuses on is a transactional app, which is defined as a branded shopping app that focuses mainly on
transacting business between the customer and the company, e.g. scheduling appointments, purchasing products, and processing payments. A typical example of a transactional app is the ABN Amro app, which consumers can use to manage their banking and make transactions. The last execution style is a location-based app, which is defined as a branded shopping app that focuses on providing time- and location sensitive information to customers, e.g.
information about nearest stores, local actions etc. For example, the Dutch company Eet.nu developed an app which provides users with actual information, e.g. visiting hours, addresses, reviews, pay methods, of restaurants that are close to their current location. Figure 2 provides an overview of the four earlier mentioned types of branded mobile shopping apps. In the next paragraphs the antecedents of the use of the four different types of branded mobile shopping apps will be discussed, starting with consumers’ online purchase frequency, type of device and gender. Figure 3 provides the conceptual model with the corresponding hypotheses of this study.
19
Figure 2: Summary different types of branded mobile shopping apps.
4.2 Online purchase frequency
Based on Rogers’ technology cluster concept (1995), which states that the compatibility affects whether people adopt a new technology, it is likely that past experiences with distant shopping methods positively influence shopping via a mobile device. Consistently, Aldás-Manzano, Ruiz-Mafé and Sanz-Blas’s (2009) findings suggest that consumers who previously shopped via the Internet are more likely to engage in mobile shopping. Thus, it may be
assumable that individuals with Internet shopping experiences are more likely to engage in mobile shopping via branded apps than those who have never used Internet as a shopping channel. Therefore, it is also assumable that the more often consumers shop via the Internet, the more they use branded mobile apps for different shopping purposes. Thus, this study expects that the frequency of Internet purchasing has a positive influence on the use of different types of branded mobile shopping apps, i.e. informational, experiential, location-based, and transactional apps. Based on this expectation, the following hypothesis is formulated:
H1: The more consumers shop via the Internet, the more they use a) informational, b)
experiential, c) location-based, and d) transactional shopping apps, and vice versa. Informational shopping apps: provide informational content to customers. Experiential shopping apps: provide enjoyable experiences to customers by the
presence of sensual elements.
Location-based shopping apps: provide time- and location sensitive information to customers.
Transactional shopping apps: transact business between customer and companies.
20
4.3 Differences between smartphone and tablet shopping
Furthermore, since compatibility affects whether people adopt a new technology, the mobile device(s), i.e. smartphone and/or tablet, people own may determine the extent of branded mobile shopping app use. Tablets are more used for buying products and services than smartphones, which may be due to the size of the screen (Adobe, 2012), security issues and the quality, i.e. speed, of the Internet connection (Forrester, 2012). Therefore, tablets may be more compatible to desktops than smartphones. Based on earlier findings, consumers who previously shopped via a desktop may be more likely to engage in mobile shopping via a tablet than via a smartphone. Furthermore, since smartphones are probably compatible to tablets, it is assumable that consumers who shop via branded apps on their tablets are also more likely to engage in mobile shopping via branded apps on their smartphones. Consumers that own a tablet, next to their smartphone, are therefore expected to use branded mobile shopping apps more than consumers that only own a smartphone. Thus, tablet ownership may positively influence the use of different types of mobile shopping apps, i.e. informational, experiential, location-based, and transactional apps. In order to test this assumption, the next hypothesis is drawn:
H2: Consumers that own both a tablet and a smartphone use a) informational, b) experiential,
c) location-based, and d) transactional shopping apps more than consumers that only own a smartphone.
4.4 Gender differences in shopping app use
According to a research of uSamp (2012), it appears that men are more likely to engage in mobile shopping than women. Furthermore, men more often write reviews of a product they purchased via their mobile device than women. The results also indicate that men more often made a mobile purchase or used their mobile device for payments in store than women.
21
Additionally, Bellman et al. (2011) find that males may have an increased likelihood of generating personal connections with the brand when using informational apps in comparison to experiential apps. The finding of different results for men and women may indicate gender differences in engagement with branded mobile apps. Therefore, this study assumes that men more often use informational and transactional shopping apps than women do.
H3: Men more often use a) informational- and b) transactional shopping apps than woman.
4.5 Perceived self-efficacy
Consumers’ perceived self-efficacy has been empirically supported as an important
antecedent in the adoption of new technologies (Dabholkar & Bagozzi, 2002; Yang, 2012). Perceived self-efficacy is defined as the belief in someone’s ability to confidently achieve positive task outcomes by performing a specific behavior (Compeau & Higgins, 1995). In addition, Dabholkar and Bagozzi (2002) find that consumers’ perceived self-efficacy can facilitate their beliefs of the functions of the new technology as funtouse. Moreover, the higher consumers’ degree of perceived self-efficacy, the more likely they are in perceiving functions and features of mobile app shopping as funtouse. The ability to use a new
technology may therefore positively affect the degree of use of the technology, thereby evoking the adoption of this technology (Compeau, Higgins & Huff, 1999). Thus, consumers’ perceived self-efficacy may positively affect the use of different types of branded mobile shopping apps. Hereby, the next hypothesis is drawn:
H4: The higher consumers’ degree of perceived-efficacy is, the more they use a)
informational, b) experiential, c) location-based, and d) transactional shopping apps, and vice versa.
22 4.6 Consumers’ level of innovativeness
Based on preliminary findings, it appears that another important antecedent to adopt new mobile technologies include the level of consumer’s innovativeness (Cowart, Fox & Wilson, 2008; Pagani, 2004; Yang, 2010). Consumers tend to be innovative by selectively adopting new products based on attributes they find interesting (Bartels & Reinders, 2011). Preliminary studies reveal confounding results about antecedents of innovativeness. Hoffmann and Soyez (2010) find two important limitations: earlier research conceptualized innovativeness on various degrees of abstraction and did not take into account the effect of the product category. These authors argue that constructs of innovativeness on a high level of abstraction may possibly fail to forecast certain behavior because innovative consumers of a certain product may not be willing to adopt other products. Therefore, various scholars suggest to measure innovativeness within a domain-specific level of interest (Ajzen, 1991; Hoffmann & Soyez, 2010; Lastovicka & Joachimsthaler, 1988). Goldsmith & Hofacker (1991) defined domain-specific innovativeness as the propensity to learn about and adopt innovations within a particular product category. This study adopts this definition for consumers’ level of
innovativeness since branded mobile apps belong to a specific product category. In particular, it is likely that adopting branded apps may not be comparable to adopting products from another product category.
As mentioned earlier, preliminary studies find that consumers’ level of innovativeness positively affects the behavioral intentions to adopt a new product (Cowart, Fox & Wilson, 2008; Pagani, 2004; Yang, 2010; Yang, 2012). In particular, consumers’ level of
innovativeness appears to have a positive effect on the intention to adopt mobile shopping (Aldás-Manzano, Ruiz-Mafé & Sanz-Blas, 2009; Lu & Su, 2009). Thus, individuals that possess a high level of innovativeness are more likely to engage in mobile shopping than
23
consumers with a low level of innovativeness. Highly innovative people generally favor technology and the use of high-tech products (Dabholkar & Bagozzi, 2002), such as different types of branded mobile shopping apps. Consequently, the higher the level of a mobile app user’s innovativeness is, the more they perceive branded shopping apps as favorable, whereby they are more willing to explore and adopt them for shopping purposes. In contrast, the lower consumers’ level of innovativeness is, the less they perceive branded shopping apps as
favorable, whereby they prefer not to explore and adopt them for shopping purposes as easily. Generally, mobile users with a higher level of innovativeness will probably use branded mobile shopping apps more than users with a lower level of innovativeness. Thus, this
research expects consumers’ level of innovativeness to have a positive effect on the use of and preferences for branded shopping apps.
H5: The higher the level of consumers’ innovativeness is, the more they use a) informational,
b) experiential, c) location-based, and d) transactional shopping apps, and vice versa.
According to Rogers (1995), there are five user segments based on the propensity to adopt new technologies, i.e. innovators, early adopters, early majority, late majority and laggards. According to Pagani’s findings (2004), the innovators include the first customers that adopt brand-new mobile technologies, such as mobile shopping. These customers are mainly young people (18-24) that are looking for low costs and convenience. They are highly innovative and optimistic about mobile technologies and feel secure and comfortable adopting it. Early adopters are mainly innovative and optimistic professionals, i.e. managers and
entrepreneurs. They adopt mobile technologies based on its usefulness. Their level of
insecurity and discomfort may be high in case of adopting new mobile services. The segment early majority includes persons who are not very optimistic and innovative regarding adopting new mobile technologies but they do feel comfortable and secure adopting it. The late
24
majority is pessimistic about their competence to obtain value from new mobile technologies. Their level of innovativeness is low, while their level of insecurity and discomfort is high. The last group, i.e. laggards, is in general very critical and not interested in new mobile technologies. Their level of insecurity and discomfort is high in case of adopting new mobile services. Moreover, they are not innovative by nature and very pessimistic regarding new technologies.
Since low innovative people are generally pessimistic about obtaining value from new mobile technologies and their level of insecurity and discomfort is high in case of adopting new technologies (Pagani, 2004), they probably prefer apps that offer them clear benefits, e.g. hedonic and aesthetic value, and security, whereby no monetary transactions are involved. Thus, this study expects that mobile users with a low innovative character prefer
informational and experiential branded apps more than they prefer transactional and location-based branded apps. Based on these assumptions, the next hypothesis is formulated:
H6: Consumers with a low innovative character prefer informational and experiential
branded apps more than they prefer transactional and location-based branded apps.
4.7 Perceived benefits of branded mobile app shopping
Besides the importance of consumers’ level of innovativeness as a driver of mobile shopping adoption, preliminary studies also find that the perceived value of mobile shopping is an important construct for understanding consumer responses to mobile shopping (Ko, Kim & Lee, 2009). The distinct attributes of mobile shopping may affect its value perceived by a consumer through the benefits yielded from mobile shopping. Distinct attributes of mobile shopping may be ubiquity, mobility, convenience, personalization, flexibility, and
dissemination (Yang, 2010). Based on preliminary research, mobile shoppers may associate several benefits with buying on the Internet and via mobile devices, e.g. fast shopping, low
25
costs in terms of time, money and effort, privacy, safety, shopping enjoyment, and high product quality (Table 1). To attract consumers to mobile shopping, consumers should have the feeling that they can get a better deal (Wang, Yeh & Jiang, 2006). The perceived benefits of a shopping innovation over traditional shopping channels is a strong predictor of whether people are willing to engage in the new shopping channel (Eastlick & Lotz, 1999). Since Koyuncu and Bhattacharya (2004) state that the frequency of mobile shopping may increase with consumers’ perceived benefits of it, it is assumable that the more consumers experience certain benefits of shopping via branded mobile apps, the more they use these apps.
Contrarily, the less benefits consumers experience, the less likely they are in using branded mobile shopping apps. In other words, the perceived benefits of mobile shopping via branded apps may have a positive effect on the use of these apps. Hereby, the next hypothesis is formulated:
H7: The more benefits consumers experience regarding mobile app shopping, the more they
use branded mobile shopping apps, and vice versa.
Table 1: Overview perceived benefits of mobile shopping
Authors Study focus Method Relevant findings
Keeney (1999) The pros and cons of using E-commerce from a customer perspective. Individual and group interviews (N = 100) across 20 countries
Product quality, cost, delivery time,
convenience, time spent, privacy, shopping enjoyment, safety, and
environmental impact are the most fundamental values for Internet shopping. Koyuncu & Bhattacharya (2004) The effects of quickness, price, payment risk, and delivery issues on online shopping
Survey data set collected by Georgia Institute of Technology (N = 1842) Internet shopping provides better prices and ensures faster shopping than other shopping alternatives.
26
Contrary, Internet shopping requires longer delivery time for products bought on the Internet and involves more risks with regard to payments via Internet. Overby & Lee
(2006)
The relevancy of value dimensions for
online shopping and the relationship between value dimensions, preference toward the Internet retailer, and intentions
Web-based survey among Internet users (N = 817)
Consumers shop online primarily for utilitarian reasons, i.e. price
savings and convenience.
Pagani (2004) Critical factors that may influence the adoption of mobile services
24 focus groups with mobile phone users in six
different countries and phone
questionnaires among
Italian mobile users (N = 1000).
The most important determinants of multimedia mobile services are
respectively usefulness, ease of use, price, and speed of use.
Wang, Yeh & Jiang (2006)
Fundamental objectives of Internet shopping
Online web survey among online users (N = 712)
Privacy, safety and product quality were the three most important Internet shopping values perceived by the respondents. Yang & Kim (2012) Differences in
mobile shopping motivations and the driving motivations to use the mobile shopping channel.
Online survey among
mobile service users (N = 400)
The mobile shopping motivations suggest that the main benefits of mobile shopping are
shopping enjoyment, flexibility,
personalization, and saving time, money and effort.
27
4.8 Consumers’ attitude toward branded mobile app shopping
Based on Petty, Unnava & Strathman (1991) findings, this study defines consumers’ attitude toward branded mobile app shopping as consumers’ global and relatively consistent
evaluations, feelings, and tendencies toward shopping via branded mobile apps. According to the TAM, the attitude toward a specific new technology, predicts the individual’s use of the technology (Davis et al., 1989). In line with this, earlier research shows that attitude has a positive effect on the intention to shop via a mobile device (Aldás-Manzano, Ruiz-Mafé & Sanz-Blas, 2009; Yang, 2010). Therefore, it is assumable that the more positive consumers’ attitudes are toward branded mobile shopping apps, the more likely they are in using these apps. On the contrary, the less positive consumers’ attitudes are toward branded shopping apps, the less likely they are in using these apps. Moreover, consumers’ attitude toward branded mobile shopping apps positively affects consumers’ use of shopping apps. Based on this assumption, the following hypothesis is formulated:
H8: The more positive consumers’ attitude toward branded mobile shopping apps is, the more
they use a) informational, b) experiential, c) location-based, and d) transactional shopping apps, and vice versa.
Lastly, this study expects that consumers’ attitude toward branded mobile app shopping, besides a direct effect on the use of different types of branded mobile apps, also has a mediating-effect on the effect of consumers’ perceived self-efficacy, their level of
innovativeness, and perceived benefits of mobile app shopping. Therefore, the last hypothesis is formulated:
H9: Consumers´ attitude toward mobile app shopping mediates the influence of consumers’
perceived self-efficacy, level of innovativeness, and perceived benefits of mobile app shopping on the use of branded mobile apps.
28
Figure 3: Conceptual model with corresponding hypotheses.
5. Method
In order to investigate whether and how consumers could differ in terms of their preferences and use of different types of mobile shopping apps and what the main factors are in driving them, an online survey in Dutch was conducted. An online survey was the most appropriate method, since this study had to be accomplished within a timeframe of five months and had a very limited budget. In this way it was possible to collect data from a broad spectrum of individuals. The online survey made it possible to measure a large amount of variables without increasing the time.
29 5.1 Selection
Considering the time pressure and the small budget of this master thesis, the subjects were recruited from the researcher’s acquaintances and business relatives, i.e. a combination of snowball sampling and convenience sampling. The subjects were approached to participate in the online survey via social media, i.e. Twitter, Facebook, and Linkedin, and by e-mail from May 1 till May 12, 2014. To increase the external validity of this study, the researcher
requested the participants to share the invitation for the survey on their social media sites or to send the invitation forward by e-mail. Hereby, subjects outside the direct range of the
researcher’s acquaintances were recruited, which lead to a larger diversity in subjects. Furthermore, three small companies from different industries, i.e. recruitment, IT, and
agriculture, spread the survey among their employees and business relatives. Since this way of sampling only recruits subjects from within a restricted network the generalization of this research is limited.
5.2 Sample
In order to determine how many respondents this study needs to get results that reflect the target population, i.e. mobile shopping app users, as needed, the sample size is calculated. In particular, there are currently 16.850.174 people living in Holland (CBS, 2014), whereby, based on the findings of Telecompaper (2013), approximately 72% of these people are smartphone/tablet owners. It appears that 38% smartphone users and 46% of the tablet users made a purchase in the past three months via an app (Adobe, 2012). Thus, this study was aiming for a minimum sample size of 271 respondents (confidence level = 0,90, error marge = 0,05, dispersion = 0,50), i.e. mobile shopping app users. In total 561 Dutch Internet users filled in the online survey, whereby 463 of these consumers actually completed the survey. Incomplete responses were deleted in this research, since these responses did not contain
30
useful data for further analysis. Only 1,3 % of all respondents did not own a mobile device at all. These respondents were deleted. It appeared that 71,5 % (N = 331) of the mobile device owners were mobile shopping app users. Thus, the sample size of this study reflects the target population, which increases the impact of the findings and the external validity. Therefore, it will also benefit the applicability of the findings in management practice.
In total, this study contained 40,5 % smartphone owners, 1,5 % tablet owners, and 58 % smartphone and tablet owners. It appeared that these individuals mostly used Apple (58,2 %) or Samsung smartphones (35,1 %), whereby 9,0 % used multiple smartphones of different brands. Furthermore, the demographics showed that tablet owners mostly used an Ipad (79,6 %) or Samsung tablet (11,4 %), whereby 15,2 % of the tablet owners posses more than one tablets of different brands, whereby practically every multiple tablet owner had an Ipad (96,7 %). Practically the entire sample had at least once bought a product/service online (95,8 %), whereby 19,4 % bought nothing in the last month, 48,2 % one or two times, 23,0 % three or four times, and 9,4 % more than five times. Approximately half of the final sample were women (n = 170), i.e. 51,7%, which allowed this study to test whether the gender of
consumers had an impact on the use of and preferences for certain types of mobile shopping apps. Ages ranged from 15 till 62 years. It appeared that 57,6 % of all the mobile shopping app users were between 22 and 28 years. Furthermore, 65 % of the respondents were highly educated (HBO/university), against 24,5 % that were low educated (MBO). The rest of them (10,5 %) only went to primary or secondary school, which this study defined as uneducated. Based on the gained data with regard to the respondents’ age, this sample may not be
representative for the entire mobile shopping app population. Therefore, the external validation of this study is limited. In the next table an overview is given regarding the demographics of the respondents of this research.
31 Table 2: Demographics (N = 331)
Variables Categories Responses
Percent
Device owner Smartphone 42,5%
Tablet 1,3%
Smartphone and tablet 54,9% Type of brand of
the smartphone
Apple 58,2%
Samsung 35,1%
Other smartphone brands 6,7% Type of brand of
the tablet
Apple 79,6%
Samsung 11,4%
Other tablet brands 9,0%
Online purchaser Yes 95,8%
No 4,2% Online purchase frequency <1 time 19,6% 1-2 times 48% 3-5 times 23,0% >5 times 9,4% Gender Male 48,3% Female 51,7% Age 15-21 3,3% 22-28 57,6% 29-35 11,4% 36-42 10,5% 43-49 10,2% 50+ 7,0% Education Uneducated 10,5% Low 24,5% High 65,0% 5.3 Research tool
The respondents received a personalized invitation with the request to participate in this research by e-mail or via social media. Two reminders were sent to increase the response rate. A short introduction was given with a link that directly led to the online survey in Qualtrics. Qualtrics is an online survey and data collection tool, which can be used for designing and conducting online surveys. First of all, respondents were exposed to an informed consent, whereby they were informed about the nature of the study, the duration, and the
confidentiality of the survey. As soon as they agreed with this information, they could start answering the questions. Participants were able to click back and forward throughout the
32
whole survey so that they could change their answers, if necessary. All the questions had forced responses to prevent missing values, which increased the internal validity. The
questions were sorted from general to very specific. The survey started with general questions about whether people owned (a) mobile device(s), the brand(s) of the device(s), their purchase behavior via Internet, and whether they used mobile shopping apps via their mobile device(s). A short definition of branded mobile shopping apps was provided in the survey to make sure that the respondents fully understood what these apps contained. The survey included multiple skip logics in order to led respondents directly to the questions that were applicable to them. People who did not own a smartphone or tablet and/or did not use mobile shopping apps via their mobile device were directly directed to the end of the survey. Questions were all asked in the same direction to preclude confusion, which increased the internal validity of the survey. The survey ended with four demographical questions. In order to increase the response rate and the internal validity, the quality of the survey was assessed by piloting it with a small group of respondents (N = 12) in a real life setting (Fan & Yan, 2010). The response rate of this study was approximately 65%, which is acceptable for an online survey (Manfreda, Bosnjak, Berzelak, Haas & Vehover, 2008).
5.4 Measurements
This research contains four different independent variables, i.e. consumers’ level of
innovativeness, their perceived self-efficacy, perceived benefits of mobile shopping apps, and their attitude toward those apps and two dependent variables, namely consumers’ use of different types of mobile shopping apps and their preferences for these shopping apps. This study starts with discussing the measurements of the independent variables, continues with the two dependent variables, and ends with the control variables. Table three contains all the measurement properties for the multi-item constructs. An overview of the questionnaire and
33 the corresponding items can be found in Appendix A.
5.5 Measures perceived self-efficacy
The independent variable perceived self-efficacy measures consumers’ belief about someone’s ability to confidently achieve positive task outcomes by performing a specific behavior (Compeau & Higgins, 1995). This variable is measured by the mean of three 7-point Likert items, which ranged from 1 (totally disagree) till 7 (totally agree), adapted from the perceived self-inefficacy dimension of the lifestyle construct of Brengman, Geuens, Weijters, Smith and Swinyard (2005). The higher respondents score on this scale, the more they agree with the statement. An example of an item is: “I know how to use shopping apps on my smartphone or tablet”. The scale appeared to be reliable with a Cronbach’s Alpha of 0,73 (M = 4,96, SD = 1,18), which is comparable to the reliability test of the attitude scale of
Brengman et al. (α = 0,79).
5.6 Measures level of innovativeness
The variable level of innovativeness measures consumers’ propensity to learn about and adopt mobile shopping apps on the basis of six items adapted from Goldsmith, Freiden and
Eastman’s (1995) study, e.g. “In comparison to my friends, I own a lot of shopping apps on my mobile phone or Ipad”, “I know shopping apps of companies and brands before other people know them”. The response categories of the Likert scale items vary between 1 (totally disagree) and 7 (totally agree). The higher respondents score on the scale for the variable consumers’ level of innovativeness, the more innovative they are. A factor analysis using a varimax rotation found two components with an initial eigenvalue higher than 1,00, but indicated an acceptable fit for a five-factor structure with one component of the scale
34
in my group of friends that knows about the newest shopping apps”. This component had an explained variance of 56,28 %. The scale appeared to be highly reliable with a Cronbach’s Alpha of 0,80 (M = 3,65, SD = 1,16), which is comparable to the reliability tests of the innovativeness measures of Goldsmith, Freiden and Eastman (1995).
5.7 Measures benefits of branded mobile shopping apps
The independent variable benefits of mobile shopping apps measure the amount of consumers’ perceived advantages of mobile shopping apps by asking the respondents the following question: “Which of the following factors do you consider as an advantage of shopping via an app on your smartphone or tablet?”. Based on findings of earlier research (table 1), fourteen items, i.e. predetermined advantages, on a multiple item scale were given, such as privacy, seeing and experiencing new things, and saving time.The more benefits respondents ticked, the more benefits they experienced of mobile app shopping (M = 5,51, SD = 2,69).
5.8 Measures attitude toward branded mobile shopping apps
This variable refers to consumers’ evaluations of mobile shopping apps. The attitude toward mobile shopping apps is measured by the mean of four 7-point semantic-differential items adopted from Bellman et al.’s study (2011), anchored by “bad/good”,
“unfavorable/favorable”, “unpleasant/pleasant”, and “unlikeable/likeable”. The higher respondents score on this scale, the more positive their attitude toward mobile shopping is. The four-item scale was factory-analyzed with a varimax rotation to determine the scale’s dimensionality. One single factor accounted for the majority of the variance in the scale (80,42 %) with an eigenvalue larger than 1,00. The scale appeared to be highly reliable with a Cronbach’s Alpha of 0,92 (M = 4,67, SD = 1,29), which is comparable to the reliability test of
35 the attitude scale of Bellman et al. (α = 0,94).
5.9 Measures use of different types of branded mobile shopping apps
This dependent variable refers to consumers’ use of different types of mobile shopping apps by asking the respondents the question: “To what extent do you engage in the following activities via a mobile app of a certain company or brand?”. This variable is based on four dimensions: informational, experiential, location-based, and transactional apps. Based on Chong’s (2013) and Bellman et al.’s (2011) findings regarding these four different types of apps, this study used two to three items to measure consumers’ use of each type of these apps. Examples of these items are: “Searching for product information” (informational shopping apps), “Looking at photo galleries of new or exclusive products” (experiential shopping apps), “Searching for offers/coupons/actions that are currently close to your present geographical location” (location-based shopping apps), and “Scheduling appointments” (transactional shopping apps). The response categories of these items varied between 1 (never) and 5 (always), i.e. a five point Likert scale. The higher respondents score on this scale, the more often they use (a) certain type(s) of mobile shopping apps. The scales to measure consumers’ use of informational (M = 3,11, SD = 0,85), experiential (M = 2,30, SD = 0,81), and
transactional shopping apps (M = 3,21, SD = 0,88), were all reliable with a Cronbach’s Alpha of respectively 0,81, 0,71, and 0,68. The scale for the use of transaction shopping apps
appeared to have a larger internal consistency if one item was deleted, i.e. “Schedule
appointments” (α = 0,82). Only the scale for the use of location-based apps (M = 2,61, SD = 0,76) appears to have a lower reliability than the other scales (α = 0,54). Since the variable consumers’ use of branded mobile shopping apps is built on the previously described
dimensions, a factor analysis with varimax rotation was used in order to determine the scale’s dimensionality. One single factor accounted for the majority of the variance in the scale
36
(53,26 %) with an eigenvalue larger than 1,00. The scale appeared to be reliable with a Cronbach’s Alpha of 0,71 (M = 2,81, SD = 0,60). In Table 3 an overview is provided of the earlier mentioned multi-item constructs.
Table 3: Measurement properties for multi-item constructs (N = 331).
Variable Mean Standard
deviation Cronbach’ s Alpha Perceived self-efficacyb 4,96 1,18 0,73 Level of innovativenessb 3,65 1,16 0,80
Attitude toward mobile shopping appsc
4,67 1,29 0,94
Use of different types of mobile shopping apps a
2,81 0,60 0,71 Use of informationala shopping apps 3,11 0,85 0,81 Use of experientiala shopping apps 2,30 0,81 0,71 Use of location-baseda shopping apps 2,61 0,76 0,54 Use of transactionala shopping apps 3,21 0,88 0,68 Notes: a
Scale items were based on five-point Likert-type scales (1=“totally disagree”, 5=“totally agree”).
b
Scale items were based on seven-point Likert-type scales (1=“totally disagree”, 7=“totally agree”).
c
Scale items were based on seven-point semantic-differential scales.
5.10 Measures preferences for different types of branded mobile shopping apps
This dependent variable measures whether consumers have certain preferences for particular type(s) of mobile shopping apps. This study tries to measure this variable by asking the respondents the following question: “To what extent do you agree with the following propositions with regard to the use of apps of companies/brands via your smartphone or tablet?”. One item per type of mobile app is used to indicate whether consumers have certain preferences, e.g. “I like to use apps of a company/brand for informative purposes”. The
37
response categories of these items vary between 1 (totally disagree) and 7 (totally agree), i.e. a seven point Likert scale. The higher respondents score on the scale for consumers’ preference for informational (M = 4,69, SD = 1,55), experiential (M = 3,90, SD = 1,70), location-based (M = 4,35, SD = 1,68), and transactional shopping apps (M = 4,88, SD = 1,54), the more they prefer certain type(s) of mobile shopping apps.
4.11 Measures control variables
This study contains nine control variables. Some of these variables were transformed into dummy variables, since they made it possible to represent the first nominal-level control variables in e.g. linear regression analysis and independent samples t-tests, which avoided severe limitations. Three of control variables were demographic variables, i.e. gender (M = 1,00, SD = 0,50), age (M = 31,16, SD = 9,73), highest level of education, i.e. primary school, mavo, MBO, havo, VWO, HBO, and university (M = 6,00, SD = 1,65). The level of education was recoded in low educated (primary and secondary school) and high educated (HBO and university), i.e. dummy variable. Furthermore this research included the variable device owner, i.e. whether people owned a mobile device, (smartphone, tablet, smartphone and tablet, none of these). The respondents that did not own a mobile device were excluded from this research (M = 3,00, SD = 0,98).Another control variable related to the device people owned, were the type of brand of the tablet (Apple, Samsung, Sony, Asus, Archos, Acer, other, namely). Since by far most of the tablet owners possessed an Apple tablet (79,6 %), this study changed this variable into the dummy variable Ipad owner (M = 1,00, SD = 0,38), i.e. whether people owned an Ipad (yes/no). The variable type of brand of the smartphone was measured by seven dummy variables, i.e. Iphone, Samsung, Blackberry, Nokia, HTC, Sony, other, namely.This variable was split into the two new variables Iphone smartphone owner and Samsung smartphone owner, since practically the whole sample owned either an Apple or
38
a Samsung smartphone (93,3 %).This study also included the control variable online purchaser (M = 1,00, SD = 0,20), i.e. whether people had ever bought a product/service online, (yes/no), and the online purchase frequency (M = 2,00, SD = 0,87), i.e. how many times people bought something online in the last month (<1 time, 1-2 times, 3-5 times, >5 times). The online purchase frequency was recoded into low intensity purchasers (0-2 times and high intensity purchasers (more than 2 times).
6. Analyses and results
6.1 Effect online purchase frequency
The variable online purchase frequency consists of two categorical unrelated groups, i.e. low intensity online purchasers and high intensity online purchasers. In order to test H1, the means
between those two groups should be compared on the continuous dependent variable
‘consumers’ use of the four different types of branded mobile shopping apps’. Therefore, this study chose to perform an independent samples t-test for each one of these types of shopping apps. Before testing the hypotheses, this study verified whether the assumptions for the corresponding analyses were met, which increased the internal validity. By testing the
corresponding assumption of the independent samples t-test, there were no significant outliers found in the two categories of the online purchase frequency. The normality of the use of the four different types of apps for both the two categories of the online purchase frequency was assessed by using skewness and kurtosis with values between -1,0 and +1,0 (Chong, 2013). It appeared that also the assumption for normal data was met. Levene’s test showed that there was also homogeneity of variances, whereby the last assumption was met. By testing H1a, the
independent-samples t-test showed that there was a negative statistically significant difference between informational app use for consumers that often shop via the Internet and consumers
39
that shop little via the Internet (t (328) = -4,03, p < 0,001). In other words, consumers that often shop via the Internet (M = 3,37, SD = 0,75) use informational shopping apps more often than consumers that shop little via the Internet (M = 2,98, SD = 0,87). Thus, H1a was hereby
confirmed. Contrary, no statistically significant difference was found between experiential app use for consumers that often shop via the Internet and consumers that shop little via the Internet (t (328) = -1,21, p = ,228), whereby H1b was rejected. There was also no statistically
significant difference found between location-based app use for consumers that often shop via the Internet and consumers that shop little via the Internet (t (328) = -1,25, p = ,212). Thus, H1c
was rejected. Lastly, an independent-samples t-test indicated that there was a negative
statistically significant difference between transactional app use for consumers that often shop via the Internet and consumers that shop little via the Internet (t (328) = -5,23, p < 0,001). In other words, consumers that often shop via the Internet (M = 3,56, SD = 0,76) use shopping transactional apps more often than consumers that shop little via the Internet (M = 3,03, SD = 0,89). Moreover, H1d was confirmed. Thus, the more consumers shop via the Internet, the
more they use informational and transactional shopping apps, and vice versa. An overview of these results can be found in Table 6. More extensive tables with the output of every
hypothesis are provided in Appendix B.
6.2 Effect device owner
The variable device owner consists of two categorical unrelated groups, i.e. smartphone owners, and both tablet and smartphone owners. In order to test H2, the means between those
two groups should be compared on the continuous dependent variable ‘consumers’ use of the four different types of branded mobile shopping apps’. Since all the earlier mentioned
assumptions were met, this study chose to perform an independent samples t-test for each one of these types of shopping apps. The independent-samples t-test indicated that there was a
40
negative statistically significant difference between informational app use for consumers that both own a tablet and a smartphone and consumers that only own a smartphone (t (324) =
-2,37, p = ,018). In other words, consumers that both own a tablet and a smartphone (M = 3,19,
SD = 0,88) use informational shopping apps more often than consumers that only own a smartphone (M = 2,97, SD = 0,80). H2a was hereby confirmed. Contrary, there was no
statistically significant difference found between experiential app use for consumers that both own a tablet and a smartphone and consumers that only own a smartphone (t (324) = -0,06, p =
,956). Therefore, H2b was rejected. The same counted for H2c: There was no statistically
significant difference found between location-based app use for consumers that both own a tablet and a smartphone and consumers that only own a smartphone (t (324) = -0,34, p = ,733).
Lastly, an independent-samples t-test indicated that there was a negative statistically
significant difference between transactional app use for consumers that both own a tablet and a smartphone and consumers that only own a smartphone (t (324) = -3,00, p = ,003). This
means that consumers that both own a tablet and a smartphone (M = 3,33, SD = 0,87) use transactional shopping apps more often than consumers that only own a smartphone (M = 3,03, SD = 0,89), which confirmed H2d. Thus, consumers that own both a tablet and a
smartphone use informational and transactional shopping apps more than consumers that only own a smartphone.
6.3 Gender-effect
The variable gender consists of two categorical unrelated groups, i.e. men and women. In order to test H3, the means between those two groups should be compared on the continuous
dependent variable ‘consumers’ use of informational and transactional shopping apps. Since the earlier mentioned assumption for the independent samples t-test were met, this study chose to perform independent t-tests for each one of these types of shopping apps. Based on
