Master Thesis
HALMSTAD
UNIVERSITY
Master’s Programme in Strategic Entrepreneurship for International Growth, 60 Credits
E-commerce in Greece and Sweden: A cross-country investigation of consumer privacy attitudes and behaviours
Business Administration, 15 credits
Halmstad 2021-05-25
Michaela Kanaan , Maria Zoi Damanaki
ii Acknowledgement
We would like to express our gratefulness to the people that made this Master’s thesis possible. Without the 232 respondents of the study, it would not be possible to reach our final conclusions and contribute to the research field. Furthermore, we would like to show our appreciation to our families and friends that have been supportive throughout the process.
Particularly, the dog named Maxi. Moreover, we are grateful for coffee that has always been there for us when we needed more energy. Lastly, we would like to thank our supervisor, Klaus Solberg Söilen for his valuable feedback.
Thank you all,
Michaela Kanaan & Maria Zoi Damanaki.
iii Abstract
Purpose: The purpose of this study is to contribute with a cross-country approach to the research about consumers’ online privacy concerns by investigating Greek and Swedish consumers’ privacy attitudes and behaviours in the digital marketplace.
Theoretical framework: The study’s theoretical framework is based on the Power- responsibility Equilibrium Theory (PRE) perspective and three of Hoftsede’s cultural dimensions; uncertainty avoidance, power distance, and individualism.
Methodology: The study has a deductive approach with a quantitative research method. The data was collected by a questionnaire in each country which resulted in 232 answers in total.
The answers were then analyzed in Microsoft Excel 16.41, SPSS 26.0, and SmartPLS 3.3.3.
Findings: Findings indicate that the impact of power-holders on consumer privacy concerns, privacy empowerment and trust is more prominent for Swedish than for Greek consumers.
Moreover, the impacts of privacy concerns, privacy empowerment and trust on consumers’
power-balancing strategies are more substantial for Greek than for Swedish consumers.
Implications: The study contributes to the marketing field and academics studying consumers’ behaviour as it adds a cross-cultural approach to the investigation of consumers’
attitudes on e-commerce from a PRE theory perspective. Additionally, they can have practical implications for marketers and policymakers in the e-commerce sector as they can gain some insight into how essential corporate privacy responsibility and regulations are to gain the trust of consumers and decrease their privacy concerns.
Limitations: Since the study covers Greek and Swedish consumers, it is limited to countries with similar characteristics. Moreover, the majority of the respondents were under 35 years old and their opinions contradicted, not allowing most of the effects to be statistically significant enough to be considered.
Originality/Value: Consumer privacy attitudes and behaviours on e-commerce have not been studied from both a power-responsibility equilibrium and a cultural dimensions perspective. This study adds a cross-country approach by studying Greek and Swedish consumers, while at the same time combining two theoretical perspectives.
Keywords: Online privacy concerns, Privacy e-commerce, Cross-country online behaviour, Power-responsibility equilibrium, Hofstede’s cultural dimensions
iv Table of Contents
1 Introduction 1
1.1 Background 1
1.2 Problem Discussion 2
1.3 Research Questions 3
1.4 Purpose 3
2 Theoretical background 4
2.1 Hofstede’s cultural dimensions 4
2.1.1 Uncertainty Avoidance 4
2.1.2 Power Distance 4
2.1.3 Individualism 5
2.1.4 Greece and Sweden Indexes 5
2.2 Power-Responsibility Equilibrium 6
2.2.1 Consumer privacy concerns 7
2.2.2 Privacy empowerment 8
2.2.3 Consumer trust 8
2.2.4 Corporate privacy responsibility 8
2.2.5 Regulatory protection 9
2.3 Research hypotheses 9
2.3.1 Null hypothesis 9
2.3.2 Impact of privacy concerns and UA 10
2.3.3 Impact of privacy empowerment and PD 10
2.3.4 Impact of trust and IDV 11
2.3.5 Impact of corporate privacy responsibility and IDV 12
2.3.6 Impact of regulatory protection and PD 12
2.3.7 Structural model of hypotheses 13
3 Method 15
3.1 Research design 15
3.2 Choice of countries to investigate 15
3.3 Data collection 16
3.4 Data analysis 18
3.5 Reliability and Validity 18
3.6 Ethical Considerations 19
v
4 Results 20
4.1 Partial least squares - structural equation modelling (PLS-SEM) analysis 20
4.1.1 Internal consistency reliability testing 21
4.1.2 Convergent Validity testing 22
4.1.3 Discriminant Validity Testing 22
4.2 Hypotheses testing 23
4.3 Mediation analysis 23
5 Discussion 25
5.1 Structural models’ evaluation 25
5.1.1 Greece 25
5.1.2 Sweden 26
5.2 Hypotheses’ Evaluation 27
5.3 Cultural Dimensions’ Influence 30
5.4 Qualitative data 30
6 Conclusion 33
6.1 Summary of findings 33
6.2 Implications and contributions 34
6.3 Limitations 34
6.4 Future research 35
References 36
Appendix 41
Appendix 1 - Structural model 41
Appendix 2 - Outer loadings 42
Appendix 3 - Composite Reliability (CR) 44
Appendix 4 - Average Variance Extracted (AVE) 44
Appendix 5 - Heterotrait-Monotrait Ratio (HTMT) 45
Appendix 6 - Path coefficients (direct effects) 46
Appendix 7 - Indirect effects 47
Appendix 8- Total effects 48
Appendix 9 - Questionnaire 49
1 1 Introduction
In this introductory chapter, an overview of the topic of privacy concerns in the e-commerce context is presented, followed by a clarification of the fact that cultural characteristics can affect consumers’ concerns in each country. Moreover, the problem of this study is discussed, leading to the research gap, which leads to the research questions and purpose of the study.
1.1 Background
Analyzing consumers’ behaviours in stores in order to assist decision making is not a new phenomenon. Since stores are a public place, consumers appear not to care that information about them is being collected, as they assume that this is done with anonymity (Caudill &
Murphy, 2000). As the world is digitizing and more and more transactions take place on the internet, consumers may feel that they are not in a public place when taking actions online.
Therefore, concerns with regards to the online gathering of their information grow (Miltgen
& Smith, 2015) and consumer online privacy is discussed by several authors (Allen & Peloza, 2015; Anic et al., 2019; Bandara et al., 2020a,2020b; Bandyopadhyay, 2009; Caudill &
Murphy, 2000; Chen et al., 2013; Dinev et al., 2006; Fox & Royne., 2018; Goldfarb &
Tucker, 2011; Gurău & Ranchhod, 2009; Krishen et al., 2017; Lwin et al., 2007; Mohammed
& Tejay; 2017; Rapp et al., 2009; Van Dyke et al., 2007). Privacy can be referred to as an individual right and a phase where no one else is present (Cambridge Dictionary, 2021a).
Despite this explanation, consumers agree that they have lost control over their privacy and what companies collect and use about them (Mosteller & Poddar, 2017).
According to Van Dyke et al. (2007), the increased privacy concerns of consumers, in combination with the absence of trust, can reduce their desire to complete online transactions.
The decreased willingness to transact online can be tackled by empowering consumers and giving them control over their personal information (Akhter, 2014; Van Dyke, 2007). Apart from giving consumers control over their personal data, online privacy concerns could be declined by an increase in online regulatory control since consumers perceive it as weak (Anic et al., 2019). The effects of consumers not trusting regulatory control and corporate privacy responsibility actions are defensive behaviours, such as fabrication of data and withholdment of personal information. Bandyopadhyay (2009) states that this could also lead to rejection of e-commerce and lack of willingness to use the internet.
E-commerce has risen over the years, and the growth is forecasted to be up to 510 billion US dollars in 2024 in Europe (Statista, 2020). This is derived from the increasing usage of mobile phones, applications and social media (Bandara et al., 2019). Nevertheless, the rise is no longer increasing at the same rate as it did a few years ago. Consumers in the US have reduced their online transactions due to consumer privacy issues, and consumers in Australia have expressed their concerns about their privacy online (Bandara et al., 2019). These phenomenons intend that consumers in the world behave differently.
Consumers’ cultural characteristics and environment can significantly impact their behaviours and attitudes, causing differences in online private information concerns between
2
countries (Bandyopadhyay, 2009). Moreover, cultural differences can influence trust and privacy concerns and thus, affect online purchasing decisions, leading to the need to address the distinct cultural elements with suitable actions (Dinev et al., 2006). National and cultural privacy issue differences exist in mobile e-commerce as well, as claimed by Gurău and Ranchhod (2009), who discovered that the country of origin influences consumers’
perception concerning their privacy concerns.
Bandara et al. (2020b), who examined privacy issues in the e-commerce context, discovered that insufficient corporate privacy responsibility and regulatory protection could make consumers feel they have less control of their privacy empowerment. This can harm consumer trust and thus increase privacy concerns and defensive responses. Altogether, their study supports earlier research, but it was only tested in Australian consumers, causing limitations in the generalizability of the findings due to the homogeneity of the sample.
Therefore, their suggestion for further research is to do a cross-country investigation of consumer privacy attitudes and behaviours in the e-commerce context. In that way, cultural differences can be connected to consumers’ attitudes and behaviours regarding their digital privacy concerns. Anic et al. (2019) also suggest a cross-cultural study using Hofstede’s cultural dimensions to reveal new findings related to consumer privacy in the e-commerce context.
Even though Gurău and Ranchhod (2009) conducted a comparative study of the privacy issues of consumers from three different nationalities, their focus was mobile commerce. A similar study by Chen et al. (2013), which compared information privacy concerns in relation to consumer characteristics between two distinct cultures, focused on mobile commerce as well. In addition, the study of Dinev et al. (2006), which also explored cross-cultural differences in the e-commerce context, did not examine the issue from a power-responsibility equilibrium theory perspective.
Therefore, a gap was found in the literature since consumer privacy attitudes and behaviours on e-commerce have not been studied from both a power-responsibility equilibrium and a cultural dimensions perspective.
1.2 Problem Discussion
Hence, this study aims to be a contribution to the research field, connecting studies of consumers’ privacy concerns in the e-commerce context to cross-country differences. Since the aim was to further expand the study of Bandara et al. (2020b) by adding a cross-cultural dimension, the authors decided to use theories and concepts as close as possible to the ones utilized in the mentioned article. Thus, this study is conducted from a power-responsibility equilibrium (PRE) theory perspective, and the extended PRE framework of privacy that was presented by Bandara et al. (2020) is used.
Furthermore, in addition to the study conducted by Bandara et al. (2020b), the authors tried to connect the privacy concerns of consumers to cross-country differences and explain them.
For this reason, three cultural dimensions that were established by Hofstede (1980) are used
3
as the basis. The dimensions that are made use of are uncertainty avoidance, power distance, and individualism.
Moreover, the countries that were chosen for the cross-cultural analysis are Greece and Sweden. The reason Greece and Sweden were the two countries selected for this study is the substantial difference in their cultural characteristics, which can be seen in Hofstede’s (1980) cultural dimension indexes. For this reason, consumers in the two countries are expected to have different attitudes and behaviours in regards to their online privacy. The choice of countries to investigate will be further explained in the method chapter.
Lastly, this study’s findings are signifficant as online consumer privacy concerns have not been studied from both a PRE and a cultural dimensions perspective. The authors believe that a cross-country investigation of this aspect can help researchers understand how the cultural elements can impact consumers’ attitudes and behaviours in the digital marketplace. Further, this study aims to be a contribution to e-commerce and marketers as it highlights the importance of corporate privacy responsibility in increasing trust and empowering consumers in order to decrease their online provacy cocnerns. Additionally, the influence of regulatory protection on consuemers’ privacy concerns is identified and could be of help to policymakers.
1.3 Research Questions
In order to connect consumers’ behaviours and attitudes regarding online privacy concerns to national and cultural differences, the following questions will be researched.
RQ1: What is the difference in the impacts of power holders (i.e. corporate privacy responsibility and privacy regulations) on consumer privacy concerns, privacy empowerment and trust between Swedish and Greek consumers?
RQ2: What is the difference in the impacts of privacy concerns, privacy empowerment and trust on consumers’ power-balancing strategies (i.e. defensive behaviours) between Sweden and Greece?
1.4 Purpose
The purpose of this study is to bring a cross-country approach to the research about online privacy concerns. This will be done by investigating Greek and Swedish consumers’ privacy attitudes and behaviours in the digital marketplace from a power-responsibility equilibrium theory perspective. Additionally, the study aims to explore the differences in the concerns between the two distinct nationalities and connect them to three of Hofstede's (1980) cultural dimensions; Uncertainty Avoidance, Power Distance and Individualism.
4 2 Theoretical background
In this chapter, the theoretical framework is presented. The chosen cultural dimensions from Hofstede are discussed, followed by the Power-Responsibility Equilibrium theory and the associated factors. Lastly, the hypotheses are presented.
2.1 Hofstede’s cultural dimensions
Hofstede (1980), studying 40 different nations, investigated the dissimilarities in actions and thinking in a social context. He suggested that people from different cultures are mentally programmed to think and act in a certain way, defined by the family, school, and organizations in each culture. In an attempt to identify the elements of each culture, he developed four primary dimensions. Hofstede later added more dimensions. In this paper, three out of the four primary dimensions will be used: uncertainty avoidance, power distance, and individualism. The fourth dimension, masculinity, was chosen to be excluded since the authors want to focus on consumers in general and not on a particular sex. Also, masculinity was considered inappropriate for the study since the connection to the power-responsibility equilibrium was more evident with uncertainty avoidance, power distance, and individualism.
2.1.1 Uncertainty Avoidance
Uncertainty Avoidance (UA) is one of the cultural dimensions described by Hofstede (1980).
It reflects the extent to which people in a society do not feel comfortable with ambiguous and unstructured circumstances. The UA index evaluates the significance of standards and rules for a culture and refers to the degree to which people in a society try to reduce anxiety by eliminating uncertainty (Dinev et al., 2006; Mohammed & Tejay, 2017).
Members of high UA cultures usually attempt to avoid unusual and unknown situations, apply rules and regulations, and rely on information systems (Bandyopadhyay, 2009;
Mohammed & Tejay, 2017). At the same time, they lower their risk -taking and are expected to repel innovative ideas and practices (Diven et al., 2006). In contrast, people in low UA cultures react positively to unstructured circumstances and changeable surroundings, while at the same time, they are anticipated to not depend on rules (Mohammed & Tejay, 2017). Their overall behaviour tends to be pragmatic, relativistic, and tolerant (Bandyopadhyay, 2009).
Regarding privacy concerns, members of high UA cultures are more protective of their personal information. They are less likely to disclose the information if they are uncertain of how it is going to be used and by whom (Bandyopadhyay, 2009). Therefore, consumers in countries with a high UA index are expected to have increased defensive behaviours concerning their online privacy than consumers in countries with a low UA index.
2.1.2 Power Distance
The cultural dimension of Power Distance (PD), as explained by Hofstede (1980), concerns the issue of human inequality and the level of attachment to formal authority. Mainly, it measures the degree to which the less powerful members of a society accept and expect the unequal distribution of power (Dinev et al., 2006; Mohammed & Tejay, 2017).
5
In cultures where PD is high, people possess an increased acceptance of authority and power inequality (Bandyopadhyay, 2009). The members of such a society tend to prefer centralized powers (Dinev et al., 2006), and subordinates acknowledge the power of people in higher hierarchical positions (Mohammed & Tejay, 2017). Opposed to high PD cultures, low PD societies perceive each other as equals, despite their formal positions. Moreover, power relations tend to be consultative, and subordinates are not afraid to get involved in the decision making processes (Mohammed & Tejay, 2017).
With regards to privacy concerns, people in high PD societies are more likely to be tolerant of revealing information than people in low PD societies where power is more equally distributed (Bandyopadhyay, 2009). Thus, it is expected that in high PD cultures, consumers demand less responsibility from the power holders regarding their online privacy than in cultures where the PD score is low.
2.1.3 Individualism
Individualism (IDV) is the third cultural dimension described by Hofstede (1980) which describes the association between the individual and the collectivity in a society, which also can be referred to as collectivism. The individualism-collectivism is a behaviour and mindset which can be linked to society’s norms (Hofstede, 1980). The IDV index refers to the extent to which people in a society want to be a part of a group identity. Countries with a high IDV index tend to be more dedicated to personal time, individual working tasks and freedom, while countries with a low IDV index tend to be more concerned about group responsibility and family (Bandyopadhyay, 2009; Hofstede, 1980).
People in individualistic societies (high IDV) are expected to defend themselves and their direct family. In contrast, people in collectivistic societies (low IDV) behave as members of a lifelong group with an extended family which is trusted and defended at all times (Mohammed & Tejay, 2017). Furthermore, the consciousness of self, individual identity and the belief in individual decisions emboss individualistic societies (Hofstede, 1980) while the group interest is always the priority in collectivistic societies (Dinev et al., 2006).
Members of high IDV cultures tend to be more aware of and concerned for their privacy than members of cultures with a low IDV index (Bandyopadhyay, 2009). Furthermore, it is anticipated that high IDV consumers have less trust in power holders than consumers in countries with a low IDV index.
2.1.4 Greece and Sweden Indexes
The countries that will be studied and compared are Greece and Sweden. The countries were chosen because of their different index in Hofstede’s cultural dimensions (Table 1).
Therefore, it is of interest to explore the differences between Greece, which has a high UA (112) and PD (60) index but a low IDV (35) index, and Sweden, which has a low PD (31) and UA (29) index but a high IDV (71) index (Hofstede, 1980).
6
Index UA PD IDV
Greece 112 60 35
Sweden 29 31 71
Table 1: Indexes from Hofstede (1980) 2.2 Power-Responsibility Equilibrium
In the power-responsibility equilibrium (PRE) theory, it is argued that a connection exists between social power and social responsibility. In particular, the more powerful member has the responsibility to make the less powerful partner feel that they are in a relationship of trust and equality (Caudill & Murphy, 2000; Martin & Murphy, 2017). According to the theory, when a company abuses its power and does not promote equality, it may benefit in the short term but will eventually lose its power (Caudill & Murphy, 2000; Lwin et al., 2007).
Lwin et al. (2007) used as a starting point the PRE theory and developed a framework that studies the relationship between corporations and their customers concerning internet privacy. They suggested that on the one side, there are the power holders - meaning corporations and governments - and on the other side, consumers, the individuals who provide information (Bandara et al., 2020b). Consumers expect that power holders (corporations and government) will use the information responsibly provided to them.
Consequently, when consumers feel that corporations and governments are not acting responsibly regarding the exchange of information, they will proceed to defensive actions, such as fabrication, protection and withholding (Bandara et al., 2020b; Krishen et al., 2017;
Lwin et al., 2007).
Bandara et al. (2020b), after a comprehensive literature review and a qualitative study, recognized four themes connected to consumers’ privacy concerns in the online context.
Hence, they extended the PRE framework of privacy developed by Lwin et al. (2007), integrating corporate privacy responsibility, regulatory protection, consumer trust, and consumer privacy empowerment.
This study, aiming to connect consumers’ online privacy concerns to consumers’ cultural characteristics, incorporates three of Hofstede’s (1980) cultural dimensions into Bandara’s et al. (2020) framework (Figure 1). The dimensions that were considered suitable for the study were uncertainty avoidance (UA), power distance (PD), and individualism (IDV).
7
Figure 1: Extended PRE framework of privacy from a cultural perspective 2.2.1 Consumer privacy concerns
The concept of privacy has changed its meaning over the years. In the online context, privacy has to do with the attempt to find the balance between the advantages and disadvantages of revealing personal information when completing transactions in the digital marketplace (Anic et al., 2009). Privacy concerns mainly represent consumers’ uncertainty of the kind of information that third parties are collecting and how this information is utilized (Anic et al., 2009; Bandara et al., 2019; Bandyopadhyay, 2009; Lwin et al., 2007).
Since e-commerce transactions have been increasing in numbers, the methods for data gathering have been multiplied. Consequently, a substantial amount of consumers’ personal data is being collected and used by various parties, such as companies and data aggregators (Bandara et al., 2020b; Bandyopadhyay, 2009). This kind of information is either being revealed by consumers at the process of online transactions or involuntarily and unknowingly acquired (Bandyopadhyay, 2009). The fact that the collection of private information nowadays is not dependent on direct interaction has made it almost impossible to track the kind of information being disclosed and the way it’s being used. Therefore, consumers’
privacy concerns have risen (Bandara et al., 2020b; Liao et al., 2011).
If those in charge of protecting data and privacy policies do not take proper measures and the consumers feel vulnerable, consumers will proceed to defensive behaviours (Lwin et al., 2007). Avoiding shopping online, leaving transactions, providing incorrect information, and applying privacy-enhancing technologies are some of the reactions to privacy concerns (Van Dyke et al., 2007; Wirtz & Lwin, 2009).
Even though consumers’ concern about online privacy is a fact, several cultural characteristics can influence the degree to which consumers in each country are concerned.
Hence, online privacy concerns can not be managed in the same way between different nationalities (Mohammed & Tejay, 2017).
8 2.2.2 Privacy empowerment
The concept of empowerment has been primarily used in the contexts of employee empowerment and consumer empowerment. The main idea in both of the terms is that the individual gains control or power that has been transferred to them from higher levels of authority (Van Dyke et al., 2007). Particularly, the concept of consumer empowerment regards the change of the balance of power from the power holders, who are the service providers, to consumers who seemingly possess less power. Wathieu et al. (2002) argue that one of the components that make consumers feel empowered is control of choice set composition. They suggest that the perception of empowerment does not come from the number of choices but from defining and controlling the choices.
Likewise, having control over private information is highly significant when defining consumer privacy empowerment (Bandara et al., 2020b). Privacy empowerment is also associated with the consumers’ perception about the degree to which they can manage the use and distribution of their personal information (Hunter & Garnefeld, 2008; Van Dyke et al., 2007). Even though having control of one’s information is crucial, it is not adequate to describe privacy empowerment. A more general explanation of privacy empowerment is the beliefs and behaviours of consumers that help avoid undesired consequences regarding their information usage (Bandara et al., 2020a).
2.2.3 Consumer trust
Despite the growing numbers in e-commerce, consumers’ trust has appeared to be a concern (Akhter, 2014; Bandara et al., 2020b). Trust is defined as “to believe that someone is good and honest and will not harm you, or that something is safe and reliable” (Cambridge Dictionary, 2021b). In e-commerce, trust implies that the consumer believes that the marketer will not abuse or harm the consumer (Lwin et al., 2016) and that consumers trust e-commerce to satisfy their privacy concerns (Bandara et al., 2020b).
Trust and privacy have shown to have a great relationship in e-commerce as studies have shown that consumers who trust an online company tend not to be as concerned about their privacy as consumers who do not trust the online company. Therefore, consumer trust has become an essential factor in e-commerce strategies (Akhter, 2014; Kim et al., 2008). Caudill and Murphy (2000) discuss that consumer trust regarding privacy might be the factor needed for e-commerce to keep growing, and Liao et al. (2011) point out that it is also the reason for its success. Furthermore, trust attitudes and risk perception has also shown to have a great relationship, while culture is also associated with trust and risk. Collectivistic cultures tend not to trust third parties as much as individualistic cultures, which indicates that culture plays an important part in consumers’ trust attitude and behaviour (Mou et al., 2017).
2.2.4 Corporate privacy responsibility
For the last 30 years, companies have been making an effort to be socially responsible by exercising policies to address various issues (Pollach, 2011). Consumers perceive corporate privacy policies as the way companies are obliged to handle their power over the use of consumers’ personal data (Bandara et al., 2020b; Lwin et al., 2007). These policies are
9
critical for corporations and manifest the procedures for gathering and protecting customers’
private information (Martin & Murphy, 2017). Moreover, consumers' actions, which are triggered by their privacy concerns, are related to their impression of each company’s policies (Lwin et al., 2009).
Apart from not being aware of how their data are being used due to insufficient transparency (Bandara et al., 2020b), consumers are struggling with understanding privacy policies as well, since they have become less readable and longer (Martin & Murphy, 2017). Therefore, consumers do not provide information free of value but expect that organizations will act ethically regarding personal data (Krishen et al., 2017). Still, the degree to which each company is being discreet about consumers’ privacy is connected to corporate morality (Pollach, 2011). Hence, aside from the fact that privacy-related activities can improve a firm’s reputation (Allen & Peloza, 2015; Pollach, 2011), companies have legal and ethical obligations to protect customers' privacy (Ferrell, 2017).
2.2.5 Regulatory protection
Regulatory protection refers to how governments and policies protect and handle consumers’
collected online data and information. It can also be referred to as the responsibility of governments to protect consumers where imbalance occurs (Lwin et al., 2007). The PRE theory argues that power holders are expected to create and contain trust for consumers and their privacy (Bandara et al., 2020b). Though, it is of importance that consumers are aware of their regulatory protections. Miltgren and Smith (2015) argue and prove that a higher level of knowledge about the regulation will lead to a higher level of perceived privacy regulatory protection. A perceived privacy regulatory protection refers to the trust that consumers have that higher power holders protect their data and information.
Studies have shown that in societies where consumers’ have great privacy concerns, the lack of regulatory protection often is a matter (Lwin et al., 2007). Also, lack of regulatory protection deprives consumers’ privacy empowerment which damages consumers’ trust, triggers privacy concerns and causes defensive behaviour. Therefore, regulatory protection becomes an essential and important factor for consumers.
2.3 Research hypotheses
In this chapter, the hypotheses will be presented. The hypotheses are created and based on Hofstede’s cultural dimensions (1980) and suitable research, as well as on the research of Bandara et al. (2020b).
2.3.1 Null hypothesis
As previously mentioned, it has been found that cultural characteristics of consumers can influence their attitudes and behaviours regarding their privacy concerns in the e-commerce context (Bandyopadhyay, 2009; Chen et al., 2013; Dinev et al., 2006; Gurău & Ranchhod, 2009). In this study, cultural differences are defined and explained with the help of cultural dimensions, which were introduced by Hofstede (1980). In particular, the cultural dimensions
10
of uncertainty avoidance (UA), power distance (PD), and individualism (IDV) are being used.
Consumers in cultures with a high UA index tend to be more defensive of their personal information (Bandyopadhyay, 2009) and thus, are expected to be more concerned about their online privacy than consumers in cultures with a low UA index. Continuing, in high PD cultures, consumers are more likely to surrender their privacy to power holders (Bandyopadhyay, 2009) and hence, they are anticipated to be less privacy concerned than low PD consumers. Lastly, in low IDV cultures, consumers are likely to be more used to their privacy being invaded by groups (Bellman et al., 2004). Therefore, they are expected to be less privacy concerned than consumers in societies with high IDV.
Since the scores of Greece and Sweden in Hofstede’s (1980) cultural dimension indexes are significantly different, the authors anticipate that consumers in each country have dissimilar attitudes and behaviours concerning their online privacy. It is therefore hypothesized that:
H0. Greek and Swedish consumers do not present the same attitudes and behavioursconcerning their online privacy in the digital marketplace.
2.3.2 Impact of privacy concerns and UA
According to the PRE theory, consumer privacy concerns connected to power holders do, as mentioned, affect their defensive behaviour. When consumers do not feel protected by those in charge of the data, they feel vulnerable and act defensively (Lwin et al., 2007; 2016) and protect their privacy rights (Choi et al., 2018). Defensive behaviour can, in this context, be avoiding e-commerce, not fulfilling transactions, rejecting cookies or even providing incorrect information to others (Van Dyke et al., 2007; Wirtz et al., 2007). Cancelling online purchases due to online privacy issues has also shown to be a common action taken (Ranganathan & Grandon, 2002), while refusing to share data when consumers feel threatened is also a frequent defensive behaviour (Choi et al., 2018). This is costly for e- commerce but can be an efficient approach for the consumers to get the power holders to take actions (Bandara et al., 2020b; Gauzente & Ranchhod, 2001).
Greek consumers have a high UA index (112) set against Swedish consumers (29). As mentioned, members of high UA cultures tend to act more protective of their personal information and do not want to hand out personal information if something is uncertain (Bandyopadhyay, 2009). Defensive behaviour is therefore more expected from countries with high UA regarding their online privacy concerns. Hence, the study’s H1 is structured as followed:
H1. Privacy concerns in the high UA Greek consumers have a more positive effect on defensive behaviour than in the low UA Swedish consumers.
2.3.3 Impact of privacy empowerment and PD
The construct of privacy empowerment represents the intention of entrusting an amount of power over personal information to the consumer. If consumers consider themselves privacy-
11
empowered, their privacy concerns will decline (Van Dyke et al., 2007). Hence, privacy empowerment is expected to negatively affect consumers’ privacy concerns (Bandara et al., 2020b; Van Dyke et al., 2007). Furthermore, Akhter (2014) argues that consumers with high efficacy concerning their internet and technology skills are more confident in their potential to protect their privacy. The perception of being able to manage their privacy can lead to decreased defensive behaviours.
Consumers in high PD cultures tend to be more willing to disclose personal information and thus, are anticipated to be less online privacy concerned than consumers in low PD cultures (Bandyopadhyay, 2009). Therefore, it is expected that consumers from Sweden, which possesses a low score of 31 in Hofstede’s cultural dimensions PD index, will have increased privacy concerns related to consumers from Greek, which has a high score of 60 in the same index. Similarly, the defensive behaviours of Swedish consumers are expected to be more increased than those of Greek consumers. Consequently, the study’s H2a and H2b hypotheses are:
H2a. Privacy empowerment in the low PD Swedish consumers has a more negative effect on privacy concerns than in the high PD Greek consumers.
H2b. Privacy empowerment in the low PD Swedish consumers has a more negative effect on defensive behaviours than in the high PD Greek consumers.
2.3.4 Impact of trust and IDV
Consumer trust related to privacy concerns has been well researched and proved important within e-commerce (Akhter, 2014; Bandara et al., 2020b; Lwin et al., 2016; Van Dyke et al., 2017). Consumers that do not have trust connected to e-commerce have proven to have defensive behaviour, protecting their privacy and not doing transactions or business online (Akhter, 2014). Van Dyke et al. (2007) also indicate this behaviour, meaning that increased privacy concerns in combination with trust, leads to a reduced desire to complete online transactions. Also, increased awareness of privacy threats has shown to reduce trust, leading to consumers demanding control (Olivero & Lunt, 2004). Furthermore, Caudill and Murphy (2000) argue that trust is the key to success within e-commerce and is something that marketers need to have in mind.
Members of high individualistic cultures such as Swedish consumers (71) tend to have greater privacy concerns, and the trust for power holders are less than members of low individualistic cultures such as Greek consumers (35) (Bandyopadhyay, 2009). Furthermore, high individualistic cultures act in a defensive way against people that are not a direct family, while low individualistic cultures act as members of a larger society which is trusted (Mohammed & Tejay, 2017). Also, low individualistic cultures tend not to trust third parties as much as cultures with high individualistic character, which can be explained by their higher trust in their own society (Mou et al., 2017). Therefore, the hypotheses H3a and H3b are:
12
H3a. Trust in the low IDV Greek consumers has a more negative effect on privacy concerns than in the high IDV Swedish consumers.
H3b. Trust in the low IDV Greek consumers has a more negative effect on defensive behaviours than in the high IDV Swedish consumers.
2.3.5 Impact of corporate privacy responsibility and IDV
As explained by the PRE theory, the more powerful member of a social interaction, in this case, corporations, is responsible for making consumers who are less powerful feel that a relationship of trust and equality exists between them. How successfully a firm communicates with its customers can have an impact on the perceptions of their overall corporate social responsibility (Stanaland et al., 2011). Corporations are therefore responsible for reducing consumers’ privacy concerns (Bandara et al., 2020b; Caudill & Murphy, 2000;
Lwin et al., 2007). Consumers' privacy concerns and behaviours are influenced by their impression about corporate policies and the way their data are being handled (Lwin et al., 2007). Moreover, research has shown that corporate privacy policies have an impact on consumers’ trust in the company (Martin & Murphy, 2017). Hence, corporate privacy responsibility is expected to affect privacy concerns negatively but, at the same time, positively impact trust. In addition, transparency and responsibility in handling consumers personal information can make consumers feel empowered and in control of their online privacy (Bandara et al., 2020b; Martin & Murphy, 2017). It is therefore anticipated that privacy empowerment is positively affected by corporate privacy responsibility.
As previously mentioned, consumers of cultures with a high IDV index are more likely to have increased privacy concerns and less trust in corporations regarding the protection of their personal information than consumers of low IDV cultures (Bandyopadhyay, 2009;
Dinev et al., 2006; Mohammed & Tejay, 2017). Consequently, the low IDV Greek consumers tend to trust more corporate policies and are likely to be less concerned for their privacy than the high IDV Swedish consumers. Likewise, the feeling of having control can be more enhanced in the trustful and less concerned Greek consumers than in the high IDV Swedish consumers. Hence, the hypotheses H4a, H4b and H4c are:
H4a. Corporate privacy responsibility in the low IDV Greek consumers has a more positive effect on trust than in the high IDV Swedish consumers.
H4b. Corporate privacy responsibility in the low IDV Greek consumers has a more negative effect on privacy concerns than in the high IDV Swedish consumers.
H4c. Corporate privacy responsibility in the low IDV Greek consumers has a more positive effect on privacy empowerment than in the high IDV Swedish consumers.
2.3.6 Impact of regulatory protection and PD
Regulatory protection has a wide impact on consumers. According to the PRE theory, power holders are expected to create and obtain trust for consumers (Bandara et al., 2020b). As mentioned, regulatory protection can be referred to as how governments and policies protect
13
consumers in societies (Lwin et al., 2007). Lwin et al. (2007) intend that strong regulatory protection can lead to reduced privacy concerns. Also, it is important that consumers know of their regulatory protection and feel protected. If consumers feel protected and that their data and information are in good custody, they build up trust for power holders (Miltgren &
Smith, 2015). Research has also shown that strict laws and regulations enhance consumers’
empowerment. Consumers’ empowerment is important to know about since it can show imbalances in markets (Kucuk, 2016). Altogether, regulatory protection is an essential factor for consumers since a lack of regulatory protection deprives consumers of their privacy empowerment which triggers trust issues and privacy concerns.
Consumers in high PD societies tend to accept the unequal balance of power holders more than consumers in low PD societies (Bandyopadhyay, 2009; Dinev et al., 2006; Mohammed
& Tejay, 2017). Centralized power can be preferred in high PD societies, and the power that people have in high positions is accepted and known (Mohammed & Tejay, 2017). High PD societies even tend to surrender their privacy to power holders (Bandyopadhyay, 2009).
Furthermore, they also tend to be more focused on belonging to an in-group than focusing on self-interest (Dinev et al., 2006). As a result, trust is expected to have for power holders while privacy concerns are less expected to have in societies with high PD, such as Greek consumers (60).
H5a. Regulatory protection in the high PD Greek consumers has a more positive effect on trust than in low PD Swedish consumers.
H5b. Regulatory protection in the high PD Greek consumers has a more negative effect on privacy concerns than in low PD Swedish consumers.
H5c. Regulatory protection in the high PD Greek consumers has a more positive effect on privacy empowerment than in low PD Swedish consumers.
2.3.7 Structural model of hypotheses
As previously mentioned, corporate privacy policies can influence consumers’ trust (Martin
& Murphy, 2017), privacy concerns (Lwin et al., 2007), and privacy empowerment (Bandara et al., 2020b). In particular, corporate privacy responsibility can have a positive impact on trust, a negative impact on privacy concerns and lastly, a positive impact on privacy empowerment. Moreover, the cultural dimension of IDV is expected to be of influence regarding the degree to which the trust, privacy concerns and privacy empowerment of each country’s consumers will affect (Bandyopadhyay, 2009; Dinev et al., 2006; Mohammed &
Tejay, 2017) by a company’s social responsibility practices. Similarly, regulatory protection can have an impact on consumers’ trust (Miltgren & Smith, 2015), privacy concerns (Lwin et al., 2007), and privacy empowerment (Kucuk, 2016). In this case, the impacts are also positive, negative, and positive respectively. Additionally, the impact of regulatory protection is most likely to be influenced by the cultural dimension of PD as each country’s consumers have different reactions towards the power holders concerning their privacy (Bandyopadhyay, 2009).
14
Furthermore, consumers' trust is expected to determine their defensive behaviours (Akhter, 2014) and privacy concerns (Bandara et al., 2020b). Specifically, trust has likely a negative impact on both defensive behaviours and privacy concerns. Since people in high IDV societies tend to have increased privacy concerns and trust for power holders than people in low IDV societies (Bandyopadhyay, 2009), IDV is anticipated to be the cultural dimension that will affect each country’s attitudes. The concept of privacy empowerment can influence consumers’ privacy concerns (Van Dyke et al., 2007) and defensive behaviour (Bandara et al., 2020b). This concept is expected to also negatively affect both defensive behaviours and privacy concerns. The cultural dimension of PD is thought of as a determinant of the extent to which each country’s consumers behaviours will be affected by privacy empowerment.
Finally, consumers' privacy concerns can have an impact on their defensive behaviours (Choi et al., 2018; Lwin et al., 2007; 2016; Van Dyke et al., 2007; Wirtz et al., 2007). Since members of cultures with high UA are more likely to be protective of their personal information than members of low UA culture (Bandyopadhyay, 2009), they are also expected to have increased defensive behaviours. Consequently, the cultural dimension of UA is most likely to influence how intense the defensive behaviours of each country’s consumers will be.
All of the correlations between the hypotheses, whether their impact is positive or negative, and which cultural dimension influences the intensity of consumers’ responses in each country can be seen in Figure 2.
Figure 2: Structural model of hypotheses
15 3 Method
In the third chapter, the methodology is presented. The research design will start the chapter, followed by the choice of countries investigated in the study. The data collection and data analysis are also discussed, followed by the reliability and validity of the study.
3.1 Research design
This thesis focuses on consumer privacy attitudes and behaviours in the e-commerce context with a cross-country approach. Since the aim was to discover consumers’ cultural differences regarding their privacy concerns, the study is of descriptive nature. A descriptive approach was used as it can help describe the characteristics of a group and explain the topic of interest (Sekaran & Bougie, 2016). Descriptive studies can either be qualitative or quantitative. In this case, the study is based on quantitative research conducted by primary collected data, which were gathered through a created online survey. Nevertheless, the online survey also contained an open comments section, letting the respondent leave comments about the topic in case they had any other thoughts that were not covered in the survey. This section adds a qualitative approach to the study. The method of surveys was chosen because it has the flexibility to permit the collection of both quantitative and qualitative data regarding the consumers’ perception (Sekaran & Bougie, 2016). The aim was to get online consumers within the e-commerce context to answer the survey and to get a wide age range in or der to have the ability to generalize the study.
The study has a deductive approach, meaning that the theory was chosen before the data was compiled (Bryman & Bell, 2017). In this way, suited data could be collected in order to answer the research questions. Particularly, a hypothetico-deductive approach was used. The authors began with the general theory that cross-cultural differences affect consumers’
privacy behaviours and attitudes in the e-commerce context and then developed specific hypotheses which could be tested. Afterwards, specific data was collected to test the hypotheses and, by extension, the general theory (Sekaran & Bougie, 2016).
The research questions were developed from Bandara et al.’s (2020b) article and set with a cross-country approach between Greece and Sweden. The fundamental theoretical background for this study was built using two theories and concepts. The first was the extended PRE framework which was developed by Bandara et al. (2020b) using Lwin et al.’s (2007) framework, which was based on the power-responsibility equilibrium (PRE) theory.
The authors further developed this framework by adding Hofstede’s cultural dimensions (Hofstede, 1980). Hofstede’s cultural dimensions were the second concept that was utilized in the attempt to answer the research questions.
3.2 Choice of countries to investigate
Greece and Sweden are the countries that constituted the cross-country part of the study.
Greek and Swedish consumers were chosen for this study because of the differences of the countries in Hofstede’s cultural dimensions scores (Hofstede, 1980). Uncertainty avoidance, Power distance and Individualism are the three dimensions that were considered for this
16
study. The three dimensions score differently to a large extent in Greece and Sweden. The cultural characteristics, habits, behaviour, perceptions, and attitudes vary substantially between Greece and Sweden, making them two culturally distinct countries and, therefore, suitable for the study. Because of the authors’ connections to both countries, the aim was to collect equally numbered responses in order for each country to be represented equivalently regarding its consumers’ privacy concerns.
3.3 Data collection
The research is, as mentioned, based on primary data collected for this study through online surveys using questionnaires. A questionnaire is a set of questions that have been formed in advance and give the opportunity to the respondents to document their answers within predefined options (Sekaran & Bougie, 2016). This specific method of data gathering was used in order to be able to collect a large volume of data in a short period of time and to adjust the process to Covid-19 pandemic’s restrictions. In addition, electronic questionnaires were chosen as the aim of the survey was to cover consumers of two geographically distant countries.
The survey was based on Bandara et al. (2020b) since the aim is to interpret their findings through a cross-country perspective. The survey included statements that originate from the literature that is connected to the hypotheses. A Likert scale was used for the respondents to answer the survey in order to investigate how strongly the participants agree with several statements. The Likert scale helps explore the differences of consumers in terms of their attitudes (Sekaran & Bougie, 2016). The authors used a five-point Likert scale, in contrast with the original questionnaire that used a seven-point Likert scale, as it was considered the most suitable in order not to confuse the participants with too many choices but also to have a variety of views regarding the statements. Furthermore, the questionnaire also included an open comments section part at the end where the participants had the chance to add any comments, reflections, or thoughts regarding their online privacy concerns. The survey answers were later the basis for the analysis and were used to accept or reject the hypotheses or explain the findings.
The questionnaires were designed using Google Forms and were distributed to the authors’
network through e-mails and social media channels using the snowball effect (Alvehus, 2013). In order to collect as many answers as desired, the authors chose to translate the original survey to each country's language, not to delimit the respondents as English is not their native language, and to be sure that the respondents did not misinterpret the questions.
Afterwards, the responses were translated to English to proceed with the analysis. The process of collecting data lasted one month, from the end of March 2021 until the end of April 2021.
From the survey, 232 responses were overall gathered, and 26 of them left comments in the comments section. Specifically, 126 Greek and 106 Swedish consumers participated in the survey, while 16 and 10 of them expressed their opinions in the open comments section, respectively. The demographic profile of the respondents is stated in Table 2. To determine
17
the amount of responses that were needed to be collected for the research to be statistically significant, the sample size was calculated with a confidence level of 95% and a 6% margin of error.
Secondary data was used for the background and theoretical part of the study in order to support the findings. The secondary data was collected through Halmstad’s University databases. In order to have a high quality in the study, it was of importance that all secondary data could be found in journals that were ranked in the Chartered ABS Journal Guide. To find the secondary data, search terms such as “Online consumer privacy”, “Privacy e-commerce”, and “Cross-country online behaviour” were used in the databases.
Dimension Category Percentage (%)
GR SE
Age 18-24 years 24.8 38.7
25-34 years 40.0 43.4
35-44 years 12.8 2.8
45-54 years 15.2 9.4
55-64 years 7.2 4.7
Above 65 years 0.0 0.9
Gender Female 67.2 63.2
Male 32.0 36.8
Other 0.0 0.0
Prefer not to say 0.8 0.0
Employment Employed full-time 60.8 51.9
Employed part-time 10.4 1.9
Casual work 4.0 2.8
Self-employed 2.4 5.7
Student 12.0 35.8
Unemployed / Unable to work
2.4 0.0
Retired 1.6 1.9
Other 6.4 0.0
Education 9th grade 1.6 2.8
High school 12.0 20.8
Technical training 4.8 4.7
Some University completed
13.6 17.9
Undergraduate degree 44.0 30.2
Postgraduate degree 18.4 23.6
Other 5.6 0.0
Internet use 3-4 years 2.4 0.0
5-6 years 1.6 1.9
7-8 years 19.2 1.9
9-10 years 12.8 4.7
More than 10 years 64.0 91.5
Online shopping frequency 1-2 times per year 16.8 0.9
1-2 times per half year 18.4 6.6
1-2 times per three months 27.2 36.8
1-2 times per month 22.4 50.0
1-2 times per week 11.2 5.7
Almost everyday 4.0 0.0
Table 2: Sample demographic profile
