Blockchain technology: to foster or to fear?
Blockchain technology usage and its effect on consumer perceptions; mediated by technology anxiety and moderated by blockchain expertise and blockchain familiarity
– an experimental study in three industries.
MSc Business Administration – Digital Business Track
First and last name: Jeanine van der Bas Student number: 11399716
Submission date: 22 June 2018 (final version) Word count: 16.677
Statement of originality
This document is written by student Jeanine van der Bas 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.
Table of Contents
Abstract ... 5 1. Introduction ... 6 2. Literature review ... 9 2.1 Blockchain technology ... 9 2.1.1 Finance industry ... 11 2.1.2 Energy industry ... 11 2.1.3 Healthcare industry ... 12 2.2 Digitalization and consumer behavior ... 14 2.3 Consumer perceptions ... 15 2.3.1 Trust ... 15 2.3.2 Transparency ... 17 2.3.3 Security ... 18 2.3.4 Convenience ... 19 2.4 Technology anxiety ... 20 2.5 Consumer knowledge; familiarity and expertise ... 22 2.6 Industries ... 24 2.7 Conceptual framework and hypotheses ... 27 2.7.1 Hypotheses ... 273. Data collection and methodology ... 30
3.1 Research design and setting ... 30
3.2 Sample and procedure ... 31
3.3 The variables ... 33
3.3.1 The operationalization of the stimuli (independent variables blockchain usage and industries) ... 33
3.3.2 The measurement of the dependent variables ... 34
3.3.3 The measurement of the independent variables ... 36
3.4 Pre-test ... 38
3.5 Data collection and statistical procedure ... 38
3.5.1 Normality tests of variables ... 39
3.5.2 Factor analysis ... 40
3.5.3 Correlations ... 41
4.1 Direct and moderation effects of conditions on dependent variables ... 45
4.1.1 Effect of blockchain usage on perceived trust, transparency, security and convenience (H1a, H2a, H3a and H4a) ... 45
4.1.2 Effect of channel on mediator, moderators and convenience ... 48
4.1.3 Moderation effect of independent variable industries on dependent variables (H1b, H2b, H3b and H4b) ... 54 4.2 Mediation and moderation effects on dependent variables ... 57 4.3 Mediation effect of technology anxiety (M) ... 58 4.4 Moderation effects of blockchain expertise (Z) ... 59 4.4.1 Trust (H5a) ... 59 4.4.2 Transparency (H5b) ... 59 4.4.3 Security (H5c) ... 60 4.4.4 Convenience (H5d) ... 60
4.4.5 Technology anxiety (H5e) ... 61
4.5 Moderation effects of blockchain familiarity (W) ... 61
4.5.1 Trust (H6a) ... 61
4.5.2 Transparency (H6b) ... 61
4.5.3 Security (H6c) ... 62
4.5.4 Convenience (H6d) ... 63
4.5.5 Technology anxiety (H6e) ... 65
4.6 Improvement on conceptual model ... 65 4.7 Summary of results ... 67 5. Discussion ... 69 5.1 General discussion ... 69 5.2 Limitations and academic recommendations ... 75 5.3 Managerial contributions ... 77 6. Conclusion ... 80 Appendices ... 91 Appendix 1. Questionnaire ... 91 Appendix 2. Experiment conditions (stimuli) ... 102 Appendix 3. Cronbach’s alpha of scales ... 105 Appendix 4. Model summaries of interaction effects mediation and moderation ... 106
Abstract
The technological innovation blockchain is rapidly gaining ground in its importance for consumers and organizations in a wide range of industries. The blockchain technology literature- strong in some respects, but still underdeveloped in others- currently provides insufficient attention to consumer perceptions. Blockchain technology creates opportunities for organizations and consumers in the light of trust, transparency, security and convenience. This study aimed to examine the effect of organizations disclosing the usage of blockchain technology on consumer’s perceived trust, transparency, security and convenience among three industries. The constructs technology anxiety, blockchain familiarity and blockchain expertise were included to assess their effects on the above-mentioned relation. Data was collected from 295 participants by means of an online experiment. The results showed that participants who are highly familiar with blockchain technology, whilst being confronted with an organization that uses blockchain technology, report higher scores of perceived security. Moreover, similar results were found for blockchain familiarity to affect the relation between blockchain technology usage and perceived convenience. On the other hand, technology anxiety and blockchain expertise appeared to have no effect on perceived trust, transparency, security and convenience, nor did these perceptions differ among the three industries that were included in the experiment.
Keywords: blockchain technology, technology, consumer perceptions, trust,
transparency, convenience, security, technology anxiety, consumer knowledge, expertise, familiarity.
1. Introduction
“The biggest fear is whether banks, or other financials, get Amazon-ed” (Van
Steenis, 2018): a title of an article published in the Financial Times earlier this year that clearly reflects one of the potential effects of the new technology underlying the well known Bitcoin: blockchain. The impact of blockchain technology on the finance industry but also on other industries becomes increasingly more evident (Münsing, Mather & Moura, 2017; PWC, 2015). Blockchain technology, which is a shared ledger that provides transparency and efficiency amongst a network of users (Nguyen, 2016), has become a topic of increasing interest among scholars. It creates a peer-to-peer network and therefore affects many organizations that have a business model involving a third party to realize trust and transparency (Beck & Müller-Bloch, 2017). Within the financial industry the implications of blockchain technology are becoming increasingly apparent, whilst in the healthcare and energy industry the technology is still in its infancy (Aitzhan & Svetinovic, 2016; Orcutt, 2017).
Surprisingly; the current research field has mainly touched upon the legal-, disruptive- and technical implications of the technology and not yet on the consumer side of blockchain technology (Lindman, Tuunainen & Rossi, 2017). Blockchain technology brings opportunities for organizations as well as its consumers with respect to increased trust, transparency, security and convenience (Aitzhan & Svetinovic, 2016; Mengelkamp et al., 2018). However, a clear understanding on what blockchain technology will induce with respect to consumer perceptions has not been studied yet. An element that has shown to affect consumer perceptions of technologies to a substantial degree is technology anxiety (Liu, 2012). Especially in
the field of mobile environments, the concept of technology anxiety has proven to be apparent (Yang & Forney, 2013).
Hence, in this study, the effect of organizational blockchain technology usage on consumer perceptions of this organization in the light of perceived trust, transparency, security and convenience is examined. The following research question was formulated in pursuit of filling the current gap in the literature:
“What is the effect of blockchain technology usage on consumer perceived organizational trust, transparency, security and convenience and how is this effect mediated by technology anxiety?”
Examining this research question will result in gaining a deeper understanding of the constructs that play a role in the perceptions of consumers. For both academics and management, this study is relevant. Academics have not yet examined the implications for consumers and more so have not yet discovered their perceptions towards this technology (Lindman et al., 2017). Moreover, this study provides insights in the role consumer knowledge (both familiarity and expertise) plays in consumer levels of technology anxiety as well as their perceived trust, transparency, security and convenience. Management can use the findings of this study in understanding what influence blockchain technology usage has on the perception of an organization. Accordingly, this study generates new scientific knowledge whilst broadening the understanding and implications of blockchain technology for its consumers.
In the following section of this research, blockchain technology will be further described. Hereafter, the promises of blockchain in the three above mentioned industries (finance, healthcare and energy) will be explored, followed by the
topics perceived trust, transparency, security and convenience will be discussed as well as technology anxiety. The literature review is concluded by the review of studies in the field of consumer knowledge. Following the literature review, the conceptual model is presented with all hypotheses, which forms the foundation for the data collection and methodology section. Subsequently, the main findings are reported in the results section. The final section of this study contains a discussion of the results that also covers recommendations for future research and the study’s limitations and will be closed out by means of a conclusion.
2. Literature review
This literature review starts with the exploration of the blockchain technology concept together with its characteristics and breadth of implementation opportunities. Hereafter, the most recent literature on consumer behavior and digitalization will be evaluated. Moreover, consumer perception constructs of organizations that align with the characteristics of blockchain will be discussed. This examination led to the formulation of hypotheses. Lastly, the concepts of technology anxiety, familiarity and expertise are explored to formulate the final hypotheses. The conceptual model with all hypotheses is then presented, which forms the foundation for the methodology and data collection section.
2.1 Blockchain technology
With its first appearance in 2008 as the technology behind the well-known Bitcoin (Nakamoto, 2008), blockchain technology is gaining ground (Matilla, 2016). It is perceived by many as a new paradigm shift within digital networks (Beck, Czepluch, Lollike, & Malone, 2016; Matilla, 2016) and provides an innovative way to digitalize the ownership of assets (Lindman et al., 2017). Yuan & Wang (2016) describe blockchain as a shared ledger where chained and chronological blocks that are encrypted ensure a peer-to-peer network that fosters verifiability and synchronized data. This ledger is maintained by a network of participants, which is in line with a predefined set of rules to ensure global agreement (Sikorski, Haughton & Kraft, 2017).
blockchain technology. The second and third generation of blockchain technology is apparent in the digitalization of asset ownership, smart contracts, identity verification and intellectual properties (Swan, 2015). With this second and third generation, the technology has a large potential to disrupt industries beyond the financial industry as well (PWC, 2015; Schlegel, Zavolokina & Schwabe, 2018; Aste, Tasca & Di Matteo, 2017), whilst having the potential to completely reshape the entire economy (Nguyen, 2016). The underlying principle of blockchain technology is having decentralized, transparent and instant access to data that creates a multitude of opportunities in various industries (Mettler, 2016). On the contrary, Iansiti & Lakhani (2017) describe blockchain more as a foundational technology rather than a disruptive innovation, as it will take time before it will be fully embedded into the existing infrastructure. As such, scholars agree on the fact that blockchain will affect many businesses, either on the short term or longer-term. However, the wide adoption of blockchain is dependent on various challenges and drawbacks as denoted by Swan (2015). These challenges mainly touch upon the technical elements of the technology, namely the limited amount of transactions that can be processed per second, the deferral of a blockchain network and an attack danger of more than fifty percent (Swan, 2015). Effective programming and implementation can overcome many of these challenges, however the security element still has to be overcome in another manner that has not been discovered yet. Subsequently, the proof-of-work scheme that is necessary for dealing with the security of blockchain technology as well as the verification of transactions is computer intensive and therefore demands high levels of energy consumption (Vranken, 2017). However, large steps have been made in the past months, which have reduced the level of energy consumption to a significant degree. Moreover, the second and third generation of blockchain technology enables an opportunity for
blockchain networks to be only partly decentralized, relaxing the need for proof-of-work which affects the energy consumption simultaneously (Vranken, 2017).
2.1.1 Finance industry
Within blockchain technology’s first generation, the finance industry was greatly affected. In this industry, blockchain technology brings along various opportunities of which decreased costs, a faster and more efficient process of transactions and transparency are most apparent (Nguyen, 2016). The finance industry has also faced various risks that blockchain technology brings along, among these are a decreased level of competitiveness between banks as the ledger is shared by all users whilst regulatory and legal measures limit the real break-through of crypto-currencies (Nguyen, 2016). The latter has also been described by Iansiti & Lakhani (2017), referring to the collapse of Bitcoin in 2014 and to various other crypto-currencies that were hacked recently. This has an effect on the level of security of data, as the technology is relatively sensitive to hacks or breaches of privacy.
With the second and third generation of blockchain technology applications, enabling smart contracts to be established, the breadth of applications are numerous in other industries and are not limited to the finance industry (Burger, Kuhlmann, Richard & Weinmann, 2016). Two of these industries are the energy- and healthcare industry.
2.1.2 Energy industry
One of the industries that is expected to be significantly affected by blockchain technology is the energy industry (PWC, 2015). Blockchain technology has been identified as being part of the fourth industrial revolution (Industry 4.0)
(Sikorski, Haughton & Kraft, 2017). This blockchain based energy infrastructure can ensure a transparent, secure and distributed network where even the smallest transaction can be traceable and kept track of (Xu et al., 2016; Mengelkamp et al., 2018). From a consumer perspective in this industry, they can decide upon their producers of energy and what sources the energy is being generated from (Aitzhan & Svetinovic, 2016; Mengelkamp et al, 2018). Subsequently, Burger et al. (2016) state that blockchain enables consumers to have access to more immediate and faster information streams whilst obtaining more precise and transparent information provision on the energy origin. It focuses on business models that foster secure connections between entities to trade energy that is being generated (Burger et al., 2016).
By the integration of renewable energy sources into blockchain technology, an efficient system that fosters reliability and sustainability can be created whilst providing economic benefits to its consumers (Lv & Ai, 2016; Mengelkamp et al., 2018). Next to the energy industry, the healthcare industry will be greatly influenced by blockchain technology; changing the way consumers manage their data and interact with healthcare institutions as delineated in the next paragraph.
2.1.3 Healthcare industry
The healthcare industry was traditionally shaped by means of three data sharing mechanisms: push, pull and view (Halamka & Ekblaw, 2017). With blockchain as new technology, a fourth mechanism is introduced whereby medical data can be kept securely whilst being shared among providers. The initial three mechanisms will be briefly touched upon to get an understanding of how blockchain technology can potentially provide a secure and burdensome solution to data sharing in the medical sector. As Halamka & Ekblaw (2017) state, push, pull and view can all
be seen as sharing data in an ad hoc manner, whereby consent and permissions are completed in an informal way and where no standardized audit trail is created. Blockchain technology within the healthcare industry can create an infrastructure for data integrity, standardized audit trails and contracts for having access to data. Moreover, Yue et al. (2016) confirm this by stating that healthcare data is currently scattered among various providers and systems, making data sharing inefficient whilst putting the patients’ data privacy at risk. As Schlegel et al. (2018) identify: one of the key merits of blockchain technology is record management, thus playing a key role in the healthcare industry. With sharing data, this technology makes the medical system smarter and increases the level of service provided to consumers. Moreover, it provides a cheap and efficient system to store patient data across institutions. A blockchain technology based system can result in patients owning their own medical data without having to compromise on either privacy or the level of healthcare service. Patients will always know who accesses their data and therefore do not rely on a third party to trust their data with in a blockchain technology driven industry (Yue et al., 2016). At this stage, blockchain technology driven solutions for medical industries are still in its infancy; various prototypes of blockchain based data infrastructures and patient identifiers are currently being explored (Orcutt, 2017).
Blockchain technology has thus showed to have a significant effect on the means by which organizations operate in the finance, energy and healthcare industry. The end-users, also referred to as consumers, will also be affected by these changes, hence looking into consumer behavior and how they perceive digitalization at large will aid in understanding the implications of blockchain technology.
2.2 Digitalization and consumer behavior
Consumer behavior is described as the process involved of individuals or groups responding to services or products (Solomon, Russell-Bennett & Previte, 2012), and addresses where and why consumers spend their resources. Developments in information and communication technologies over the past decade have resulted in structural changes in business strategies and affect consumers in their social and economic lives (Corrocher & Ordanini, 2002; Bharadwaj, El Sawy, Pavlou & Venkatraman, 2013). This is often referred to as digitalization or a digital economy. By means of these technologies, organizations are focusing more on enhancing the level of customer service (Bharadwaj et al., 2013). This is supported by a study on customer interactions with organizations; customers who perceived a below standard service experience were 40% less likely to stay involved with such an organization (Dougherty & Murthy, 2009). Within the financial industry it is evident that by the increased use of technology (by the introduction of ATMs and internet banking i.a.) there are fewer human-to-human interactions and rather increased standardized processes (Dabholkar, Michelle Bobbitt & Lee, 2003; Ganguli & Roy, 2011). Blockchain technology integration in the finance industry has shown to impact these interactions as well, as a peer-to-peer network is created where a third party becomes redundant (Davidson, De Filippi & Potts, 2016).
Whilst most organizations take measures to maintain a high customer service level, research on blockchain technology have so far mainly focused on the technical issues, legal structures and disruptiveness of the innovation rather than on the implications for its consumers (Lindman et al., 2017; Schlegel et al., 2018). Lindman et al. (2017) continued by stating that current studies focused on the features of blockchain applications and that its effect on consumer adoption is lagging behind.
More precisely, the characteristics of blockchain affecting consumer usage and what contextual elements make consumers decide what platform to use are areas that need exploration to complement current research conducted (Lindman et al., 2017).
2.3 Consumer perceptions
To gain deeper understanding of what drives consumers to accept technologies, previous studies regarding technology acceptance and what elements play a role in consumer perceptions to newly introduced technologies were examined. The perception measurements that have been taken into account are the ones that are relevant to blockchain technology; using the technology’s promises as delineated above for the three industries as starting point. The brand perception measurements that seemed most applicable to this technology are: trust, transparency, security and convenience.
2.3.1 Trust
The importance of trust as foundation for business relationships has been described by many scholars, among which Hart & Saunders (1997) describe it as a fundamental element. Mayer, Davis & Schoorman (1995, p. 712) defined trust as
“willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other part”. A similar definition
has also been used by Gambetta (2000), however Mayer et al. (1995) have added the willingness to be vulnerable to the definition. Trust and risk have proven to be interlinked as trust can be seen as the willingness to assume risk. When the propensity
a risk taking business relationship (Mayer et al., 1995). Stewart (1999) conducted research in the field of adoption of financial transactions via the Internet and denotes that trust is two-sided: the technology used and the institution adopting the technology. The experiment showed that perceived technology trust (in Stewart’s research organizations’ sales via the Internet) is influenced by consumer’s perceived organizational trust. Whereby there is a carryover effect from trusted organizations to the technology they are adopting.
However, according to Hoffman, Novak & Peralta (1999), many consumers are still lacking trust towards organizations engaging in web activities (i.e. e-commerce related activities) as consumers feel a lack of control over the data that these organizations own. This is closely interlinked with their privacy concerns regarding data and information usage as will be discussed in the security section below.
The study by Joseph, Sekhon, Stone & Tinson (2005) identified that when making use of new technology offered by a financial institution, confidence in the bank was of crucial importance when evaluating their electronic banking services. The study concerning e-shopping found that organizational trust plays a key determining factor in consumer attitudes towards the technology (Ha & Stoel, 2009).
With blockchain technology, consumers do not need to trust a third party to control and manage their data as the technology enables them to control this themselves, however in stead they need to trust the platform used by the organization. According to the studies mentioned above, research have only touched upon the carryover effects from trust in an organization to having trust in technology. However, to explore how consumers perceive organizations that use blockchain in their level of perceived trust it is hypothesized that:
H1a. An organization using blockchain technology is perceived more trustworthy.
2.3.2 Transparency
Organizations are increasingly pushed towards more transparency, which is mainly driven by the speed of communication via the Internet among other advances in technology communications (Carter & Rogers, 2008; Bhaduri & Ha-Brookshire, 2011). The term transparency has been described by many scholars, whom agree upon transparency referring to organizations sharing information with consumers (Vishwanath & Kaufmann, 2001; Carter & Curry, 2010). This sharing of information can be by means of four types: supply chain, cost/price, organization and technology according to Hultman & Axelsson (2007). Whereby their perceived transparency refers to the degree an organization discloses information regarding its organizational practices and technology usage to its stakeholders. For this research the following definition of transparency is used: “the visibility and accessibility of information
especially concerning business practices” (Bhaduri & Ha-Brookshire, 2011, p. 136).
In accordance with this definition, organizations that do not provide information regarding business practices or provide inaccurate or irrelevant information are perceived to be not transparent. Transparency can be viewed as two-fold, whereby it poses benefits for both the consumer and the organization; both are described below.
Firstly, transparency limits the amount of information asymmetries between the consumer on the one side and the organization on the other (Grewal, Iyer, Krishnan & Sharma, 2003). Enabling the consumer to enjoy greater levels of empowerment (Siminitiras, Dwivedi, Kaushik & Rana, 2015) through the availability of information whilst it increases the level of consumer welfare (Carter & Curry, 2010). In its turn, it enables organizations that operate in a transparent manner to have
Subsequently, Carter & Curry (2010) found that consumers value organizational transparency over opaqueness.
As was denoted in section 2.1, blockchain technology enables organizations to be more transparent in its applications in various industries, hence it is hypothesized that:
H2a. An organization using blockchain technology is perceived more transparent.
2.3.3 Security
Security is among others an important factor that determines consumers’ perceptions of organizations and affects the likeliness of using and accepting technological interfaces. Security has been defined by Whitworth, Bañuls, Sylla & Mahinda (2008, p. 788) as “to protect against unauthorized entry, misuse or
takeover”. More specifically, Chang & Chen (2009) defined perceived security as “the extent to which a potential customer believes that the interface is secure for transmitting sensitive information”. A definition posed by Fang, Chan, Brzezinski &
Xu (2005, p. 138) is most closely related the blockchain as it focuses on authorization of data access: “The extent to which a user believes that using a particular
application will not expose his or her private information to any unauthorized party”,
therefore this definition will be used for this study.
Within various industries and applications, the influence of perceived security on consumer perceptions has been researched. Joseph et al. (2005) identified that security is among one of the most important factors for consumers when making use of ATMs: the technology introduced in the financial industry. Subsequently, Chang & Chen (2009) denoted that a lack of perceived security is a main reason for consumers to be reluctant in using online shopping offerings. Perceived security is shaped by
their perceptions of the risks involved by disclosing sensitive personal details and bank account details via the Internet or online interfaces. A main concern of these consumers that engage in online shopping is the level of security of their transactions via the Internet.
Within blockchain technology applications, according to Kosba, Miller, Shi, Wen, & Papamanthou (2016) the present form of blockchain technology and smart contracts is lacking in its degree of transactional privacy. Despite the fact that entities can create pseudonyms to ensure their anonymity, recent studies have shown that hackers engage in de-anonymization by analyzing cryptocurrency transaction graph structures (Kosba et al., 2016).
As blockchain technology is still in its infancy with it being sensitive to hacks in its current development stage, and consumers placing great value on security as posed above, it is hypothesized that:
H3a. An organization using blockchain technology is perceived less secure.
2.3.4 Convenience
Studies addressing the adoption of technologies in numerous industries have highlighted the importance of perceived convenience in consumer’s valuation of technology (Joseph et al., 2005; Ha & Stoel, 2009). Consumer convenience is a broad concept, but has been defined by various scholars by time spent and effort made by consumers to obtain a service (Berry, Seiders & Grewal, 2002; Collier & Sherrell, 2010). Berry et al. (2002) make a distinction between perceived product and service convenience, whereby it is perceived to be convenient when either it saves time or it limits the emotional, cognitive and physical burdens of consumers.
Studies in the field of marketing highlighted the critical role a consumer’s desire towards convenience plays. In general, the greater the effort made or time spent by consumers, the lower the consumer perceives the level of convenience (Berry et al., 2002). Meuter, Ostrom, Roundtree & Bitner (2000, p. 55) categorized consumers in three segments covering convenience trough ‘When I want’, ‘Where I want’ and
‘Saved time’. Similar to Berry et al. (2002), Collier & Sherrel (2010) denoted that
perceived convenience is shaped by situational factors, encompassing cognitive and physical efforts made. Studies on the adoption of online shopping (Kim, Chan & Gupta, 2007) and RFID (Radio Frequency Identification) (Hossain & Prybutok, 2008) found that perceived convenience directly affects the adoption of technologies or systems. Hence, perceived convenience plays a crucial role in the consumer perceptions of technologies at large.
As blockchain technology enables consumers to have instant access to information and fosters a more efficient process of transactions or information streams, it is hypothesized that:
H4a. An organization using blockchain technology is perceived more convenient.
2.4 Technology anxiety
A psychological aspect that has been associated with consumer’s intentions to use technologies and actual experiences of technologies is technology anxiety. Meuter, Ostrom, Bitner & Roundtree (2003, p. 900) defined technology anxiety as
“The fear, apprehension and hope people feel when considering use or actually using all forms of technology”. Moreover, Yang & Forney (2013) denoted that anxiety is
among transactions is not clear in a technology driven environment. Technology anxiety may be enlarged as a result of consumers being exposed to an environment where information or monetary funds can be lost due to a mistake or failure of the technology (Yang & Forney, 2013).
Liu (2012) stated that technology anxiety is an intermediate construct that affects consumer’s perceptions of technologies. Technology anxiety can be viewed as a construct that affects consumer’s behavioral intentions in light of cognitive theories (Yang & Forney, 2013).
As blockchain technology is viewed as a technology that changes the means by which organizations operate and is disruptive in its nature, it is expected to have an effect on the level of technology anxiety of consumers. It is therefore hypothesized that it has a significant effect on how consumers perceive organizations with respect to the elements described in the previous sections. The formed hypotheses are delineated below:
H1c: The relationship between blockchain technology usage and perceived trust is
negatively mediated by technology anxiety.
H2c: The relationship between blockchain technology usage and perceived
transparency is negatively mediated by technology anxiety.
H3c: The relationship between blockchain technology usage and perceived security is
H4c: The relationship between blockchain technology usage and perceived
convenience is negatively mediated by technology anxiety.
2.5 Consumer knowledge; familiarity and expertise
Yang & Forney (2013) and Venkatesh, Thong & Xu (2012) state that facilitating conditions could aid in consumers overcoming barriers to the usage of technologies, especially in the early technology adoption stage. In addition, Venkatesh, Morris, Davis & Davis (2003) state that social influences determine the means by which consumers accept technologies and what their perceptions are. Whereby consumers that have a high degree of technology anxiety are more likely to depend on peer’s views and opinions on technologies as this high degree of anxiety is directly related to a lack of confidence and knowledge. Consumers that have a low degree of technology anxiety depend less on peer’s opinions and depend more on their own understanding of technology and experience a higher level of confidence (Venkatesh et al., 2003). Jacques, Garger, Brown & Deale (2009) state that consumers that experience technology anxiety towards components of an unfamiliar technology are less likely to have an intention to adopt such a technology.
Subsequently, consumers’ judgments about things they are unknowledgeable about are shaped by information that is made available via cues according to the inference theory (Baker, Parasuraman, Grewal & Voss, 2002; Chang & Chen, 2009). Thus, consumer knowledge directly affects consumers’ judgments as well as that this affects their level of technology anxiety. The topic of consumer knowledge has been studied by many scholars; Moorman, Diehl, Brinberg & Kidwell (2004) have described the term consumer knowledge to encompass two elements. A distinction is made between actual knowledge and consumer’s perceptions of their knowledge
(Moorman et al., 2004). In other words: the former referring to objective knowledge or expertise and the latter referring to subjective knowledge or experience. Articles have shown a significant positive correlation between these two, as reported by Brucks (1985). Moreover, as described by Cordell (1997), knowledge affects not only the means by which people attempt to gather information/cues but more so that knowledge affects consumer judgments. Cordell (1997) made a similar distinction as Moorman et al. (2004), and categorized knowledge in familiarity, objective expertise and subjective expertise. Whereby familiarity is stated to be similar to the term experience as used by Moorman et al. (2004). Objective expertise is measured by a testing procedure and covers the understanding and application of the potential of a product. Subjective knowledge, referred to as familiarity, reflects an individuals’ evaluation of their knowledge of a product (Cordell, 1997). A study conducted by Allgood & Walstad (2016) address familiarity and expertise in light of financial literacy, whereby they tested the factual knowledge by means of test questions with correct and incorrect answer possibilities as well as the subjective evaluation of one’s knowledge by means of self-assessment questions. These both appeared to have a significant influence on (financial) behaviors.
Hence, as consumer knowledge (both expertise and familiarity) has a significant influence on both technology anxiety as well as judgments and behaviors, the following hypotheses have been formulated:
Blockchain expertise
H5a: Blockchain expertise will moderate the relationship between blockchain usage
and perceived trust.
H5c: Blockchain expertise will moderate the relationship between blockchain usage
and perceived security.
H5d: Blockchain expertise will moderate the relationship between blockchain usage
and perceived convenience.
H5e: Blockchain expertise will moderate the mediation effect of technology anxiety
between blockchain usage and perceived trust/transparency/security/convenience. Blockchain familiarity
H6a: Blockchain familiarity will moderate the relationship between blockchain usage
and perceived trust.
H6b: Blockchain familiarity will moderate the relationship between blockchain usage
and perceived transparency.
H6c: Blockchain familiarity will moderate the relationship between blockchain usage
and perceived security.
H6d: Blockchain familiarity will moderate the relationship between blockchain usage
and perceived convenience.
H6e: Blockchain familiarity will moderate the mediation effect of technology anxiety
between blockchain usage and perceived trust/transparency/security/convenience.
2.6 Industries
As mentioned in section 2.1, the industries that are greatly affected by blockchain technology are the finance, healthcare and energy industry. Looking into consumers’ current perceptions of these industries will provide an insight in the as-is scenario to validate whether there are any differences in perceptions of organizations
in these industries. Please note that for this section, management reports were used because academic literature does not cover nor address industries as differentiating variable.
Research conducted by Stylianides (2014) from PWC shows that consumer trust levels in financial institutions is clearly lacking, as on average only twenty-five percent of consumers have indicated that they have trust in retail/investment banks or financial advisors. On the contrary, Stylianides’ (2014) research shows that the healthcare industry has gained consumer trust with 77.5%. Consumers indicated that privacy and security issues are among the foundational elements affecting their perceptions of financial institutions. Whereas the Edelman Trust Barometer on healthcare has shown that its level of trust is relatively low, only the finance industry has reported lower levels of perceived trust (Edelman, 2017). Whereby the protection of patient data and quality control have been identified as most important areas to focus on, if it wants to increase its levels of trust. Edelman’s Trust Barometer within the energy industry has shown to be mainly driven by an organization’s level of transparency with regard to its practices as well as safeguarding consumer data (Edelman, 2017).
The industries where blockchain technology is expected to be of significant influence thus appear to be currently evaluated differently with respect to their perceived trust, transparency and security levels. As blockchain technology addresses these elements as delineated in section 2.1, it is hypothesized that industries will act as a moderator on its effect on consumer perceptions of organization that adopt blockchain technology:
H1b: Industries will moderate the relationship between blockchain usage and
perceived trust.
H2b: Industries will moderate the relationship between blockchain usage and
perceived transparency.
H3b: Industries will moderate the relationship between blockchain usage and
perceived security.
H4b: Industries will moderate the relationship between blockchain usage and
2.7 Conceptual framework and hypotheses
The theories and key study results mentioned in the literature review contribute to the conceptual framework of this research study. The conceptual framework of the effect of organization’s blockchain usage on consumer perceptions is shown in figure 1 and encompasses all hypotheses proposed in the literature review section.
Figure 1: Conceptual Framework 2.7.1 Hypotheses
H1a. An organization using blockchain technology is perceived more trustworthy. H2a. An organization using blockchain technology is perceived more transparent. H3a. An organization using blockchain technology is perceived less secure. H4a. An organization using blockchain technology is perceived more convenient.
H1b: Industries will moderate the relationship between blockchain usage and
perceived trust.
H2b: Industries will moderate the relationship between blockchain usage and
perceived transparency.
H3b: Industries will moderate the relationship between blockchain usage and
perceived security.
H4b: Industries will moderate the relationship between blockchain usage and
perceived convenience.
H1c: The relationship between blockchain technology usage and perceived trust is
negatively mediated by technology anxiety.
H2c: The relationship between blockchain technology usage and perceived
transparency is negatively mediated by technology anxiety.
H3c: The relationship between blockchain technology usage and perceived security is
negatively mediated by technology anxiety.
H4c: The relationship between blockchain technology usage and perceived
convenience is negatively mediated by technology anxiety.
H5a: Blockchain expertise will moderate the relationship between blockchain usage
and perceived trust.
H5b: Blockchain expertise will moderate the relationship between blockchain usage
and perceived transparency.
H5c: Blockchain expertise will moderate the relationship between blockchain usage
and perceived security.
and perceived convenience.
H5e: Blockchain expertise will moderate the mediation effect of technology anxiety
between blockchain usage and perceived trust/transparency/security/convenience.
H6a: Blockchain familiarity will moderate the relationship between blockchain usage
and perceived trust.
H6b: Blockchain familiarity will moderate the relationship between blockchain usage
and perceived transparency.
H6c: Blockchain familiarity will moderate the relationship between blockchain usage
and perceived security.
H6d: Blockchain familiarity will moderate the relationship between blockchain usage
and perceived convenience.
H6e: Blockchain familiarity will moderate the mediation effect of technology anxiety
between blockchain usage and perceived trust/transparency/security/convenience.
These hypotheses form the basis for the data collection and methodology. The next section will elaborate on what method has been used to examine these
3. Data collection and methodology
This section will provide the setting for the experiment, the operationalization of the variables, the sampling and data collection plan and lastly the methodology used to analyze the outcomes of the data gathered.
3.1 Research design and setting
This experiment was conducted in the online survey platform Qualtrics. The link obtained via Qualtrics was used to distribute the survey via social media channels (i.e. Facebook, e-mail and LinkedIn) and hence the sample was obtained by making use of convenience sampling. The measurement of the response rate was not possible via these channels as this information is not readily available. Moreover, the platform Mechanical Turk was used to collect a proportion of the participants needed for this study (33% of all participants).
The majority of the data collection was conducted in the Netherlands. Most of the participants were students, employed part- or full-time. It was assumed that the vast majority of the participants would be of Dutch nationality as this was the main location of data gathering as well. Moreover, a large proportion would be of American nationality as this was set as one of the preconditions for the data collection via Mechanical Turk, as these are assumed to have a high proficiency of the English language. Generally, the sample consisted of participants of all age ranges and backgrounds as this would give a most representative image of any type of consumer that would come across organizations sharing technology usage to its consumers.
The experiment was a between-subjects experimental design, which enabled to examine causal links between variables. Subsequently, the nature of this study is
explanatory as it strives to establish relationships between variables (Saunders, Lewis, Thornhill, 2009). The data collection was conducted by means of six conditions which represented the independent variables: blockchain usage and industries. (1) Blockchain usage and (2) industries were manipulated in two and three levels respectively. Hence, participants were randomly allocated to one of the six conditions of a 2 (blockchain usage: yes/no) x 3 (industries: finance, healthcare, energy) between subjects design as shown in figure 2.
Figure 2. Online experiment conditions
3.2 Sample and procedure
Six hundred and fifteen participants entered the survey via Qualtrics of which three hundred twenty were not entirely completed or did not answer the quality check questions (for the Mechanical Turk platform) correctly and hence were not included in the analysis. Subsequently, two hundred ninety five participants completed the online survey. The participants were randomly allocated to one of six conditions (blockchain usage/ industries). The amount of participants and its distribution among the six conditions is visualized in figure 3.
Figure 3. Online experiment conditions and participant distributions
The age of the participants ranged from 17 to 76 years old with a mean age of 35.83 years (SD= 14.87), more than half of the participants were between 20 and 31 years old. The participants consisted of 130 males and 166 females, 43.9 and 56.1 percent respectively. Among these participants, 93 were from the United States (31.4%) and 162 from The Netherlands (54.7%), the remaining 13.9% were from European, Asian or African nationalities. The majority of the participants were students (37.2%), employed part-time (11.5%) or employed full-time (34.5%), the remaining 16.8% were either unemployed looking for work (4.1%), unemployed not looking for work (4.4%) or retired (8.4%). Moreover, more than half of the participants (53.4 %) had obtained their bachelors’ degree.
From this final sample, all variables were checked on missing data using frequency tests, which resulted in 0% missing data. No missing data were observed due to the option ‘force response’ for all questions posed via the questionnaire in Qualtrics.
When the participants commenced the survey they were asked to give permission to use their data and responses confidentially for research purposes only. Hereafter they were presented with the description of TMO (fictitious organization), after which they were presented with numerous statements regarding their perceptions of this organization in the light of trust, transparency, security and convenience. Moreover, questions examining their general technology anxiety levels and prior knowledge on blockchain were posed. Finally, questions regarding the control variables, such as demographics, were asked. To conclude the survey, the participants were thanked for their participation and could view their test score on blockchain expertise. They were given the opportunity to provide their e-mail address in case
they wanted to receive the outcome of the study and were provided with contact details in case they had any questions regarding the survey or the study in its totality. The complete survey can be found in appendix 1.
The experiment has a between-subjects design with five dependent variables: (1) trust, (2) transparency, (3) security, (4) convenience and (5) technology anxiety, and two independent variables next to the stimuli: (6) blockchain familiarity and (7) blockchain expertise. Technology anxiety (5) was examined on its mediating effect whilst (6) blockchain familiarity and (7) blockchain expertise on its moderating effects on the relation between the independent variables and dependent variables 1, 2, 3, 4 and 5.
The following section will provide an explanation and overview of the various variables that have been taken into consideration as well as their operationalization.
3.3 The variables
This section will elaborate on how the independent variables were operationalized as conditions for the experiment as well as the operationalization of the dependent and independent variables of the model. Moreover, it will discuss the control variables that have been taken into account for this study.
3.3.1 The operationalization of the stimuli (independent variables blockchain usage and industries)
In developing the stimuli for this study, each of the conditions was visualized: the participant was presented with an organization either using blockchain technology or using technology (not specified what technology) and in what industry this
of organization TMO. All participants were randomly allocated to one of six conditions. The stimuli used in this study can be viewed in appendix 2. In accordance with the experimental design as described in the previous section, the participants were exposed to a condition where blockchain is used by the organization or is not used by the organization and the type of industry it concerned; either energy, finance or healthcare. The stimuli were manifested in the usage of text (e.g. for blockchain usage: … is a blockchain technology company…in the energy/ finance /healthcare sector; and for no blockchain usage: … is a technology company … in the energy/ finance /healthcare sector) and images (fictitious logo for TMO, consistent among all stimuli). Hence, creating a realistic scenario for this experiment as this increases the degree of generalizability within experimental settings (Berkowitz & Donnerstein, 1982; Lynch, 1982). Other than the changes in technology usage and industry specifications, the fonts, company logo, texts and background color have been kept constant to eliminate the chance of other factors influencing participant’s perceptions. Moreover, various organizations in the finance, energy and healthcare industry have been looked at to verify what information they share on their technology/blockchain usage. This was the main source of information to create an as realistic as possible stimuli for the experiment.
3.3.2 The measurement of the dependent variables Trust
Sheinin, Varki & Ashley (2011) three items (⍺ = .82) on trustworthiness were used to measure participant’s perceived trustworthiness in the presented organization. The scale consisted of three items, such as “TMO is dependable”. Participants were
asked to give their opinion on a 7-point Likert scale (1= strongly disagree – 7= strongly agree).
Transparency
Rawlins’ (2008) measure on transparency was combined with Vaccaro & Madsen’s (2009) measure of perceived transparency. These scales were combined, as some of the items were not relevant for this study and its context. The items were slightly altered to fit with the scope of this study of which “This organization is likely to provide information that is useful to people like me for making informed decisions” is an example of an adapted item from Rawlins’ measure and “This organization is likely to disclose relevant information regarding organizational practices” is an example of an adapted item from Vaccaro & Madsen’s study. The scale has five items and was validated in this study with a Cronbach’s ⍺ = .83. Responses to all items were recorded on a 7-point Likert scale (1= strongly disagree – 7= strongly agree).
Security
Salisbury, Pearson, Pearson & Miller (2001) measure on perceived security was used to assess the participant’s perceived security of a technology. Only slight changes have been made to the items to match the context of this study. An example of this is: “I would feel secure disclosing sensitive information to this organization”. The third item of the scale showed a lower factor loading than the other three, however its effect on the scale’s Cronbach alpha would not increase significantly when the item could be removed (<.10). Hence it was decided to keep the scale as is –
Responses to the all four items were recorded on a 7-point Likert scale whereby 1= strongly disagree – 7= strongly agree.
Convenience
Convenience was measured by using Yoon & Kim’s (2007) measure on perceived convenience. The four items posed by Yoon & Kim were slightly rephrased to fit with this study. Yoon & Kim’s (2007) measure was validated in this study with a Cronbach’s ⍺ = .90. Responses to the four items were measured on a 7-point Likert scale (1= strongly disagree – 7= strongly agree). An example of an item is “This organization is likely to enable me to accomplish a request at a time that is convenient for me”.
Technology anxiety
Collier & Sherell’s (2010) scale on technology anxiety (⍺=0.83) has been used to measure the participant’s level of technology anxiety in general. It encompasses three items that have been measured on a 7-point Likert scale whereby 1= strongly disagree – 7= strongly agree. An example of one of the items is: “I hesitate to use technology for fear of making a mistake I cannot correct”.
The Cronbach’s alpha for each of these scale items in case it would be deleted can be viewed in the table in appendix 3. The variables included in this table are trust, transparency, security, convenience, technology anxiety and blockchain familiarity.
3.3.3 The measurement of the independent variables Blockchain expertise
A questionnaire with seven questions was designed to assess the participant’s exposure to and knowledge of blockchain. The questions derived from Grewal-Carr & Marshall (2016) paper where the elements that are common to all blockchain applications are delineated. These questions were posed with a multiple-choice answer model, in which only one answer was correct for each question. The amount of questions answered correctly resulted in a ‘score’ for each of the participants. Subsequently, each participant scored between zero and seven on their total score. The mean score among all participants was 4.57 and the SD=1.55.
Blockchain familiarity
Blockchain familiarity was measured by means of Dursun, Kabadayı, Alan & Sezen’s (2011) three items on familiarity. Their measure was used to assess respondents’ familiarity with a ‘store brand’, hence for this study the three items were adapted to blockchain technology. The items were measured on a 7-point Likert scale whereby 1= strongly disagree – 7= strongly agree. An example item is: “I am experienced with blockchain technology”. The scale was validated in this study with a Cronbach’s ⍺ = .91.
Control variables
In order to test the robustness of the proposed relationships, the effects and to rule out any extraneous influences, various control variables were included in this study. In this study, the following control variables were taken into account: age, gender, nationality, education level, occupation and the channel through which they completed the survey (i.e. researcher’s own network or Mechanical Turk). The latter
responses derived from the Mechanical Turk platform as opposed to 67% via the researcher’s own network.
3.4 Pre-test
A pre-test was executed in Qualtrics among four participants to verify whether the stimuli used were clearly formulated. Moreover, the items used were checked upon their readability and understanding. As a result of this pre-test, two slight adjustments were made to the survey. Firstly, a statement highlighting the fact that the company presented (TMO) is a fictitious organization was added in bold to the introduction text, as this was not entirely clear to the participants. Moreover, the participants felt as if they needed more information regarding the implications of the usage of (blockchain) technology to be able to answer the questions posed. Disclosing information regarding blockchain technology’s implications cannot be added as this might be an additional source of information that could influence the participants in their perceptions. Hence, the sentence “They use technology in the field of data storage and accessibility” was added, which was kept the same for the blockchain technology usage stimuli as well as the technology usage stimuli. This aided in the understanding of the application of technology within the presented organization TMO for the participants as this was cross-validated with the participants of the pre-test.
3.5 Data collection and statistical procedure
IBM SPSS statistics version 24 was used to analyze the outcomes of the experiment. Firstly, the data gathered via the Qualtrics survey was checked for
missing data or outliers, whereby one outlier was detected within the dataset and was excluded from further analysis. Dummy variables were created for the independent variables blockchain usage and industries. The counter-indicative item from the security construct was reverse coded in SPSS and all measures were validated on their reliability. As one scale showed a significant change in reliability scores based on one item, a principal axis factoring analysis was performed. As one of the preconditions for such an analysis is to verify the normality of variables used, normality checks were conducted for the moderators, mediator and dependent variables.
3.5.1 Normality tests of variables
To test for normality, the Kolmogorov-Smirnov tests as well as skewness and kurtosis scores were examined. Firstly, the Kolmogorov-Smirnov tests were performed to test whether the distributions of the variables were approximately normal. The Kolmogorov-Smirnov tests and inspections of the stem and leaf plots showed that none of the variables were normally distributed, as all tests were significant p<.005. The main results regarding the skewness and kurtosis can be viewed in table 1.
A general rule of thumb states that any value of kurtosis and skewness between -1 and 1 is acceptable. As can be derived from the data in table 1, all variables comply with this rule except for the blockchain familiarity measure. Blockchain familiarity has a platykurtic distribution, indicating that the distribution is flatter than a normal distribution (Field, 2009). It was examined whether a transformation of this variable would create a normal distribution of the construct (Field, 2009). This was not the case for the blockchain familiarity construct, however with large sample sizes (more than 40 participants per treatment) skewness or kurtosis will not affect the analysis substantially (Field, 2009). As within this study each condition contained more than forty participants, the risk of having distorted results was kept to a minimum.
3.5.2 Factor analysis
Hereafter, a principal axis factor analysis (PAF) was conducted on the scales used. The Kaiser-Meyer-Olkin measure verified the sampling adequacy for the analysis, KMO=.862. Bartlett’s test for sphericity χ2
(231) = 3653.827, p<.001, indicated that correlations were sufficiently large for a PAF. An initial analysis was run to obtain eigenvalues for each component in the data. Six components had eigenvalues above Kaiser’s criterion of 1 and in combination explained 71.76% of the variance. In accordance to Kaiser’s criterion, examination of the scree plot showed a leveling off after the sixth factor. Thus, six factors were retained and were rotated with an Oblimin with Kaiser normalization rotation. Table 2 shows the factor loadings after rotation. The items that cluster on the same factor suggest that factor 1 is Trust, factor 2 is Transparency, factor 3 is Security, factor 4 is Convenience, factor 5 is Blockchain familiarity and factor 6 is Technology anxiety.
Table 2: Factor loadings using principal axis factoring 3.5.3 Correlations
To initially assess the size and direction of the relations between the variables included in this study, a bivariate Pearson’s product-movement correlation (r) was computed (Field, 2009). Table 3 provides the overview of all correlations, means and reliability scores of the variables used. For this study the independent variables (i.e. conditions blockchain usage and industries), dependent variables (trust, transparency, security and convenience), mediator (technology anxiety), moderators (blockchain familiarity and blockchain expertise) as well as control variables (age, gender, education, nationality, occupation and channel) were included. The analysis showed some significant correlations, which will be briefly touched upon in the following
As can be seen in table 3, the bivariate relations between the independent variable blockchain usage and the dependent variables trust, transparency, security and convenience does not show correlations. However, the moderating variables do correlate with one another. Blockchain expertise positively correlates with blockchain familiarity (r=.411, p<.01), this was also expected according to current research as mentioned in the literature review. Moreover, channel (i.e. own network or Mechanical Turk) correlates with technology anxiety (r=.197, p<.01), blockchain familiarity (r=.217, p<.01), blockchain expertise (r=.173, p<.01) and convenience (r=.229, p<.01). These correlations give the impression that there is a relation between these variables. Hence, this will be looked into in the analysis section to verify whether there is a significant difference in mean scores on the variables technology anxiety, blockchain familiarity, blockchain expertise and convenience for the two levels of channel (Mechanical Turk and own network).
3: M ea ns , S ta nda rd D evi at ions , and C or re la ti ons
3.5.4 Methodology and analysis
In order to test the effect of the two conditions of blockchain technology usage on the dependent variables, four univariate analyses of variance (one-way ANOVAs) were conducted. These one-way ANOVAs gave insights in mean score differences among the four dependent variables for both conditions. Thereby, addressing the direct effect that was hypothesized by H1a, H2a, H3a and H4a.
Furthermore, to examine the moderating effect of industries on the direct relationship between blockchain technology usage and the four dependent variables, PROCESS model 1 (Hayes, 2013) was used for the dependent variables trust, transparency, security and convenience (H1b, H2b, H3b and H4b).
Moreover, Hayes’ PROCESS model 10 (Hayes, 2013) was used to examine the mediation effect of technology anxiety between blockchain usage and the dependent variables trust, transparency, security and convenience as well as the moderating roles of blockchain familiarity and blockchain expertise. Thus addressing H1c, H2c, H3c and H4c with the mediation effect of technology anxiety, H5a, H5b, H5c, H5d and H5e with the moderating effect of blockchain expertise and H6a, H6b, H6c, H6d and H6e with the moderating effect of blockchain familiarity.
4. Results
4.1 Direct and moderation effects of conditions on dependent variables
4.1.1 Effect of blockchain usage on perceived trust, transparency, security and convenience (H1a, H2a, H3a and H4a)
For the following one-way ANOVA analyses, preliminary checks were conducted to ensure that there were no violations of the assumptions of normality, homogeneity of variances, independence and all dependent variables being at the interval level. The Levene’s test statistic is mentioned for each of the ANOVAs.
Trust (H1a)
The scale mean of this variable was used in a one-way ANOVA to measure the difference in mean score on trust for the two conditions of blockchain technology usage (independent variable). Levene’s test for homogeneity of variances was insignificant (p=.689), thus equality of variances can be assumed.
As can be seen in table 4 and 5 there was no significant effect of blockchain usage on levels of perceived trust, F(1, 294) = 0.00, p=.96. Thus, there is no significant difference in perceived trust between the two conditions of blockchain technology usage.
Table 5: Descriptive statistics for one-way ANOVA
Transparency (H2a)
To examine whether there is a statistically significant difference between the two conditions with respect to their levels of perceived transparency, a one-way ANOVA was conducted that compared the means of perceived transparency for both conditions. Levene’s test for homogeneity of variances was insignificant (p=.354), thus equality of variances can be assumed.
As can be seen in table 6 and 7 there was no significant effect of blockchain usage on levels of perceived transparency, F(1, 294) = 0.38, p=.54. Thus there is no significant difference in perceived transparency between the two conditions of blockchain technology usage.
Table 6: One-way ANOVA (N=295)
Security (H3a)
A one-way ANOVA was conducted to examine whether there was a difference in mean scores between the two conditions blockchain technology usage on perceived security. The Levene’s test for homogeneity of variances was insignificant (p=.789), hence equality of variances can be assumed.
As can be viewed in table 8 and 9, there was no significant effect of blockchain usage on perceived security, F(1,294) = 0.23, p=.63. Thus, there is no significant difference in perceived security between the two conditions of blockchain technology usage.
Table 8: One-way ANOVA (N=295)
Table 9: Descriptive statistics for one-way ANOVA
Convenience (H4a)
Lastly, a one-way ANOVA was conducted for the relation between blockchain technology usage and perceived convenience. The mean scores for both conditions were compared on their perceived level of convenience. The Levene’s test for homogeneity of variance was insignificant (p=.719), thus equality of variances
As table 10 and 11 show, there was no significant effect of blockchain usage on perceived convenience, F(1,294)= 1.14, p=.29. Thus there is no significant difference in perceived convenience between the two conditions of blockchain technology usage.
Table 10: One-way ANOVA (N=295)
Table 11: Descriptive statistics for one-way ANOVA
4.1.2 Effect of channel on mediator, moderators and convenience
The correlation matrix showed moderate relations between the channel through which the Qualtrics survey was accessed and the variables technology anxiety, blockchain familiarity, blockchain expertise and convenience. Therefore, one-way ANOVAs were conducted to verify whether there is a significant difference in mean scores for the variables mentioned above.
Technology anxiety
The Levene’s test for homogeneity of variances was insignificant (p=.306), hence equality of variances can be assumed. As can be viewed in tables 12 and 13, there was no significant effect of channel used on levels of technology anxiety
