How purchase intention is affected by mobile advertising,
consumer’s risk attitude and the level of price discount.
An experimental study focused on mobile advertising with the physical store and
mobile as the sales channels.
Student name: Laura van Dokkum Student number: 10540016
Version: Final
Submission date: March 17, 2018 Thesis supervisor: Drs. Frank Slisser
MSc. in Business Administration – Digital Business Track University of Amsterdam – Amsterdam Business School
Statement of originality
This document is written by Student Laura van Dokkum who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
Mobile advertising makes unique and personalized exchange of information possible. However, consumers often use their mobile devices for the search of information, while the product is bought in the physical store. This is due to mobile purchases that come with uncertainties and risks. Risk-averse consumers avoid taking these risks, while risk-seekers are more focused on the potential of taking the risk. Therefore, consumer behavior is affected by the type of risk attitude as well as the level of price discount offered in mobile advertising. In order to research this, the following research questions were set up: “How is a consumer’s risk attitude affecting the relationship between mobile advertising with physical store or mobile as the sales channel and purchase intention? And what level of price discount is necessary to overcome the barrier created by the sales channel and consumer’s risk attitude?”. Dutch respondents participated in an online experiment (N = 245). The results of moderation tests indicated that purchase intention is higher for mobile advertising with the physical store as the sales channel compared to mobile advertising with mobile as the sales channel. Consumers with a low level of risk attitude (i.e., being risk-averse) are more likely to purchase the product in the physical store instead of with their mobile device after being exposed to mobile advertising. Risk-seekers, consumers with a high level of risk attitude, have approximately the same purchase intention for both sales channels. Furthermore, the level of price discount consumers would like to get in order to increase their purchase intention is indifferent between the sales channels and the risk attitude. Price discounts in mobile advertising can increase purchase intention in both sales channels, but since the level is indifferent, price discount cannot overcome the risk barrier of the mobile sales channel.
Table of Content Abstract 3 1. Introduction 6 2. Literature Review 9 2.1 Mobile advertising 9 2.2 Purchase intention 11
2.2.1 Physical store purchase intention 11
2.2.2 Mobile purchase intention 12
2.2.3 The effect of mobile advertising on purchase intention 14
2.3 Risk 16
2.3.1 Risk society 16
2.3.2 Risk in decision making 17
2.3.3 Consumer’s risk attitude 19
2.4 Price discount 23
2.4.1 The relationship between mobile advertising, price discount and consumer’s risk
attitude 24 2.5 Conceptual framework 26 3. Methodology 28 3.1 Sample 28 3.2 Manipulation 29 3.2.1 Pre-test 30 3.3 Measurements 32 3.3.1 Control variables 32
3.3.2 Consumer’s risk attitude 32
3.3.3 Purchase intention 33
3.3.4 Level of price discount 34
3.4 Statistical procedure 35
4. Results 36
4.1 Manipulation check 37
4.2 Correlation 38
4.3 Moderation effect 40
4.3.1 Effect of consumer’s risk attitude on the relationship between mobile advertising
4.3.2 Effect of consumer’s risk attitude on the relationship between mobile advertising
and level of price discount 44
5. Discussion 46 5.1 Theoretical contributions 47 5.2 Limitations 50 5.3 Practical implications 52 6. Conclusion 53 References 55 Appendices 60
1. Introduction
Mobile devices are becoming part of the daily life. The sophistication of the Internet goes along with the booming growth of e-commerce, the business conducted through electronic devices. In the US, e-commerce grew over the last years with twenty-three percent (Wallace, 2016). Along with this growing business, an expected trend of 2017 was the increasing importance of mobile smartphones. By the end of 2017, it was expected that 4.77 billion people over the world would use smartphones (Steinberg, 2016). The use of mobile devices has increased the last years and make up over half of all web traffic. Especially, the development of a mobile application in order to deliver an optimized experience to consumers increased in importance (Steinberg, 2016). Because of the modern wireless communication technology nowadays, consumers are enabled with new types of e-commerce transactions, also described as mobile commerce (m-commerce) (Lin & Wang, 2006). Mobile commerce can be defined as the use of wireless technology or a wireless telecommunication network, particularly handheld mobile devices and mobile Internet to facilitate the search for information, transactions and user task performance (Hung, Yang, & Hsieh, 2012). Because of this growing market, consumers are harder to win, easier to lose and fussier on price and user experience (Allen, 2017). This emphasizes the importance of mobile applications, which is not only delivering an optimized experience but for organizations also to offer the best deals (Steinberg, 2016).
As part of mobile shopping, mobile services make unique and personalized exchange of information possible. These services are becoming increasingly important for consumers but also for organizations, because of ubiquitous, universal, and unison access to information and services. Examples of these m-services are mobile advertising, text messaging, gaming, contact, and payment. It has several characteristics, such as person-to-person interactive services when sending a text message or with contact services. For payment with a mobile device, the m-service can be characterized as goal-directed services. In these cases, mobile
services such as mobile advertising connect consumers with organizations (Nysveen, Pedersen, & Thorbjørnsen, 2005). This connection between consumers and e-commerce retailers can be seen as part of mobile shopping (m-shopping).
This connection empowers consumers with the ability to gather information anywhere and anytime from multiple sources. Consumers can check on product availability or special offers and alter the selection they make at any point along the path to purchase (Lai, Debbarma, & Ulhas, 2012). Nowadays, the path to purchase represents the process of searching for information, purchasing the product and engaging in post-purchase activities across different channels, manufacturers and retailers (Shankar, Inman, Mantrala, Kelley, & Rizley, 2011). In the pre-purchase phase or the phase where consumers search for product information, consumers form an expectation about the (online) transaction with the retailer. The higher the expectation is for a certain retailer, the more a consumer is willing to make a transaction through that retailer’s store, website or mobile application (D. Kim, Ferrin, & Rao, 2009). In the search and purchase phase, consumers interchangeably and seamlessly use different channels, therefore showrooming or webrooming can occur (Verhoef, Kannan, & Inman, 2015). Showrooming refers to consumers searching for information in physical stores while simultaneously searching on their mobile device for other information and better prices. On the opposite, webrooming is searching for information online while purchasing it in a physical store.
The reason why webrooming occurs is that over half of the consumers are not happy with purchasing goods or services with a mobile device, therefore they prefer purchasing on a PC or even in physical stores. The reason why consumers do not use their mobile device for a purchase could be because consumers perceive mobile purchases as a risky activity or because of privacy concerns (Grewal, Bart, Spann, & Zubcsek, 2016; Korgaonkar & Karson, 2007). The decision of taking this risk of adopting mobile shopping can be determined by the
consumer’s risk attitude (Hubert, Blut, Brock, Backhaus, & Eberhardt, 2017). A consumer’s risk attitude influences the purchasing behavior of a consumer. However, no research has focused yet on the moderating role of a consumer’s risk attitude on the purchasing behavior of a consumer and the level of price discount when being exposed to mobile advertising. Therefore, this research is trying to fill this gap. Based on that, this research is trying to answer the following research questions: “How is a consumer’s risk attitude affecting the relationship between mobile advertising with physical store or mobile as the sales channel and purchase intention? And what level of price discount is necessary to overcome the barrier created by the sales channel and consumer’s risk attitude?”.
The findings of this research contributed to the academic literature because it focused on the moderating effect of consumers’ risk attitude on the relationship between mobile advertising and purchase intention and between mobile advertising and level of price discount. Besides, no research yet has focused on the risk attitude of consumers within the mobile commerce. Moreover, other studies have not focused on making a distinction between physical store and mobile purchases within the mobile commerce.
The results provide managers information whether consumers still prefer purchasing products in the physical store instead of with their mobile device. This can be important for the mobile advertising strategy executed by organizations. In addition, the results indicate what level of price discount is necessary to overcome the barrier created by the sales channel and consumer’s risk attitude. Especially, for risk-averse consumers in the mobile sales channel, knowing what level of price discount helps them to overcome the perceived risk of mobile purchases can be of great value for organizations. This information could be used by organizations in their marketing strategy as well.
In the next paragraph, earlier findings of mobile advertising were discussed followed by physical store and mobile purchase intention, consumer’s risk attitude and the level of price
discount. In the paragraph thereafter, the methodology is elaborated including the sample, manipulations, the measurements, and the procedure. The paragraph thereafter contained the analysis of the correlation and the moderation tests that were done in order to test the hypotheses. In addition, the most important findings were given as well as discussed. This is followed by the conclusion consisting the answers to the research questions.
2. Literature Review
In the literature review, mobile advertising in mobile commerce was discussed first. Mobile advertising leads to purchase intention, which was discussed as well. In addition, there was made a distinction between physical store and mobile purchase intention. Thereafter, risk in mobile commerce and consumer’s risk attitude were elaborated as well as price discounts that both could influence the purchasing behavior of consumers when being exposed to mobile advertising. This is followed by the contributions and the conceptual model.
2.1 Mobile advertising
As already mentioned, m-commerce and e-commerce as well have been growing in importance (Steinberg, 2016). Along with this growing importance was the rise of mobile advertising, or m-advertising, which has become a booming business. One dominant reason for this growth is the worldwide spread and adoption of smartphones and other mobile devices (Grewal et al., 2016). Organizations can, in order to increase sales, offer products or services through mobile advertising which could influence consumer purchasing behavior (Bart, Stephen, & Sarvary, 2014; Dehghani & Tumer, 2015). Mobile advertising as part of mobile services come in different types: video, text advertisements, multimedia, short message service (SMS) and mobile display advertising (Hongyan & Zhankui, 2017). Organizations can either push or pull mobile advertising. Push-based advertising are offers often delivered via SMS or mobile apps
if consumers agreed to receive push messages. Mobile advertising delivered as browser-based or within-app advertisements refers to pull-based mobile advertising (Grewal et al., 2016). One type of mobile advertising, a mobile display advertisement, is a small banner image displayed on a mobile phone’s screen either in a mobile web browser or in an application (Bart et al., 2014). They contain minimal information such as a logo, slogan or a very short message. Consumers are often exposed to the advertisement when they are on the move, distracted or attending to other stimuli so that they pay not much attention to it. However, with mobile advertising, organizations can reach the right consumers at any time and anywhere (Wong, Tan, Tan, & Ooi, 2015). For example when organizations use trajectory-based mobile advertising. In this case, “trajectory” refers to the physical behavioral trace of a consumer’s individual movement. Based on that, consumer preferences are analyzed and used for personalized mobile advertising strategy (Ghose, Li, & Liu, 2014).
Other factors influencing the effectiveness of mobile display advertisements are contextual factors such as location and weather. The reason for that is that each factor activates different goals, which could influence its effectiveness (Grewal et al., 2016). The study of Bart et al. (2014) found that product characteristics also matter in mobile display advertising effectiveness. The most effective mobile display advertisements are offers with high involvement and utilitarian products. In addition, consumer context is important for the effectiveness of mobile advertising as well. Consumers being exposed to the mobile advertisement could be in different phases of the customer journey. A consumer may already be searching for information about the offered product, or the advertisement stimulates consumers’ recognition of an unmet need (Grewal et al., 2016).
Moreover, the content information of the advertisement the marketer chooses differs each mobile display advertising campaign. This could influence the effectiveness too. The content information could be sweepstakes, buy-one-get-one non-price promotions or price
promotions (Grewal et al., 2016). The content could also include that, for example, the sale is time-limited meaning that it is only available for today or for tomorrow. Besides, it could be limited to a specific channel such as online or offline. In turn, this sales channel could influence the physical store or mobile purchase intention of consumers.
2.2 Purchase intention
As stated before, the (mobile) shopping process consists of three crucial stages: the information search, the (online) purchase of the product or service and the post-purchase phase where a consumer can engage with the organization (Shankar et al., 2011). Because showrooming and webrooming occur, it is for organizations important to understand if consumers are willing to buy a specific product with their mobile device or prefer to buy it in a physical store.
2.2.1 Physical store purchase intention
In general, purchase intention is widely studied and is seen as a predictor of subsequent purchase (Grewal, Krishnan, Baker, & Borin, 1998). When consumers are intended to make a purchase, they face uncertainty about the consequences of the decision (Flanagin, Metzger, Pure, Markov, & Hartsell, 2014). Multiple factors can influence this uncertainty such as the vendor, product, brand or mode of purchase which in turn, could influence the intention to make a purchase. Since the rise and growing importance of omnichannel experience, consumers use different channels simultaneously in their path to purchase (Verhoef et al., 2015). For example, positive word-of-mouth is found to encourage purchases for consumers who perceive the purchase as a high risk. For offline commercial transactions, studies have found that perceived product quality positively correlates with purchase intention (Chang & Wildt, 1994; Flanagin et al., 2014). In addition, promotions involving price discounts increase store traffic and stimulate purchase (Grewal et al., 1998). This study has also shown that brand
name, store image, perceived value, and store name positively influence purchase intention. The study of Flanagin et al. (2014) has shown that in the context of e-commerce, product quality, website quality, ratings, and perceived credibility of electronic word-of-mouth messages all have a positive impact on purchase decisions.
These studies have shown the determinants of purchase intention in general whether this was offline or online purchases. However, prior studies have not focused on making a distinction between channels. What determining factors are for consumers for mobile purchase intention is being discussed next.
2.2.2 Mobile purchase intention
As stated before, nowadays, mobile shopping is becoming more important which is illustrated by, for example, showrooming. Showrooming refers to consumers searching for information in physical stores while using their mobile device simultaneously for searching information, better prices and eventually for mobile shopping. Mobile shopping refers to any monetary transaction related to purchases of goods or services through mobile smartphones that are enabled with Internet or over the wireless telecommunication network (Groß, 2015). Because of the ability to make mobile purchases, consumers can buy anytime anywhere and they are no longer obliged to visit the physical store to make a purchase because the purchase can be delivered at home (Pantano & Priporas, 2016). Until today, mobile shopping is mostly used in service industries for purchasing tickets such as public transportation, while financial products are still not bought often through mobile devices (Hubert et al., 2017).
Existing literature studying mobile purchase intention has, as a theoretical basis, made use of the Technology Acceptance Model (TAM), Unified Theory of Acceptance and Use of Technology (UTAUT), theory of planned behavior (TPB), expectancy confirmation model (ECM) and perceived value theory (Gao, Waechter, & Bai, 2015). According to this literature,
factors that influence mobile purchase intention are ease of use, usefulness. and enjoyment of the mobile application (Ko, Kim, & Lee, 2009). Another study has stated that trust, flow, and perceived usefulness positively determine mobile purchase intention (Gao et al., 2015; Zhou, 2013). In addition, the study of Groß (2015), who used the modified TAM model, has shown that perceived enjoyment and trust in the mobile vendor affect the intention of consumers to engage in mobile shopping. This was in addition to the traditional TAM factors, which are perceived ease of use and perceived usefulness (Groß, 2015; Pantano & Priporas, 2016).
However, one reason for consumers not to buy products or services (again) with their mobile devices could be lack of trust in mobility. Trust can be developed by the party being trusted (trustee) through creating the perception that the trustee possesses characteristics that are beneficial for the trustor (the party placing trust) (Li & Yeh, 2010). To reduce the uncertainty and risks of mobile transactions, such perception is important. Trust is about the willingness to take a risk rather than not taking the risk. It helps to address concerns over privacy and security essential to online transactions and it becomes especially pivotal in selecting products or services that are already perceived as risky (Korgaonkar & Karson, 2007). Also, compared to offline commerce, mobile purchases involve great uncertainty and risk because of the vulnerability of a hacker attack and information interception (Gao et al., 2015). This concern results in consumers doubting whether organizations can effectively protect personal information, location privacy and payment security from potential problems. Consumers, who perceive a mobile purchase as too risky, avoid the mobile purchase and instead buy it offline. Another reason for avoiding mobile purchases could be that consumers are very sensitive to services that involve monetary transactions because they worry about both money and information loss (Kleijnen, De Ruyter, & Wetzels, 2007). In that sense, mobile purchases can also be seen as a risk that consumers need to take in order to receive the service or product.
2.2.3 The effect of mobile advertising on purchase intention
Existing literature has shown that a substantial proportion of consumers that are exposed to mobile advertising are more likely to purchase the advertised brand (Bart et al., 2014). However, price promotions in mobile advertising attract low-income consumers, because these consumers are highly price sensitive. High-income consumers, on the other hand, are less sensitive to price discounts, but more sensitive to the quality of mobile targeting (Ghose et al., 2014). Besides income, the attitude of consumers towards mobile text advertising determines purchase intention as well. If the consumer is satisfied with one type of mobile advertising, they are likely to accept the included information and to purchase the offered product or service (Hongyan & Zhankui, 2017).
However, these findings focused on the relationship between mobile advertising and purchase intention in general. Since there are differences between mobile or physical store purchase intention, this distinction should be made. One study has found that organizations that use trajectory-based mobile advertising increase the physical store purchase intention (Ghose et al., 2014). As stated above, “trajectory” refers to the physical behavioral trace of a consumer’s individual movement. Based on that, consumer preferences are analyzed and used for personalized mobile advertising strategy (Ghose et al., 2014). This study has shown that there is a significant and positive direct relationship between trajectory-based mobile advertising and the revenues of the physical store. This could be the consequence of studies stating that mobile devices still are most relevant in the search phase of the path to purchase, but that consumers still prefer to buy the product in the physical store (Gupta & Arora, 2017; Verhoef et al., 2015).
In this sense, mobile advertising provides consumers with relevant information and promotions, which in turn could influence their purchasing behavior. However, mobile
advertising can also induce negative consequences, because consumers might believe that their privacy is being invaded (Grewal et al., 2016). Such privacy invasion results in negative attitudes towards the organization or brand and influences consumer behavior as well. This could lead to a greater sense of vulnerability and a reduction of click-through rates. Organizations can overcome this problem by first, being careful with mobile advertising strategies (Grewal et al., 2016). The acceptance of mobile marketing increases significantly when organizations use mobile advertising carefully. It should provide consumers with personalized and contextual-related information, ensuring consumers’ privacy and avoid spamming. In this way, purchase decisions are encouraged as well (Groß, 2015). This is also shown by another study that has found that the likelihood of a mobile purchase increases as a function of crowdedness. When consumers are in a more crowded environment, they may be more deeply immersed in their mobile devices, which leads to a higher likelihood of making a mobile purchase (Andrews, Luo, Fang, & Ghose, 2015). However, as mentioned before, consumers are often exposed to mobile advertising when they are on the move, distracted or attending to other stimuli, so that they pay not much attention to it (Bart et al., 2014). Therefore, consumers that are in a crowded environment, could be distracted and pay less attention to mobile advertising. In turn, this could decrease their purchase intention.
Nevertheless, consumers who were low spenders before adopting mobile shopping, increase their amount of mobile orders and place larger orders as soon as they adopt mobile shopping. These consumers often order habitual products that they have purchased before (Shankar et al., 2016; Wang, Malthouse, & Krishnamurthi, 2015). Moreover, privacy concerns regarding mobile shopping can be overcome as soon as an organization is being trusted by consumers (Korgaonkar & Karson, 2007). Especially for products that are perceived as risky, trusting the organization results in a higher willingness of consumers to take the risk and to make the mobile purchase. However, the willingness of taking this risk depends, besides trust,
also on personal characteristics such as consumer’s risk sensitivity that is discussed below (Hubert et al., 2017).
Based on the factors mentioned before, it is expected that mobile advertising has more impact on physical store purchase intention than mobile purchase intention, because of the uncertainties and risks involved in mobile purchases. Therefore, this research hypothesized the following:
Hypothesis 1: Mobile advertising with the physical store as the sales channel has a stronger positive effect on purchase intention compared to the effect of mobile advertising with mobile as the sales channel on purchase intention.
2.3 Risk
This paragraph is committed to risk which is discussed based on the risk society as a sociological point of view and risky decisions. This latter consists of two different points of view at risk, namely psychological and economical. In the end, this leads to the risk attitude of consumers which is being discussed as well.
2.3.1 Risk society
Before getting to the risk attitude of consumers, it is necessary to get a clear understanding of risk in general. According to the sociologist Ulrich Beck, there has taken place a process of individualization where agents are better able to reflexively create themselves and the societies in which they live (Ritzer & Stepnisky, 2014). Beck also states that today the central issue is the risk and how it can be prevented, minimized, or channeled instead of wealth which was the central issue before. Nowadays, the ideal in the society is safety and no longer equality as before. Although the risk is the central issue of the modern society, it is produced by individuals themselves, especially by the sources of wealth in modern society. Therefore, the risk is not
restricted to place or time (Ritzer & Stepnisky, 2014). This effect of individuals or nations producing risk and coping with risk at the same time is called the “boomerang effect” according to Beck (Ritzer & Stepnisky, 2014). Next to risk, the modernization produces reflexivity too. Reflexivity involves that individuals themselves, who are victims of the risks in the society, begin to reflect on those risks as well. These individuals become experts by observing and collecting data on risks and its consequences. Therefore, they question the modernization and its dangers. It is a consequence of individuals who no longer can rely on scientists to do it for them (Ritzer & Stepnisky, 2014). Since mobile purchases are still seen as a risk, individuals nowadays are becoming experts by observing and collecting data on this risk. This has the consequence that they question its danger which could lead to a higher probability of adopting mobile shopping.
2.3.2 Risk in decision making
The information above shows that nowadays consumers live in a risk society where they are aware of risks and where they are trying to prevent, minimize or channel these risks (Ritzer & Stepnisky, 2014). When looking at consumers individually, at the micro-level, consumers face daily situations that require them to decide between actions that differ in level of risk (Figner & Weber, 2011). There are two different points of view by which consumers can look at these decisions that involve risk.
First, the risky decision-making process from a psychological point of view consists of processes that can be either “hot” or “cold” as Figner and Weber (2011) stated. These processes affect the decision-making process of consumers. When decisions are made in hot, affective processes and emotions, their decision can be influenced by (a) directing attention to different characteristics of choice options; (b) by influencing the translation of outcomes and probabilities into subjective values and (c) by influencing the process of choice itself more
directly through, for example, resisting temptation or succumbing to it (Figner, Mackinlay, Wilkening, & Weber, 2009; Figner & Weber, 2011). The affection in the “hot” decision making can be either integral, deriving from the decision or choice options at hand, or incidental as in a source unrelated to the decision such as enjoying a bonus that may lead to speeding on the highway. This process is also called System 1 and can be characterized as automatic, largely unconscious, and relatively undemanding of computational capacity (Stanovich & West, 2000). It can also be seen as the peripheral route in the decision making of the Elaboration Likelihood Model (ELM) (Petty & Cacioppo, 1986).
Logically, the other process next to System 1 is System 2. This latter system refers to the decision-making process based on cold, deliberative processes or the central route of ELM. It conjoins the various characteristics that have been viewed as typifying controlled processing where decisions are being made on the behalf of deliberation (Figner & Weber, 2011; Stanovich & West, 2000). This psychological view at risk involves psychological models that construct risk and return as psychological constructs, which is in contrast with the economic point of view that is discussed below. One other important difference between these processes of decision making is that they can lead to different decisions. For example, when a consumer gave birth, the consumer might be overwhelmed by emotions. These emotions could have the consequence that via the peripheral route, the consumer make a mobile purchase when being exposed to mobile advertising.
Second, from an economic point of view, some high risky options might come with greater returns because of the high level of riskiness. Risky decision making involves the result of the tradeoff between the expected return of an option and the perceived risk of an option (Figner & Weber, 2011). This tradeoff model originally comes from the field of finance, where expected benefits and risk are objective measures. In the mobile commerce, mobile purchases involve making risky decisions as well, therefore consumers might make a tradeoff between
the expected returns of the purchase and the risks. Greater expected return makes an option more attractive and leads to a greater approach of consumers, while a greater perceived risk of an option makes it less attractive and leads to greater risk avoidance (Figner & Weber, 2011). According to this study, differences in these risk-taking levels depend on individual differences. First of all, because of differences in the perceptions of expected benefits, secondly because of differences in the perceptions of the risks. In turn, these perceptions of risk and return are subjective and vary across situational contexts and across decision makers (Figner & Weber, 2011). Finally yet importantly, differences in risk-taking levels occur due to differences in how much risk consumers are willing to accept in exchange for a specific return. This latter characteristic is described as an individual’s risk attitude.
2.3.3 Consumer’s risk attitude
The decision consumers make regarding purchasing a product or service in the physical store or with their mobile device, is affected by characteristics such as consumer’s risk sensitivity (Hubert et al., 2017). Risk sensitivity refers to the probability of an individual choosing a riskier or less risky option (Weber, Shafir, & Blais, 2004). In turn, the willingness of consumers to accept risk in exchange for a specific return is called the risk attitude of consumers (Figner & Weber, 2011). This risk attitude is commonly considered as a personal trait and can be placed on the continuum of risk aversion to risk seeking (Weber, Blais, & Betz, 2002). A consumer shift on this continuum from one to another depending on the domain where decisions have to be made in (Weber et al., 2002). For example, when making an organizational decision, the employee might be more risk-taking because it is the organization’s monetary decision the employee has to make. On the other hand, for personal decisions regarding recreational activities such as skydiving, the individual might be more risk-averse. These examples indicate that there are multiple domains where decisions have to be made in. These domains include
first of all financial decisions, including both investing decisions, such as investing money in Bitcoin, and gambling decisions such as gambling money on the winner of a soccer game. Other domains are recreational decisions such as bungee jumping, and personal decisions. This personal domain can be broken down into smaller categories such as health/safety (e.g., smoking), social (e.g., confronting one’s coworker), and ethical decision (e.g., cheating on an exam) (Weber et al., 2002).
This study of Weber et al. (2002) has shown that an individual’s risk attitude is highly domain-specific and that woman, for example, are more risk-averse than men in all domains except social risk. Another study has found that older adults are more risk-seeking than younger adults (Weber et al., 2004). However, the study of Figner and Weber (2011) has shown that adolescents are willing to take more risk than children or adults, but only when the individuals use the hot, affective system of decision making. These findings indicate that the level of risk a consumer is willing to take depends on the psychological process the consumer uses, the age and which domain the choice has to be made in.
In order to determine a consumer’s risk attitude, different methods can be used. One method that was used in the study of Figner and Weber (2011) is the Columbia Card Task (CCT), which creates a hot and cold version of risky making task. This decision-making process is focused on behavior. Since this research focuses on the attitude of consumers, a scale is needed that assesses risk attitude in every domain situations using self-report questionnaires (Appelt, Milch, Handgraaf, & Weber, 2011). Therefore, the method of the Domain-Specific Risk-Taking (DOSPERT) Scale is more appropriate than CCT. It observes the risk-taking levels among consumers, where a low level of risk attitude refers to being risk-averse and a high level of risk attitude indicates a risk-seeking consumer (Weber et al., 2002).
2.3.3.1 The relationship between consumer’s risk attitude and purchase intention within the mobile commerce
Compared to traditional offline and online environments, consumers perceive more uncertainties and risks in the mobile environment. This is due to the lack of visual representation of the product, the product description and the payment system, which are displayed on the small screen of mobile devices (Gao et al., 2015). These risks in the mobile environment can be classified into three risk facets: financial risk, performance risk and security risk (Hubert et al., 2017). According to this study, financial risk refers to the money loss when using a mobile application. Performance risk is the possibility of flaws in the mobile application and whether the app does not work as it is intended to work. Lastly, security risk refers to the possibility of losing control of personal information (Kleijnen et al., 2007). Because of these risk facets, consumers may perceive mobile purchases as a risk.
The relationship between consumer’s risk attitude and either physical store purchase intention or mobile purchase intention can be understood on behalf of two different risk attitudes. Existing literature has shown that a consumer’s risk attitude can be put on a continuum involving two distinctive behaviors: risk-taking and risk-averse behavior (Gupta, Su, & Walter, 2004; Kahneman & Tversky, 1979).
Risk-averse individuals prefer brick-and-mortar retailers for personal and direct interfaces with the retailers (S. H. Kim & Byramjee, 2014). These risk-averse consumers want to stay away from the negative potentials of uncertainties by keeping the status quo (Chernev, 2004). Although, when facing these uncertainties, they want to reduce it by evaluating more products, therefore they are more willing to pay premium prices (Gupta et al., 2004). Because, as stated before, mobile purchases go along with uncertainties and risks, it is expected that these consumers avoiding risks, also avoid making mobile purchases regardless of price
discounts. They prefer to buy a premium price for a product in a physical store (Pantano & Priporas, 2016). One reason for consumers to be more risk-averse than others and to avoid mobile purchases is that these consumers have never used mobile marketing before (Izquierdo-Yusta, Olarte-Pascual, & Reinares-Lara, 2015).
On the other hand, risk takers weigh benefits of mobile shopping more heavily than the potential harms from e-commerce. They are willing to benefit from the positive potentials of uncertainties in such a way that they are rather entrepreneurial and promoting growth. These types of consumers depart from the current state (Chernev, 2004). Among these risk takers, there are risk seekers who deliberately seek for risky options such as gamblers (Gupta et al., 2004; Kahneman & Tversky, 1979). These types of consumers are aware of the uncertainties of mobile purchases but are willing to take the risk, therefore it is expected that these consumers purchase with their mobile device.
According to prior research, a consumer’s risk attitude influences the purchasing behavior of a consumer. However, no research has focused yet on the moderating role of a consumer’s risk attitude on the purchasing behavior of a consumer when being exposed to mobile advertising. Therefore, this research is trying to fill this gap. Based on the findings, the following hypotheses are set up:
Hypothesis 2a: The positive relationship between mobile advertising with physical store as sales channel and purchase intention is moderated by a consumer’s risk attitude, so that this relationship is stronger for a low level of consumer’s risk attitude (i.e., being risk-averse) compared to a high level of consumer’s risk attitude (i.e., being risk-seeking).
Hypothesis 2b: The positive relationship between mobile advertising with mobile as sales channel and purchase intention is moderated by a consumer’s risk attitude, so that this relationship is stronger for a high level of consumer’s risk attitude (i.e., being risk-seeking) compared to a low level of consumer’s risk attitude (i.e., being risk-averse).
2.4 Price discount
Because mobile shopping is still seen as a risky activity, consumers might avoid purchasing products with their mobile device. One stimulant organizations can use, are price discounts promoted in mobile advertising. This form of sale promotions delivers an incentive from the organization to the consumer to purchase a brand’s product. It is defined as a short-term reduction of the listed price when all buyers are equally eligible for the price reductions. Moreover, price promotions strongly, positively correlate with purchase intention (Alford & Biswas, 2002; Nusair, Jin Yoon, Naipaul, & Parsa, 2010).
Price discount can be framed in several ways by organizations. The framing of a discount is the use of the buyer’s decision-related information to evaluate a product or service relative to a reference point (Kahneman & Tversky, 1979). An organization can frame its price discount in euro terms (€ off), percentage terms (percentage off), or a combination of the two frames. Presenting the discount in absolute terms (€ off) yields higher response levels compared to percentage off. However, in service industries, most consumers prefer sale promotions in percentages (Bitta & Monroe, 1981; Nusair et al., 2010). Besides, price discounts presented in euros and cents are more effective for high-priced products when the discount size is small. For large discount sizes, the price discount can be given in absolute and relative terms. Percentage presentation of the price discount is more effective for low-priced products when the discount is large. When the discount size is small both absolute and relative terms can be effective (McKechnie, Devlin, Ennew, & Smith, 2012). Moreover, organizations have to make a decision of what to present. They can present the reference price and the sale price for example. When presenting only the sale price, the perceived savings of consumers are significantly lower than presenting reference price information with the sale price (Bitta & Monroe, 1981).
2.4.1 The relationship between mobile advertising, price discount and consumer’s risk attitude Price discounts can be seen as an element of mobile advertising (Grewal et al., 2016). As stated above, organizations can use multiple ways to frame their price discounts. When consumers respond differently to various descriptions of the same decision problem, framing effect occurs (Nusair et al., 2010). In this sense, price framing can influence consumers’ decision behavior and purchase intention (Biswas & Grau, 2008).
Besides the presentation of the price discount in mobile advertising, the amount of the price discount is maybe even more important. In general, a higher discount level significantly reduces search intention and increases value perceptions and buying intention. It is an effective tool for organizations to increase sales and product trial (Alford & Biswas, 2002; Nusair et al., 2010). Mobile advertising involving price discount, promotional activity and brand information properly formatted for each mobile device, provides consumers with detailed product information in a quick way. Receiving and viewing the personalized advertising content leaves a good impression and increase consumers’ desire to make a purchase (Izquierdo-Yusta et al., 2015).
However, there is also criticism of its effectiveness. Several studies have shown that price discounts can lead to undermining of the perceived quality of the discounted product which leads to more negative consumer perceptions. In line with this, actual purchases of products that are offered with discount are perceived by consumers as poorer quality products (Grewal et al., 1998). This, in turn, decreases the probability of future purchases or repurchases (Nusair et al., 2010). A discount that is too large may be perceived as suspicious and leads to more negative perceptions (Alford & Biswas, 2002; Bitta & Monroe, 1981). According to this study of Bitta and Monroe (1981), retailers should offer price discounts of approximately 15 percent in order to attract consumers to buy a product.
Again, these findings of previous studies have not made clear the differences between the effect of mobile advertising with price discounts on different sales channels. However, focused on mobile purchases, price discounts are price savings which are a factor for consumers to adopt mobile shopping. Consumers who are highly sale prone, meaning highly focused on price savings, are expected to shop through non-traditional retail formats such as mobile shopping when being exposed to mobile advertising (Gupta & Arora, 2017). In this sense, price discount can convince consumers to adopt mobile purchases. However, it also could be that consumers still perceive the mobile purchase as too risky and avoid adopting mobile shopping, regardless the price discount. Therefore the following hypothesis was set up:
Hypothesis 3: Mobile advertising with mobile as the sales channel has a stronger positive effect on the level of price discount compared to mobile advertising with the physical store as the sales channel.
In addition, this research focused on the effect of a consumer’s risk attitude on the level of price discount when being exposed to mobile advertising. In other words, the focus is on what level of price discount consumers would need in order to overcome the barrier created by the sales channel and consumer’s risk attitude. Existing literature has not focused on this yet. However, expected is that for mobile purchases risk-averse consumers would like to get a higher level of price discount compared to risk-seeking consumers since they perceive mobile purchases as a greater risk than risk seekers. Besides, with a physical store as the sales channel, it is expected that averse consumers need a lower level of price discount compared to risk-seeking consumers because risk-averse consumers might perceive mobile commerce also as a risk or an uncertainty. When facing these uncertainties more products are evaluated and a higher level of price discount could not influence them instead, they prefer to buy premium prices.
Hypothesis 4a: When being exposed to mobile advertising with mobile as the sales channel, consumers with a low level of consumer’s risk attitude (i.e., being risk-averse) need a higher level of price discount compared to consumers with a high level of consumer’s risk attitude (i.e., being risk-seeking) in order to purchase the laptop.
Hypothesis 4b: When being exposed to mobile advertising with the physical store as the sales channel, consumers with a low level of consumer’s risk attitude (i.e., being risk-averse) need a lower level of price discount compared to consumers with a high level of consumer’s risk attitude (i.e., being risk-seeking) in order to purchase the laptop.
2.5 Conceptual framework
Because of the growing importance of mobile shopping, this research was focused on mobile advertising and its effect on purchase intention. In particular, there was made a distinction between two sales channels, namely physical store and mobile. With this distinction, it can be investigated whether consumers avoid mobile purchases or were more likely to take the risk of mobile shopping. Moreover, what the moderating effect was of a consumer’s risk attitude on this relationship between mobile advertising and purchase intention was investigated. A consumer’s risk attitude can be described as the willingness of consumers to take a risk in exchange for a specific return (Figner & Weber, 2011). Since no research has focused on this moderating effect of consumers’ risk attitude on purchase intention when being exposed to mobile advertising with physical store or mobile as the sales channel, this research was trying to fill this gap. In addition to that, consumers might need a certain level of price discount in order to adopt mobile shopping. Therefore, this research was also focused on the level of price discount needed to overcome the barrier created by the sales channels and consumer’s risk attitude. Again, the moderating effect of consumer’s risk attitude on this relationship was investigated in this study as well, since it was not researched yet. Therefore, the following
research questions were set up: “How is a consumer’s risk attitude affecting the relationship between mobile advertising with physical store or mobile as the sales channel and purchase intention? And what level of price discount is necessary to overcome the barrier created by the sales channel and consumer’s risk attitude?”.
The findings of this research contributed to the academic literature because it is trying to fill in the gaps stated above. Besides, no research yet has focused on consumers’ risk attitude within the mobile commerce. Furthermore, this research has made a distinction between physical store and mobile sales channel in mobile advertising which existing literature have also not focused on yet.
The findings of this study are relevant for managers as well. The results confirm if webrooming still occurs. It also shows whether consumers still prefer buying in the physical store after being exposed to mobile advertising or that these mobile services can influence the mobile purchasing behavior. This can be important for the mobile advertising strategy executed by organizations. Also, the results of this study will show if physical store or mobile as the sales channel in mobile advertising will lead to higher purchase intention. In addition, it proves what type of risk attitude can be more influenced by mobile advertising with physical store or mobile as the sales channel. Furthermore, the results indicate what type of risk attitude and sales channel lead to a higher level of price discount for consumers after being exposed to mobile advertising. This information could, in turn, be used by organizations in their marketing strategy as well.
Based on the findings of prior literature and the literature gaps, the conceptual framework was made and given in Figure 1 including the hypotheses that are mentioned earlier.
Figure 1: Conceptual model.
3. Methodology
This paragraph contains the elaboration of the research design including information about the sample, the manipulation, the pre-test, the measurements, and the procedure of the research.
3.1 Sample
A cross-sectional experimental online survey in Qualtrics was used to collect data from respondents. The respondents were gathered through non-probability, convenience sampling. Dutch students and acquaintances were asked to participate in the research through email, WhatsApp, LinkedIn, and Facebook. In total 328 respondents started the online survey however, 75 respondents have not completed it. These were excluded from the data (completion rate of 77%). This leaves a total of 253 respondents. However, five respondents indicated to be exposed to mobile advertising with Laptop A while it was Laptop B (or vice versa). These respondents were excluded from the dataset as well. Another three respondents were excluded from the sample because of misunderstanding of the additional question regarding the level of price discount. In this sample of 245 respondents, 63% were women and
Mobile Advertising H2 Consumer’s Risk Attitude Purchase Intention Sales channel:
physical store vs. mobile Level of
Price Discount
H1 H3
37% were men. The average age of the respondents was 29 years (M = 29.02, SD = 11,73, range: 18-64). From all the 245 respondents, 58% were 24 years or younger. Moreover, 37% indicated to follow or have completed their ‘HBO’-degree, followed by 25% that follow or have completed their ‘MBO’-degree. The majority had a yearly income of less than 10.000 euros (37%) followed by 28% who indicated to have a yearly income between 20.000 and 29.999 euros.
3.2 Manipulation
This research consisted of an experiment executed through an online survey in Qualtrics. There were two different versions of the online survey (see Appendix C and D).
According to the literature, a mobile display advertisement is most effective for a high involvement, utilitarian product (Bart et al., 2014). Besides, for assessing the risk-taking behavior of consumers, a product should be used that is perceived as risky. Therefore, a high involvement product, namely a laptop, was used. This type of product generally has a higher price, carries greater risk and has more important personal consequences which lead consumers to think more carefully, deliberately and deeply when evaluating them (Bart et al., 2014; Bhasin, 2017).
In both versions of the survey, respondents were introduced to mobile advertising with the imagination that the respondent was searching for a new laptop. Right after this introduction, the respondent was exposed to a visualization of a hand that was holding a mobile device, which indicated the respondent being exposed to mobile advertising. In order to give some idea of the value of the laptop, some specifications of the laptop were given. This information was the same for both versions. It included that the laptop involved i5 processor, 1TB hard disc and a battery of six hours (see Appendix C). Besides, both versions presented the same price that was determined by the pre-test, namely 500 euros. The difference between
the two versions included the difference in the sales channel. One offer of the laptop, which was indicated by Laptop A, had to be purchased with the respondent’s mobile device. Laptop B’s offer applied to purchases in the physical store. These difference between the versions indicated by Laptop A or Laptop B represented the manipulation of this research. Respondents were randomly assigned to either one of the two versions.
After the respondents were exposed to the mobile advertising, a control question was asked. This was done in order to check whether the respondent had read the mobile advertising with the corresponding laptop (A or B). Besides, based on this question respondents could be categorized in the right experimental group.
3.2.1 Pre-test
For this research, a pre-test in Dutch was done in order to determine the price for the laptop respondents were exposed to in the mobile advertising. Therefore, the same mobile advertising as in the research itself was used for the pre-test, including the same specifications. However, the pre-test differed from the survey in the presented name. In the pre-test the name was “15-inch laptop” instead of Laptop A or Laptop B. Besides, the pre-test advertising did not include a price since this price had to be determined by the pre-test.
Four questions were asked in order to determine the price respondents were willing to pay for the laptop. These questions were adapted from the Van Westendorp method (Kivit, 2007). The method consisted of four open questions (see Appendix A). An example was as followed: “At what price would you begin to think the product is too expensive to consider?” (Kivit, 2007). The answers to the questions determined the laptop’s point of marginal cheapness (PMC), point of marginal expensiveness (PME), optimum price point (OPP) and the indifference price point (IPP) (see Graph 1).
participated in the pre-test. These respondents were gathered by asking acquaintances to participate. The pre-test was executed in Qualtrics so that after gathering the data, it could be exported to Excel. Thereafter, the data were categorized into four categories: too inexpensive, inexpensive, expensive, too expensive. For each category, the cumulative percentages were calculated and were plotted on a graph. However, in contrast with the other categories, the first two categories were plotted inversed (see Appendix B). In order to analyze the graph more detailed and to determine the price, the ranges of the axis were decreased and the result is shown in Graph 1.
Graph 1: Determining the Laptop price
According to Graph 1, respondents were willing to pay between 445 euro (PMC) and 600 euros (PME) for the laptop. The optimal price point (OPP) was 509 euros and the indifference price point (IPP) equaled 500 euros. This difference between OPP and IPP was minimal, but since the price at the IPP could be seen as “normal”, this price was used in this research. Besides, at this point of 500 euros, approximately 17% of the respondents perceived
0% 5% 10% 15% 20% 25% 400 450 500 550 600 C um ul ati ve pe rc ent ag es Price in Euro
Van Westendorp Price Sensitivity Meter
Too inexpensive Inexpensive Expensive Too expensive PMC PME IPP OPP
this price as not expensive and as not inexpensive. This was deliberately more compared to four percent of the respondents who perceive 509 euros as not expensive and not inexpensive.
3.3 Measurements
In order to measure the variables of the conceptual model, several measurements were used. The control variables, the measurements of consumer’s risk attitude, purchase intention and level of price discount were discussed below. For all the items that were used per variable in the survey, see Appendix D.
3.3.1 Control variables
In this research, the control variables were gender, age, education and yearly income. These were measured in order to identify the characteristics of the respondents and to assure that this group was representative of the Dutch population. Gender was a dummy variable meaning that a male respondent was coded as zero and a female as one. Therefore, this was measured by a dichotomous question. Age was collected by an open question and education and yearly income of the respondents was gathered through ordinal questions.
3.3.2 Consumer’s risk attitude
The measurement of the consumer’s risk attitude was done through the already named, Domain-Specific Risk-Taking (DOSPERT) scale (Blais & Weber, 2006). This psychometric scale assessed to what extent consumers were willing to engage in domain-specific risky activities which were conducted in Part I. An optional Part II assessed perceptions of the magnitude of the risks and expected benefits of the activities judged in Part I. For this research, Part II was excluded from the survey, since the focus of this research was only on the willingness to take the risks and not on how consumers perceived the risks and benefits of the
risky situations. This original DOSPERT scale consisted of 40 items, but in order to apply it to a wider range of people, the scale was revised and this version consisted of 30 items, which were used in this research. The scale was translated into different languages such as into Dutch. Because the survey was in Dutch, the Dutch version of the DOSPERT scale was used. These 30 items of the scale had a Cronbach’s Alpha of .86 meaning that there was internal consistency.
The 30 items of the scale assessed the risk-taking behavior of consumers in five domains consisting of six items each: financial decisions (separately for investing and gambling), health/safety, recreational, ethical, and social decisions. Risk-taking behavior in investing decisions was measured by for example “Investing 10% of your annual income in a new business venture.” and for gambling decisions, an example was “Betting a day’s income on the outcome of a sporting event.”. For taking risks in health/safety decisions, statements as “Sunbathing without sunscreen.” were used. An example of a recreational risky statement was “Bungee jumping off a tall bridge.”. Risk-taking behavior in ethical decisions was measured by statements as “Having an affair with a married man/woman.”. And for social decisions statements such as “Admitting that your tastes are different from those of a friend.” were used. Respondents indicated how likely they would take the risk and execute the risky activity if the chance would occur. They gave an answer on a seven-point Likert scale ranging from 1 (very unlikely) to 7 (very likely). In that sense, the higher the overall score of respondents, the higher the likelihood of taking risks and being a risk seeker. The lower the scores, the more risk-averse the respondent was. Please refer to Appendix D for all the survey questions.
3.3.3 Purchase intention
To measure purchase intention, a validated three-item scale was used (Shaouf, Lü, & Li, 2016). The three items used a five-point Likert scale ranging from 1 (totally disagree) to 5 (totally
agree). Because the survey was in Dutch, the items were translated according to the translation-back-translation procedure (Brislin, 1970). This holds that the items first were translated into Dutch and then were translated back into English in order to check discrepancies. Furthermore, the three items of this scale were adjusted according to the advertisement the respondent was exposed to. Although the adjustments, the scale had a Cronbach’s Alpha of .88 which indicated that it was reliable. One example of an item for mobile advertising with mobile as the sales channel was “After viewing the web advertisement, I will probably purchase the offered laptop with my mobile device.”.
3.3.4 Level of price discount
Depending on the answers respondents gave to the items of purchase intention, an additional question was exposed to the respondent. This question was only asked when respondents indicated to ‘totally disagree’ or ‘disagree’ with either one or more items that measured purchase intention. Only these respondents get the additional question because when respondents indicated a higher level of purchase intention, it implied that those already had the intention to purchase the laptop. Furthermore, it indicated that they would not specifically need a certain level of discount for having the intention to purchase it. The question involved what level of minimum price discount the respondents would desire for the offered laptop. It was based on a single-item that simply asks the respondents to enter the level of discount in an open question (Hicks, 2003). Again this item was translated into Dutch according to the translation-back-translation procedure in order to check discrepancies. It was also adjusted so that it was aligned with the mobile advertisement. This open question stated: “What is the minimum discount (in %) which you desire for the offered laptop?”. The reliability could not be tested since this level of price discount was measured by one item.
3.4 Statistical procedure
For this research, data was gathered by participants responding to items that use self-reported measures in an online survey. This survey was administered in Dutch between the period of 18 January 2018 till 24 January 2018. Participants were invited to fill in the survey through a link which led them to the introduction of the survey. First, participants were informed by the aim of the study, namely the focus on consumers’ (mobile) purchase behavior. The given information was kept as neutral as possible so that it did not influence the behavior of the respondents in advance. Furthermore, it was emphasized that there were no wrong or right answers and that their responses were treated anonymously. After that, participants answered the first series of questions that measured the consumer’s risk attitude. Then the participant was exposed to either one of the two versions of mobile advertising (Laptop A or Laptop B) with the corresponding sales channel. Participants were randomly assigned to one of the two versions. Thereafter, participants answered which version of the laptop they were exposed to in order to check if they had read the advertisement and to categorize them in the right experimental group. Then the participant was asked to answer a series of questions regarding the mobile or physical store purchase intention for the laptop. And if the participant indicated to (totally) disagree with one or more of the items of the purchase intention, the open question regarding the level of price discount was asked. In the end, participants answered questions about demographics namely gender, age, education and yearly income.
After collecting the data, it was exported to SPSS where it was screened by checking frequencies so that any missing values could be identified. To deal with one missing value that was found, the cases were excluded from the data on a pairwise basis so that only the cases that had no missing data in any variable were analyzed. One item of the DOSPERT-scale was counter-indicative and was reversed coded. In order to use gender as a dummy variable, it was recoded including men indicated by zero instead of one. Women were recoded from two to
one. Also, the independent variable mobile advertising was recoded: version one with Laptop A with mobile as the sales channel was recoded from one to zero so that version two with Laptop B and physical store as the sales channel was changed from two to one.
Besides, based on the frequency test, any errors and outliers could be found in the data. There were no errors found, although there were two outliers. After excluding these outliers from the data, a normality check was done by checking skewness, kurtosis, and the Kolmogorov-Smirnov test (see Appendix E). Based on the results of these tests, consumer’s risk attitude was normally distributed, while purchase intention was not normally distributed. However, this research consisted of more than 200 participants that could be seen as a large sample. Therefore, skewness won’t make a substantive difference in the analysis (Tabachnick & Fidell, 2007). The risk of kurtosis also was reduced because of this large sample.
With this dataset, a manipulation check was done in order to see if the manipulation in the mobile advertising was successfully executed. This is done by comparing the means of the two conditions in an independent-sample t-test. Thereafter, reliability tests were done as well as computing the correlations between the variables. Furthermore, moderation tests were done with Process, an external macro developed by A.F. Hayes (2012). The results of these tests aimed to understand the moderating effect of a consumer’s risk attitude and the level of price discount on purchase intention when being exposed to mobile advertising.
4. Results
The next paragraph consists of the results of the tests that were executed. First, the manipulation of this research was checked. Then, the reliability test and correlation test were executed. Thereafter, the results of the moderation tests were analyzed. This led to supporting or rejecting the hypotheses.
4.1 Manipulation check
Before running the independent samples t-test, the control question of the mobile advertisement was checked. After reading the mobile advertisement, participants were asked which version they were exposed to which was Laptop A or Laptop B. To check if there were any errors, indicating misreading the mobile advertisement, a frequency test was done. The results showed that in the sample (N = 253) there were three participants who indicated to see Laptop B while they were exposed to Laptop A. And vice versa, there were two participants who stated that they were exposed to the mobile advertising with Laptop A, while it was Laptop B. These participants’ answer could not be used therefore, they were excluded from the dataset. This meant that the sample included 248 participants. Additionally, another three respondents were excluded from the dataset because they misunderstood the additional question regarding the level of price discount. Their answer could not be used in the analysis as well. In the end, the sample size included 245 participants.
With this sample, two independent samples t-tests were done in order to check if the means of the two conditions were statistically different from each other. This was done two times since the first model in this research predicted purchase intention, while the second model predicted the level of price discount. The first results of the independent samples t-test showed that participants that were exposed to the mobile advertisement of Laptop A with mobile as the sales channel had a slightly lower purchase intention (N = 120, M = 2.16, SE = .91) compared to those who were exposed to Laptop B with the physical store as the sales channel (N = 125, M = 2.49, SE = .94). This included that respondents who had to buy the laptop in the physical store had a higher intention to make a purchase compared to those who had to buy it with their mobile device. The independent t-test found this pattern to be significant, indicating that the two conditions were statistically different from zero, t(243) = -2.74, p < .05. This suggested
that mobile advertising with purchase in the physical store or purchase with a mobile device both affected the participant’s purchase intention.
The results of the second independent samples t-test indicated that the level of price discount was slightly higher for participants exposed to mobile advertising with mobile as the sales channel (N = 96, M = 25.32, SE = 14.19) compared to mobile advertising with the physical store as the sales channel (N = 75, M = 24.27, SE = 13.75). However, this pattern was not significant, indicating that the two conditions were not statistically different from zero t(169) = .49, p > .05. Apparently, there was not a significant difference between the average level of price discount of both conditions.
4.2 Correlation
Before the correlation test was executed, the Cronbach’s Alpha’s, means and standard deviations for consumer’s risk attitude and purchase intention were computed. The reliability of the variables mobile advertising and level of price discount could not be examined since these scales consisted of only one measurement. For the other variables, all Cronbach’s Alpha’s were above .70 which implied that all measures had high internal consistency and were reliable (see Table 1). For purchase intention, all items had a corrected item-total correlation score of above .30, therefore no items were removed from the scale. However, the variable consumer’s risk attitude had nine out of the 30 items with a corrected item-total correlation below .30. This score indicated that nine items did not have a good correlation with the total score of the scale. However, deleting these items did not change the Cronbach’s Alpha with more than .10. Therefore, there was no reason for excluding these items from the scale.
In order to examine the correlations of the variables, the scale means per variable were computed. Since the independent variable, the mobile advertising was a fixed variable, the scale mean could not be assessed. Also, the scale mean of the level of price discount and the
