Blockchain-enabled services facilitate electronic voting (e-voting) experiences that are secure, immutable, transparent, and reliable. However, extant research on blockchain-based e-voting services has predominantly focused on its technical aspects, often overlooking the influence of digital literacy factors (namely, cognitive, technical, and socio-emotional dimensions) on adoption behavior. This study investigates the effects of these dimensions of digital literacy on citizens’ adoption intentions, subsequent adoption behavior, and outcome variables, such as satisfaction and referral behavior, while accounting for the mediating role of perceived trust. Data were collected from a sample of 315 U.S. voters with experience using blockchain technology via MTurk. The analysis using SmartPLS revealed that only the technical dimension of digital literacy directly influenced users’ adoption intentions toward blockchain-based e-voting. However, all dimensions of digital literacy exerted indirect effects on adoption intentions through perceived trust. Additionally, adoption behavior positively predicted outcome variables, including satisfaction and referral behavior. These findings advance the theoretical understanding of blockchain-enabled e-voting services by underscoring the critical roles of digital literacy and trust in shaping adoption behavior and its outcomes while also offering practical implications for technology managers and policymakers.
Blockchain is a transformative technology characterized by robust cryptographic features with several societal, environmental, and economic implications. As a result, blockchain technology has attracted significant research and managerial attention (Benchis et al., 2025; Gorkhali & Shrestha, 2020; Hajian et al., 2025; Shukla et al., 2024). Blockchain is employed for global payment, digital currency (Bitcoin), platform-based applications, and smart contracts. Blockchain has been applied in financial services to ensure reliable and secure transactions, sharing services (e.g., Uber and Airbnb), credit information management to facilitate data transactions and processing, accounting, trade management to expedite and secure logistics, and the Internet of Things to enhance decision-making (Benchis et al., 2025; Gorkhali & Shrestha, 2020; Hajian et al., 2025; Hassani et al., 2018; Shukla et al., 2024). Its adoption varies across industries, and its application to practice remains at an early stage of development (Gorkhali & Shrestha, 2020; Saed et al., 2025; Zhu et al., 2022).
Initially, blockchain technology gained widespread attention as a platform for virtual currencies such as Bitcoin. Subsequently, scholars examined its application across various disciplines (e.g., logistics, marketing, and management) as well as its leveraging qualities (e.g., security and anonymity) (Benchis et al., 2025; Hajian et al., 2025). Using the Technology–Organization–Environment framework, Benchis et al. (2025) compared blockchain adoption across public and private sectors, identifying key drivers such as technological readiness and barriers, including regulatory challenges. Their study revealed variations in adoption levels across contexts, offering implications for policy development in digital public services, such as electronic voting (e-voting). E-voting is one of the most prominent applications of blockchain technology in e-governance (Albashrawi et al., 2025; Benchis et al., 2025; Gorkhali & Shrestha, 2020). E-voting systems can utilize the key properties offered by blockchain, i.e., verification of cryptographic transactions and transparency of the distributed ledger. These properties secure privacy and anonymity and provide access to auditable, public, network-based records. Additionally, risks associated with misuse of tokens are minimized, while transparency further increases system security and reliability.
Research suggests that the characteristics and attitudes of end users are critically important when designing and implementing blockchain-based applications, such as e-voting. A recent literature review by Zhu, Bai and Sarkis (2022) found that 24.20 % of the articles analyzed were specifically related to the adoption of blockchain technology, aiming at identifying and understanding the determinants influencing adoption. These factors have been examined across various service contexts, including healthcare, financial services, supply chain management, public services (Benchis et al., 2025), e-voting services (), and remittance services (Abbasi et al., 2025; Albashrawi et al., 2025; Hajian et al., 2025). Drawing on the Unified Theory of Acceptance and Use of Technology (UTAUT; Venkatesh, Thong & Xu, 2012), scholars identified several factors contributing to the adoption of blockchain technologies, including transparency, convenience, social influence, performance expectancy, perceived credibility, cost, and effort expectancy. However, the application of UTAUT and UTAUT2 in blockchain research has yielded inconsistent findings (Chang et al., 2022; Khazaei, 2020; Prados-Castillo et al., 2025), suggesting that a single model may not adequately explain e-voting behavior.
Despite providing some promising results, a deeper understanding of the determinants and user characteristics (i.e., citizens) that facilitate and impede adoption of blockchain-enabled services like e-voting is needed. Most previous research on the use of blockchain has focused on organizational, technological, environmental, and regulatory factors that stimulate or inhibit blockchain adoption (Benchis et al., 2025; Hajian et al., 2025), neglecting to a large degree to investigate how citizens utilize blockchain-based G2C e-voting service platforms to cast their votes. Albashrawi et al. (2025) were the first to study digital literacy as a higher-order mediator of the relationships among UTAUT2 constructs and citizens' intention when using e-voting services. Thus, how user-level factors, such as digital literacy dimensions (e.g., social-emotional, technical, and cognitive abilities), influence adoption decisions in G2C services remains an understudied topic. To overcome this deficiency, this study aims to investigate the impact of digital literacy dimensions (i.e., cognitive, technical, and socio-emotional) on users' adoption intentions, behavior, and outcome variables (i.e., satisfaction and referral behavior) mediated by perceived trust.
Digital literacy has received comparatively limited attention in the business and technology literature. Digital literacy refers to a set of skills associated with competence and confidence in using digital technologies for work, leisure, learning, and communication (Eurostat, 2019; Liang et al., 2025; Ng, 2012). Information technology, information, media, technological, and visual literacies are all considered components of digital literacy. The European Commission (2003) characterizes digital literacy as “fast becoming a prerequisite for creativity, innovation, and entrepreneurship, and without it, citizens can neither participate fully in society nor acquire the skills and knowledge necessary to live in the 21st century” (p. 3). Digital literacy “has been one of those key concepts whose relevancy and weight as a key element for a digital citizenship have shifted from being recommended to essential” (Pérez-Escoda et al., 2019, p. 233). Johnson (2019) argued that digital literacy is a significant concern in the implementation of blockchain-based e-voting, as it may influence citizens’ ability to exercise their democratic rights. Thus, integrating digital literacy perspectives with technology adoption perspectives may facilitate a more comprehensive understanding of blockchain adoption in the e-voting context and help address previously identified inconsistencies in the literature.
Digital literacy encompasses socio-emotional, technical, and cognitive competencies in the use of digital technologies (Ng, 2012). However, research examining the role of these three digital literacy dimensions in blockchain-enabled e-voting adoption remains limited. Furthermore, existing literature examining the outcomes of adopting blockchain-enabled technologies has primarily focused on organizational capabilities, sustainability, information processing, and transaction costs (Zhu et al., 2022) while overlooking the role of these technologies in shaping users’ satisfaction with the service experience and referral behavior. Moreover, the e-voting literature is replete with discussions of the advantages and disadvantages or risks of blockchain-enabled systems (Albashrawi et al., 2025), while the antecedents and outcomes of adopting and using these technologies are overlooked.
While digital literacy reduces functional risks (e.g., difficulty of use), thereby enhancing technology usage, trust helps to mitigate the remaining risks (e.g., security threats) and further increases technology adoption. In the technology acceptance models, digital literacy positively relates to perceived ease of use and trust, as well as perceived usefulness (Kim & Eom, 2025). Research has demonstrated that high digital literacy is related to both low privacy concern and high online trust, and these directly explain users' intentions to use e-government services (Benchis et al., 2025). Technological literacy has been identified as the main driver of trust in e-services (Barisat, 2012). Trust is assumed as essential for any electronic services since it's a basic element in every economic or social activity. Therefore, it is a crucial factor in the success of any electronic services, especially in the blockchain-based e-voting services. Research shows that a blockchain-based tracing system can improve management processes and supply chain transparency and thus increase consumer trust (Wang et al., 2021). For e-voting services, citizens are mainly worried about the privacy of their votes and the complicated automated processes (Gorkhali & Shrestha, 2020). Because citizens may be confused about how to use e-voting systems, trust could be a prerequisite for using a blockchain-based e-voting system.
Although numerous studies have examined the various applications of blockchain (e.g., Abbasi et al., 2025; Albashrawi et al., 2025), a significant gap in research is evident regarding the G2C e-voting domain. Most of the research conducted in the G2C e-voting context focused either on technological or organizational perspectives, and little attention was paid to behavioral outcomes (e.g., user satisfaction and referral behavior). Crucially, the mediating effect of trust and its relation with different dimensions of digital literacy (e.g., cognitive, technical, and socio-emotional) on the adoption of blockchain-based G2C e-voting services are yet to be researched.
Therefore, the present study aims to investigate the antecedents and outcomes of adopting blockchain-enabled G2C e-voting service platforms while considering the mediating role of trust. Specifically, the objectives of this study are as follows: (1) to examine how digital literacy dimensions influence the adoption of blockchain-enabled technologies within the G2C e-voting context; (2) to explore how the adoption of such platforms affects users’ satisfaction and referral behavior; and (3) to investigate the mediating role of trust between digital literacy and the adoption of blockchain-enabled e-voting services. To achieve these objectives, the study integrates digital literacy perspectives (Ng, 2012), trust theory (Singh & Sirdeshmukh, 2000), and technology adoption frameworks (Davis, 1989; Venkatesh et al., 2012) to propose a comprehensive conceptual model.
This study contributes to the literature on governmental citizen services, blockchain-based e-voting, innovation, and services marketing in several ways. First, the study delineates three dimensions of digital literacy—namely, cognitive, technical, and socio-emotional—as key individual-level determinants of blockchain-enabled G2C service adoption, such as e-voting. Second, it identifies trust as the primary mediating mechanism linking citizen capabilities (cognitive, technical, and socio-emotional literacy) to adoption behavior, thereby offering deeper insights into trust formation in government-led digital environments. Third, it extends post-adoption G2C service outcome research by associating adoption behavior with citizen satisfaction and referral behavior—outcomes that have been overlooked in prior e-voting studies. Fourth, the integration of post-adoption outcomes with technology adoption perspectives advances the understanding of how G2C services create value and achieve sustainability. Collectively, these contributions advance theoretical integration by combining digital literacy, trust, and technology adoption perspectives, while also providing practical guidance for innovation managers and policymakers in promoting citizen-centered, trustworthy, and inclusive blockchain-based public services.
This study proceeds by outlining the conceptual framework and hypotheses, followed by the research methodology and empirical results. Finally, it concludes with key findings, implications (theoretical and practical), limitations, and future work.
Conceptual development – an integrative frameworkBlockchain-enabled E-voting as an innovative serviceTraditional G2C vote services require people to rely entirely on electoral authorities for gathering, storing, and tallying votes. E-voting through the use of blockchain has altered the delivery of G2C services by employing a decentralized, distributed ledger rather than a central database to record and store vote details. Citizens are required to put complete trust in a single entity when using a conventional e-voting system, whereas blockchain disperses vote details across all network nodes. Voting data cannot be modified without the consensus of the entire network. Transparency is achieved as any user on the network can verify vote tallies, and voter anonymity means vote identity cannot be determined (Abo-Akleek et al., 2025). These three features, which distinguish blockchain-based e-voting, have benefits over other public service models in terms of immutability, process transparency, and voter anonymity. Blockchain-based e-voting provides four benefits to citizens: accessibility (enabling remote voting and eliminating geographic barriers), inclusivity (multilingual interfaces and accessibility protocols extend democratic rights), verifiability (allowing citizens to confirm that their votes have been accurately counted), and systemic trustworthiness (where decentralized architectures mitigate single-point-of-failure risks) trust (Shaikh et al., 2025). A pilot voting conducted in West Virginia that employed blockchain technology for remote mobile voting showed 99.9 % accuracy without any fraud incidences (Solaiman, 2018), whereas in the case of Oman using Ethereum-based blockchain voting, the citizens gave a 100 % verification confidence rate, and the confidence of citizens in the system rose by 40 % when they were given a chance to verify their own votes (Shaikh et al., 2025). In view of all this, blockchain-based e-voting constitutes an advancement of public services with specific applications and thus an innovative measure to combat fraud, increase confidence of voters, overcome lack of trust and engagement, single points of failure, citizen apathy, and central vulnerabilities in the voting system (Benchis et al., 2025; Shaikh et al., 2025).
Theoretically, the two paradigms of consumers' reaction to innovations are resistance and adoption. The resistance to innovation paradigm explains user characteristics, negative emotions, technology perceptions, and firm stimuli as factors leading to resistance (Tsiotsou et al., 2023). Instead, the innovation diffusion paradigm considers the positive side of adoption and includes well-established models like the Theory of Reasoned Action, the Technology Acceptance Model (Davis, 1989), and UTAUT (Venkatesh et al., 2012). A recent study has integrated the Technology Acceptance Model with trust theory and the experience paradox in explaining blockchain adoption for digital natives by defining trust, perceived utility, and perceived risk as determinants of blockchain adoption (Prados-Castillo et al., 2025). Among widely accepted technology adoption models for explaining blockchain adoption, UTAUT was frequently used in blockchain-enabled technology research (Zhu et al., 2022). However, research on organizations using UTAUT was unable to provide strong evidence on the acceptance of blockchain technology in SMEs (Khazaei, 2020) and has questioned its applicability in e-government services (Dwivedi et al., 2017).
There is an inconsistency in e-voting adoption research across contexts (Chalabi et al., 2025; Zhu et al., 2022). These inconsistencies might result from unique theoretical issues of the blockchain e-voting services, like cognitive complexity (citizens must understand decentralized consensus and cryptographic verification) (Gorkhali & Shrestha, 2020), duality of trust requirement (trust in the technological architecture and trust in democratic institutions) (Wang et al., 2021); and digital capabilities obstacles (technical, cognitive, and socio-emotional skills influence their ability to use the system) (Ng, 2012; Johnson, 2019). These problems may suggest generic technology adoption models could lack context-specific individual-level antecedents. Blockchain voting requires certain literacy components different from common performance/effort expectancy. Therefore, this study adopts individual-level constructs like digital literacy and trust (Wang et al., 2021) along with widely established outcomes (adoption intention and adoption) as dependent variables to explain the use of blockchain technology in e-voting and its consequences. This research contributes to the e-voting service adoption theory by extending it to propose that digital literacy (cognitive, technical, and socio-emotional) is one of the direct antecedents mediated by trust and by providing an explanation for the citizens' adoption behavior.
Digital literacy and its dimensionsLiteracy is a fundamental human requirement for realizing democratic potential by encouraging participation and fostering a robust civic culture. As political and civic activities shift online, a second-level digital divide—the disparity in user capabilities—undermines the democratic potential of the internet. In line with this reasoning, the advancement of digital literacy reinforces intercultural understanding and knowledge, participatory civic societies, long-term international peace, freedom, democracy, and good governance (Julien, 2019). Jaeger et al. (2012) conceptualize digital literacy as “encompassing the skills and abilities necessary for access once the technology is available, including a necessary understanding of the language and component hardware and software required to successfully navigate the technology” (p. 3). Thus, Jaeger et al. (2012) adopt a broader perspective by emphasizing that digital literacy includes both the capacity to access and effectively utilize technology. Digital literacy has also been defined as “critical and confident use of IS (information systems), including an ability to participate in social networking applications and in collaborative environments, an awareness of security threats and risks, and the ability to use IS for creative and innovative purposes, irrespective of the context” (Ng, 2012, p. 1067). Subsequently, digital literacy was defined as “a set of knowledge, attitudes, and skills required to gain digital information in an effective, ethical, and efficient manner” (Julien, 2019). UNESCO (2018) defines digital literacy as “the ability to access, manage, understand, integrate, communicate, evaluate, and create information safely and appropriately through digital technologies for employment, decent jobs, and entrepreneurship. It includes competencies such as computer literacy, ICT literacy, information literacy, and media literacy.”
Several theoretical approaches to digital literacy have been proposed over time to facilitate the analysis of different digital competencies (Table 1). These approaches have primarily concentrated on delineating the necessary expertise or competencies. For example, the “skills-based theoretical framework” proposes five key dimensions—photo-visual, branching, information, reproduction, and socio-emotional—which were later extended to include real-time thinking (Eshet-Alkalai, 2012). According to Bawden (2008), digital literacy encompasses skills in online searching, hypertext navigation, knowledge assembling, and content evaluation. Focusing on Web 2.0 and ICTs, Area and Pessoa (2014) identified five dimensions of digital literacy: instrumental, cognitive-intellectual, sociocultural, axiological, and emotional. Recently, Martínez-Bravo, Chalezquer and Serrano-Puche (2022) used content analysis of international reference frameworks to identify six dimensions of digital literacy: cognitive, emotional, critical, operative, social, and projective. Nevertheless, the literature lacks a comprehensive understanding of the digital literacy dimensions as a universal construct, largely due to context-specific research (e.g., in education, Ng, 2012; in Web 2.0 and ICTs, Mohammadyari & Singh, 2015). In addition, Ng’s (2012) tri-dimensional model is adopted in this study due to its balanced integration of cognitive (e.g., ethical evaluation of blockchain), socio-emotional (e.g., safe social interactions in decentralized systems), and technical (e.g., operational proficiency) dimensions, making it particularly suitable for analyzing second-level digital divides in service contexts such as e-voting. Compared to more expansive models (Martínez-Bravo et al., 2022), this framework avoids redundancy while enabling clear links to trust and adaptability to new literacies—both of which are central to our manuscript (please see Table 1 for a comparison).
A comparison between digital literacy frameworks.
| Framework | Dimensions/Areas | Strengths | Limitations | Suitability for Our Study |
|---|---|---|---|---|
| Eshet-Alkalai (2012) | Photo-visual, branching, information, reproduction, socio-emotional (later real-time thinking) | Skills-focused, adaptable to multimedia contexts | Narrowly skill-based, lacks broader socio-ethical integration | Useful for technical competencies but insufficient for trust analysis |
| Bawden (2008) | Online searching, hypertext navigation, knowledge assembling, content evaluation | Practical for information handling | Limited to Web 1.0-era skills, omits social/emotional aspects | Relevant baseline but outdated for blockchain interactions |
| Area and Pessoa (2012) | Instrumental, cognitive-intellectual, sociocultural, axiological, emotional | Comprehensive Web 2.0/ICT coverage | Overly expansive (five sets), context-specific to education | Strong socio-cultural fit but less parsimonious for e-voting focus |
| Martínez-Bravo et al. (2022) | Cognitive, emotional, critical, operative, social, projective | Derived from global frameworks via content analysis | Six dimensions risk dilution in empirical application | Broadly applicable but complex for user-level digital gap analysis |
| Ng (2012) Tri-Dimensional | Cognitive (critical thinking/ethics), socio-emotional (communication/safety), technical (ICT operations/troubleshooting) | Parsimonious (three dimensions), progressive, balances skills with relational trust | Requires contextual adaptation for emerging tech like blockchain | Optimal: Directly addresses user-level gaps in blockchain e-voting, linking literacy to trust, security perceptions, and adoption (e.g., Gorkhali & Shrestha, 2020) |
Therefore, the tri-dimensional approach (cognitive, socio-emotional, and technical dimensions) proposed by Ng (2012) represents a parsimonious model of digital literacy. The cognitive component pertains to critical thinking in the identification, evaluation, and generation of digital information. It encompasses the capacity to evaluate and select appropriate technologies, as well as a comprehensive understanding of the ethical and legal dilemmas associated with their use. The socio-emotional dimension of digital literacy encompasses the responsible use of the internet for communication, social interaction, and education, while adhering to appropriate language norms, ensuring personal safety, and identifying potential risks. The technical dimension encompasses the requisite technical and operational abilities for utilizing ICT, including the ability to connect and operate input and peripheral devices, an understanding of functional components, and proficiency in troubleshooting. However, identifying the features driving the digital divide at the user level is essential, particularly in relation to variations in usage across different applications, such as blockchain-based e-voting systems.
Previous research has linked computer literacy to voters’ difficulties in installing and using the e-voting system, as well as to their perceived trust in its security, vulnerability, and usage intentions (Gorkhali & Shrestha, 2020). Individual differences in cognitive capacity may account for variations in digital competencies or dimensions. Digital literacy is cumulative in nature, as it builds upon previously acquired expertise and information. In theory, digitally literate individuals are better positioned to adapt to emerging technologies and acquire new symbolic languages of communication. Higher levels of digital literacy facilitate smoother transitions to evolving forms of “new literacies” (Ng, 2012). However, differing levels of competence across these dimensions may either facilitate or hinder the use and adoption of certain technologies, such as blockchain-enabled e-voting.
The role of trust in blockchain-based E-votingPerceived trust is defined as “the psychological state leading to accepting the vulnerability of a trustor, based on positive expectations of the trustee’s actions” (Singh & Sirdeshmukh, 2000, p. 154). Trust can be broadly defined as being a general tendency (dispositional), being a result of experience (learned), or being dependent on the performance context (performance-based). Perceived trust in a technology means the citizen is willing to accept a vulnerability based on a positive expectation from a technology (e.g., a security, performance, or benevolent expectation), which has been shown to be multidimensional (competence, predictability, and integrity) (Schuetz et al., 2025). There are differences between interpersonal (human to human), institutional (human to organization), and technological (human to automated systems) trust, wherein technological trust relates to reliance on a non-human agent under conditions of uncertainty. Factors that contribute to technology trust have been shown to be technological (e.g., usability, reliability, and feedback mechanisms), user-specific (e.g., prior experience and self-confidence), and contextual factors (e.g., task demands) (Schuetz et al., 2025).
There are three different forms of trust in e-voting: institutional, technological, and process trust, with each form targeting different potential failure points (Prados-Castillo et al., 2025; Shaikh et al., 2025). Institutional trust focuses on the trust citizens put in electoral authorities (government, officers, etc.) to have adequate capacity and integrity to conduct and count votes (Shaikh et al., 2025). Technology trust, on the other hand, refers to citizen trust in the technology components (cryptography, smart contracts, digital wallets, etc.) to fulfill their programmed task properly (Oliveira et al., 2017), while process trust concerns the reliance on the procedural integrity (consensus protocol, tabulation, auditing procedure, etc.) of the electoral system to ensure vote accuracy and integrity (Shaikh et al., 2025).
This study operationalizes technology trust (Oliveira et al., 2017) for multiple reasons. The whole trust landscape of e-voting is redefined by blockchain as it replaces the human-centered, institutionalized electoral authorities with decentralized cryptographic mechanisms. Unlike conventional e-voting, which expects citizens to trust one electoral authority, in a blockchain system votes are distributed among multiple nodes in the network, and only cryptographic mechanisms (zero-knowledge proofs, immutability, etc.) provide assurance of integrity without any human intervention. In this context, instead of the institutional domain, the technology domain appears to be the source of trust, making the institutional domain relatively irrelevant, as there is minimal reliance on the central institution and the domain of process trust more closely relates to system administrators or auditors, rather than the end-users of the system. Trust in the technology—as citizens trust in cryptographic mechanisms, smart contracts, and decentralized structures to ensure accuracy, security, and transparency of vote recording—is thus the most relevant and theoretically sound construct for citizen adoption of a blockchain-based G2C e-voting service.
Immutability serves to ensure verifiability and trust in blockchain-based e-voting, which addresses and rectifies weaknesses of conventional e-voting systems like vote rigging or coercion via cryptographic mechanisms such as zero-knowledge proofs. Transparency and auditability are discussed in the literature to be key drivers of trust, whereas elements like scalability concerns and complexity of user interface can become deterrent factors that influence end results, like voter turnout. Literature reviews indicate that trust in technology is one of the most important determinants of system adoption, as it can foster perceived integrity, whereas field studies reveal that system interoperability remains a problem (Schuetz et al., 2025).
Hypothesis developmentDigital literacy comprises three dimensions (e.g., technical, socio-emotional, and cognitive), which are specified and examined as antecedents in the proposed conceptual model (Fig. 1). The technical dimension refers to the possession and application of technical and operational skills (Ng, 2012) in the use of specific technologies and services. Technological considerations influencing blockchain adoption include factors such as technological maturity, complexity, security, and scalability (Kouhizadeh et al., 2021). Therefore, individuals with higher levels of digital literacy are expected to possess the necessary skills to effectively use various technologies. Furthermore, individuals with higher levels of technical digital literacy are more likely to develop intentions to use and implement blockchain-based e-voting services. Accordingly, the following hypothesis is proposed:
H1: Technical digital literacy positively predicts the adoption of blockchain-enabled e-voting services.
The socio-emotional dimension of digital literacy relates to the possession of “netiquette” and the ability to understand emotional cues (Ng, 2012), as well as to receive emotional support from social networks, including friends, family, tutorial videos, and the internet. The socio-emotional dimension may be associated with users’ expectations regarding others’ reactions to their adoption of specific services or technologies, such as online banking, tourism services (Huang, Tsiotsou & Liu, 2023), and blockchain technology used in supply chain coordination (Zhu & Kouhizadeh, 2019), management, and data sharing (Ertz, Maravilla & Cao, 2025). Accordingly, a positive relationship is expected between the accumulation of favorable viewpoints within users’ immediate social networks and their intentions to adopt blockchain-enabled e-voting systems. Thus, the following hypothesis is proposed:
H2: Socio-emotional digital literacy positively predicts the intention to adopt blockchain-enabled e-voting services.
The cognitive dimension refers to the capacity for critical thinking, assessing, and creating digital information, requiring individuals to be proficient in using digital resources and interpreting information (Huang, Tsiotsou & Liu, 2023; Ng, 2012). Cognitive structures have the potential to influence the comprehension and integration of knowledge, which are critical for the adoption of new technologies and services. This is because such structures facilitate meaning construction, activation of existing knowledge, and the ability to condense and integrate information (Neumeyer et al., 2020). Consequently, individuals with higher levels of cognitive competencies are more likely to develop intentions to use blockchain-based e-voting services. Thus, we hypothesize that:
H3: Cognitive digital literacy is positively associated with the intention to adopt blockchain-enabled e-voting services.
According to Kamarulzaman et al. (2021), technological considerations and the establishment of trust in the implementation of blockchain technology are related within government organizations. The technical dimension encompasses knowledge and understanding of technological concepts and tools. Given the complexity and specialized nature of blockchain technology, comprehension of the underlying processes of blockchain-based e-voting systems is more likely among individuals with higher levels of technical literacy. Such comprehension can enhance individuals’ understanding of the security, transparency, and immutability features of blockchain (Gorkhali & Shrestha, 2020), thereby strengthening their trust in the technology (Wang et al., 2021). Accordingly, the following hypothesis is proposed:
H4: Technical digital literacy positively predicts citizens’ perceived trust in blockchain e-voting.
Socio-emotional digital literacy promotes positive online interactions and fosters participant trust (Ng, 2012). Research using the e-Qual 4.0 scale, which is primarily used to differentiate interactive users from non-interactive users based on trust dimensions and empathy factors in e-government, has found that respect-based relationships and empathy facilitate the construction of trust (Fawcett et al., 2017). Exhibition of proper conduct, respect, and empathy in online discourse and interactions fosters a perception of reliability and security in the online environment. In blockchain-based e-voting, those who actively participate in courteous and constructive online discourse are inclined to place more faith in the integrity and impartiality of the electoral process. A healthy online environment fosters the facilitation of open and meaningful discussions and constructive participation, eventually contributing to the development of a sense of trust. This notion leads to the following hypothesis:
H5: Socio-emotional digital literacy positively predicts citizens’ perceived trust in blockchain e-voting.
The cognitive dimension of digital literacy includes the acquisition of knowledge and the comprehension of digital resources. Research indicates a correlation between cognitive biases and individuals' inclinations toward e-voting systems, specifically concerning trust, as trust may be associated with the cognitive state of e-voting electors. Qian and Papadonikolak (2021) confirmed a positive association between cognition-based factors and trust in building supply chains facilitated by blockchain technology. Thus, in blockchain-based e-voting, individuals with advanced cognitive digital literacy are more inclined to comprehend blockchain technology’s underlying principles and mechanics. This comprehension empowers individuals to grasp the advantages of blockchain technology regarding its enhanced security measures, increased transparency, and immutable nature. Through an in-depth understanding of blockchain technology’s technical intricacies and fundamental principles, users may cultivate a more profound understanding of e-voting processes, fostering heightened levels of perceived trust. Thus, the following hypothesis is generated:
H6: Cognitive digital literacy positively predicts citizens’ perceived trust in blockchain e-voting.
Blockchain technology enables the e-voting process to be “detectable and traceable on the ledger” and reduces the risk of data fraud and manipulation (Kouhizadeh et al., 2021). Therefore, blockchain technology may enhance trust among users, which, in turn, can facilitate the adoption of blockchain-enabled e-voting services (Manrai et al., 2021). Using UTAUT, Chang et al. (2022) demonstrated the impact of trust transparency on performance, effort expectation, and social influence, but they did not validate trust’s influence on behavioral intention. Zhu et al. (2021) exhibited that trust can significantly determine users’ intentions to embrace e-voting services in developing countries. The following hypothesis was thus proposed:
H7: Perceived trust positively predicts citizens’ intentions to embrace blockchain e-voting services.
According to Fishbein and Ajzen (1975), additional evidence is required to support the notion that a comprehensive assessment of attitudes toward an object will consistently and accurately predict a given behavior. In conjunction with perceived control and standards, attitudes significantly predict our intentions. Attitudes constitute a central component of the planned behavior theory in predicting consumer behavior. Additionally, Fishbein and Ajzen’s (1975) persuasion theory elucidates the interconnectedness between beliefs, attitudes, intentions, norms, and actions, positing that an individual’s behavior is contingent upon their desire to engage in such activity. In line with this reasoning, research supports an association between positive beliefs and the user adoption of e-voting websites in Jordan (Alomari, 2016). Therefore, the following hypothesis was derived:
H8: Citizens’ intention to cast a vote using a blockchain-enabled e-voting service is positively associated with adopting blockchain-enabled e-voting.
Recent findings support the notion that the use of blockchain technology across several sectors, such as electronic agriculture, digital marketing, and the sharing economy, might enhance customer well-being and happiness (Rabby et al., 2022). Therefore, people who have embraced blockchain e-voting are more inclined to express satisfaction with the e-voting services. Therefore, the following hypothesis can be generated:
H9: Citizens’ adoption positively predicts satisfaction with the e-voting service.
Individuals exhibit skepticism toward novel technologies or services. However, using referral credentials, such as ratings or direct referrals from acquaintances, can enhance their inclination toward adopting blockchain technology (Robb et al., 2021). According to Choi (2018), blockchain technology can enhance users’ loyalty and foster engagement and referral behavior. The study by Mazambani and Mutambara (2019) explored customer referral behavior concerning the adoption of cryptocurrencies. Based on their findings, we posit that individuals who use the blockchain e-voting service are inclined to recommend it to their acquaintances, such as relatives or family members. The following hypothesis can thus be generated:
H10: Citizens’ adoption positively predicts referral behavior to adopt a blockchain-enabled e-voting service.
Since trust can significantly influence users’ adoption in new and uncertain situations (Koroma et al., 2022), we propose that perceived trust mediates the relationships between the three dimensions of digital literacy and citizens’ intention to embrace blockchain-enabled e-voting services. When trust exists between users and the blockchain e-voting technology, the users are more likely to adopt this service. For example, trust in cryptocurrency has been shown to mediate the relationships between technology attachment and users’ behaviors in the adoption of blockchain cryptocurrency (Koroma et al., 2022). Wang et al. (2021) report that while using blockchain technology, consumer trust acts as a mediator between supply chain transparency and sales and product returns. Furthermore, trust mediates the relationships between security and attitude in adopting a blockchain-based system (Shrestha et al., 2021). The following hypotheses can thus be generated:
H11: Perceived trust serves as a mediator between technical digital literacy and citizens’ intention to use the blockchain e-voting service.
H12: Perceived trust serves as a mediator between socio-emotional digital literacy and citizens’ intention to use the blockchain e-voting service.
H13: Perceived trust serves as a mediator between cognitive digital literacy and citizens’ intention to use blockchain e-voting services.
MethodThe blockchain-enabled E-voting as the study contextWith the rapid advancement of the internet and information communication technologies, several traditional offline services (e.g., voting, mail, and payments) have transitioned to online platforms. E-voting refers to the process of casting and counting votes electronically, which is used by electors and election officials (Abo-Akleek et al., 2025; Shaikh et al., 2025). E-voting allows electors to exercise their voting rights electronically from anywhere in the world while election officials collect votes. E-voting saves time and effort and is inclusive, while offering excellent efficiency and flexibility, gaining popularity as an alternative to traditional voting. In the 2014 State elections in Victoria, Australia, for example, the vVote verifiable voting technology allowed blind, partially sighted, and mobility-disabled voters, voters speaking languages other than English, and voters in distant areas to cast completely secret votes in a verifiable way (Burton, Culnane & Schneider, 2016). Such initiatives demonstrate how technologically mediated voting systems can address accessibility and participation challenges inherent in traditional electoral processes.
E-voting is an efficient, inclusive, and cost-effective method of executing a voting procedure using real-time data, requiring high security. Security is a significant issue in e-voting to ensure the secrecy of the ballot and the integrity of the election procedure. Therefore, several e-voting systems have been developed and improved over time. For example, vVote, Scantegrity, and choreographed distributed electronic voting using quantum proxy signatures have been developed to construct secret electronic voting, protecting the privacy and anonymity of all electors and guaranteeing unconditional security (Burton, Culnane & Schneider, 2016). Currently, end-to-end solutions such as vVote and end-to-end verifiability (ESIV) are gaining attention. ESIV is a computer security solution specifically built for elections that allows highly reliable detection of vote loss, damage, or fraud (Abo-Akleek et al., 2025; Shaikh et al., 2025).
Moreover, blockchain technologies have been introduced to e-voting technologies to reduce voter fraud and increase voter access to elections. Eligible voters can use a computer or smartphone to vote anonymously. E-voting systems powered by blockchain technologies use an encrypted key and tamper-proof personal IDs (Prados-Castillo et al., 2025; Shaikh et al., 2025). These technologies provide robust protection for voter and server-side security, guaranteeing verifiability, fairness, and immunity to bribery, coercion, and collusion among election officials. At the same time, blockchain-enabled e-voting upholds election integrity by maintaining eligibility, anonymity, privacy, and trust (Gorkhali & Shrestha, 2020).
Given the growing interest in blockchain-based electoral technologies, blockchain-based e-voting demonstrates its significance by enabling large-scale, secure, and accessible elections across multiple countries and high-threat environments. As evidenced by Voatz’s successful delivery of 140 elections and service to >5 million voters, blockchain infrastructure enhances transparency, auditability, and trust while maintaining the integrity of the election (Voatz, 2024). By combining blockchain with identity verification and biometrics, such systems expand voter access through remote, mobile participation without compromising security. This marks a critical step toward rebuilding public confidence and modernizing democratic processes globally.
The study focuses on U.S.based voters for several substantive reasons. The U.S. has one of the most documented case histories of pilots involving blockchain-enabled e-voting, from nationally publicized initiatives such as West Virginia's Voatz-powered mobile voting (Solaiman, 2018) to many smaller, local trials, making this an empirically rich and relevant population for an examination of citizen adoption behavior. Moreover, the U.S. has reasonable homogeneity in terms of institutional, regulatory, and technological structures of the electoral process as compared to voters from different countries with widely different electoral processes and levels of infrastructure, making this a theoretically acceptable single-country analysis, rather than an indiscriminate cross-national sample.
Data collectionAn online survey method was employed for data collection purposes during 2024. MTurk was chosen for data collection owing to its established validity for technology adoption and behavioral intention studies, cost-effectiveness, scalability, quality control (>95 % and attentiveness check), and access to American participants (Aguinis et al., 2021). Despite the accessibility advantages associated with using MTurk as a data collection platform and worker diversity, some limitations include non-attentive respondents, potential bot behavior, and demographic predominance of technologically fluent populations. To prevent the possibility of inaccurate responses caused by non-attentive participants, attentiveness check questions were embedded within the survey, and any respondents who failed these were excluded from the final analysis.
Blockchain e-voting is still relatively novel and has been introduced in only a handful of elections, such as in the 2024 Mexican presidential election, the 2022 Ontario municipal election, the 2019 Denver municipal election, West Virginia mid-term elections, U.S. mid-term elections, and certain U.S. general elections (e.g., Presidential 2020 elections and Utah 2020 elections) (Voatz, 2024), and yet large-scale populations for studying usage are difficult to identify and access. Thus, data were predominantly gathered from U.S. individuals that have either voted in a U.S. election or are currently residing in the U.S. and are likely to have used blockchain e-voting in their home countries due to the efficiency that the technology provides in this context. In terms of context, the United States is suitable for studying adoption of blockchain e-voting. Not only is the nation equipped with superior digital technology (95 % broadband penetration), but it has been a frontrunner in electronic voting innovations such as the blockchain mobile voting pilot launched by West Virginia to assist overseas military voters in the 2018 elections using Voatz technology. This experience, combined with existing federal discourse (e.g., UOCAVA reforms), renders the voting U.S. population as the most ideal candidate to explore individual adoption of blockchain e-voting compared to less developed contexts.
Prior to data collection via M-Turk, a power analysis was carried out in G*Power (Faul et al., 2007) to calculate the minimum sample size required for testing the proposed model (Fig. 1). The analysis revealed that 129 respondents are required to produce a power of 0.95 for a medium effect size (0.15), alpha = 0.05, and four predictors, which are the variables included for the PLS-SEM analysis. A purposive sampling method was used to determine the relevant population experience and background knowledge relevant to blockchain e-voting and e-voting technology to be utilized in this study. Although this approach cannot result in a statistically representative sample of the U.S. voter population, it is a justified and methodologically sound approach that targets individuals who have relevant personal experience of the phenomenon in question, which is blockchain e-voting, a rather specific and developing concept that is likely to be encountered only by a limited subset of the population at this stage. The sampling methodology applied for this research, thus, falls under a purposive, criterion-based approach and not a probability or population-representative-based approach. To this effect, participants must: 1) reside in the U.S. and have cast a vote using a blockchain e-voting system to ensure experience with the technology and to provide contextual validity; 2) have voted at least once in a blockchain e-voting system to serve as a criterion screening for user experience with the technology; and 3) fall within generations X, Y, or Z that have been more exposed to e-voting technologies and applications. The following standard description was used to brief all participants prior to the survey to achieve conceptual alignment: “Blockchain-based e-voting is defined as a secure online voting system using blockchain technology. Its main features and benefits are the following: (1) Vote immutability: vote contents can no longer be altered after they have been added to the blockchain, (2) Transparency: vote totals are publicly auditable without revealing how each individual voted, (3) Voter anonymity: voter identification and ballot are kept separately using cryptographic principles, and (4) Decentralization: the blockchain system is not controlled by a single entity.” Overall, 315 valid responses were obtained through M-Turk, and the descriptive information of the participants is provided in Table 2 below.
Respondents’ profile.
The data for the study were obtained from a single source, making it crucial to accurately identify common method bias (CMB). To address CMB from single-source data, we conducted Kock’s (2017) full collinearity test using SmartPLS 4.0, which is used to calculate full collinearity VIFs for all latent variables by regressing each against all global model predictors (Kock, 2017; Hair et al., 2022). All full collinearity VIFs ranged from 1.42 (socio-emotional literacy) to 2.87 (adoption intention), well below the conservative threshold of 3.3, confirming no CMB or pathological collinearity.
Measurement modelThe items used to examine the dimensions of digital literacy (e.g., technical, socio-emotional, and cognitive) were adapted from a study by Ng (2012) and modified in the context of blockchain e-voting. The items covering the aspect of perceived trust from the study of Oliveira et al. (2017), adoption intention from a study by Venkatesh et al. (2012), adoption behavior from a study by Ramírez-Correa et al. (2019), satisfaction with the e-voting service from a study by Lee et al. (2002), and referral behavior from a study by Abbasi et al. (2022) were taken and modified from the current study’s perspective (Appendix). It is important to note that, given the purposive sampling criterion requiring all participants to have prior experience with blockchain-based e-voting, the study operates within a post-adoption framework. Accordingly, adoption intention captures citizens’ continued use intentions—that is, their forward-looking willingness to use the system in future elections—while adoption behavior reflects self-reported actual use during their prior voting experience. This sequencing is consistent with post-adoption models in which continued use intention, shaped by accumulated experience and trust, serves as a psychological antecedent to sustained behavioral engagement (Venkatesh et al., 2012; Ramírez-Correa et al., 2019).
Partial least squares structural equation modeling (PLS-SEM) was employed to test the proposed research model using SmartPLS 4.1.1.7. PLS-SEM was chosen over covariance-based SEM (CB-SEM) due to its suitability for the study’s objectives (prediction-oriented research), data characteristics, and flexibility in emerging research contexts (Hair et al., 2022). The study aimed to examine novel and underexplored relationships and mediation effects using an exploratory and predictive approach rather than strict theory confirmation (Aw et al., 2023). Additionally, two constructs (e.g., satisfaction and referral behavior) were measured using single-item indicators. While CB-SEM typically requires multiple indicators per latent construct for model identification and reliable estimation, PLS-SEM can accommodate single-item constructs when they are conceptually concrete and well-defined (Hair et al., 2022). Furthermore, PLS-SEM imposes fewer distributional assumptions and is appropriate for survey data that may deviate from multivariate normality, while emphasizing explained variance and path coefficient estimation. Accordingly, the analysis followed a two-stage approach, which comprises the assessment of the measurement model and the evaluation of the structural model.
ResultsEstimating the measurement model (Reflective models)The study model, illustrated in Fig. 1, comprised exclusively reflective constructs. According to Hair Jr et al. (2022), the reliability and validity of reflective constructs should be assessed using previously established reliability indicators (e.g., Cronbach's alpha, composite reliability, and Dillon-Goldstein's rho values should be between 0.60 and 0.70). Convergent validity, measured through the average variance extracted (AVE), should exceed 0.50. As shown in Table 3, all reflective constructs in this study met or surpassed the recommended threshold values for both reliability and validity.
Measurement Model.
| Study’s Constructs, Items | loadings |
|---|---|
| Technical Dimension of Digital Literacy adapted Ng (2012) from α=0.76; rho_A = 0.77; CR=0.85; AVE=0.59 | |
| DL.TD1: I know how to vote using a blockchain e-voting service.DL.TD3: I keep up with important technology/updated features of blockchain e-voting service.DL.TD4: I have the technical skills to use blockchain e-voting service to cast a vote.DL.TD5: I have good blockchain-enabled e-voting service skills to cast a vote. | 0.84 |
| 0.76 | |
| 0.71 | |
| 0.75 | |
| Social-emotional Dimension of Digital Literacy adapted from Ng (2012) α=0.63; rho_A = 0.64; CR=0.84; AVE=0.73 | |
| DL.SE1: Regarding the blockchain e-voting service, I can get help from friends, family, tutorial videos, and the internet.DL.SE2: Blockchain e-voting service enables me to cast a vote without hassle. | 0.88 |
| 0.83 | |
| Cognitive Dimension of Digital Literacy adapted Ng (2012) from α=0.69; rho_A = 0.70; CR=0.87; AVE=0.77 | |
| DL.CD1: I am confident I can vote with my blockchain e-voting service skills.DL.CD2: I am familiar with issues (technology complexity or failure and internet connection) related to using blockchain e-voting services. | 0.89 |
| 0.86 | |
| Perceived Trust adapted from Oliveira et al. (2017) α=0.81; rho_A = 0.81; CR=0.88; AVE=0.64 | |
| PTrus1: I think I can trust using blockchain e-voting service to cast a vote.PTrus2: Blockchain e-voting service can be trusted to cast a vote faithfully.PTrus3: In my opinion, casting a vote through a blockchain e-voting service is trustworthy.PTrus4: I value the reliability of the blockchain e-voting service for casting my vote. | 0.80 |
| 0.77 | |
| 0.78 | |
| 0.84 | |
| Adoption Intention adapted from Venkatesh et al. (2012) α=0.79; rho_A = 0.79; CR=0.88; AVE=0.71 | |
| AI1: I intend to continue using blockchain e-voting services to cast a vote in the near future. | 0.87 |
| AI2: I will always try to use blockchain e-voting services to cast my vote. | 0.78 |
| AI3: I plan to continue using blockchain e-voting services when casting a vote. | 0.87 |
| Adoption Behavior adapted from Ramírez-Correa et al. (2019) α=0.77; rho_A = 0.78; CR=0.87; AVE=0.69 | |
| AdoptB1: When casting a vote, I tend to use a blockchain e-voting service to cast a vote.AdoptB2: On average, I spend a lot of time using blockchain e-voting services.AdoptB3: I would take advantage of using a blockchain e-voting service to cast a vote. | 0.86 |
| 0.76 | |
| 0.86 | |
| Consumer Satisfaction adapted from Lee et al. (2002) (*single item scale includes the value of 1) | |
| How do you generally feel about using a blockchain e-voting service to cast a vote? Please indicate your satisfaction level with the e-voting service. | 1 |
| Referral Behavior adapted from Abbasi et al. (2022) (*single item scale includes the value of 1) | |
| Referral: I recommend blockchain e-voting services to my friends and relatives. | 1 |
In addition, discriminant validity using the heterotrait-monotrait (HTMT) ratio was also assessed. HTMT ratios of 0.90 or lower are sufficient for establishing the absence of discriminant validity concerns. The HTMT ratios confirmed that the study model is free of any discriminant validity concern, as shown in Table 4.
Discriminant validity (HTMT) analysis.
Note: DL>CD-Cognitive Dimension of Digital Literary; DL>SE-Social emotional dimension of digital literacy; DL>TD-Technical dimension of digital literacy.
After estimating the reflective measurement model, we moved forward to estimate the structural model for hypothesis testing, as shown in Fig. 1. Using SmartPLS 3.3.9, we tested the proposed hypotheses using the bootstrap distributions of 10,000 subsamples and discussed the findings in Table 4. The findings revealed that the technical dimension of digital literacy, with a t-value of 3.070, a p-value of 0.000, and an effect size of 0.060, is an important antecedent for the adoption intention of blockchain e-voting services. Thus, H1 is supported. However, the socio-emotional and cognitive dimensions of digital literacy fail to influence voters’ intention to adopt blockchain e-voting directly. Hence, H2 and H3 are not supported. By contrast, the digital literacy dimensions comprising technical, socio-emotional, and cognitive positively impact voters’ perceived trust in blockchain e-voting and indirectly affect voters’ adoption intention of blockchain e-voting; therefore, H4, H5, H6, and H7 are supported. Adoption intention, with a t-value of 13.730, p-value of 0.000, and an effect size of 0.800, also positively explains voters’ adoption behavior of blockchain e-voting, which subsequently plays a positive role in predicting voters’ satisfaction with the e-voting service, with a t-value of 11.100, a p-value of 0.000, an effect size of 0.510, and referral behavior with a t-value of 6.700, a p-value of 0.000, an dan effect size of 0.190. These findings confirm hypotheses H8, H9, and H10. Furthermore, R² and Q2 were used to assess our model's explanatory and predictive (in-sample) power. According to Table 5, the R² and Q2 values for perceived trust are 0.540 and 0.340, respectively; for adoption intention, they are 0.550 and 0.380; for adoption behavior, they are 0.440 and 0.300; for satisfaction with the e-voting service, they are 0.340 and 0.330; and for referral behavior, they are 0.160 and 0.150. According to the findings, the study model has adequate explanatory and predictive power.
Structural Model (Direct Effects/Hypotheses).
Note: AdoptB-Adoption Behavior; AdoptI-Adoption Intention.
Next, we estimated the mediating mechanism of voters’ perceived trust in blockchain e-voting between digital literacy dimensions (e.g., technical, socio-emotional, and cognitive) and adoption intention. To test the mediating hypotheses, we examined the indirect effect alongside a bias-corrected confidence interval to accept or reject the hypothesis. SmartPLS 3.39 was employed to test the mediating hypotheses, and the results presented in Table 6 demonstrated that all mediating hypotheses, H11, H12, and H13, are significant and supported (Fig. 2).
Mediation results.
Using PLSpredict, we estimated the items of endogenous constructs (e.g., perceived trust, adoption intention, adoption behavior, satisfaction, and referral behavior). We compared the PLSRMSE value to the standard (i.e., LMRMSE). The predictive strength is considered strong if the items have exclusively negative values in the outcome variables; medium if 50 % or more of the items have negative values; and low if >25 % or at least one of the items has negative values. Table 7 demonstrates that adoption intention has a high prediction, perceived trust has a medium prediction, whereas adoption behavior has a low prediction. No prediction was recorded for satisfaction and referral behavior, but all outcome variables attained predictive power because Q² values exceeded zero.
PLSpredict.
The study findings indicate that the technical dimension of digital literacy is positively associated with the adoption of blockchain e-voting services (H1 supported). This finding is consistent with Ng’s (2012) research on the significance of the technical dimension of digital literacy in an e-learning context and has profound implications for G2C digital service delivery. These results suggest that individuals could be more open to using new government services, given that they are provided with the technical and operational capacity to operate on platforms powered by new technologies such as blockchain e-voting. Highly digitally literate citizens are more likely to understand the use of technology and digital environments and be very comfortable/at ease when working on government digital platforms. Thus, the citizens, who have a more profound understanding of digital tools, are likely to use digital platforms without any hesitation. In the case of blockchain e-voting technology, a cornerstone G2C service, people who have higher technical digital literacy are more likely to feel at ease with the technology involved, including the voting platform and associated digital interfaces. This perceived competence and comfort level increase their likelihood of adopting and utilizing blockchain e-voting services.
In contrast, the study results suggest that the socio-emotional dimension of digital literacy is not significantly related to the adoption of blockchain e-voting services (H2 rejected). This finding aligns with those of Jain et al. (2022) and indicates that individuals with “netiquette” to interpret emotional cues (Ng, 2012) and seek emotional support from friends, family, tutorial videos, and the internet do not necessarily adopt blockchain e-voting services. A possible reason could be that individuals may need more support for e-voting services among their friends and family. Furthermore, blockchain e-voting services may not require substantial socio-emotional literacy to influence adoption intentions directly. Unlike specific social platforms or online communities that heavily rely on emotional connections and support, blockchain e-voting, as a critical G2C democratic service, usually takes place privately and primarily focuses on the technical aspects of voting integrity, transparency, and security. Consequently, the socio-emotional dimension of digital literacy may have limited relevance regarding blockchain e-voting services. This difference is important for digital transformation projects in the public sector because it shows that civic participation services may have different adoption dynamics than digital platforms aimed at consumers.
The lack of relationship between cognitive digital literacy and adoption of blockchain e-voting services (H3 is rejected) contradicts Ng’s (2012) findings in the e-learning context. The rationale might be that the cognitive dimension focuses on analytic ability, information evaluation, and integration of information. As blockchain technology is a highly technical one based on cryptography, consensus algorithms, and distributed ledger (Ølnes et al., 2017), people may find it difficult to grasp its complexity, explaining the weak correlation between cognitive digital literacy and adoption of blockchain e-voting services. Complexity might act as a barrier to adoption despite cognitive digital literacy. This issue is an obstacle for the public digital transformation process, as if the citizens cannot fully understand G2C service technology even with a high level of analytical and evaluative skills, they might be reluctant to engage. Another possible explanation might be that access to comprehensive and easily understandable information about blockchain technology and its application in e-voting may be limited. The cognitive dimension of digital literacy relies on the availability of relevant resources and educational materials that enable individuals to acquire knowledge and understand new technologies (Tomczyk et al., 2020). If there is a lack of accessible information or if available resources are too technical or challenging to comprehend, individuals may struggle to develop the necessary cognitive understanding of blockchain e-voting, leading to a diminished connection between cognitive digital literacy and adoption intentions. These findings suggest that government digital transformation initiatives should invest in public education and accessible documentation to demystify complex technologies and enhance citizen digital literacy. The study findings regarding the socio-emotional and cognitive aspects of digital literacy indicate that their role in adoption intentions of blockchain-enabled services is context-dependent.
As expected, the three dimensions of digital literacy (i.e., technical, cognitive, and socio-emotional) are positively associated with perceived trust (H4, H5, and H6 supported). Specifically, the technical dimension of digital literacy encompasses skills and competencies related to the use of technology and its operation. Individuals who are digitally literate can operate digital services, understand technical concepts, and use digital platforms effectively. Technical skills and knowledge allow individuals to develop perceived trust in digital systems, and by extension, perceived trust in blockchain e-voting. Their familiar and comfortable interactions with technology can foster their trust in the blockchain e-voting service, assuring them of its security, authenticity, and proper functioning.
The cognitive dimension of digital literacy relates to analytical understanding, information evaluation, and knowledge assimilation. Individuals who possess high cognitive digital literacy are able to critically analyze the information, comprehend the technical dimension of blockchain e-voting, make cognitive judgments, and be conscious of the system's advantage and benefit in voting. The socio-emotional dimension of digital literacy refers to interpersonal relations that individuals can establish online, emotions people can understand and control while interacting with others in online settings, and social interaction skills on digital platforms. These individuals can have meaningful interactions online, share support, and form social connections that can foster trust in the system of blockchain e-voting. The cognitive interaction people have and the information they get can lead to a perception of higher trust in the blockchain e-voting system.
The study findings indicate that, for achieving public digital transformation, perceived trust is the factor that connects individuals' digital literacy with their decision to engage with government digital services. Governments implementing G2C digital services must recognize that building citizen trust requires simultaneous attention to technical accessibility, transparent information provision, and community engagement—each connected to different dimensions of digital literacy.
As for the mediation effects, perceived trust mediates the relationships between three dimensions of digital literacy (i.e., technical, cognitive, and socio-emotional) and the adoption intentions of blockchain e-voting services (H11, H12, and H13 supported). The research findings suggest that perceived trust bridges the three dimensions of digital literacy and the adoption intentions of blockchain e-voting services. This finding is particularly important to justify the insignificant relationship between the socio-emotional dimension of digital literacy and the adoption of blockchain e-voting services, as the connection between them is established through perceived trust. A possible explanation is that the mediation effect of perceived trust is validated through the emotional closeness and connection aspect of digital literacy, which allows users’ emotional support to increase the trust level and further facilitate the adoption of blockchain e-voting services. These results align with those of previously published studies where perceived trust was discovered to act as a mediator in the connection between emotional support and behavioral intention in a social commerce context (Al-Tit et al., 2020). This mediation effect has substantial implications for public digital transformation. It suggests that even when citizens lack strong socio-emotional engagement with G2C services, their trust in these services—built through technical competence and reliable information—can facilitate adoption. Therefore, governments should prioritize trust-building mechanisms that work across multiple dimensions of digital literacy.
Similarly, the finding that perceived trust mediates the relationship between the cognitive dimension of digital literacy and the adoption of blockchain e-voting services is particularly crucial in understanding the significant relationship between these two variables. Through the mediating role of perceived trust, the way the cognitive dimension indirectly affects the adoption of blockchain e-voting services may be explained. The cognitive dimension of digital literacy might not directly affect the adoption of blockchain e-voting services; however, it indirectly influences people's judgment and choice through perceived trust. The link between the cognitive dimension of digital literacy and perceived trust indicates that greater cognitive digital literacy can foster individual acceptance towards the technology behind blockchain e-voting and, consequently, increase perceived trust towards the system. A higher level of perceived trust will, in turn, enhance the individual's intention to adopt blockchain e-voting services, as they are more confident and secure when using it to vote. In other words, while the cognitive dimension may not directly influence adoption intentions, it indirectly shapes individuals’ perceptions and beliefs through perceived trust, ultimately impacting their intentions for blockchain e-voting adoption. This finding aligns with that of Punyatoya (2019), who demonstrated that both cognitive and affective trust serve as mediators between website-related factors (such as quality, security, and reputation) and customer satisfaction, which, in turn, enhances loyalty intentions in online retailing. The study results on digital literacy’s technical, socio-emotional, and cognitive elements reveal that their influence on adopting blockchain-enabled services is context-dependent, with trust being a vital prerequisite for developing these intentions.
As expected, perceived trust is positively associated with the intention to adopt blockchain e-voting services and the adoption behavior of blockchain e-voting services (H7 and H8 supported). The findings are consistent with those of Queiroz and Wamba (2019), who found that trust is a vital indicator of behavioral intention for blockchain adoption, and Nath et al. (2022), who found that perceived trust is significantly related to the adoption of blockchain in the apparel logistics network. By trusting the blockchain e-voting service, uncertainty can be minimized, and transparency can be empowered. These results emphasize that in the public digital transformation context, establishing and sustaining citizens’ trust in G2C services should not only be a related issue but also a determining factor for both democratic participation and effective service provision in government.
In addition, the results show that using a blockchain e-voting service will increase e-voting satisfaction and referral intention (supported H9 and H10), confirming previous research findings (Manrai et al., 2021). The convergence between the study findings and prior research investigating technology and service adoption demonstrates generalizability across contexts. Moreover, the results are significant in public digital transformation, as satisfied citizens of G2C services could become advocates of e-government services, influencing others and thus strengthening citizens’ usage behavior of digital services.
Overall, these findings indicate that once citizens use the blockchain e-voting system, they tend to experience satisfaction from using the service and are inclined to make recommendations to their friends and relatives. When individuals choose to adopt the blockchain e-voting service, they tend to be satisfied with the e-voting experience. This adoption of blockchain e-voting means that individuals consider this service as beneficial, trustworthy, and useful. Further, the findings indicates that adoption of blockchain e-voting is not limited to personal use but can lead to recommendations to friends and relatives. Once citizens use the blockchain e-voting service, they have a greater tendency to propose using the service to their friends and relatives. Therefore, positive word-of-mouth would help promote digital transformation, suggesting that the implementation of G2C services can enhance the self-reinforcing effects of public adoption of these services.
ImplicationsTheoretical implicationsThis research offers several important theoretical contributions to the literature on governmental citizen services, blockchain-based e-voting, innovation, and services marketing. The study provides a theoretically significant advancement in understanding technology adoption within G2C services by integrating three distinct theoretical traditions: digital literacy theory, trust mechanisms, and technology adoption perspectives. The research expands the notion of e-government service adoption (Benchis et al., 2025) by illustrating how digital literacy and trust collaboratively reduce domain-specific risks in critical citizen interactions, such as blockchain e-voting. In government, literacy helps people use complicated systems (like cryptographic verification), which makes it easier for them to do their jobs. Technology trust, on the other hand, helps people deal with institutional uncertainties such as data sovereignty, making them more likely to use decentralized platforms instead of centralized bureaucracy. This two-part system improves G2C service design models, such as blockchain e-voting, that are sensitive to vulnerabilities.
The study makes a critical contribution by demonstrating that generic technology adoption models require substantial modification for complex, decentralized systems like blockchain e-voting. For example, while UTAUT assumes that performance expectancy, effort expectancy, social influence, and facilitating conditions operate uniformly across contexts (Venkatesh et al., 2012), this research shows that blockchain adoption in the G2C context requires context-specific individual-level characteristics as primary antecedents. Furthermore, the finding that only technical digital literacy directly predicts adoption intention while cognitive and socio-emotional dimensions fail to have direct effects challenges UTAUT's universalistic assumptions. This suggests that generic technology acceptance frameworks must be augmented with domain-specific competency measures when applied to technologies requiring specialized knowledge of cryptographic principles and decentralized architectures. Thus, understanding blockchain e-voting services requires integrated theoretical perspectives (digital literacy + trust + technology adoption outcomes) rather than a single generic model.
Additionally, the results enhance digital literacy from a marginal variable into a key theoretical construct explaining technology adoption behavior in G2C services. By following Ng's (2012) three-dimensional framework, the study identifies that digital literacy in G2C services acts as an enabler of trust formation instead of a direct antecedent of adoption intention. This difference is theoretically meaningful because whereas technical digital literacy has a direct impact on adoption intention (i.e., those users with technical expertise are likely to adopt the system), the cognitive and socio-emotional components have an impact on adoption intention through the mediation of perceived trust. Theoretically, this finding suggests that capability and confidence are distinct psychological constructs; a user's competence (technical skills) may directly lead to his intention, whereas cognitive understanding and the emotional response toward the technology may lead to intention only when they contribute to citizens' confidence (trust) in the system.
By positioning digital literacy in G2C services as an important antecedent, the study links to the broader theoretical constructs of digital divide and digital citizenship (Prez-Escoda et al., 2019). Our findings that technical literacy plays the most direct effect on adoption intention imply that a second-level digital divide—inequality in knowledge and skill to use a system, rather than a first-level digital divide—equality in the ability to access a system—is likely to be more relevant in the adoption of sophisticated technology applications like blockchain-based e-voting. The implication for democratic theory (Benchis et al., 2025; Shaikh et al., 2025) is that if blockchain e-voting is inclusive, a government should be concerned not only about access issues but also about competency levels to ensure that vulnerable people with low technical literacy will actually benefit from these technologies.
Our study confirms trust as a significant mediator of all three dimensions of digital literacy toward adoption intention, demonstrating trust as the key psychological construct linking citizen capabilities to behavioral intentions in G2C services. It shows a significant deviation from traditional technology adoption models (Venkatesh et al., 2012), since trust is not explicitly tested as a mediator in previous studies. The direct impacts of the three dimensions on adoption intention, mediated through perceived trust, show what might be called a "capability-confidence-intention" pathway. This implies that by having competence in digital literacy, users build their confidence in using the technology effectively; this confidence in turn helps them to determine to adopt the technology. In the case of blockchain e-voting, it helps to answer citizens' legitimate concern about cryptographic security and the integrity of their votes (Abo-Akleek et al., 2025) without depending solely on their technical expertise.
Furthermore, we treated trust as technology trust (Oliveira et al., 2017), instead of institutional or process trust. This operationalization t is theoretically significant because technology such as blockchain can reduce or substitute the roles of human institutions and processes with cryptographic security and guarantee. Thus, trust is shifted from trusting an institution to trusting an algorithm and cryptographic certainty. This contributes to the broad theoretical literature of trust in that different technological frameworks might be defined with different theoretical understandings of trust. Traditional e-voting systems demand institutional trust, whereas blockchain e-voting requires technology trust.
Finally, by treating blockchain e-voting (Benchis et al., 2025; Shaikh et al., 2025) as a G2C service, this study enriches both the literature of services marketing and public management in how digital literacy affects citizens' participation, satisfaction, and continued use of digital public services, in the sense that digital literacy influences citizens' intention to adopt technology and their satisfaction with it by providing technology trust. It indicates that for public-sector innovation, where institutional trust level is generally high, technology trust is indeed an essential component connecting citizen readiness to behavior outcomes. Adoption intention and satisfaction are employed as outcome variables in this study to reflect the behavior after adoption for blockchain-enabled G2C services; thus, the multi-layer behavior analysis reflects a more profound understanding of the post-adoption behavior.
Practical implicationsIn addition, the findings provide valuable practical implications that benefit governments, policymakers, innovators, and technology managers in understanding the impact of digital literacy dimensions. The findings of this research provide actionable insights for designing effective G2C services during public digital transformation. First, governments must prioritize technical accessibility and user interface design, recognizing that technical digital literacy is the strongest direct predictor of service adoption. Second, trust emerges as the critical mediating mechanism across all dimensions of digital literacy, requiring governments to invest in transparency, security assurance, and clear communication about service functionality. Third, understanding that adoption leads to satisfaction and positive referral behavior suggests that early adopter experiences are a crucial investment in the successful implementation of flagship G2C services, which can catalyze broader digital transformation. Finally, the context-contingent nature of different digital literacy dimensions in predicting adoption suggests that a one-size-fits-all approach to public digital service design is insufficient; instead, governments should employ segmented strategies that account for varying citizen digital capabilities while maintaining universal accessibility standards.
For instance, governments should pay more attention to accommodating channels that increase citizens' technical literacy, which is the most influential factor in adopting and using blockchain e-voting. While the other two dimensions of digital literacy, namely, socio-emotional and cognitive, are essential, they must be associated with trust and context (e.g., e-learning and e-voting). Thus, trust can be a proxy for signifying digital literacy that could lead to broader adoption and engagement in using blockchain e-voting. Moreover, when promoting a blockchain-enabled service, emphasis should be given to the digital literacy aspect(s) relevant to the specific context, and efforts should be directed to its/their enhancement.
Furthermore, the results of this research may offer valuable insights to enhance the design, advancement, and deployment of blockchain-based e-voting technology. Specifically, these findings can inform the incorporation of elements that promote information assessment, knowledge absorption, emotional intelligence, and social and technical abilities in the system. Integrating such elements into the blockchain e-voting system would facilitate its acceptance and increase its usage intention.
In addition, since governments and firms must evaluate the use of blockchain e-voting services, two factors must be assessed regularly: service satisfaction and referral behavior. Such factors would most often provide more confidence in deploying blockchain e-voting to a large-scale population. Overall, government officials can work with industry experts to develop a comprehensive strategy aimed at improving digital literacy across the public sector and raising the trusting atmosphere around blockchain e-voting to boost its adoption and usage.
Limitations and future research recommendationsAlthough the study produced significant results on citizens’ attitudes toward new technologies in e-voting, it has several limitations. First, this research gathered data solely from U.S. participants, hence constraining the applicability of the results to the country’s unique socio-cultural, technological, and political milieu. The comparatively elevated digital adoption rates and e-governance infrastructure in the U.S. may not accurately reflect nations with varying digital maturity, cultural perceptions of technology-government interactions, or electoral system frameworks. The cultural elements within the U.S., such as low uncertainty avoidance and high individualism, may influence the digital literacy and trust relationship more strongly compared to collectivist and high uncertainty-avoidance nations. Future studies should replicate and extend these findings to other nations—especially those where blockchain e-voting is widely adopted (e.g., Estonia, Switzerland, and Sierra Leone)—to determine the cross-cultural generalizability of this framework. Cross-national comparative studies can provide particular insight into the ways in which institutional context affects the relationships among digital literacy, trust, and adoption behavior.
Second, though MTurk allowed us to reach a heterogeneous population on the internet, it likely disproportionately selected digitally literate, computer-savvy individuals, leading to self-selection bias. Self-selection bias was exacerbated by our decision to use purposive sampling to recruit participants that had some level of prior interaction with the system of blockchain-based e-voting, which intrinsically excludes voters who had never experienced such systems. As such, this sample cannot represent the average U.S. voter, and the findings should be understood as such. Future studies using a probability-based sampling frame or alternate panel providers (e.g., Prolific) can help ensure greater external validity and provide a sense of how the findings compare across all levels of the population. Third, though purposive sampling can offer theoretical relevance in the study of the experiences of the sample population, it restricts the generalization of the findings across the full U.S. voting population. Specifically, our sampling criteria (U.S. residents with experience with blockchain-based e-voting, reporting a basic level of comfort and familiarity, and belonging to Gen X, Y, or Z) excluded older Americans (Baby Boomers), rural populations, and low internet penetration individuals, which could be considered groups that may be more susceptible to institutional trust and digital literacy barriers. Future research must actively include such segments in order to provide a comprehensive picture of adoption dynamics.
Fourth, despite ensuring that participants had experience with blockchain-based e-voting and providing participants with a standard description of the technology, this study did not explicitly ask participants to elaborate on the degree to which they were aware of the specific workings of the blockchain utilized for the particular election in question. Future research should examine to what extent (if any) election authorities inform their voters of the features of blockchain-based e-voting (e.g., through training materials or interface design) and whether this level of contextual knowledge influences the way digital literacy relates to adoption behavior. Fifth, the e-voting service used in the context of the study is inadequate to address all of blockchain technology’s complicated principles and concerns. We solely prioritized digital literacy competencies and trust in our research since these are crucial factors among the unique qualities of blockchain. There are unresolved concerns with blockchain technology that we have yet to account for, such as security and transparency. Furthermore, considering the psychology of voters, experimental designs could be used in the future to capture the psychological changes in individuals before and after using blockchain-enabled e-voting. Also, the study did not consider demographic differences, given that younger generations are more proficient in using new technologies. Thus, future research should examine the influence of demographics on blockchain adoption in e-voting. Furthermore, larger samples, data from other countries, or e-voting settings (e.g., unions and universities) could help replicate and confirm current findings.
Ethics approvalAll procedures performed in studies involving human participants were conducted in accordance with the ethical standards of the institutional and/or national research committee, as well as the 1964 Helsinki Declaration and its subsequent amendments or comparable ethical standards.
Informed consentInformed consent was obtained from all participants whom we reached out for data collection.
Availability of dataData is made available upon a reasonable request.
CRediT authorship contribution statementRodoula H. Tsiotsou: Writing – original draft, Conceptualization. Amir Zaib Abbasi: Project administration, Methodology, Formal analysis. Lin Li: Conceptualization. Mousa Albashrawi: Writing – review & editing.
All authors declare that they have no conflicts of interest.
The authors would like to express their sincere gratitude to King Fahd University of Petroleum & Minerals (KFUPM), Dhahran, Saudi Arabia, for the generous financial support that made this research possible. This study was supported under grant number INFE2113.
