What drives the use of mobile devices in different phases of shopping : do drivers differ in different phases of shopping?

(1)

1

What Drives the Use of Mobile Devices in

Different Phases of Shopping: Do Drivers Differ

in Different Phases of Shopping?

Author: Zouhair Ouroui Ouriaghli Student Number: 10983279 University of Amsterdam

Faculty of Economics and Business

Msc. Business Administration-Marketing Track Supervisor: Dr. U. Konuş

Date of submission: January 28th₂₀₁₆

(2)

2 Acknowledgements

I want to thank the people that have supported me through this process and gave me advice when needed. First, I want to thank my girlfriend for being there for me when I needed it and to listen to me. Second, I want to thank Maarten and Bob for helping me with the grammar and their insightful comments. Lastly, and most importantly, I want to thank Umut Konuş who supported me through this process and was always clear with his insights. Thank you all!

(3)

3 Statement of originality

This document is written by Student Zouhair Ouroui Ouriaghli 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.

(4)

4

5.4.1 Hypothesis 1 46 5.4.2 Hypothesis 2 46 5.4.3 Hypothesis 3 47 5.4.4 Hypothesis 4 47 5.4.5 Hypothesis 5 48 5.4.6 Hypothesis 6 48 5.4.7 Hypothesis 7 49 6. Discussion 52 6.1 Discussion 52 6.1.1 Age 52 6.1.2 Time consciousness 53 6.1.3 Shopping enjoyment 53 6.1.4 Risk perception 54 6.1.5 Convenience 54

(5)

5

6.2 Managerial implications 55

6.3 Limitations and further research 56

5. References 58

Appendix I - Survey 63

(6)

6

Abstract

The rise of various mobile devices has created new challenges for research, including understanding how consumers choose mobile devices with respect to their search, purchase and after-sales support behaviour. This research considers tablets and

smartphones as different channels and examines the choice of mobile device within the shopping process, which is mainly driven by consumer characteristics. This is relevant because literature perceives the mobile channel currently as one channel while there are indicators that the behaviour within the mobile channel differs between multiple mobile devices. This is also relevant for managers as they can differentiate in marketing initiatives in the mobile channel. The author (1) examines whether mobile shoppers of device differ in the search, purchase and after-sales support stages of the shopping process; (2) investigates whether demographic, perceived benefits and costs, psychographic drivers and behavioural drivers influence this particular choice of consumers. Results have shown that age, time consciousness, shopping enjoyment, risk perception, convenience, and offline channel usage in the information search phase are significant drivers for tablet usage. They have also shown that time consciousness, risk perception, and convenience are significant drivers for smartphone usage.

(7)

7

1. General Introduction

The introduction of new information channels has caused retailing to evolve significantly. With the advent of the mobile channel, social media, tablets and the integration of these new channels in online and offline retail, the retail landscape continues to change (Verhoef, Kannan & Inman, 2015). These changes also bring new insights and opportunities for business and science. Mobile shopping (M-shopping) is a relevant and important topic that has gained a lot of attention from multiple industries and scholars.

The imposing number of smartphone users is growing every year; eMarketer (2014) predicted that there will be a rise of 0.25 billion smartphone users between 2015 and 2016, which is a total of 2.16 billion smartphone users in the world. This is more than a quarter of the global population. Husson, Ask, Overby, Parrish, Mullen and McCarthy (2013) predict that there will be a tipping point of online traffic through mobile device; more than 50% will be done through mobile devices. An example of the increase in mobile sales is Single’s Day, the biggest one-day shopping event in China and in the world, which reached a mobile shopping peak in 2014. Alibaba’s sales made via mobile on that day totalled 43% of the company’s $9.3billion in sales in 2014

(PriceWaterCooper, 2015).

But this could be just the beginning of mobile commerce. Forrester Research (Mulpuru, Evans, Roberge & Johnson, 2013) predicts that mobile shopping will grow significantly. The projection of overall retail annual growth rate is 4% for 2015 through 2016, while mobile commerce is estimated to grow at a staggering number of 21-29%. Deloitte Consulting (2012) predicts that the retail revenues transacted through mobile devices will be $31 billion in 2016. Next to transaction through mobile devices,

(8)

8 customers also use them to search information such as finding directions to the store or looking for better deals on another website (Google Shopper Marketing Council, 2013). It can be concluded that M-shopping is a fruitful area for marketers, but how could this study provide new knowledge?

The focus of this research is to investigate whether the use of mobile devices differs in the different phases of the shopping process (search, purchase and post-purchase support), and which factors (demographic, perceived benefits and costs, psychographic and behavioural drivers) influence the device choice and use along the different phases of shopping. This study will specifically test whether the consumers’ motivation, preference and use of mobile device vary in every phase of the shopping process. It will also test how these drivers differentiate. This means that consumers could have certain motivations to use smartphones while they are searching for

information, which is the first phase of the shopping process. The same consumers could have specific motivations regarding tablets when they purchase the product/service, which is the second phase of the shopping process. Another possibility is that there are consumers who use one device through every phase of the shopping process but with different motivations. This research will explore the consumer characteristics related to the choice of these mobile devices in the three phases of the shopping process.

It is important to know the differences of consumer behaviour caused through mobile devices from a scientific perspective and from a manager’s perspective.

While research already has shown that mobile is an important channel in comparison to other channels, no research has been done on the differences between different devices for mobile shopping and their influence on shopping behaviour. However, it is an

(9)

9 important subject that should not be overlooked. This research will contribute to the existing knowledge about the mobile channel.

The research results could also benefit managers and companies. Companies invest in their mobile applications and websites and want to see results caused through these channels. This research will provide the underlying perceived benefits of

smartphones and tablets in every phase of the shopping process. Companies can implement this knowledge in their strategies for applications and make a distinction between their tablet and smartphones applications/websites. They could create different shopping environments in these channels and focus, for example, more on smartphones in the information search phase and on tablets in the purchase phase. This implementation could lead to an increase in efficiency in their mobile applications. This information could provide a roadmap for companies that will help them to allocate their marketing budgets effectively.

This research paper provides a literature review that includes the main pillars. This literature review exists of the chosen concepts and their definitions, and includes the links between concepts. Furthermore, it also consists of previous work on this topic and the gaps in existing research. Secondly, this paper will provide a conceptual

framework and the accompanying hypotheses. Lastly, this paper will also have a research description in which I will elaborate on the methods that will be used to answer the hypotheses.

(10)

10

2. Literature Review

2.1 Mobile marketing in the multichannel environment

Mobile marketing has been defined as “the two-way or multi-way communication and promotion of an offer between a firm and its customers using a mobile medium, device or technology” (Shankar and Balasubramanian, 2009). This channel is growing rapidly in the multichannel environment (Neslin and Shankar, 2009). Wang, Milthouse and Krishnamurti (2015) are also showing that shopping through mobile devices increases the amount of orders placed within one year. One could argue that the reason for this rapid growth is because the interaction is possible at all times, since consumers have their mobile devices continuously with them. This is changing the retail environment drastically as retailers can access the consumer’s environment at all times. Retailers want to satisfy customer needs profitably in this mobile environment; therefore they engage in a number of mobile marketing practices. These practices include: mobile website creation and maintenance, mobile emailing and messaging, mobile advertising, mobile couponing, mobile customer service and mobile social network management (Shankar, Venkatesh, Hofacker & Naik, 2010). Although companies are trying to engage in these practices, it seems that many don’t know how to do this successfully. Shankar and Balasubramanian (2009) argue that the focus should be on consumer behaviour and then on return on investment (ROI). They argue that this is important because

consumer behaviour is changing very quickly, which is dangerous for retailers if they “bet” too much on a certain practice. In this dynamic environment it is hard for companies to find an effective mobile marketing strategy. Shankar, O’Driscoll, and Reibstein (2003) argue that it is important to know the drivers of device and service adoption in the shopping process to cultivate a powerful mobile marketing strategy.

(11)

11 This argument makes a point, because to be effective with a mobile marketing strategy, companies need to know why consumers are choosing for certain devices when online shopping.

2.2 Shopping process

The shopping journey is a process that has been researched extensively. There are multiple ways to explain this process. The journey of customers is composed of the familiar stages of the customer-decision making process; need recognition, information search, alternative evaluation, purchase/transaction, after-sales support (Jobber & Ellis-Chadwick, 2013; Kotler & Armstrong, 2010). Table 1 elaborates on each phase.

Table 1

The five phases of the shopping process

Phase Definition

Need recognition Identifying a particular need to purchase a product Information search Seeking information about the need and how it can be solved

Alternative evaluation Evaluating the different alternatives found in the information search phase

Purchase Decision to purchase and the purchase itself

After-sales support Consumer will check whether the product is in line with his original needs

Balasubramanian, Raghunathan and Mahajan (2005) argue that the customer’s purchase process for moderate and high involvement products is composed of three distinct stages; (a) forming a consideration set; (b) choosing a product; and (c) buying the product. Chen & Chang (2002) argue that three phases are relevant for a shopping

(12)

12 environment; (1) information search; (2) transaction; and (3) after-sales support. This research will use these three phases of the customer-decision making process.

2.2.1 Information search

The information search stage is the first step in the shopping process. Consumers need to find information for their need. Consumers rely in this phase on internal information and external information. Internal information is information that is already existent in the memory of consumers. External information is information that is gained through browsing (Kotler, 2003). In this review we focus on how external information is gathered.

Sands, Ferraro, and Luxton (2010) have found that “prior studies about

information search behaviour have focused on multiple issues, including its extent and duration, how many and what kind of sources of information consumers use (search depth) and how much time consumers devote to each source when purchasing a particular product or service.” An example of an information source would be the different channels that are used by consumers. It is important to investigate which channels have more influence on consumers whilst searching for external information. Cheema and Papatla (2010) found that the relative importance of offline information is higher for hedonic products. This means that consumers would like to experience hedonic products in an offline setting instead of searching for the information online. However, one could argue that when consumers have experienced certain products often, they would not have the need to experience it again. Therefore they could search for certain information online, as this would be easier for them. Bennet, Hartell and McColl-Kennedy (2009) support this argument as they found that extensive product experience impacts the importance and depth of information search.

(13)

13 Research has proven that the information search phase is used most through the online channel in comparison to brick-and-mortar and catalogue channels (Verhoef & Neslin, 2007). M-shopping users heavily use this stage because it is convenient to search with a mobile device. Results indicate that mobile search is in some ways more effective than online browsing (Church, Smyth, Cotter & Bradley, 2007). Results from another study suggest that search usage is much more focused for the average mobile-user than for the average computer-based user (Kamvar, Kellar, Patel & Xu, 2009).

Information search has become a significant phase in the M-shopping customer journey. There is however little research based on the choice of mobile device in this phase. There are a lot of information sources that customers can use during online browsing, and they can do this through multiple mobile devices. Church et al. (2007) have found that the devices that have been used most for search are those that have better displays and especially a better resolution and a larger screen size. However they also found that none of the devices had an extended keyboard despite the information search phase involves higher need for text-entry support. Kamvar, Kellar, Patel and Xu (2009) have found that the usage intensity of smartphones is similar to that of the PC while searching for information. This could relate to the easiness of searching

information because of the continuous availability of smartphones.

2.2.2 Purchase

When customers have chosen their product it is time to purchase the product through their preferred channel. The purchase decision that has been made by consumers is based on the information found in the earlier phase. This information is based on product’s features, perceived value and capabilities that they deem meaningful (Kotler, 2003).

(14)

14 However, consumers need the last push in the purchase phase. The information found in the previous phase is sometimes not enough to let them purchase products. Levin, Levin and Weller (2005) argue that “high-touch products require an offline presence at least in the purchase phase.” This is a sound argument, as consumers want to experience certain products before they decide whether they will buy it or not. Zeithaml, Parasuraman, and Malhotra (2000) support this argument as they found that one of the consumers’ motivations to purchase offline is access to the product. Other motivations were assurance, reliability, responsiveness and customization. Therefore, one could argue that the ‘human touch’ makes the offline channel relevant when consumers want to buy certain products. However, technology improves continuously and so does the online channel. Research has proven that the online channel can also generate this ‘human touch’ through various means. This ‘e-human touch’ proves to be effective on consumers’ consumption motives (Sivaramakrishnan, Wan, and Tang, 2007). One could conclude that the online channel is copying the features of the offline channel and therefore becomes a very effective channel for purchase behaviour of consumers. Furthermore, Brasel and Gips (2014) have shown that touchscreen on mobile devices increase the perceived product ownership of consumers and therefore also the purchase intentions. This shows that the e-human touch has positive effects for purchase intentions.

Chen and Chang (2003) found that customers deem two factors important while purchasing online. Firstly, value, which consists of price and quality factors. Secondly, convenience, which is determined by location, time and variety. Mobile devices do not differentiate on value but they do on the convenience factor. Salesforce (2014) shows smartphones are used most between 08 a.m. and 6 p.m. and that tablet usage increases

(15)

15 after 6 p.m. This indicates that tablets are used more frequently when consumers are at home in a convenient location. Findings from Koyuncu and Lien (2003) confirm this as they find that online users prefer to purchase when they are home as they see this as a private and secure place. However, one could argue that if consumers purchase more often that the perceived risk would go down and that they would purchase in a place where they would feel less secure. However this hypothesis was rejected in the research of Pires, Stanton and Eckford (2004).

2.2.3 After-sales support

After-sales support is the feeling customers have after the purchase. That feeling becomes stronger after a longer use of the channel or product. The channel that will be adopted after repeated satisfaction in this stage would be the channel that customers are most experienced with. However if the experience of consumers is mediocre or unpleasant than they will repeat the shopping process but delete the channel from their choices. Customers that are comfortable with the Internet could choose repeatedly for the online channel, whereas for other customers this could be the offline channel (Frambach et al., 2007).

Companies can influence this feeling through acting accordingly when consumers have questions or complaints about their bought products. To act effectively on the questions and complaints of consumers, firms need to know which channel(s)

consumers will use. Day and Landon (1977) argue that consumers can take direct action when they have a complaint and vent their frustration with the service provider. Mattilla and Wirtz (2004) found that consumers do this through remote channels (e-mail) when they wanted to show their frustration of the service or product. However, when

(16)

16 Store, Social media). This means that firms need to adjust their communication with customers to reach a good result in this phase. It is also important to segment

consumers in this phase as this results in customized responses. Keyser, Schepers and Konuş (2015) have found that consumers “that are young, with a high average revenue and a high level of involvement” tend to be research shoppers and use both online and offline for their after-sales support. They argue that they use multiple channels in this phase because the product is important for them. This means that those important products that contribute to the self-image need to have good after-sales support in all channels.

Figure 1. The shopping process

2.3 Mobile devices

2.3.1 Mobile shopping behaviour

The previous chapter has shown that there are differences between channels in the shopping process. This chapter will elaborate on research focused on the mobile shopping behaviour of consumers and their underlying drivers. Previous research has focused on the adoption of the mobile channel in a shopping context. Most studies made use of the technology acceptance model (TAM) (Davis, 1989) to examine the drivers of

(17)

17 the adoption of mobile technology in a shopping context. TAM showed that the

perceived ease of use, perceived usefulness, and attitude towards a system (mobile technology) had a significant influence whether the user would use this system. Lu and Su (2009) have used the TAM to examine which drivers cause purchase intention on mobile shopping websites. They found that usefulness, compatibility and enjoyment had a positive effect on the behavioural intentions while anxiety had a negative effect on intention to use. Petina, Amialchuk and Taylor (2011) proved that interactive

engagement through a mobile interface also leads to higher intention of online shopping. The research aforementioned shows that the mobile shopping behaviour

depends on various factors. It is important to notice that usefulness and compatibility are factors that are perceived differently by consumers. Therefore one could argue that context would play an important role in the mobile shopping behaviour literature. So would there be a difference in these results when tested for various mobile devices? Smartphones and tablets are the most common mobile devices used by consumers and it is therefore interesting to find out whether there are differences between these two devices in a mobile shopping context. The following chapter will discuss these two mobile devices.

Table 2

Drivers mobile channel use

Author Drivers

Lu & Su (2009) Usefulness (+), Compatibility (+), Enjoyment (+) Anxiety (-)

(18)

18 2.3.2 Smartphones and tablets

“Mobile devices are becoming ubiquitous. There is an explosion in the use of handheld electronic communication devices, such as mobile phones, digital music players, and handheld Internet access devices” (Shankar et al., 2010). This creates a new playing field for marketers. Mobile devices have certain important characteristics. These

characteristics are ultra-portability, location sensitivity, and untetheredness (Shankar & Balasubramanian, 2009). In this research two mobile platforms will be used,

respectively smartphones and tablets. Before discussing these platforms, overviews of the platforms are presented in table 3.

Table 3

Overview of platforms

Platforms Definitions

Smartphones A mobile phone that performs many of the functions of a computer, typically having a touchscreen interface, Internet access, and an operating system capable of running downloaded apps.

Tablets A small portable computer that accepts input directly on to its screen rather than via a keyboard or mouse.

2.3.3 Smartphones

The ownership of smartphones versus desktop has reached a tipping point in 2014. There are more consumers owning a smartphone than desktop computers

(Smartinsights, 2015). It is also one of the most popular devices used to search the Internet; 80% of users searches through smartphones compared to 91% that uses a

(19)

19 desktop PC (Global Web Index, 2015). This shows that the usage intensity of

smartphones are almost the same as the desktop PC, this is only going to grow more in the next years. One of the reasons why smartphone users shop online is because that “it was close at hand at the time of the purchase” (Statista, 2015). One could argue that consumers develop a certain need which they want fulfilled directly, and therefore use the smartphone to purchase as it is easily accessible.

2.3.4 Tablets

The ownership of tablets has rapidly increased the last few years. According to Global Web Index (2013) tablet usage has increased 282% between Q1 2011 and Q1 2013. Tablets commerce has grown significantly the last few years. Tablets are the ideal devices to use for online shopping as you can lean back, or do some power shopping sessions. Their larger screens make it easy to see the product details on websites (Business Insider, 2013). This is also confirmed in scientific research as Chae and Kim (2007) also found that screen size significantly affects the navigation behaviour of users. “When it comes to making purchases online, 69% of desktop shoppers versus 34% of tablet shoppers made at least one purchase in Q2 2013” (Internet Retailer, 2013). This shows that consumers’ usage of tablets are not that high as desktop pc. This could be the case because tablets are most used at home and the desktop pc is also within reach. 2.3.5 Why does and should device matter

Although smartphones and tablets are both seen as mobile devices within one mobile channel, there are some crucial differences. It is imperative that researchers and managers understand these differences between smartphones and tablets for multiple reasons. Firstly, smartphones and tablets consist of differentiating and exclusive aspects, and therefore they need to be treated as distinctive platforms within the mobile channel

(20)

20 (Forrester, 2012). Secondly, consumers use different mobile devices in the various phases of the shopping process (Holmes, Byrne and Rowley, 2013). This means that consumers demonstrate disparate behaviour within the shopping process. Thirdly, unfolding this mobile device will help managers to build up effective mobile marketing strategies. Forrester (2012) argues that marketers need to optimize for both the

smartphone and tablet experience, as the content and services accessed by consumers is very different. Lastly, while researching this different behaviour, unfolding the

underlying drivers of mobile device choice will help researchers to build on the found conclusions. An overview of the key characteristics of smartphones and tablets are presented in Table 4.

Table 4

Key characteristics of smartphones and tablets

Dimension Smartphones Tablets Sources

Location Everywhere At home Forrester, 2012;

Holmes et al, 2013

Personalization 1 User only Multiple users Salesforce, 2014; Forrester, 2012 Customer base High income Low/High income Salesforce, 2014;

Age 18-34 Age 35-44 Holmes et al,

2013

Location. One of the greatest differences between smartphones and tablets is the location where they are being used. “Tablets are portable, while smartphones are

pocketable” (Forrester, 2012). While one would think that tablets are mobile devices so they could be used everywhere, research shows that tablets are used most at home

(21)

21 (Holmes et al, 2013). Smartphones, on the other hand, are used at any time of day

because they are easily accessible (Salesforce, 2014). With this knowledge one could argue that consumers with accessibility to the Internet at all times would search for information more easily. Consumers perceive the risks of purchasing and transacting more when they are outside compared to being at home or at work (Salesforce, 2014; Koyuncu and Lien, 2003). Therefore, location creates a certain differentiation in choice of device for consumers whilst shopping online.

Personalization. Another difference between smartphones and tablets is the degree of personalization. Personalization is the concept of one-to-one marketing where a business tries to provide tailored content or services based on individual needs and preferences (Kim, 2002). Studies have shown that the degree of personalization affects the information search behaviour in a positive way (Peterson and Merino, 2003). However, effective personalization can only occur as the marketer gets sound input. Consumers use their smartphone individually, whereas they share their tablets with others in a household (Salesforce, 2014). Therefore one could argue that personalized content and services work more effectively for smartphones than for tablets.

Customer base. Customers who shop online through smartphones are different than those who use tablets. Salesforce (2014) and Forrester (2012) show that people with a high income own the most tablets compared to those with a low income. They also found that most tablet users are between 35 and 44 years old. This is rational, as tablets are seen as a luxury products, which are harder to obtain for younger consumers

(22)

22

2.4 Drivers that influence choice of device

Previous studies identified that various drivers have an influence in channel choice and whether consumers would shop online or not (Chang, Cheung and Lai, 2004). For example, one study found that young people are more predisposed to M-commerce adoption than old people (Bigne, Ruiz and Sanz, 2005). Chang et al. (2004) found various drivers that relate to the intention or usage of online shopping. Another study found that the current location of consumers plays a role in choice of device while online shopping (Holmes, Byrne and Rowley, 2014). However, no research exists on the drivers that influence choice of mobile device in the shopping process.

Consumers differ in their preferences, attitudes and behaviour and this can influence their choice of mobile device (Neslin and Shankar, 2009). This study will include certain expected benefits and costs, behavioural drivers, psychographic and demographic attributes of consumers and measure whether and how this influences the choice of mobile device in three phases of the shopping process. The following section will lay down the descriptions of these drivers that may influence the choice of mobile device.

2.4.1 Demographics

Research shows that demographics have impact in multichannel environments (Konuş et al., 2008). Consumers could perceive the various mobile devices as different channel in the shopping process. Therefore it is also important to know whether and how

demographics have an influence on the choice of mobile devices in the shopping process. Furthermore, demographic attributes of consumers are important for researchers as it can hint to deeper structure variables (Chang et al., 2004) and the outcomes of these

(23)

23 variables could lead to new research. The following demographic characteristics of consumers have been chosen: age and income.

Age

Age plays a role in the choice of mobile device because of several factors. This research will make the distinction in age between young (18-54) and old (55+) consumers. Salesforce (2014) shows that younger consumers tend to spend more time on their smartphones than older consumers (3.75 hours versus 2 hours). They also show that there were younger consumers having a smartphone than older consumers (21.5% versus 13%). Bigne et al. (2005) found that younger users tend to use more M-commerce than older users. This could explain why they spend more time on their smartphones than older users. This is however in contradiction with a recent research from Telecom Paper (2012). They found that middle-aged users tend to shop more mobile than purchase more through the mobile channel than younger people. This is in line with the PriceWaterCooper (2015) report in which they argue that older consumers have more money and therefore consume more through the mobile channel than

younger consumers. Younger consumers tend to browse more via the mobile channel than older consumers, this could explain the time spend on smartphones being larger than older consumers. However, previous research does not explain whether age makes a difference in the choice of device. Therefore another factor should be taken in

consideration.

It seems that smartphone ownership is much higher with younger consumers than older consumers (71% versus 37%) (Deloitte, 2014). Another important research shows that the ratio of tablet versus smartphone ownership is significantly different

(24)

24 between the two age groups. Pew Research Center (2015) finds that smartphone versus tablet ownership is balanced with old consumers (27% versus 25%) while there are more young consumers that own a smartphone but no tablet (85% versus 48%). This means that most young consumers do not have the choice between mobile devices when online shopping while old consumers do. Therefore, it is expected that the age of

consumers has an impact on mobile device choice. Young consumers have a higher usage rate of smartphones when online shopping than old consumers and old consumers have a higher usage rate of tablets when online shopping than young consumers. This leads us to the following hypotheses:

H1a: Age has significant negative impact on smartphone usage in the shopping process H1b: Age has significant positive impact on tablet usage in the shopping process

Income

It is important to know whether income plays a role in choice of device because managers can then decide on which devices they can communicate their high- or low-value propositions. Forrester and Salesforce (2012; 2014) show that it is more likely that consumers with a high income own tablets than consumers with a low income. This is because tablets are not seen as a necessity while smartphones are. Therefore

consumers with a low income would choose owning a smartphone before they would buy a tablet. Therefore we assume that consumers with a high income intend to use tablets significantly more than consumers with a low income. This leads us to the following hypothesis:

(25)

25 H2: Income has significant positive impact on tablet usage in the shopping process

2.4.2 Psychographics

Psychographic attributes are important in the field of marketing research. They are linked to the interests, opinions, attitudes, values, lifestyles and personalities of consumers (Ganguly, Dash, Cyr & Head, 2010). Previous studies have shown that

psychographic attributes have an influence in channel choice (Konuş et al., 2008; Chang et al., 2004). Furthermore, managers could use the relevant psychographic attributes to develop effective marketing communications. This research will investigate three psychographic factors: time consciousness, shopping enjoyment and innovativeness.

Time consciousness

Speed and time are benefits that are commonly communicated about the usage of

mobile devices. George and Jones (2000) argue that the experience of time may differ for people, as it is inherent to an intrinsic personality characteristic; they conceptualize it as “an intrinsic part of consciousness”. Kleijnen, Ruyter and Wetzels (2004) define time consciousness “as the extent to which consumers are aware of the passing of time and how they spend it.” This impacts the choice of mobile device in certain phases. As

previously mentioned smartphones are used mostly “on the go” while tablets are mostly used at home. This means that consumers do not have the access to use tablets

whenever they want. It is however possible for them to use smartphones as they always carry this device with them. Consumers tend to look for information when a need is created; high time conscious consumers want to spend their time effectively.

Furthermore, if high time conscious consumers want to purchase a product they do not want to wait until they get home. Therefore we argue that high time conscious

(26)

26 consumers will use their smartphones more than low time conscious consumers in the information search phase and in the purchase/transaction phase. This leads us to the following hypothesis:

H3a: Time consciousness has significant positive impact on smartphone usage in the information search phase

H3b: Time consciousness has significant positive impact on smartphone usage in the purchase/transaction phase

Shopping enjoyment

Shopping enjoyment is about entertainment and benefits that are gained through shopping (Kepmerman, van Delft &Borgers, 2015). Konuş et al. (2008) found that shopping enjoyment is related to channel selection. Therefore one could wonder if shopping enjoyment is also related to device selection. It is already argued that tablets are being used most at home between 6pm and 9pm when sitting on the couch, while smartphones are being used all the time because they are easily “accessible” (Salesforce, 2014). It seems that tablets are being used at a moment of tranquillity after a day of engaging in various activities. Because of this consumers also have less distractions and more time to engage in online shopping. This causes them to enjoy online shopping more than consumers that shop online during daytime. Therefore one could argue that consumers with a need for shopping enjoyment intend to use tablets significantly more than smartphones. This leads us to the following hypothesis:

(27)

27 H4a: Shopping enjoyment has significant positive impact on tablet usage in the

information search phase

H4b: Shopping enjoyment has significant positive impact on tablet usage in the purchase/transaction phase

2.4.3 Expected benefits and costs

Following the line of reasoning from Konuş et al. (2008) it is argued that consumers decide which mobile device they will use anticipating on the utility they gain from the usage in the multiple phases of the shopping process. To increase their utility they need to evaluate certain perceived benefits and costs of the mobile devices in the shopping process. In this research we have chosen one benefit and one cost that distinguishes smartphones from tablets. These factors are respectively risk perception of the mobile device and convenience of the mobile device in the shopping process.

Risk perception

The mobile channel tends to magnify some concerns involved with the shopping process. Consumers perceive a higher level of risk when they shop through a mobile channel instead of an online channel (Makwana & Rahaman, 2013). There are however different kinds of perceived risk. Forsythe, Liu, Shannon and Gardner (2006) argue that time/convenience risk in one of them. They argue that “time/convenience risk includes the inconvenience during online transactions, often resulting from difficulty of

navigation and/or submitting orders, or delays receiving products.” This concept is also applicable to mobile devices. A major factor related to this is the portability of the mobile device. Koyuncu and Lien (2003) found that online users prefer to purchase at home as they see this as a private and secure place. As stated before, smartphones are

(28)

28 being used everywhere while tablets are mostly being used at home. This is because smartphones are easier to carry while tablets are heavier and bigger, which increases the difficulty of purchasing while “on the go”. Therefore, one could argue that tablets are being used significantly more in the purchase/transaction phase as consumers. This leads us to the following hypothesis:

H5: Risk perception has significant positive impact on tablet usage in the purchase/transaction phase.

Convenience

Yale and Venkatash (1986) state that convenience can be grouped in multiple

categories, where portability and accessibility are two of those categories. These are categories that fit in the mobile channel. Users of the mobile channel seem to find convenience a relative advantage over other channels. Bathnar and Gose (2004) have found that convenience is perceived as a benefit while shopping online. In this research we frame convenience as the accessibility to a product/service when consumers desires to do so. This relates to the choice of mobile device in an important manner. Firstly, the portability of devices plays a major role in this. As stated before, it is more convenient to access smartphones anytime while this is more difficult for tablets as they are heavier. Secondly, when consumers are looking for information it is more convenient to use their smartphones than tablets because the content provided is more personalized. Therefore, one could argue that smartphones are perceived as more convenient in comparison to tablets while searching for information. This leads to the following hypothesis:

(29)

29 H6: Convenience has significant positive impact on smartphone usage in the information search phase

2.4.4 Behavioral drivers

This research will also examine which channel (online/offline) consumers use in the different phases of the shopping process. This is done because it is also important to examine if this shopping behaviour has an influence on the use of a certain mobile device in another phase of the shopping process. Previous studies have examined the concept of research shopping where shoppers research in one channel and then

purchase in another one (Verhoef et al., 2007). Showrooming is one example of this type of behaviour. Showrooming is an act of consumers when they search information on an offline channel but then purchase on an online medium where the prices are lower (Zimmerman, 2012). It is important for this research to find out whether there is a preference for smartphones or tablets when searching information offline. Marketing Land (2013) has found that most consumers that are searching for information on an offline channel purchase the product through their smartphones. They argue that location plays a significant role because smartphones are used on the go while tablets are mostly used at home. This leads to the following hypothesis:

H7: Offline channel usage in the information search phase has significant positive impact on smartphone usage in the purchase phase.

(30)

30

3. Conceptual Model

Previous research has provided knowledge about the shopping process and channel choice, but relevant research gaps remain. Firstly, consumer behaviour has been researched between the online and mobile channel (Holmes et al., 2014), but no

research has been done about consumer behaviour within the mobile channel. Secondly, studies have focused on M-shopping drivers (Ruiz et al., 2005), but they did not include why consumers choose certain mobile devices. Thirdly, one may argue that empirical research is needed to find out whether mobile devices are different. And if so, what drives consumers to choose certain mobile devices. This is becoming relevant as recent reports are stating that there needs to be a distinction between mobile devices within the mobile channel (Forrester, 2012). Therefore, this thesis will contribute to the scientific literature. To fill the gaps stated above and to add to the insights of consumer behaviour within the mobile channel the following research question has been

formulated:

“Whether/how mobile shopper users of device (smartphones/tablets) differ in different stages of the shopping process? Are there different factors that have an impact on smartphone or tablet use during the different phases of the shopping process?”

The goal of this research is to examine the effect of certain consumer characteristics on the choice of mobile devices within the shopping process. This research will examine the usage of mobile devices within the shopping process. Furthermore, the goal is to create a deeper understanding into the consumer characteristics that drive the choice of mobile

(31)

31 devices. A conceptual framework is presented in Figure 2 to answer this research

question.

Figure 2. Conceptual Framework

This framework is based on previous studies on channel usage and consumer behaviour. It suggests that the choice of smartphones and tablets in the shopping process are

influenced through certain customer characteristics. The framework visualizes the drivers that are chosen. The research question is to be answered through the previously stated hypotheses. Furthermore, the mobile device usage in three phases of the

shopping process is also being examined. Measuring the usage percentage and usage frequency of tablets and smartphones in the information search, purchase/transaction, and after-sales support phase will add a relevant insight of how the customer journey is experienced within the mobile channel.

(32)

32

3.1 Hypotheses

Multiple hypotheses are developed, based on reviewing past literature. The hypotheses that are being tested in this research are presented in Table 5.

Table 5 Hypotheses Hypotheses Statements

H1a,b Age has significant negative impact on smartphone usage in the shopping process Age has significant positive impact on tablet usage in the shopping process H2 Income has significant positive impact on tablet usage in the shopping process H3a,b Time consciousness has significant positive impact on smartphone usage in the

information search phase

Time consciousness has significant positive impact on smartphone usage in the purchase/transaction phase

H4a,b Shopping enjoyment has significant positive impact on tablet usage in the information search phase

Shopping enjoyment has significant positive impact on tablet usage in the purchase/transaction phase

H5 Risk perception has significant positive impact on tablet usage in the purchase/transaction phase

H6 Convenience has significant positive impact on smartphone usage in the information search phase

H7 Offline channel usage in the information search phase has significant positive impact on smartphone usage in the purchase phase

(33)

33

4. Research Design

4.1 Research scope

The objective of this research is to examine whether mobile shoppers’ choice of device differs in the different stages of the shopping process. Furthermore, this research will also test if specific consumer characteristics influence this choice. The opportunity to study mobile shopping behaviour in practice through empirical research is great. ‘Chapter 2-Literature review- found sufficient evidence on the need for new insights because it had identified a gap within current research about the choices of mobile devices in the shopping process. The aforementioned objectives take this field of

research one step further in the right direction. An important contribution of this master thesis will be the analysis of empirical data on which channel/devices consumers use through the shopping process and the consumer characteristics that drive this choice. By comparing theory with empirical data the researcher will get a fuller comprehension about the differences within the mobile channel and whether/how it is driven. This useful information will contribute to the consumer-firm relationship as firms can personalize their mobile websites and applications. This section -Research Methods- will provide the details of the research strategy chosen to discuss the objectives determined above. It will also discuss how data was collected, the means for sample selection, and the method for the data analysis.

4.2 Research strategy

Saunders, Lewis and Thornhill (2012) argue that it matters, which research strategy, is fitting for your research. The objectives of this research wants to find out “Whether/how mobile shopper users of device (smartphones/tablets) differ in different stages of

(34)

34 shopping process and which perceived customer characteristics have an impact on this differentiation?” Which research strategy is fitting for these objectives?

The empirical research in this Master Thesis is curious about quantitative data (Whether/how mobile shoppers…) within authentic surroundings. Which research strategy is appropriate for a study where a large amount of data needs to be gathered in a short amount of time?

As stated before, this study requires a quantitative approach, not qualitative, where it relates to “research that uses data that do not indicate ordinal values.” (Nwiki, Nyamongo & Ryan, 2001), since it is used most in anthropology, sociology, and

humanities to study in-depth events.

The researcher adopts a positivist view, which is the belief that certain causes determine effects or outcomes. Therefore theories can be made and they can be measured and tested to understand the world (Creswell, 2013). This perspective is in line with the researcher’s views and objectives to which consumer characteristics drive the choice of mobile device.

This quantitative research will be executed through a cross-sectional survey design because this sorts out whether and how the drivers influence the choice of

mobile devices at one given point of time. A cross-sectional survey is a study of particular phenomena at a particular point in time (Saunders, Lewis, and Thornhill, 2012). This design is chosen for this study for multiple reasons. Firstly, through a survey design it is possible to reach an extensive range of various respondents and it is possible to gain a high quantity of data (Saunders et al., 2012). This is necessary for this study, as many respondents are needed to generalize the findings about the mobile device behaviour. Secondly, this design makes it possible to measure numerous constructs in one single

(35)

35 event, which is relevant for this study as we are trying to measure multiple consumer characteristics (Saunders et al., 2012). Thirdly, this study does not aim to manipulate certain variables because the objective is to measure actual behaviour of consumers. Therefore a cross-sectional survey design is the optimal design for this study.

A cross-sectional survey design has limitations that need to be addressed. Firstly, a certain point in time is measured within this design and therefore it is possible that another measurement would provide different results. Secondly, a survey design might make it difficult for the researcher to find causality. However, to remove these

limitations the researcher has done two important things. Firstly, a large group of respondents have been gathered to take away the problem of reliability. Secondly, the researcher devised the framework in such a way that causality is already guaranteed. Consumer characteristics are traits within consumers; therefore consumer

characteristics drive the choice of mobile device and not otherwise.

4.3 Data collection

A relevant subject within the Methods chapter is the procedure of data collection. The following section will explain the sample selection and data collection. Firstly, the sampling technique used in this study will be explained. Secondly, the survey method used for this research will be discussed. Lastly, the research sample will be presented.

4.3.1 Sample selection

Convenience sampling was used to choose consumers for this study. This form of

sampling has been chosen for three specific reasons. Firstly, it is convenient because the objectives of this study are looking at all Dutch consumers with a smartphone and/or tablet. As stated in Chapter 2 -Literature Review- device ownership is very high in the Netherlands. Secondly, this study also considers the consumers that do not shop online

(36)

36 and it is also therefore useful to include those consumers in this study. Therefore, all Dutch consumers can be included within this study, which is convenient. Lastly, convenience sampling has also been chosen because of time limits. To ensure generalization a minimum of 200 respondents are required for this study.

4.3.2 Data collection technique

The survey design was built on one certain data collection technique: an internet-mediated questionnaire. Questionnaires are an effective way of gathering quantitative data, and frequently used in survey research. Saunders et al. (2012) believe that “questionnaires will enable you to identify and describe the variability in different phenomena,” because most questionnaires are a searching for explanatory insights (as this study is) and they can supply insights in a large population (behaviour of many consumers). This technique, although the confidence is low that the right person has responded, is very time-saving. Furthermore, the sample size can be very high and it can be geographically dispersed which is good for generalization. Additionally, the

distribution of the questionnaire finds place through the Internet, which makes it very appropriate for this research (this study focuses on online shopping). Lastly, an internet-mediated questionnaire has closed questions to measure certain variables, which is fitting for this study, as it wants to measure the actual behaviour and characteristics of consumers. Therefore one could conclude that the internet-mediated questionnaire is a suitable technique for this Master Thesis.

There are multiple resources that provide ways to distribute internet-mediated resources; this study will use the software program Qualtrics. Qualtrics is convenient as it lets the researcher export data directly into SPSS and has an advanced user-friendly

(37)

37 dashboard. Furthermore, to ensure a good amount of respondents, three follow-ups will be executed through Qualtrics. These follow-ups will be executed after 3, 5 and 10 days.

An online-mediated questionnaire has limitations that need to be addressed. A big limitation is that the researcher cannot include an abundant number of items as this will lead to weariness from a respondent’s perspective. Therefore the number of items needs to be concise and they need to be well designed to measure the right variables. Therefore this study incorporates items that are used in previous well-known studies. Furthermore, items within the questionnaire could be interpreted in multiple ways, which is bad for the content validity of the study (Saunders et al., 2012). Therefore, a pilot test should be executed before the actual questionnaire. The pilot test was

executed through “respondent debriefing”. In this debriefing four persons were asked about their comprehension and interpretation of the items. Furthermore, they provided their opinion about difficulty and time. The respondents were satisfied with the

questionnaire except for one particular item. This item was changed in the final questionnaire.

4.3.2.1 Variable measurements

Selecting the right variable measurements has the same amount of weight as selecting an effective research strategy. The full survey can be found in Appendix I. As stated before, the items have been extracted from previous studies. Therefore, some items have been adjusted to this specific study so that content validity can be preserved. The first part of the questionnaire measures the preference and actual use of channel choice with three different consumer goods: mobile telecom, clothing and daily foods. This choice has been made to create cross-validation. These items can be answered through a “Yes or No” and are perceived as category questions (Saunders et al., 2012).

(38)

38 The psychographic variables and benefits and costs variables are measured through statements. Respondents answer these statements through a 5-point type scale ranging from “Strongly Disagree”(1) to “Strongly Agree”(5). A 5-point Likert-type scale has been chosen instead of a 7-point Likert-Likert-type scale because respondents seem to “find it difficult to distinguish between values on rating scales of more than five points (Saunders et al., 2012).

Two psychographic variables in this study have been chosen, time consciousness and shopping enjoyment. To measure the degree of time consciousness, three items from Kleijnen, Ruyter and Wetzels (2007) have been incorporated and adjusted accordingly for this study. The items for the second psychographic variable, shopping enjoyment, have been extracted from multiple sources (Barnes, Bauer, Neumann and Huber, 2007; Dawson, Scott, Bloch and Ridgway, 1990; Babin, Barry, Darden and Griffin, 1994).

The perceived benefits and costs variables relevant for this research are risk perception and convenience. The items for both variables have been simplified and adjusted from Forsythe, Liu, Shannon and Gardner (2006).

The behavioural variable in this study is measured through asking the amount of time on the Internet through a category question. Finally, the demographic variables such as income, gender, age and education have been included in the questionnaire as they are part of the conceptual framework. Table 6 presents all the variables and the items by which they are measured.

(39)

39 Table 6

Variables and Items (translated) Variables Items

Actual Info Search

● Which of the following options did you use when looking up information about mobile telecom (subscriptions/phone/accessories etc.)? You can choose more than one option.

● Which of the following options did you use when looking up information about clothing? You can choose more than 1 option.

● Which of the following options did you choose when looking up information about food? You can choose more than 1 option. Preferred Info

Search

● Which of the following options do you prefer when looking up information about mobile telecom (subscriptions/phone/accessories etc.)? You can choose more than 1 option.

● Which of the following options do you prefer when looking up information about clothing? You can choose more than 1 option

● Which of the following options do you prefer when looking up information about food? You can choose more than 1 option

Actual Purchase ● Which of the following options did you use when buying a mobile product (subscriptions/phone/accessories etc.)? You can choose more than one option.

● Which of the following options did you use when buying clothing? You can choose more than one option.

● Which of the following options did you use when buying food? You can choose more than one option.

Preferred Purchase

● Which of the following options do you prefer when buying a mobile telecom product (subscriptions/phone/accessories etc.)? You can choose more than 1 option.

● Which of the following options do you prefer when you buy clothing? You can choose more than 1 option

● Which of the following options do you prefer when you buy food? You can choose more than 1 option

Actual After-

Sales ● Which of the following options did you use when you had questions or complaints about your mobile product (subscriptions/phone/accessories etc.)? You can choose more than one option.

● Which of the following options did you use when you had questions or complaints about your clothing? You can choose more than one option. ● Which of the following options did you use when you had questions or

complaints about your food? You can choose more than one option. Preferred

After-Sales

● Which of the following options do you prefer when you had questions or complaints about your mobile telecom product

(subscriptions/phone/accessories etc.)? You can choose more than 1 option.

● Which of the following options do you prefer when you had questions or complaints about your clothing? You can choose more than 1 option ● Which of the following options do you had questions or complaints about

your food? You can choose more than 1 option

Convenience ● I consider smartphones more convenient than tablets because I can shop whenever I want.

(40)

40 ● I consider smartphones more convenient than tablets because I get more

personalized offers. Time

Consciousness

● I like to make to-do lists that help me to bring order in my activities. ● I think about how to schedule my time effectively.

● I don’t think about how I schedule my time

Risk Perception ● I consider smartphones riskier than tablets because it is more complex to place an order

● I consider smartphones riskier than tablets because the websites has no good format.

● I consider smartphones riskier than tablets because it takes too long before the pictures are loaded.

Shopping Enjoyment

● I like shopping

● I take my time when shopping ● Shopping is exciting

● I enjoy when I am shopping Age ● What is your age?

Gender ● What is your gender? Income ● What is your yearly income?

Education ● What is your highest achieved education?

4.4 Framework for data analysis

This research will use a multiple linear regression to analyze the data. This technique will be used because there are multiple predictors that need to be analyzed together. Furthermore, the dependent variable is continuous and therefore a multiple regression is the only applicable analysis in this situation.

All hypotheses will be tested through this technique because we are looking for significant drivers. The independent variables will be Age, Income, Time Consciousness, Shopping Enjoyment, Risk Perception, and Convenience for all phases. The independent variable offline channel usage in the information search phase will be used in the

regression models for the purchase phase. Income will be divided in three groups of low, medium and high income and there will be two dummy variables whereas medium

(41)

41 income will be the baseline. The dependent variables will be the frequency of tablet and smartphone usage in the different phases of the shopping process for all three product categories. This will eventually lead to 18 different multiple regressions. Table 7

presents the 18 different multiple regressions and the hypotheses that are being tested in these regressions.

Table 7

The hypotheses that are being tested through 18 multiple regressions for smartphone (S) and tablet (T) usage

Mobile Clothing Food

Info search Purchase After-Sales Info search Purchase After-Sales Info search Purchase After-sales

S T S T S T S T S T S T S T S T S T Age H1 a H1b H1a H1b H1a H1b H1a H1b H1a H1b H1a H1b H1a H1b H1a H1b H1a H1b Low Income H2 H2 H2 H2 H2 H2 H2 H2 H2 High Income H2 H2 H2 H2 H2 H2 H2 H2 H2 Time consciousnes s H3 a H3b H3a H3b H3a H3b Shopping enjoyment H4a H4b H4a H4b H4a H4b Risk perception H5 H5 H5 Convenience H6 H6 H6 Offline channel usage in IS-phase H7 H7 H7

(42)

42

5. Results

This chapter will discuss the results of the analysis that has been performed. Firstly, the data quality has been checked before analyzing it. Secondly, general frequency tables are presented. Thirdly, a reliability analysis has been performed. Finally, the results of the hypotheses analysis will be shown.

5.1 Data quality

Before the data has been analyzed, it needed to be checked first on it’s quality. It is a possibility that the questionnaire could be incomplete or ambiguous. It is therefore necessary to delete this incoherent output (Maholtra and Birks, 2007). There were 73 respondents that did not fill in the whole survey; this was tested through measuring whether the last six questions were filled in. These respondents were excluded from the analysis through case wise deletion because their answers were biased. This resulted in a sample size of 153 respondents. Furthermore, some questions (frequency x channel usage) could not be filled in when respondents did not use that particular channel. The output was changed from no value to a value of 0 so that the analysis could be done correctly.

5.2 Descriptives

To give an overview of the respondents that participated in this study, demographics are presented in Table 4.1 of all the respondents. The sample of 153 respondents consists out of 45.8% (70) male and 54.2% (83) female respondents, resulting in a fairly equal distribution. This is quite good for the analysis because this means that the results can be generalized for both genders. Furthermore, most respondents are between 18-34 years old (82.3%) while some older respondents have also participated in this survey

(43)

43 (11.8%). Most respondents have or participate in a Bachelor’s or Master’s degree

(71.9%).

Table 8

Demographic profile of respondents (N=153)

Measure Items Frequency Percentage (%)

Gender Male 70 45.8 Female 83 54.2 Age 18-24 78 51 25-34 57 31.3 35-44 7 4.6 45-54 7 4.6 55+ 4 2.6 Education LBO/VMBO 1 0.7 MAVO 2 1.3 MBO 29 19 HAVO/VWO 11 7.2 HBO/WO 110 71.9

Respondents were also asked which channel they use in the shopping process for three different consumer goods (Mobile telecom/Clothing/Food). These results are presented in Table 8. Note that participants can use multiple channels in a single phase and that therefore the sum of the percentages could be more than 100%. This study focuses on the differences between smartphones and tablets; we will aim our attention to these two channels. It is interesting to notice that smartphones are generally being used more than tablets in all phases. However, the literature review already stated that ownership of smartphones is much higher than tablets and it is therefore explainable why these numbers are higher in all three phases.

(44)

44 Table 9

Percentages of actual usage channel in the shopping process (N=153)

Smartphones Tablets Online (PC) Offline Mobile Info search 69.9% 29.4% 91.5% 42.5% Purchase 34% 14.4% 77.8% 54.2% After-Sales 60.8% 12.4% 68% 52.9% Clothing Info search 60.8% 30.7% 84.3% 74.5% Purchase 30.1% 14.4% 77.8% 85.6% After-sales 19% 9.8% 44.4% 78.4% Food Info search 71.2% 26.8% 77.1% 61.4% Purchase 22.9% 9.2% 32% 89.5% After-Sales 0.7% 0.7% 0.7% 68.6% 5.3 Reliability Analysis

Reliability means whether the data collection techniques give the same output when used multiple times (Bryman and Bell, 2007). Various scales have been used as independent variables and they have been measured through different items. To determine whether the items within these scales and the scale have an internal

consistency, the Cronbach’s alpha analysis has been performed. Saunders et al. (2012) argue that when the Cronbach’s alpha ≥ 0.7 than the scale is considered reliable. The variables that have been tested are respectively time consciousness, shopping

enjoyment, risk perception, and convenience. The results of this analysis can be found in Table 4.3. The scales “Shopping enjoyment” and “Risk Perception” fulfill the

(45)

45 requirement of a minimum Cronbach’s alpha = .07. However, the scales “Time

consciousness” and “Convenience” do not meet this minimum requirement. The scale “Time consciousness” has α= .66, however the Cronbach’s alpha will go up with .02 if the item “I like to make to-do lists that help me to bring order in my activities”. This item is not deleted because there are only three items in this scale and they are therefore valuable for the results as a whole. According to Saunders et al. (2012), a Cronbach’s alpha between 0.60 and 0.70 can be qualified as sufficient. Therefore, both constructs are also used in the analysis and are included in this study.

Table 10

Scale reliability analysis

Construct Cronbach’s alpha(α) N of items Time consciousness .66 3 Shopping enjoyment .84 4 Risk perception .84 3 Convenience .68 2

(46)

46

5.4 Testing the hypotheses

This paragraph consists of the results of the analysis for all hypotheses. As stated in paragraph 4.4, 18 multiple linear regressions have been conducted to test the hypotheses. Firstly, the results of every hypothesis will be displayed and checked. Finally, a summary table will be shown to have an overview of all results.

5.4.1 Hypothesis 1

Hypothesis 1 tests the relationship between age and mobile device usage in the three phases of the shopping process. Firstly, the anticipation was that age has significant negative impact on smartphone usage in the information search, purchase, and after-sales phase (H1a). However, the driver age was not significant in any of these cases. Therefore, H1a is rejected.

Hypothesis 1b suggests that age has significant positive impact on tablet usage in the three phases of the shopping process. For the product category “Mobile Telecom”, age is a positive significant driver in the purchase phase (β= 1.89, p < .05) and in the after-sales phase (β= 2.18 p < .05). For the product category “’Food”, age is a positive significant driver in the after-sales phase (β= 1.19, p < .05). Therefore, H1b is partially accepted.

5.4.2 Hypothesis 2

Hypothesis 2 examines whether income has significant positive impact on tablet usage in the three phases of the shopping process. This hypothesis was also tested through the multiple regressions and two dummy variables have been made to make a

differentiation between low and high income. However, no significance has been found in all multiple regressions. Therefore, Hypothesis 2 can be rejected.

What drives the use of mobile devices in different phases of shopping : do drivers differ in different phases of shopping?