Cybercrimes such as phishing, hacking, identity theft, fraud, and misrepresentation are increasingly eroding consumer confidence in e-commerce across Sub-Saharan Africa. As a result, Cybersecurity has become a critical concern for businesses, consumers, and governments. This study examines the relationship between Cybersecurity, Uncertainty Avoidance, Digital Trust, and the Continual Usage of e-commerce platforms among consumers in Ghana. Using structural equation modeling, the study reveals that Cybersecurity significantly influences Continual Usage, Uncertainty Avoidance, and ICT Skills. It also finds that Uncertainty Avoidance mediates the relationship between Cybersecurity and Digital Trust, while ICT Skills mediate the relationship between Cybersecurity and Continual Usage. Furthermore, Uncertainty Avoidance, Digital Trust, and ICT Skills collectively explain the relationship between Cybersecurity and Continual Usage of e-commerce platforms. The findings emphasize the need for strong Cybersecurity measures with user-friendly features, user training to enhance ICT Skills, and effective regulatory policies to reduce uncertainty among the risk-averse consumers. This approach encourages greater adoption of e-commerce while mitigating the risks of cyberattacks. The insights have practical implications for e-commerce businesses, policymakers, and individuals, highlighting the importance of Cybersecurity in building Digital Trust and fostering Continual Usage. The study contributes to the existing literature by underscoring the critical role of Cybersecurity in cultivating consumer trust and loyalty within the e-commerce sector.
In today's digital age, online buying and selling of goods and services known as electronic commerce or e-commerce have become integral to global commerce, offering convenience, accessibility, and a wide range of services to users worldwide (Naatu et al., 2024; Oguta, 2024). This current global culture accelerated when measures aimed at curbing the transmission of COVID-19 were implemented to limit human interaction (Florea et al., 2022). As the reliance on these platforms grows, so does the need for robust information security practices to protect both businesses and consumers from cybercrime which is also on the increase (Gangwer and Narang, 2021; Azizah, 2021; Ogar, 2022). While the people of Sub-Saharan Africa are trying to adopt digital technology platforms, most of them lack information and communication technology (ICT) skills (Ogar, 2022). This vulnerability makes it easier for cybercriminals to exploit, especially since even experts have been reported to struggle with detecting cyberattacks—let alone individuals with limited technical knowledge (Ogar, 2022). For instance, e-commerce platforms often gather sensitive information including names, addresses, and payment details of consumers (Oguta, 2024). As a result, despite the growing role of e-commerce as an essential technology, many people remain distrustful and insecure about engaging in online transactions (Alzoubi et al., 2022). This coupled with the fact that technology adoption does not occur in cultural vacuum has led to disparities in the adoption and usage rate of emerging technologies in different cultural contexts (Oregon et al., 2008).
Culture profoundly influences how individuals perceive, adopt, and interact with each other (Okello et al., 2019). Cultural values, norms, and practices shape attitudes toward innovation, risk, and change which can either accelerate or hinder the adoption of new technologies (Okello et al., 2019; Altuncu et al., 2012). For example, in cultures that emphasize collectivism, individuals often are more open to connectivity and collaboration (Naatu et al., 2020; Okello et al., 2019). Thus, they are likely to encourage their peers, family and friends to adopt digital platforms that can help them connect. In contrast, cultures that value individualism and autonomy might prioritize technologies that offer personal control or privacy. The cultural dimension, Uncertainty Avoidance, is identified to be an influential element on individual acceptance and patronage of products and services (Sartono et al., 2024). For instance, high sense of Uncertainty Avoidance may influence a person’s trust and even their patronage of some products or services (Sartono et al., 2024).
Studies exploring the effect of Cybersecurity and cultural factors on Continual Usage of e-commerce platforms are limited in Sub-Saharan Africa. Most research have focused on aspects such as information security and consumer trust (e.g., Pieters, 2011; Barnard & Wesson, 2014), the relationship between usability and trust in information systems (Sasse, 2005), and how trust impacts the ongoing use of e-commerce platforms (Pinem et al., 2018). No study has simultaneously examined the effect of Cybersecurity on Culture, particularly Uncertainty Avoidance—and Digital Trust in e-commerce platforms. As noted by Sasse (2005) and Ghelani (2022), despite the availability of numerous security mechanisms in cyberspace (e.g., Akiddo, New Relic, and Rippling IT), many users find them overly complex, especially in regions with high illiteracy rates, widespread poverty, weak institutions, and poor infrastructure such as Sub-Saharan Africa (Alkhwaldi et al., 2023; Naatu et al., 2023).
This study seeks to explore the impact of Cybersecurity on Continual Usage of e-commerce platforms, Uncertainty Avoidance, Digital Trust and ICT Skills (i.e. how individuals leverage their ICT abilities for online shopping and selling) using the Social Exchange Theory. The research questions are:
Using Social exchange theory, the study discusses the factors that build trust in business transactions. We posit that Cybersecurity directly impacts on Uncertainty Avoidance, ICT Skills and Continual Usage of e-commerce platforms. We also posit that, Uncertainty Avoidance, Digital Trust, and ICT Skills mediates the relationship between Cybersecurity and Continual Usage of e-commerce platforms. This study is essential for bridging gaps in existing literature as it offers valuable insights into the factors shaping consumer behaviour in the digital landscape of Sub-Saharan Africa, Ghana. It provides critical market intelligence for businesses and serves as a foundation for future research and policy formulation.
The structure of the study is as follows: it begins with a review of relevant literature and the theoretical framework, followed by the development of hypotheses. The methodology is then outlined, before presentation and discussion of the results. The study concludes with a summary of the findings and their implications.
E-commerce, or electronic commerce, refers to the digital transformation of traditional commerce which allows the transaction of businesses online using information technology, Internet, computer, and other electronic devices (Oguta, 2024). It plays a vital role in industrial and economic growth, offering efficient, convenient, and fast business processes (Babu et al., 2020). It enables businesses and individuals to buy and sell goods and services online (Tofan & Bostan, 2022). E-commerce operates across various devices, including computers, tablets, smartphones, and other smart gadgets (Tofan & Bostan, 2022). Activities in e-commerce entails online payments, sales, advertising, and browsing (Kinal, 2022). Some platforms function through mobile apps, while others rely on traditional websites for transactions. Leading e-commerce businesses today include Jumia, Kikuu, and Amazon (Alkhunaizan & Ali, 2022).
The driving force behind e-commerce is the digital economy also known as the internet or web economy, which relies on digital technologies, especially, the internet (Pomeroy, 2020). In recent years, the growth of the digital economy has led to a significant increase in e-commerce users in Sub-Saharan Africa. This surge marks a major shift from the late 1990s when internet access in Africa was a luxury limited to a few (Interpol African Cyberthreat Assessment Report, 2024). The rise in e-commerce has been fueled by factors such as the COVID-19 pandemic and the African Continental Free Trade Area (AfCFTA) agreement which has reduced trade barriers among African nations (Han et al., 2023). Consequently, the number of e-commerce customers have skyrocketed, and a range of new e-products and services emerged, allowing consumers to shop conveniently from home (Alkhunaizan & Ali, 2022).
Alongside this rapid growth, there has been a growing concern regarding the rise in cybercrime. As technology advances, cybercriminals have developed highly sophisticated methods to exploit vulnerabilities (Interpol African Cyberthreat Assessment Report, 2024; Shakeri et al., 2022). Cybercrime is the term used to define a broad range of illegal behaviours that involve using digital equipments or networks (Sartono et al., 2024). Victims of cybercrime face financial, psychological, and emotional harm. Cybercrime affects individuals, governments, and businesses alike, explaining the variety of security measures in place to protect cyberspace (Sartono et al., 2024). Online security or Cybersecurity is a type of information system security which involves processes and practices designed to detect, prevent, and ensure safety on the internet (Rajaonah, 2017). Cybersecurity encompasses key principles such as confidentiality, integrity, availability, authentication, and non-repudiation (Shakeri et al., 2022). Cybersecurity is different from the information security which is broad and involves protecting all types of information, regardless of whether it's stored on digital devices or in other forms, such as paper records (Shakeri et al., 2022).
For e-commerce platforms to be successfully embraced, they must earn users' trust to encourage ongoing use, hence the need for Cybersecurity (Sartono et al., 2024; Oguta, 2024). The literature defines trust as a firm believe in the reliability or capability of a system or entity (Oguta, 2024; Liu et al., 2022). It entails confidence that a system can deliver consistent, dependable results, and is characterized by positive relationships, sound judgment, expertise, and minimal risk exposure (Oguta, 2024).
Culture also plays a vital role in consumers ability to adopt, use and continue using technologies (Alkhwaldi et al., 2023). Hofstede's (1991) study identifies four dimensions in which people from different cultures differ. One of the dimensions is Uncertainty Avoidance (Hofstede, 1991). This dimension is closely related to an individual's attitude toward risk and uncertainty. Specifically, it refers to the extent to which people feel threatened by and attempt to avoid uncertainty and ambiguity (Hofstede, 2001). In cultures with high Uncertainty Avoidance, individuals often view unfamiliar or unpredictable situations as threatening. Consequently, they tend to seek clarity and clear interpretations of events in both their personal relationships and business dealings, often searching for products or services that provide certainty and reliability (Hofstede, 2001).
However, while a wide range of Cybersecurity systems are available, many are perceived as difficult to use (Sasse, 2005; Familoni, 2024), highlighting the importance of user-friendliness for the adoption and continued use of e-commerce technologies. As Sartono et al. (2024) observed, complex technologies can introduce uncertainty, which may influence individuals' decisions to adopt or persist in using a technology. This suggests a critical link between Cybersecurity , Digit Trust, Uncertainty Avoidance, ICT Skills, and the Continual Usage of e-commerce platforms. These relationships however remain underexplored in Sub-Saharan Africa.
Research on Cybersecurity has been explored from various perspectives. For example, Rajaonah (2017) reviewed studies on trust in information system security, particularly concerning infrastructure protection, while Sartono et al. (2024) conducted an empirical analysis of the impact of sustainable digital transformation on the perceived value and adoption of Industry 4.0 in Indonesia. Other notable studies include a literature review on security challenges in cyberspace (Oguta, 2024), an empirical analysis of user acceptance and satisfaction in e-commerce (Kassim et al., 2012), and an exploration of how providing security system information influences user trust (Pieters, 2011). Pinem et al. (2018) examined the role of trust in the continued use of government-to-business online services. However, to the best of our knowledge, no study has specifically investigated how Cybersecurity affects Uncertainty Avoidance, Digital Trust, and the Continual Usage of e-commerce platforms.
Our study aims to address this gap by exploring these key influencing factors of e-commerce on Continual Usage. Cybersystems security challenges can have dire consequences, affecting critical services and their usage (Sartono et al., 2024; Familoni, 2024), hence, research in this area would be highly important. The findings will expand the current body of knowledge and provide valuable insights for stakeholders and practitioners within the e-commerce industry.e
2.1 Theoretical review
Research across multiple disciplines, including psychology, economics, marketing, and management information systems, have proposed a range of theories to explain trust (Thurik et al., 2023). One prominent theory is Rational Choice Theory. This theory suggests that trust is based on a logical evaluation of the costs and benefits involved in trusting others (Trabucchi et al., 2023). Another influential framework is the Diffusion of Innovations Theory, which focuses on trust in the context of adopting new ideas, products, or technologies. According to this theory, individuals are more likely to trust and embrace innovations when they perceive them to align with their existing values, beliefs, and social norms (Ullah et al., 2021).
In our study, we draw on Social Exchange Theory to explain the dynamics that shape individual trust in business interactions. Building on the concepts of Rational Choice Theory, Social Exchange Theory suggests that trust develops through ongoing social exchanges between individuals (Park & Kim, 2023). This theory views trust as a reciprocal process, where people base their trust on past experiences and the expectation of future benefits. It is frequently applied to analyze various forms of social relationships, such as business relationships, families, romantic partnerships, friendships, and social networks (Chapman et al., 2022). Social Exchange Theory offers valuable insights into how individuals navigate relationships, negotiate power dynamics, and evaluate the balance of rewards and costs. It sheds light on factors influencing attraction, consumer satisfaction, commitment, and trust in relationships, as well as decision-making in social interactions such as internet business transactions (Lauren et al., 2021).
2.2. Hypotheses Development
The main variables of the study are: Cybersecurity (CS), Uncertainty Avoidance (UA), Digital Trust (TR), Information and Communication Technology Skills (ICT Skills), and Continual Usage (CU) of e-commerce platforms.
2.2.1 Cybersecurity (CS)
Despite the growing importance of e-commerce, concerns over cyber theft, fraud, forgery, hacking, harassment, and other forms of cyberattacks have left many users hesitant to engage with online platforms (Solms & Solms, 2018). This apprehension is primarily due to e-commerce platforms' collection of sensitive information such as names, addresses, and payment details, which heightens fears around data privacy and security (Oguta, 2024). As a result, security has become a crucial factor in e-commerce (Oguta, 2024).
Cybersecurity and information security both play vital roles in protecting valuable data, yet, they address different aspects despite often being used interchangeably (Solms & Solms, 2018). Cybersecurity specifically focuses on safeguarding computer systems and networks from malicious actors seeking to breach or disrupt them (Rajaonah, 2017; Oguta, 2024). This includes efforts to prevent, detect, and respond to threats such as hacking and malware, while also ensuring the security of websites and digital platforms. In contrast, information security has a broader focus, aiming to protect all types of information, whether stored digitally or in physical formats, such as paper records (Solms & Solms, 2018).
Several studies, including Kassim and Abdullah (2010), emphasize that Cybersecurity is a critical prerequisite for the adoption and use of e-commerce platforms and services. According to Oguta (2024) Cybersecurity is essential for fostering trust among internet users (Oguta, 2024). Similarly, Löbbers and Benlian (2019) affirm that robust Cybersecurity positively influences individual confidence and trust in digital platforms, it allows them to continue using the platforms. Based on the above we argue that strong Cybersecurity is essential for fostering trust among internet users, encouraging them to engage with and to continue using e-commerce platforms. Therefore we hyposize that:
H1: Cybersecurity positively influences e-commerce platforms Continual usage.
2.2.4 Uncertainty Avoidance
Uncertainty is an individual's perceived inability to accurately predict an outcome or situation (Altuncu et al., 2012). It is often experienced as a state of not knowing something about ourselves or our environment, which makes it difficult to prepare for or cope with the unknown (Altuncu et al., 2012; Alkhwaldi et al., 2023). Uncertainty Avoidance, a concept from Hofstede’s (1991) cultural dimensions, refers to the extent to which individuals in a society feel threatened by uncertainty and ambiguity and take steps to avoid it (Alkhwaldi et al., 2023). Culture, in this context, refers to the collective mindset of a group of people within a specific community or society, which shapes their behaviour and distinguishes them from others (Alkhwaldi et al., 2023). In societies with low Uncertainty Avoidance, the need for security is relatively low, and people tend to accept life as it comes. These cultures are more likely to embrace new situations, diverse viewpoints, and various technologies even if they are complex or uncertain (Alkhwaldi et al., 2023).
On the other hand, high Uncertainty Avoidance cultures are characterized by a higher need for security (Altuncu et al., 2012). In these societies, individuals are more cautious and seek security measures, clear guidelines, rules, and predictable outcomes (Altuncu et al., 2012; Alkhwaldi et al., 2023). People in these cultures tend to prefer certainty and stability over ambiguity. Thus, Cybersecurity may eradicate uncertainty and minimise individual’s fear of cyber attack in e-commerce, gain their trust and lead to continual usage of e-commerce platforms. Based on the argument above, we hypothesize that:
H2: Cybersecurity is significantly related to Uncertainty Avoidance (UA)
2.2.2 Digital Trust (DT)
Digital Trust refers to an individual’s confidence in the safety, privacy, security, reliability, and ethical handling of their personal data by companies in the digital environment (Oguta, 2024). It is closely linked to the perceived value of the information exchanged (Sartono et al., 2024). In the context of e-commerce, Digital Trust reflects the confidence customers place in platforms to meet their needs, deliver positive experiences, and protect their personal information. This trust is essential for the long-term success and growth of e-commerce businesses. As Valencia-Martinez et al. (2023) showed, trust has a direct impact on users' intentions to engage with e-commerce platforms and serves as a key indicator of perceived security, which supports the hypothesis of this research. Furthermore, Saeed (2019) showed that Cybersecurity significantly enhances consumer satisfaction and trust in e-commerce, emphasizing the important role security plays in fostering trust. This research aligns with those findings, asserting that Cybersecurity positively influences Digital Trust. Hence we hypothesize that:
H3: Uncertainty Avoidance has a significant positive effect on Digital Trust
2.2.3 Continual Usage (CU)
Continual Usage refers to an individual's intention to use a technology or platform now and in the future (Rahmayanti et al., 2022; Menon, 2022). It reflects a individual's decision to adopt and continue using a particular product or service both now and in the future (Menon, 2022). In this study context, Continual Usage implies that if individuals feel assured that the e-commerce platform they are using is easy to navigate, free from fraud, and secure, it enhances their trust in the e-commerce process and platform. This may in turn motivate them to keep using the platform in the future. Trust, in this case, refers to the confidence individuals have that a process or system will function successfully. As e-commerce users' trust in the reliability and security of the platforms growth, this will likely lead to Continual Usage. This hypothesis is supported by Darmiasih and Setiawan (2020). The study hypothesizes that:
H4: Digital Trust positively influences the Continual Usage of e-commerce service platforms
H5: The relationship between Cybersecurity and Digital Trust is mediated by Uncertainty Avoidance.
2.2.4 Information and communication technology skills (ICT Skills)
ICT skill refers to the basic computer and internet skills for operating, processing, and dissemination of information including changing, accessing and using software and hardware of computer and internet devices. The ability to detect and solve problems of ICT and knowledge in digital surroundings is often attributed to the extent of ICT Skills a person has or has not (Adams & Alhassan, 2021). ICT Skills are crucial for managing information security and effectively using e-commerce platforms (Thomas, 2018). While cyberfraud may sometimes go unnoticed by even experts or individuals well-versed in information technology, those with ICT Skills are generally more capable of detecting and avoiding cyberattacks compared to those without such skills (Ben-Asher & Gonzalez, 2015). As a result, individuals with ICT knowledge are at relatively lower risk of falling victims to fraud. In contrast, individuals who lack ICT Skills are more vulnerable to fraud as they may fail to recognize cyberattacks or potential threats to information systems (Ben-Asher & Gonzalez, 2015). Research shows that people without ICT expertise are significantly more likely to become victims of fraud (Xu et al., 2022).
Moreover, ICT Skills are essential in moderating the relationship between personal trust and the continued use of e-commerce platforms (Amofah, 2022). Individuals with strong ICT Skills are better positioned to build trust and maintain their use of these platforms. A solid understanding of information and communication technologies enhances users' confidence in the security and reliability of e-commerce services. Research indicates that familiarity with a system or process is vital for fostering trust (Rezaei & Bresciani, 2020). As such, ICT Skills play a crucial role in shaping the link between personal trust and sustained engagement with e-commerce platforms. Those with ICT expertise are more adept at recognizing fraudulent alerts or security threats, enabling them to navigate platforms with greater ease and avoid cyberattacks. This ability to detect and mitigate risks not only strengthens their trust in the platform but also boosts their willingness to continue using e-commerce services. Based on this, we hypothesize that:
H6: The relationship between Cybersecurity and Continual Usage of e-commerce platforms is significantly mediated by ICT Skills.
H7: Uncertainty Avoidance, Digital Trust and ICT Skills are significant mediators of the relationship between Cybersecurity and Continual Usage of e-commerce platforms
Below is the study model showing the hypotheses from hypothesis 1 to hypothesis 9.
(Source: Authors’ Construct, 2023).
The study investigates the impact of Cybersecurity and Digital Trust, ICT Skills, and Uncertainty Avoidance on the Continual Usage of e-commerce among Ghanaians. Ghana, a Sub-Saharan African country presents a fascinating context for this research due to its progress in internet and digital technology adoption despite facing economic, institutional, and infrastructural challenges. The majority of Ghanaians possess limited ICT Skills, making this setting uniquely relevant and intriguing for exploring the causal relationship among these factors and e-commerce platform usage.
Research design: A quantitative research design was used to facilitate a rigorous and objective analysis (Naatu et al., 2024) and measurement acuracy for more precise results (Nassè, 2018) . The study employed a convenience sampling strategy, selecting respondents who were readily available and willing to participate. While this approach limits the ability to generalize the findings, it offers several advantages such as being less time-consuming and cost-effective (Fricker & Schonlau, 2002). The challenge of securing voluntary participation further justified the use of convenience sampling as the primary method for gathering responses (Naatu et al., 2024). The data collection was conducted between April 2023 and April 2024. To get participants who were genuinely interested and willing to voluntarily participate in the survey, we assured them of anonymity and avoided enticing them with gifts for participation. This helped to prevent speed answers (Naatu et al. 2022). Data collection was organized by dividing the country into three regions: the Southern, Middle, and Northern belts. Three agents were assigned to each region to distribute the questionnaires and gather responses. Initially, 117 responses were collected, with 38 from the Southern belt, 10 from the Middle belt, and 69 from the Northern belt. However, 7 responses from the Southern belt, 3 from the Middle belt, and 7 from the Northern belt were incomplete, resulting in the removal of 17 responses. This left a final total of 100 usable responses.
To determine an appropriate sample size, the study followed the "10-times rule" for Partial Least Squares Structural Equation Modeling (PLS-SEM), a widely accepted guideline for estimating the minimum sample size (Hair et al., 2011). According to this rule, the sample size should exceed 10 times the number of inner and outer model connections (Goodhue et al., 2012). Since each of the study’s constructs—Cybersecurity , Digital Trust, ICT Skills, Uncertainty Avoidance, and Continued Usage—were represented by five indicators, the minimum required sample size was calculated to be 50 (i.e., 5 indicators × 10). However, the actual sample size surpassed this requirement, with 117 responses initially collected, and 100 responses deemed valid and complete, providing a solid basis for analysis, as shown in Table 1. The respondents were individuals aged 18 and older, all of whom could read and write. Their educational backgrounds ranged from senior secondary school to PhD levels. They were selected based on their ability to comprehend and respond appropriately to the survey questions (Naatu et al., 2024). Of the respondents, 85% were single, while 15% were married (see details in Table 1).
Table 1. Demographic Distribution of Respondents
Demographic |
Characteristics |
Number |
Percentage |
||
Gender |
Male |
67 |
67 |
||
Female |
33 |
33 |
|||
Total |
100 |
100 |
|||
Age |
18 -25years |
58 |
58 |
||
36-45years |
25 |
25 |
|||
46-55years |
17 |
17 |
|||
Above 55years |
0 |
0 |
|||
Total |
100 |
100 |
|||
|
|
|
|||
Education |
SHS |
8 |
8 |
||
Degree |
75 |
75 |
|||
Masters |
16 |
16 |
|||
PhD |
1 |
1 |
|||
Total |
100 |
100 |
|||
Marital Status |
Single |
85 |
85 |
||
|
Married |
15 |
15 |
||
Total |
100 |
100 |
|||
Monthly Income |
GHC 0-GHC1000 |
69 |
69 |
||
GHC 1001- |
19 |
19 |
|||
GHC2000 above |
12 |
12 |
|||
Total |
100 |
100 |
|||
|
|||||
Occupation |
Student |
62 |
62 |
||
Entrepreneur |
11 |
11 |
|||
public servant |
17 |
17 |
|||
private/NGOs |
10 |
10 |
|||
Total |
100 |
100 |
|||
Location |
Southern belt |
31 |
31 |
||
Middle belt |
7 |
7 |
|||
Northern belt |
62 |
62 |
|||
|
Total |
100 |
100 |
||
(Source: Authors’ Construct, 2023)
4.Measurement Items
A four-step process was followed based on Li et al. (2018) to ensure the reliability and validity of the constructs. The steps included item generation, pilot testing (preliminary study), a full preliminary study, and finally, a comprehensive analysis of the hypothesized model. Items for the constructs were derived from previous literature. Items measuring Cybersecurity (CS) were adapted from Calder et al. (2009), ICT Skills from Rezaei and Bresciani (2020), Digital Trust (TR) and Continual Usage from Nambisan and Baron (2007), while Uncertainty Avoidance was taken from (Altuncu et al., 2012). Each construct was assessed using a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree).
Indicator reliability refers to the degree to which a factor explains the variance in the data. This was assessed using reflective factor loadings, which range from 0 to 1. A loading of 0.7 or higher is considered acceptable, while values below 0.7 are regarded problematic. Items with loadings below this threshold were excluded from the analysis. For example, item 1 (0.193), 3 (0.249), and 4 (0.401) from the Cybersecurity (CS) construct were removed, leaving items 2 and 5, which demonstrated adequate indicator reliability. Additionally, item 3 from the ICT Skills construct and item 2 from the Uncertainty Avoidance (UA) construct were also discarded (see details in Table 2). All remaining factor loadings were within the recommended range.
Table 2. Factor Loadings
Indicator item Cross-loading/factor loadings
Continual Usage |
Information Security |
ICT Skills |
Individual Trust |
Uncertainty Avoidance |
||
CU1 |
0.704 |
|
||||
CU2 |
0.757 |
|
||||
CU3 |
0.934 |
|
||||
CU4 |
0.858 |
|
||||
CU5 |
0.888 |
|
||||
CS1 |
|
Dropped |
|
|
|
|
CS2 |
|
0.838 |
|
|
|
|
CS3 |
Dropped |
|
||||
CS4 |
|
Dropped |
|
|
|
|
CS5 |
0.879 |
|
||||
ICT1 |
0.920 |
|
||||
ICT2 |
0.928 |
|
||||
ICT3 |
|
|
Dropped |
|
|
|
ICT4 |
0.852 |
|
||||
ICT5 |
0.897 |
|
||||
TR1 |
0.890 |
|
||||
TR2 |
0.890 |
|
||||
TR3 |
0.919 |
|
||||
TR4 |
0.925 |
|
||||
TR5 |
|
|
|
0.879 |
|
|
UA1 |
|
|
|
|
0.760 |
|
UA2 |
|
|
|
|
Dropped |
|
UA3 |
|
|
|
|
0.784 |
|
UA4 |
|
|
|
|
0.804 |
|
UA5 |
|
|
|
|
0.711 |
|
Note: The threshold of Indicator loadings is ≥ 0.708
(Source: Authors’ Construct, 2023)
To determine the internal consistency of the constructs, Cronbach alphas (α), Rho A, Composite reliability and Average variance extacted (AVE) were also used. The recommended threshold for Cronbach alphas (α), Rho A, and Composite reliability is 0.7. For the AVE, 0.50 is the minimum threshold. As shown in Table 3., all were above the minimum recommended thresholds. The Cronbach alphas (α) ranged from 0.700 to 0.942, Rho A from 0.700 to 0.943, and Composite Reliability from 0.849 to 0.956. The AVE ranged between 0.568 to 0.811 (See details in Table 3).
Table 3. Construct Reliability
Variable
|
Cronbach Alpha (α) |
Rho A
|
Composite Reliability |
Average variance extracted (AVE) |
Continual Usage |
0.887 |
0.908 |
0.918 |
0.693 |
ICT Skills |
0.921 |
0.931 |
0.944 |
0.809 |
Digital Trust |
0.942 |
0.943 |
0.956 |
0.811 |
Cybersecurity |
0.700 |
0.700 |
0.849 |
0.738 |
Uncertainty Avoidance |
0.748 |
0.752 |
0.840 |
0.568 |
Note: The threshold of Cronbach’s Alpha, Composite Reliability is ≥ 0.7, and (AVE) is ≥ 0.5
(Source: Authors’ Construct, 2023)
Table 4. Discriminant Validity- Heterotrait-Monotrait Ratio (HTMT)
Variable
|
Continual usage |
ICT Skills |
Individual trust |
Cybersecurity
|
Uncertainty Avoidance |
||
Contintual Usage |
0.832 |
|
|
|
|
|
|
ICT Skills |
0.706 |
0.803 |
|
||||
Digital Trust |
0.829 |
0.581 |
0.850 |
0.850 |
|
||
Cybersecurity |
0.800 |
0.621 |
0.810 |
|
|||
Uncertainty Avoidance |
0.704 |
0.632 |
0.802 |
0.800 |
0.754 |
||
Note: None of the constructs should measure up to 90% or 0.90 against other constructs. Also, The √AVE (in bold) should be > than the factor correlations with each other.
(Source: Authors’ Construct, 2023)
The study assessed the construct validity using measures of discriminant validity, specifically the Heterotrait-Monotrait Ratio (HTMT). HTMT measures the differentiation between a variable and other variables in a study. Henseler et al., (2015) argue that, unlike other measures such as the Fornell and Larcker, (1981) criterion, HTMT performs well in situations where the item loadings of constructs differ slightly. For valid discriminant validity, HTMT values close to zero shows good discriminant validity and scores above 0.85 shows there is a problem with discriminant validity. As Table 4 shows, the results shows sufficient fitness as they suggest that the items distinctively measure their assigned constructs (Henseler et al., 2015). HTMT values presented in Table 4 are below the threshold of 0.85. Also, the square root of the Average Variance Extracted (√AVE) should be greater than the factor correlations with each other (Naatu et al., 2024; Voorhees et al., 2016). The results confirm that we have no discriminant validity issues with the variables. This is evident in Table 4, where the √AVE values (in bold) are higher than the factor correlation values.
Table 5. SRMR
Variable
|
Original Sample (O) |
Sample Mean (M) |
95% |
99% |
Saturated Model |
0.052 |
0.051 |
0.062 |
0.067 |
Estimated Model |
0.073 |
0.066 |
0.074 |
0.070 |
Note: SRMR value of above 0.08 indicates the absence of fit.
(Source: Authors’ Construct, 2023)
Table 6. Variance Inflation Factor (VIF)
|
Cont. usage |
ICT Skills |
Ind. Trust |
Continual Usage |
|||
ICT Skills |
1.496 |
||
Digital Trust |
1.921 |
1.681 |
|
Cybersecurity |
1.761 |
1.681 |
1.000 |
Note: 1 ≤ VIF < 5: are acceptable.
(Source: Authors’ Construct, 2023)
The model's fitness was further assessed using SRMR (Standardized Root Mean Square Residual) estimations, as proposed by Hu and Bentler (1999). The SRMR score for the estimated was 0.073 is below the 0.08 threshold, indicating that there is no significant measurement or structural model misspecifications. This suggest that the model is well-fitting. Additionally, to assess potential collinearity issues, the study examined the Variance Inflation Factors (VIFs) (see Table 6). The VIF values ranged from 1.496 to 1.921, which indicates the absence of collinearity problems.
Fig2.Estimated model.
(Source: Authors’ Construct, 2023)
MODEL 1.
Table 7. Direct Regression Results
|
|
Path |
Original Sample |
Sample Mean |
Standard Deviation |
T Statistics |
P Values |
Confirmed |
|
|||
CS -> CU |
0.272 |
0.263 |
0.102 |
2.679 |
0.007 |
Yes |
||||||
2. |
CS -> UA |
0.912 |
0.914 |
0.015 |
62.322 |
0.000 |
Yes |
|||||
3. |
UA -> TR |
0.689 |
0.690 |
0.061 |
11.239 |
0.000 |
Yes |
|||||
4. |
TR -> CU |
0.438 |
0.444 |
0.092 |
4.740 |
0.000 |
Yes |
|||||
|
CS -> ICT |
0.231 |
0.235 |
0.108 |
2.145 |
0.032 |
Significant |
|||||
|
TR -> ICT |
0.401 |
0.401 |
0.109 |
3.694 |
0.000 |
Significant |
|||||
|
ICT ->CU |
0.285 |
0.288 |
0.066 |
4.347 |
0.000 |
Significant |
|||||
Note: The threshold of, t- values ≥ 1.65, P-values < 0.05 at 95% confidence interval, Std beta > 0., IS = Information Security; CU = Continual usage; TR = Individual Trust; ICT = Information and communication technology skills.
(Source: Authors’ Construct, 2023)
.
Table 7. Indirect Effect
|
Path |
Original Sample |
Sample Mean |
Standard Deviation |
T Statistics |
P Values |
Confirmed |
5. |
CS -> UCA -> TR |
0.628 |
0.631 |
0.056 |
11.129 |
0.000 |
Yes |
|
CS -> ICT -> CU |
0.066 |
0.068 |
0.037 |
1.765 |
0.078 |
No |
6. |
TR -> ICT -> CU |
0.114 |
0.115 |
0.041 |
2.772 |
0.006 |
Yes |
|
CS -> UA -> TR -> ICT |
0.252 |
0.255 |
0.078 |
3.212 |
0.001 |
Significant |
|
CS -> UA -> TR -> CU |
0.275 |
0.283 |
0.072 |
3.843 |
0.000 |
Significant |
7. |
CS -> UA -> TR -> ICT -> CU |
0.072 |
0.073 |
0.027 |
2.625 |
0.009 |
Yes |
|
UA -> TR -> ICT -> CU |
0.079 |
0.080 |
0.031 |
2.577 |
0.010 |
Significant |
|
UA -> TR -> CU |
0.302 |
0.309 |
0.077 |
3.907 |
0.000 |
Significant |
|
UA -> TR -> ICT |
0.276 |
0.279 |
0.087 |
3.166 |
0.002 |
Significant |
Note: The threshold of, t-values ≥ 1.65, P-values < 0.10 at 90% confidence interval, Std beta > 0.
The study tested a total of seven hypotheses. First, we hypothesized that, Cybersecurity has a significant positive impact on the Continual Usage of e-commerce platforms. Secondly, we proposed that Cybersecurity positively influences Uncertainty Avoidance, and thirdly, we suggested a significant positive effect of Uncertainty Avoidance on Digital Trust. The fourth proposed that Digital Trust is significantly related to Continual Usage of e-commerce platforms. All were confirmed at 95% confidence level, supporting our expectations (H1: β = 0.272, p < 0.01; H2: β = 0.912, p < 0.001; H3: β = 0.689, p < 0.001, H4: β = 0.438, p < 0.001). What this mean is that, Cybersecurity minimises uncertainty and encourages the people to trust or feel confident that they are safe to use e-commerce platforms, and leads them to continue using the platforms (Oguta, 2024). These are in line with existing studies (e.g., Sartono et al., 2024; Handoyo, 2024; Oguta, 2024). For instance, Oguta’s (2024) study maintained that, security is a critical consideration in e-commerce due to the sensitive nature of online transactions. As such, smart businesses invest in Cybersecurity technologies to safeguard customer data and build trust (Handoyo, 2024). According to Hossain et al., (2024), trust minimises perceived risks associated with online shopping and makes customers comfortable enough to have repeated interactions and long-term engagement with the platform. Digital Trust is built by ensuring perceived reliability through Cyberseurity and consistent positive encounters or experiences of the platform's services (Hossain et al., 2024).
According to Blender and Felderer, (2023), cybercrime is escalating at an alarming rate, and nearly everyone has experienced some form of cyberattack (Stzelecki and Rizun, 2022). Consequently, users are feeling increasingly insecure and are therefore seeking clear signs of Cybersecurity before proceeding with online transactions, especially in cultures where Uncertainty Avoidance are very high (Ashraf et al., 2019; Crossler et al., 2018). Users seek convincing cues to trust or believe that the environment in which they are engaging in the e-commerce has appropriate safeguards or protection before they proceed (Vance et al., 2008; Oguta 2024).
Next, hypothesis 5, 6 and 7 were also highly significant (i.e., : H5: β = 0.628, p < 0.001; H6: β = 0.114, p < 0.01, H7: β = 0.072 p < 0.01). The significance of these hypotheses demonstrates that 1. the relationship between Cybersecurity and Digital Trust is mediated by Uncertainty Avoidance (H5). This indicate that, Uncertainty Avoidance helps to explain the effect of Cybersecurity on Digital Trust. Thus, Cybersecurity minimises the likelyhood of users avoiding engagement with e-commerce platforms as it allows people to trust that they are safe in using e-commerce platforms. While no study has established the mediation role of Uncertainty Avoidance in cyberspace, several studies have shown that, Uncertainty Avoidance moderates the relationship between Digital Trust and intention to adopt digital technology (e.g., Faqih 2022; Sartono et al., 2024).
Next, H6 demonstrates that ICT Skills mediates the relationship between Digital Trust and Continual Usage. ICT Skills are crucial for managing information security and effectively using e-commerce platforms (Thomas, 2018; Adams & Alhassan, 2021). They are the basic computer and internet skills used for operating, processing, and dissemination of information as well as changing, accessing and using software and hardware of computer and internet devices (Kusumaningtyas & Suwarta, 2015). The requisite ICT Skills equips users with the knowhow to be able to appreciate whether a particular Cybersecurity is genuine or effective, hence leading to Trust in the e-commerce platform at hand (Faqih, 2022). This is consistent with Valencia-Martinez et al. (2023) and Faqih (2022) which showed that, trust serves as a key indicator of perceived security and impacts on users' intention to continue engaging with an e-commerce platform. This is not to say that, with ICT Skills one cannot be defrauded, however, it puts one who has the skills at an advantage over one without the skills as they are more likely to detect and prevent cyberattack than one without the requisite skills (Ben-Asher & Gonzalez, 2015).
Lastly, H7 shows that, Cybersecurity significantly impacts on Continual Usage through the mediation of Uncertainty Avoidance, Digital Trust, and ICT Skills. This implies that Cybersecurity minimises users doubt or uncertainty and leads to Digital Trust which in turn impacts on ICT Skills and finally encourages users to continue patronising the platform. It suggests that, although consumers are seeking information systems security, they are skeptical of readily accepting any technology claiming to offer security (Rana & Chicone 2024). For instance, according to Rana and Chicone (2024), despite the benefits of Cybersecurity , a number of ethical concerns have been raised concerning their impact on privacy. Users seek trustworthy and convincing cues before embracing such technologies (Rana & Chicone 2024). The cues include ease of use, user friendly interface, quality systems, systems with high ethical standards and checks against cyber attacks (Vance et al., 2008; Han et al. 2023; Blender & Felderer 2023). Studies have established that people are more likely to trust and engage with e-commerce platforms that implement robust security measures than those believed to have less robust security measure (e.g., Hossain et al., 2024).
Accordingly, the tech industry has created numerous security systems to mitigate risks or make attacks significantly more challenging (Blender & Felderer 2023). However, many people find these security technologies difficult to use (Sasse, 2005; Blender & Felderer 2023; Han et al., 2024). Their inability to easily navigate and use the technology irritates them and impacts on their trust in the technology (Johnson et al., 2008). This finding aligns with previous research that highlight the role of user competence in enhancing the effectiveness of security measures and trust in digital environments (Han et al., 2024; Blender & Felderer
In conclusion, the study successfully tested and confirmed seven hypotheses, demonstrating the critical role of Cybersecurity in fostering Continual Usage of e-commerce services and platforms. The findings underscores that, Cyberseurity not only directly influences an ongoing use of e-commerce platforms, but also plays a fundamental role in its affect of culture, in this case, by minimising doubt or uncertainty as a cultural trait, to build and sustain consumer Trust, which further enhances Continual Usage. This is consistent with prior literature, which highlights the importance of Cybersecurity in nurturing trust and loyalty in e-commerce contexts (Oguta 2024; Hossain et al., 2024; Sartono et al., 2024). Additionally, the study shows that ICT Skills significantly mediates the relationship between Trust and Continual Usage, reinforcing the idea that while users seek secure systems, they also require ease of use and user-friendly interfaces to maintain their trust and engagement with these platforms. The marginal impact of ICT Skills as a mediator between Cybersecurity and Continual Usage suggests that while technical competence is important, it may not be the sole factor in enhancing user trust and platform loyalty. Overall, the study underscores the necessity for e-commerce platforms to prioritize robust Cybersecurity, provide trustworthy cues including user-friendly interfaces and easy to use security measures to ensure sustained Digital Trust and Continual usage.
IMPLICATIONS
7.1 Theoretical implications
The theoretical implications of the study extends across multiple domains, particularly in the fields of information systems, e-commerce, and consumer behaviour. By confirming that Cybersecuriy significantly impacts Uncertainty Avoidance, Digital Trust and Continual Usage of e-commerce platforms, the study reinforces existing theories that emphasize the foundational role of security in digital environments. The study contributes to the broader understanding of how security measures are not just protective mechanisms but are integral to the user's decision-making process. This finding aligns with and expands upon the Social Exchange theory and Technology Acceptance Model (TAM) where Social Exchange theory sees trust as a result of a reciprocal process in which individuals trust institutions based on their previous experiences with their e-commerce platforms and their expectations of future benefits from their transactions (Park and Kim, 2023). The TAM suggests that perceived ease of use and usefulness are key factors influencing technology adoption. Here, Cybersecurity emerges as a critical component of perceived usefulness, directly affecting Uncertainty Avoidance, Digital Trust and Continual Usage of e-commerce platforms.
Additionally, the study's examination of Uncertainty Avoidance as a mediating factor between Cybersecurity and Digital Trust, as well as ICT Skills between Digital Trust and Continual Usage, adds a valuable layer of insight into the influence of culture and user competence on the development of Digital Trust. While prior research has acknowledged the importance of user competence (e.g. Vance et al., 2008; Blender and Felderer, 2023), this study highlights Culture and ICT Skills as significant facilitator of the relationship between Trust and the ongoing use of e-commerce services. This suggests that models of technology acceptance and usage must increasingly consider culture and user competence as variables that can promote the effectiveness of security measures in fostering trust.
The marginal effect of ICT Skills as a mediator between Cybersecurity and Continual Usage also invites further theoretical exploration. This finding could lead to a refinement of existing models, to provide more nuanced understanding of how user skills interact with security perceptions to influence trust and usage behaviours.
In conclusion, our findings provide a richer understanding of the dynamics that drive consumer behaviour in digital environments, offering a foundation for future research to explore the complex relationships between these variables.
7.2 Practical Implications
The study emphasized the significance of Cybersecurity , Uncertainty Avoidance, Digital Trust and ICT Skills effects on Continual Usage of e-commerce platforms in Ghana. This finding has practical implications for e-commerce businesses, policymakers, and individuals. In line with Parmer, (2024) and Maqableh et al. (2021) we recommend that e-commerce businesses must prioritize Cybersecurity to protect user data and ensure trust, investing in robust measures to safeguard against cyber threats. Moreover, Cybersecurity systems should be user-friendly to avoid frustrating users and eroding trust (Sasse, 2005; Chang and Chen, 2009). Providing clear and convincing cues of Cybersecurity , such as trust badges and secure payment gateways, can also enhance user trust. Additionally, e-commerce businesses can offer training or support to enhance users’ ICT Skills, which play a crucial role in mediating the relationship between Digital Trust and Continual Usage (Johnson et al., 2008). Regular security audits can help identify vulnerabilities and ensure the effectiveness of Cybersecurity measures. Policymakers can develop guidelines and regulations to ensure e-commerce businesses prioritize Information Security and protect user data. Educating consumers about the importance of Information Security and how to identify trustworthy cues can also help them make informed decisions when using e-commerce platforms. Ultimately, e-commerce businesses must continuously monitor user behaviour and adjust their Cybersecurity measures accordingly to ensure optimal trust and Continual Usage, maintaining a competitive edge in the market. Consumers should exercise caution before making purchases, as many shopping websites contain features that can trigger anxiety and attract criminals (Faqih, 2022; Hariharan et al., 2023). Moreover, it is crucial for consumers to improve their knowledge and ICT skills. Staying up-to-date with current security measures and becoming well-informed. Tech-savvy users are more likely to identify vulnerabilities in online platforms.
7.3 Limitations and directions for future research
This study has some limitations. First, it was conducted in Ghana, so future research should be carried out in other countries to provide a more comprehensive understanding of the impact of Cybersecurity on e-commerce across Africa. Secondly, the use of convenience sampling, along with the study's focus on a single country and a small sample size, limits the generalizability of the findings. We recommend using probability sampling across multiple countries to obtain more representative data for a broader perspective. Additionally, future studies could employ qualitative methods to explore people perceptions of security in e-commerce more deeply.
Declaration of interest statement:
This study is an original unpublished work that has not been submitted to any other journal for review. Its content has not been copyrighted, published previously, or under consideration for publication anywhere else.
Funding
This study was not funded by any grant from any funding agencies in the public, commercial, or not-for-profit sectors.